JP2004190593A - Inter-cylinder variation detecting device of internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately detect a variation in a suction air quantity between cylinders in the substantially whole operation area of an internal combustion engine. <P>SOLUTION: An inter-cylinder suction air quantity variation rate DEV (#i) is calculated on the basis of output of an air flowmeter for every one cycle; the inter-cylinder suction air quantity variation rate DEV (#i) calculated in an operation state of securing calculation accuracy of the inter-cylinder suction air quantity variation rate DEV (#i) after realizing a prescribed variation learning execution condition is learnt as an inter-cylinder variation learning value GDEV (#i); and an average suction valve lift quantity VLL at that time is learnt as a reference suction valve lift quantity GVVL. Afterwards, in a state of being unable to realize the variation learning execution condition, the inter-cylinder suction air quantity variation rate DEV (#i) according to a present average suction valve lift quantity VLL is calculated on the basis of the inter-cylinder variation learning value GDEV (#i), a reference suction valve lift quantity GVVL and a present average suction valve lift quantity VLL. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【００６６】
上式の分母は、全気筒の瞬時空気流量極大値のなまし値ＧＡＰＥＡＫＳＭ(#i)の平均値であり、Ｋ２は、瞬時空気流量極大値のばらつきを吸入空気量のばらつきに変換するための補正係数である。
【００６７】
上記［数１］式から明らかなように、各気筒の気筒間吸入空気量ばらつき率ＤＥＶ(#i)は、各気筒の気筒間吸入空気量ばらつき率なまし値ＧＡＰＥＡＫＳＭ(#i)を全気筒の気筒間吸入空気量ばらつき率なまし値の平均値で割り算して、それに補正係数Ｋ２を乗算した値である。
【００６８】
［気筒間ばらつき補正ルーチン］
図８に示す気筒間ばらつき補正ルーチンは、エンジン運転中に所定周期で実行される。本ルーチンが起動されると、まず、ステップ４０１で、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了しているか否か（気筒間ばらつき学習完了フラグＧＸＶＶＬ＝ＯＮか否か）を判定する。もし、気筒間ばらつき学習が完了していなければ、ステップ４０２以降の気筒間ばらつき補正に関する処理を実行することなく、本ルーチンを終了する。
【００６９】
一方、上記ステップ４０１で、気筒間ばらつき学習が完了していると判定された場合には、ステップ４０２以降の気筒間ばらつき補正に関する処理を次のようにして実行する。まず、ステップ４０２で、図３の気筒間ばらつき算出ベースルーチンよって最終的に決定された気筒間吸入空気量ばらつき率ＤＥＶ(#i)を読み込む。
【００７０】
この後、ステップ４０３に進み、各気筒毎に気筒間吸入空気量ばらつき率ＤＥＶ(#i)に応じた基本リフト補正量ＦＶＶＬ１(#i)を図９のマップを用いて算出する。図９のマップは、気筒間吸入空気量ばらつき率ＤＥＶ(#i)がプラス値となる領域では、基本リフト補正量ＦＶＶＬ１(#i)が減量値（マイナス値）となり、気筒間吸入空気量ばらつき率ＤＥＶ(#i)がマイナス値となる領域では、基本リフト補正量ＦＶＶＬ１(#i)が増量値（プラス値）となる。つまり、ある気筒の吸入空気量が全気筒の平均吸入空気量よりも多くなるほど、基本リフト補正量ＦＶＶＬ１(#i)による減量補正量が大きくなり、反対に、ある気筒の吸入空気量が全気筒の平均吸入空気量よりも少なくなるほど、基本リフト補正量ＦＶＶＬ１(#i)による増量補正量が大きくなる。尚、気筒間吸入空気量ばらつき率ＤＥＶ(#i)が０付近の所定領域では、基本リフト補正量ＦＶＶＬ１(#i)＝０に設定され、吸気バルブリフト量ＶＶＬが補正されない。
【００７１】
そして、次のステップ４０４で、エンジン回転速度ＮＥとエアフローメータ１４で検出した吸入空気量ＧＡ（平均空気流量）を読み込んだ後、ステップ４０５に進み、図１０に示す補正係数ＦＶＶＬ２のマップを検索して、各気筒毎に現在のエンジン運転条件（例えばエンジン回転速度ＮＥと吸入空気量ＧＡ）に応じた補正係数ＦＶＶＬ２(#i)を算出する。
【００７２】
一般に、吸入空気量ＧＡが少なくなると（吸気バルブリフト量が小さくなると）、吸気バルブリフト量補正の影響を受けやすくなるため、図１０の補正係数ＦＶＶＬ２のマップは、吸入空気量ＧＡが少なくなるほど、補正係数ＦＶＶＬ２が小さくなるように設定されている。
【００７３】
この後、ステップ４０６に進み、各気筒の基本リフト補正量ＦＶＶＬ１(#i)に補正係数ＦＶＶＬ２(#i)を乗算して、各気筒のリフト補正量ＦＶＶＬ(#i)を求める。
ＦＶＶＬ(#i)＝ＦＶＶＬ１(#i)×ＦＶＶＬ２(#i)
この後、ステップ４０７に進み、補正前の全気筒の平均バルブリフト量ＶＶＬに各気筒のリフト補正量ＦＶＶＬ(#i)を加算して、最終目標バルブリフト量ＶＶＬＭを求める。
【００７４】
この場合、クランク角カウンタＣＣＲＮＫ＝１２〜１７の期間（つまり第１気筒＃１の吸気行程に対応する期間）は、第１気筒＃１のバルブリフト補正量ＦＶＶＬ(#1)を用いて、最終目標バルブリフト量ＶＶＬＭを次式により算出する。
ＶＶＬＭ＝ＶＶＬ＋ＦＶＶＬ(#1)
【００７５】
クランク角カウンタＣＣＲＮＫ＝６〜１１の期間（つまり第２気筒＃２の吸気行程に対応する期間）は、第２気筒＃２のバルブリフト補正量ＦＶＶＬ(#2)を用いて、最終目標バルブリフト量ＶＶＬＭを次式により算出する。
ＶＶＬＭ＝ＶＶＬ＋ＦＶＶＬ(#2)
【００７６】
クランク角カウンタＣＣＲＮＫ＝１８〜２３の期間（つまり第３気筒＃３の吸気行程に対応する期間）は、第３気筒＃３のバルブリフト補正量ＦＶＶＬ(#3)を用いて、最終目標バルブリフト量ＶＶＬＭを次式により算出する。
ＶＶＬＭ＝ＶＶＬ＋ＦＶＶＬ(#3)
【００７７】
クランク角カウンタＣＣＲＮＫ＝０〜５の期間（つまり第４気筒＃４の吸気行程に対応する期間）は、第４気筒＃４のバルブリフト補正量ＦＶＶＬ(#4)を用いて、最終目標バルブリフト量ＶＶＬＭを次式により算出する。
ＶＶＬＭ＝ＶＶＬ＋ＦＶＶＬ(#4)
【００７８】
この後、ステップ４０８に進み、各気筒の吸気行程に対応して変化する各気筒の最終目標バルブリフト量ＶＶＬＭに応じて吸気バルブ２８の可変バルブリフト機構３０のモータ４１を高速駆動して、各気筒の吸気行程毎に吸気バルブリフト量を補正して各気筒の吸入空気量を補正する。これにより、気筒間の吸入空気量ばらつきを補正する。
【００７９】
以上説明した気筒間ばらつき補正ルーチンでは、気筒間ばらつき学習が完了する前は、気筒間ばらつき補正を実行しないようにしたが、気筒間ばらつき学習が完了する前は、図３のステップ１０５で算出した気筒間吸入空気量ばらつき率ＤＥＶ(#i)を用いて気筒間ばらつき補正を実行するようにしても良い。
【００８０】
尚、各気筒の吸気バルブ２８を電磁アクチュエータで駆動する構成の可変バルブ機構に本発明を適用する場合には、各気筒の最終目標バルブリフト量ＶＶＬＭに応じて各気筒の電磁アクチュエータの制御量を補正することで、各気筒の吸気バルブリフト量を補正して各気筒の吸入空気量を補正するようにすれば良い。
【００８１】
以上説明した本実施形態（１）の実行例を図１１に示すタイムチャートを用いて説明する。気筒間ばらつき補正実行条件が成立して気筒間ばらつき補正実行フラグがＯＮされている期間は、１サイクル毎にエアフローメータ１４の出力（瞬時空気流量ＧＡＦＭ）に基づいて各気筒の気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出し、ばらつき学習実行条件が成立してばらつき学習実行フラグがＯＮされているときに算出した気筒間吸入空気量ばらつき率ＤＥＶ(#i)を気筒間ばらつき学習値ＧＤＥＶ(#i)として学習すると共に、この気筒間ばらつき学習値ＧＤＥＶ(#i)を学習したときの平均吸気バルブリフト量ＶＬＬを基準吸気バルブリフト量ＧＶＶＬとして学習する。
【００８２】
そして、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了して気筒間ばらつき学習完了フラグＧＸＶＶＬがＯＮされた後、ばらつき学習実行条件が不成立となって気筒間ばらつき学習実行フラグがＯＦＦされている期間は、気筒間ばらつき学習値ＧＤＥＶ(#i)と、現在の平均吸気バルブリフト量ＶＬＬと基準吸気バルブリフト量ＧＶＶＬとの比（ＶＶＬ／ＧＶＶＬ）を用いて、最終的な気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出する。
【００８３】
以上説明した本実施形態（１）によれば、ばらつき学習実行条件が成立して気筒間吸入空気量ばらつき率ＤＥＶ(#i)の算出精度を確保できる運転領域のときに算出した気筒間吸入空気量ばらつき率ＤＥＶ(#i)を気筒間ばらつき学習値ＧＤＥＶ(#i)として学習するようにしたので、気筒間ばらつき学習値ＧＤＥＶ(#i)を精度良く学習することができる。そして、ばらつき学習実行条件が成立しない運転領域のときには、上記精度の良い気筒間ばらつき学習値ＧＤＥＶ(#i)を用いてその時点の運転領域に応じた気筒間吸入空気量ばらつき率ＤＥＶ(#i)を推定するので、エアフローメータ１４の出力波形が各気筒の吸入空気量のばらつきを精度良く反映した脈動波形にならない運転領域や、エアフローメータ１４の出力をサンプリングする時間（又はサンプリング回数）を十分に確保できない運転領域に対しても、気筒間吸入空気量ばらつき率ＤＥＶ(#i)を比較的精度良く算出することができる。これにより、ほぼ全ての運転領域で気筒間ばらつきの検出精度を向上させることができるので、ほぼ全ての運転領域で気筒間の吸入空気量ばらつきを精度良く補正することができ、気筒間のトルクばらつきや空燃比ばらつきを両方とも低減することができる。
【００８４】
《実施形態（２）》
ところで、ばらつき学習実行条件が成立したときに、吸気バルブ２８、排気バルブ２９、スロットルバルブ１５のバルブ動作条件が、必ずしも、気筒間吸入空気量ばらつき率ＤＥＶの算出に適した動作条件になっているとは限らない。
【００８５】
そこで、図１３乃至図１５に示す本発明の実施形態（２）では、ばらつき学習実行条件が成立する運転領域のときに、吸気バルブ２８、排気バルブ２９、スロットルバルブ１５のバルブ動作条件を学習時バルブ動作条件に変更して、気筒間吸入空気量ばらつき率ＤＥＶを算出し、その算出値を気筒間ばらつき学習値ＧＤＥＶとして学習するようにしている。
【００８６】
以下、この処理を実行する図１３の気筒間ばらつき算出ベースルーチンの処理内容を説明する。本ルーチンが起動されると、前記実施形態（１）で説明した図３のステップ１０１〜１０４と同じ処理により、各気筒の吸気行程に対応する期間の瞬時空気流量ＧＡＦＭの極大値ＧＡＰＥＡＫ(#i)を算出した後、ステップ１０６に進み、前記実施形態（１）と同じ方法で、ばらつき学習実行条件が成立しているか否かを判定する。
【００８７】
その結果、ばらつき学習実行条件が成立していると判定されれば、ステップ１０６ａに進み、スロットル開度ＴＡを学習時のスロットル開度ＫＴＡに変更する。この学習時のスロットル開度ＫＴＡは、全開又は吸気管圧力が大気圧付近となる開度に設定されている。これにより、スロットルバルブ１５の上流側に配置されたエアフローメータ１４で検出する吸気脈動の振幅を大きくすることができ、気筒間吸入空気量ばらつき率ＤＥＶ(#i)の算出に用いる吸気脈動の特性値（例えば極大値）を判別し易くなるため、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習に適した状態となる。
【００８８】
この後、ステップ１０６ｂに進み、吸気バルブリフト量ＶＶＬを学習時の吸気バルブリフト量ＫＶＶＬに変更する。この学習時の吸気バルブリフト量ＫＶＶＬは、所定リフト量以下又は最小値に設定されている。これにより、吸気バルブリフト量ＶＶＬが小さくなって、各気筒で目標リフト量に対する実リフト量のばらつきの割合が大きくなるため、各気筒の吸入空気量ばらつきが大きくなって、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習に適した状態となる。
【００８９】
この後、ステップ１０６ｃに進み、吸気バルブタイミングＩＮＶＶＴを学習時の吸気バルブタイミングＫＩＮＶＶＴに変更すると共に、排気バルブタイミングＥＸＶＶＴを学習時の排気バルブタイミングＫＥＸＶＶＴに変更する。図１４に示すように、学習時の吸気バルブタイミングＫＩＮＶＶＴと学習時の排気バルブタイミングＥＸＶＶＴは、吸気バルブ２８と排気バルブ２９のバルブオーバーラップがなく且つ吸気バルブ２８の開弁期間が上死点（ＴＤＣ）から下死点（ＢＤＣ）までの間に収まるように設定されている。これにより、燃焼ガスや吸入空気の吹き返しを無くすことができるので、燃焼ガスや吸入空気の吹き返しによる吸気脈動の乱れを防止することができ、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習に適した状態となる。
【００９０】
以上のようにしてスロットルバルブ１５、吸気バルブ２８、排気バルブ２９のバルブ動作条件を学習時バルブ動作条件に変更した後、ステップ１０６ｄに進み、前記実施形態（１）で説明した図７の気筒間吸入空気量ばらつき率算出ルーチンを実行して、各気筒の瞬時空気流量極大値ＧＡＰＥＡＫ(#i)に基づいて気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出する。この後、ステップ１０７に進み、上記ステップ１０６ｄで算出した気筒間吸入空気量ばらつき率ＤＥＶ(#i)を気筒間ばらつき学習値ＧＤＥＶ(#i)として学習（記憶）する。その他の処理は、前記実施形態（１）で説明した図３の各ステップの処理と同じである。
【００９１】
以上説明した本実施形態（２）では、図１５のタイムチャートに示すように、ばらつき学習実行条件が成立してばらつき学習実行フラグがＯＮされている期間は、スロットル開度ＴＡ、吸気バルブリフト量ＶＶＬ、吸気バルブタイミングＩＮＶＶＴ、排気バルブタイミングＥＸＶＶＴを、それぞれ気筒間ばらつき学習に適した学習時バルブ動作条件（ＫＴＡ、ＫＶＶＬ、ＫＩＮＶＶＴ、ＫＥＸＶＶＴ）に変更し、その状態のときに算出した気筒間吸入空気量ばらつき率ＤＥＶ(#i)を気筒間ばらつき学習値ＧＤＥＶ(#i)として学習する。これにより、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習精度を高めることができる。
【００９２】
《実施形態（３）》
次に、図１６及び図１７を用いて本発明の実施形態（３）を説明する。本実施形態（３）では、図１６に示す吸入空気量制御方式切換ルーチンを実行することで、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了するまで可変吸気バルブ制御による吸入空気量制御を禁止して、スロットルバルブ制御により吸入空気量を制御し、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了した後に可変吸気バルブ制御による吸入空気量制御を実行するようにしている。
【００９３】
この処理を実行する図１６の吸入空気量制御方式切換ルーチンは、所定周期（例えば４ｍｓ周期）で起動される。本ルーチンが起動されると、まず、ステップ５０１で、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了しているか否か（気筒間ばらつき学習完了フラグＧＸＶＶＬ＝ＯＮか否か）を判定する。もし、気筒間ばらつき学習が完了していなければ、ステップ５０２に進み、可変吸気バルブ制御による吸入空気量制御を禁止して、スロットルバルブ制御による吸入空気量制御を実行する。このスロットルバルブ制御による吸入空気量制御では、アクセル開度やエンジン運転状態等に基づいてスロットル開度ＴＡを可変して吸入空気量を制御する。この場合、吸気バルブリフト量ＶＶＬは、所定値、例えば最大値（例えば１０ｍｍ）に固定する。
【００９４】
その後、上記ステップ５０１で、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了していると判定されたときに、ステップ５０３に進み、可変吸気バルブ制御による吸入空気量制御を実行する。この可変吸気バルブ制御による吸入空気量制御中は、気筒間ばらつき学習値ＧＤＥＶ(#i)を用いて推定した気筒間吸入空気量ばらつき率ＤＥＶ(#i)に基づいて気筒間の吸入空気量ばらつきを補正しながら、アクセル開度やエンジン運転状態等に基づいて吸気バルブリフト量ＶＶＬを可変して吸入空気量を制御する。この場合、スロットル開度ＴＡは、吸気脈動騒音低減等のために吸気管圧力ＰＭを所定負圧（例えば−５ｋＰａ）に維持するように制御する。
【００９５】
以上説明した本実施形態（３）では、図１７のタイムチャートに示すように、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了する前（気筒間ばらつき学習完了フラグＧＸＶＶＬがＯＦＦのとき）は、可変吸気バルブ制御による吸入空気量制御を禁止して、スロットルバルブ制御により吸入空気量を制御する。このようにすれば、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了するまでは、スロットルバルブ制御による吸入空気量制御によって吸入空気量の制御精度を確保することができ、気筒間ばらつき学習値ＧＤＥＶ(#i)の未学習によるドライバビリティ悪化や排気エミッション悪化を回避することができる。
【００９６】
その後、気筒間ばらつき学習値ＧＤＥＶ(#i)の学習が完了して気筒間ばらつき学習完了フラグＧＸＶＶＬがＯＮされた後に、可変吸気バルブ制御による吸入空気量制御を開始する。このようにすれば、気筒間ばらつき学習値ＧＤＥＶ(#i)を用いてほぼ全運転領域の気筒間吸入空気量ばらつき率ＤＥＶ(#i)を精度良く推定できる状態になった後に、推定した気筒間吸入空気量ばらつき率ＤＥＶ(#i)を用いて気筒間の吸入空気量ばらつきを精度良く補正しながら、可変吸気バルブ制御による吸入空気量制御を実行することができ、可変吸気バルブ制御による吸入空気量制御を精度良く実行することができる。
【００９７】
《実施形態（４）》
上記実施形態（１）〜（３）では、エアフローメータ１４で検出した瞬時空気流量ＧＡＦＭを用いて気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出するようにしたが、図１８及び図１９に示す本発明の実施形態（４）では、吸気管圧力センサ１８（検出手段）で検出した瞬時吸気管圧力ＰＭＡＰを用いて気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出するようにしている。
【００９８】
エンジン運転中は、図１９に示すように、エアフローメータ１４の出力（瞬時空気流量ＧＡＦＭ）の脈動に同期して吸気管圧力センサ１８の出力（瞬時吸気管圧力ＰＭＡＰ）が脈動し、エアフローメータ１４の出力（瞬時空気流量ＧＡＦＭ）が極大になるときに吸気管圧力センサ１８の出力（瞬時吸気管圧力ＰＭＡＰ）が極小となる。
【００９９】
そこで、本実施形態（４）では、図１８に示す各気筒の瞬時吸気管圧力極小値算出ルーチンを実行して、各気筒の吸気行程に対応する期間の瞬時吸気管圧力極小値ＰＭＰＥＡＫ(#i)を求め、この瞬時吸気管圧力極小値ＰＭＰＥＡＫ(#i)を用いて気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出する。
【０１００】
図１８に示す各気筒の瞬時吸気管圧力極小値算出ルーチンは、吸気管圧力センサ１８の出力のＡ／Ｄ変換タイミング（例えば４ｍｓ周期）で起動される。本ルーチンが起動されると、まず、ステップ６０１で、吸気管圧力センサ１８で検出した瞬時吸気管圧力ＰＭＡＰが極小値であるか否かを判定する。この極小値の判定方法は、例えば瞬時吸気管圧力ＰＭＡＰの今回値を前回値と比較して、瞬時吸気管圧力ＰＭＡＰの変化方向が減少から増加に反転したか否かによって判定すれば良い。
【０１０１】
このステップ６０１で、瞬時吸気管圧力ＰＭＡＰが極小値であると判定されたときに、ステップ６０２に進み、クランク角カウンタＣＣＲＮＫ＝１２〜１７の期間（つまり第１気筒＃１の吸気行程に対応する期間）であるか否かを判定し、ＣＣＲＮＫ＝１２〜１７であれば、ステップ６０３に進み、第１気筒＃１の吸気行程に対応する期間の瞬時吸気管圧力ＰＭＡＰの極小値を、第１気筒＃１の瞬時吸気管圧力極小値ＰＭＰＥＡＫ(#1)とする。
【０１０２】
上記ステップ６０２で「Ｎｏ」と判定されて、ステップ６０４で、クランク角カウンタＣＣＲＮＫ＝６〜１１の期間（つまり第２気筒＃２の吸気行程に対応する期間）であると判定されれば、ステップ６０５に進み、第２気筒＃２の吸気行程に対応する期間の瞬時吸気管圧力ＰＭＡＰの極小値を、第２気筒＃２の瞬時吸気管圧力極小値ＰＭＰＥＡＫ(#2)とする。
【０１０３】
上記ステップ６０２、６０４で「Ｎｏ」と判定されて、ステップ６０６で、クランク角カウンタＣＣＲＮＫ＝１８〜２３の期間（つまり第３気筒＃３の吸気行程に対応する期間）であると判定されれば、ステップ６０７に進み、第３気筒＃３の吸気行程に対応する期間の瞬時吸気管圧力ＰＭＡＰの極小値を、第３気筒＃３の瞬時吸気管圧力極小値ＰＭＰＥＡＫ(#3)とする。
【０１０４】
上記ステップ６０２、６０４、６０６で全て「Ｎｏ」と判定された場合、つまり、クランク角カウンタＣＣＲＮＫ＝０〜５の期間（第４気筒＃４の吸気行程に対応する期間）である場合は、ステップ６０８に進み、第４気筒＃４の吸気行程に対応する期間の瞬時吸気管圧力ＰＭＡＰの極小値を、第４気筒＃４の瞬時吸気管圧力極小値ＰＭＰＥＡＫ(#4)とする。
【０１０５】
このようにして求めた各気筒の瞬時吸気管圧力極小値ＰＭＰＥＡＫ(#i)を用いて、前記実施形態（１）で説明した気筒間吸入空気量ばらつき率の算出方法（図７参照）と同じ方法で各気筒の気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出する。
以上説明した本実施形態（４）でも、気筒間吸入空気量ばらつき率ＤＥＶ(#i)を精度良く算出することができる。
【０１０６】
《実施形態（５）》
図２０に示す本発明の実施形態（５）では、気筒内に吸入される吸入空気量が多くなるほど、筒内圧が大きくなるという特性に着目して、各気筒の筒内圧を検出して、各気筒の圧縮ＴＤＣ前後の期間の筒内圧の極大値を求めるようにしている。これを実現するために、本実施形態（５）では、各気筒毎に、筒内圧を検出する筒内圧センサ（検出手段）が設けられている。尚、筒内圧センサは、点火プラグ２１に組み込まれたものを用いても良いし、点火プラグ２１とは別に設けても良い。
【０１０７】
本実施形態（７）では、図２０に示す各気筒の筒内圧極大値算出ルーチンを筒内圧センサのＡ／Ｄ変換タイミング（例えば４ｍｓ周期）で実行する。本ルーチンが起動されると、まず、ステップ７０１で、筒内圧センサで検出した筒内圧ＣＰＳが極大値であるか否かを判定する。この極大値の判定方法は、例えば筒内圧ＣＰＳの今回値を前回値と比較して、筒内圧ＣＰＳの変化方向が増加から減少に反転したか否かによって判定すれば良い。
【０１０８】
このステップ７０１で、筒内圧ＣＰＳが極大値であると判定されたときに、ステップ７０２に進み、クランク角カウンタＣＣＲＮＫ＝２２〜３の期間（つまり第１気筒＃１の圧縮ＴＤＣの前後９０℃Ａ以内の期間）であるか否かを判定し、ＣＣＲＮＫ＝２２〜３であれば、ステップ７０３に進み、第１気筒＃１の圧縮ＴＤＣの前後９０℃Ａ以内の期間の筒内圧ＣＰＳの極大値を、第１気筒＃１の筒内圧極大値ＣＰＳＰＥＡＫ(#1)とする。
【０１０９】
上記ステップ７０２で「Ｎｏ」と判定されて、ステップ７０４で、クランク角カウンタＣＣＲＮＫ＝１６〜２１の期間（つまり第２気筒＃２の圧縮ＴＤＣの前後９０℃Ａ以内の期間）であると判定されれば、ステップ７０５に進み、第２気筒＃２の圧縮ＴＤＣの前後９０℃Ａ以内の期間の筒内圧ＣＰＳの極大値を、第２気筒＃２の筒内圧極大値ＣＰＳＰＥＡＫ(#2)とする。
【０１１０】
上記ステップ７０２、７０４で「Ｎｏ」と判定されて、ステップ７０６で、クランク角カウンタＣＣＲＮＫ＝４〜９の期間（つまり第３気筒＃３の圧縮ＴＤＣの前後９０℃Ａ以内の期間）であると判定されれば、ステップ７０７に進み、第３気筒＃３の圧縮ＴＤＣの前後９０℃Ａ以内の期間の筒内圧ＣＰＳの極大値を、第３気筒＃３の筒内圧極大値ＣＰＳＰＥＡＫ(#3)とする。
【０１１１】
上記ステップ７０２、７０４、７０６で全て「Ｎｏ」と判定された場合、つまり、クランク角カウンタＣＣＲＮＫ＝１０〜１５の期間（第４気筒＃４の圧縮ＴＤＣの前後９０℃Ａ以内の期間）である場合は、ステップ７０８に進み、第４気筒＃４の圧縮ＴＤＣの前後９０℃Ａ以内の期間の筒内圧ＣＰＳの極大値を、第４気筒＃４の筒内圧極大値ＣＰＳＰＥＡＫ(#4)とする。
【０１１２】
このようにして求めた各気筒の筒内圧極大値ＣＰＳＰＥＡＫ(#i)を用いて、前記実施形態（１）で説明した気筒間吸入空気量ばらつき率の算出方法（図７参照）と同じ方法で各気筒の気筒間吸入空気量ばらつき率ＤＥＶ(#i)を算出する。
【０１１３】
以上説明した本実施形態（５）でも、気筒間吸入空気量ばらつき率ＤＥＶ(#i)を精度良く算出することができる。
尚、気筒間ばらつき率の算出方法は適宜変更しても良く、要は、吸入空気量、吸気管圧力、筒内圧力の所定期間毎の極大値、極小値、平均値、振幅値、面積、軌跡長等のうちの少なくとも１つに基づいて気筒間ばらつき値を算出するようにしても良い。
【０１１４】
また、上記各実施形態（１）〜（５）では、可変吸気バルブ制御による吸入空気量制御を行うシステムに本発明を適用したが、スロットルバルブ制御による吸入空気量制御のみを行うシステムに本発明を適用しても良い。スロットルバルブ制御による吸入空気量制御のみを行うシステムでは、スロットルバルブを全開又は大きく開いた高負荷領域で、吸気脈動の振幅が大きくなって、吸気脈動の特性値（極大値、極小値、面積等）を判別し易くなるため、この高負荷領域で、気筒間ばらつき学習値を学習するようにすると良い。
【０１１５】
また、本発明の適用範囲は４気筒のエンジンに限定されず、５気筒以上又は３気筒以下の複数気筒エンジンに本発明を適用しても良い。
【図面の簡単な説明】
【図１】本発明の実施形態（１）におけるエンジン制御システム全体の概略構成図
【図２】可変バルブリフト機構の正面図
【図３】実施形態（１）の気筒間ばらつき算出ベースルーチンの処理の流れを示すフローチャート
【図４】エアフローメータの出力電圧ＶＡＦＭを瞬時空気流量ＧＡＦＭに変換するマップを概念的に示す図
【図５】気筒間吸入空気量ばらつき率ＤＥＶのマップを概念的に示す図
【図６】実施形態（１）の各気筒の瞬時空気流量極大値算出ルーチンの処理の流れを示すフローチャート
【図７】実施形態（１）の気筒間吸入空気量ばらつき率算出ルーチンの処理の流れを示すフローチャート
【図８】実施形態（１）の気筒間ばらつき補正ルーチンの処理の流れを示すフローチャート
【図９】基本リフト補正量ＦＶＶＬ１のマップを概念的に示す図
【図１０】補正係数ＦＶＶＬ２のマップを概念的に示す図
【図１１】実施形態（１）の実行例を示すタイムチャート
【図１２】エアフローメータの出力の挙動を示すタイムチャート
【図１３】実施形態（２）の気筒間ばらつき算出ベースルーチンの処理の流れの一部を示すフローチャート
【図１４】学習時の吸気バルブタイミング進角位置と学習時の排気バルブタイミング進角位置を説明するためのバルブ特性図
【図１５】実施形態（２）の実行例を示すタイムチャート
【図１６】実施形態（３）の吸入空気量制御方式切換ルーチンの処理の流れを示すフローチャート
【図１７】実施形態（３）の実行例を示すタイムチャート
【図１８】実施形態（４）の各気筒の瞬時吸気管圧力極小値算出ルーチンの処理の流れを示すフローチャート
【図１９】エアフローメータの出力と吸気管圧力センサの出力との相関関係を説明するタイムチャート
【図２０】実施形態（５）の各気筒の筒内圧極大値算出ルーチンの処理の流れを示すフローチャート
【符号の説明】
１１…エンジン（内燃機関）、１２…吸気管、１４…エアフローメータ（検出手段）、１５…スロットルバルブ、１８…吸気管圧力センサ（検出手段）、２０…燃料噴射弁、２１…点火プラグ、２２…排気管、２４…排出ガスセンサ、２６…クランク角センサ、２７…ＥＣＵ（気筒間ばらつき学習手段，気筒間ばらつき推定手段）、２８…吸気バルブ、２９…排気バルブ、３０，３１…可変バルブリフト機構。[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an inter-cylinder variation detection apparatus for an internal combustion engine that detects a variation rate of intake air amount between cylinders and an intake air amount for each cylinder.
[0002]
[Prior art]
In general, in an internal combustion engine having a plurality of cylinders, the intake air amount (in-cylinder charged air amount) of each cylinder varies due to a difference in intake manifold shape of each cylinder, a variation in valve clearance of an intake valve, and the like. Such variation in intake air amount between cylinders causes variation in torque and air-fuel ratio of each cylinder. When the torque variation between the cylinders becomes large, the fluctuation of the engine torque in the cycle becomes large, and there is a possibility that an uncomfortable vibration is generated for the driver. Also, if the air-fuel ratio variation among the cylinders becomes large, the fluctuation of the air-fuel ratio of the exhaust gas flowing into the catalyst within the cycle becomes large. May decrease.
[0003]
As a countermeasure for these, some methods have been proposed for correcting torque variation and air-fuel ratio variation between cylinders. For example, as disclosed in Japanese Patent Application Laid-Open No. 62-17342, torque is detected for each cylinder by a torque sensor provided on a crankshaft, and the torque of each cylinder becomes the average torque of all cylinders. In some cases, the fuel injection amount is corrected for each cylinder.
[0004]
Further, as shown in Patent Document 2 (Japanese Patent Application Laid-Open No. 2000-220489), the air-fuel ratio of each cylinder is estimated based on the output of an air-fuel ratio sensor installed in an exhaust pipe, and the air-fuel ratio variation among the cylinders is reduced. In some cases, the fuel injection amount is corrected for each cylinder.
[0005]
In general, the amount of intake air is controlled by a throttle valve, but recently, a variable intake valve mechanism that varies the amount of lift of the intake valve is provided, and the amount of lift of the intake valve is adjusted according to the accelerator opening and the engine operating state. A technology for controlling the intake air amount by making it variable has been developed. In the intake air amount control by the variable intake valve control, by reducing the lift amount of the intake valve, the intake air amount can be reduced without narrowing the intake passage with a throttle valve, so that pumping loss is reduced and fuel consumption is reduced. There is an advantage that can be improved.
[0006]
However, in the intake air amount control by the variable intake valve control, since the lift amount of the intake valve becomes small at a low load, the variation of the actual lift amount with respect to the target lift amount in each cylinder (variation due to component tolerance or assembly tolerance of each cylinder). ) Tends to increase, and the variation in the amount of intake air between cylinders tends to increase. For this reason, the torque and the air-fuel ratio of each cylinder tend to fluctuate under the influence of the intake air amount variation between the cylinders, and the torque variation and the air-fuel ratio variation between the cylinders tend to increase.
[0007]
[Patent Document 1]
JP-A-62-17342 (page 2 etc.)
[Patent Document 2]
JP-A-2000-220489 (Pages 2 to 3 etc.)
[0008]
[Problems to be solved by the invention]
In the above-mentioned Patent Documents 1 and 2, the torque and air-fuel ratio are detected for each cylinder, and the fuel injection amount is corrected for each cylinder based on the detection result. The variation is corrected. However, if the variation in the intake air amount between the cylinders becomes large, it is difficult to correct the torque variation and the air-fuel ratio variation in each cylinder with sufficient accuracy by simply correcting the fuel injection amount. Moreover, even when a plurality of factors such as a variation in the intake air amount and a variation in the intake fuel amount among the cylinders are intertwined and a variation in torque and a variation in the air-fuel ratio occur between the cylinders, it is difficult to correct the variation with sufficient accuracy. .
[0009]
Therefore, in order to correct the torque variation and the air-fuel ratio variation between the cylinders with sufficient accuracy, it is desirable to have a function of correcting the intake air amount variation between the cylinders. It is necessary to develop a technology for accurately detecting variations in the amount of intake air between the two.
[0010]
Therefore, the present inventors have researched a technique for detecting variations in intake air amount between cylinders using an output of an air flow meter that detects an intake air amount flowing through an intake pipe and an intake pipe pressure sensor that detects an intake pipe pressure. However, since the airflow meter and the intake pipe pressure sensor are installed in the intake pipe assembly that is affected by reflected waves of intake pulsation and the intake interference of other cylinders, the output waveforms of the airflow meter and the intake pipe pressure sensor Includes noise components due to reflected waves of intake pulsation and intake interference of other cylinders. For this reason, depending on the operating region, the output waveform of the air flow meter or the intake pipe pressure sensor does not become a pulsating waveform that accurately reflects variations in the intake air amount of each cylinder due to the influence of reflected waves and intake interference, etc. There is a problem that it may not be possible to accurately detect variations in the amount of intake air.
[0011]
In addition, when the vehicle is actually running, the driving state changes every moment, and depending on the driving region, it may not be possible to secure a sufficient time (or the number of times of sampling) for sampling the output of the air flow meter or the intake pipe pressure sensor. This also causes a variation in the intake air amount between cylinders to not be accurately detected.
[0012]
The present invention has been made in view of these circumstances, and accordingly, it is an object of the present invention to make it difficult to detect information accurately reflecting variations in the intake air amount of each cylinder from the output of a detection means such as an air flow meter. Cylinders of an internal combustion engine that can accurately detect variation in intake air amount between cylinders (variation between cylinders) even in an operating region and can improve detection accuracy of variation between cylinders in almost all operating regions. An object of the present invention is to provide an inter-variation detecting device.
[0013]
[Means for Solving the Problems]
In order to achieve the above object, according to a first aspect of the present invention, there is provided an inter-cylinder variation detection apparatus for an internal combustion engine, comprising at least one of an intake air amount flowing through an intake pipe of an internal combustion engine having a plurality of cylinders, an intake pipe pressure, and an in-cylinder pressure. Detecting means for detecting the variation in the intake air amount between cylinders or cylinder-by-cylinder based on the output of the detecting means in an operating region where a predetermined variation learning execution condition is satisfied. The intake air amount is calculated, and the calculated value is learned as an inter-cylinder variation learning value. When the operating range is such that the variation learning execution condition is not satisfied, the inter-cylinder variation estimating means uses the inter-cylinder variation learning value at that time. In this case, the variation rate of the intake air amount between the cylinders or the intake air amount for each cylinder is estimated in accordance with the operation range of the above.
[0014]
In this way, when the variation learning execution condition is satisfied and the operating range is such that information that accurately reflects the intake air amount variation of each cylinder can be obtained from the output of the detection means such as an air flow meter, the inter-cylinder variation learning value In the other operation regions, the cylinder-to-cylinder intake air amount variation rate or the cylinder-by-cylinder intake air amount is estimated in the other operation regions using the accurate cylinder-to-cylinder variation learning value. For an operation region in which the output waveform of the detection means does not become a pulsation waveform that accurately reflects variations in the intake air amount of each cylinder, or an operation region in which the time for sampling the output of the detection means (or the number of times of sampling) cannot be sufficiently secured. However, the variation rate of the intake air amount between cylinders or the intake air amount for each cylinder can be calculated relatively accurately. As a result, it is possible to improve the detection accuracy of the inter-cylinder variation in almost all operation regions.
[0015]
In this case, the variation learning execution condition may be satisfied at the time of fuel cut and / or at the time of cranking. At the time of fuel cut or cranking, combustion does not occur in the cylinder and is not affected by rotation fluctuation due to combustion, so the output waveform of the detection means such as an air flow meter can accurately detect the variation in the intake air amount of each cylinder. The pulsation waveform reflects well, and the learning value of variation between cylinders can be learned with high accuracy.
[0016]
Alternatively, the variation learning execution condition may be satisfied during the deceleration operation. For example, in a system that controls the intake air amount by varying the lift amount of the intake valve, the lift amount of the intake valve decreases during deceleration operation, and the ratio of the variation in the actual lift amount to the target lift amount in each cylinder increases. Therefore, the variation in the intake air amount of each cylinder becomes large, so that during the deceleration operation, the learning value of the variation between cylinders can be learned with high accuracy. In addition, during deceleration operation, even if the valve operating conditions of the intake valve, the exhaust valve, the throttle valve, etc. are changed to valve operating conditions suitable for inter-cylinder variation learning, the driver is noticed to the accompanying torque fluctuations. There is also an advantage that it is difficult.
[0017]
Furthermore, the variation learning execution condition may be satisfied when the engine rotation speed is equal to or lower than a predetermined rotation speed and the intake air amount is equal to or higher than a predetermined amount. As the engine speed decreases, the time of one cycle increases, and the time for sampling the output of the detecting means (or the number of times of sampling) can be increased. In addition, when the intake air amount is large, the amplitude of the intake pulsation increases, and it becomes easy to determine the characteristic value (the maximum value, the minimum value, the area, etc.) of the intake pulsation used for the inter-cylinder variation learning. For these reasons, in the operating region where the engine speed is low and the intake air amount is large, the learning value between cylinders can be learned with high accuracy.
[0018]
By the way, when the variation learning execution condition is satisfied, the valve operating conditions of the intake valve, the exhaust valve, and the throttle valve are not always the operating conditions suitable for the inter-cylinder variation learning.
[0019]
Therefore, at least one of the intake valve, the exhaust valve, and the throttle valve operating condition is changed to a predetermined learning valve operating condition in the operating region where the variation learning execution condition is satisfied. Alternatively, the learning value of the variation between cylinders may be learned. With this configuration, the learning value between cylinders can be learned while the valve operating conditions of the intake valve, the exhaust valve, and the throttle valve are changed to the learning valve operating conditions suitable for learning the learning value between cylinders. Thus, the learning accuracy of the learning value of the variation between cylinders can be improved.
[0020]
Specifically, when the variation learning execution condition is satisfied in the operating range, the opening degree of the throttle valve is changed to the full opening state or the opening degree at which the intake pipe pressure is close to the atmospheric pressure. The variation learning value may be learned. With this configuration, the amplitude of the intake pulsation detected by the detection means such as the air flow meter or the intake pipe pressure sensor disposed on the upstream side of the throttle valve can be increased during the execution of the inter-cylinder variation learning. Since the characteristic values (maximum value, minimum value, area, etc.) of the intake pulsation used for the inter-cylinder variation learning can be easily determined, the learning accuracy of the inter-cylinder variation learning value can be improved.
[0021]
Further, in the operating range where the variation learning execution condition is satisfied, the lift amount of the intake valve is changed to a predetermined lift amount or less or a minimum value to learn the inter-cylinder variation learning value. Is also good. In this way, during the execution of the inter-cylinder variation learning, the lift amount of the intake valve decreases, and the ratio of the variation of the actual lift amount to the target lift amount in each cylinder increases. The variation becomes large, and the learning accuracy of the learning value of the variation between cylinders can be improved.
[0022]
Generally, when there is a valve overlap of the intake and exhaust valves (a period during which both the intake and exhaust valves are open), a phenomenon occurs in which combustion gas in the cylinder blows back to the intake pipe. If the intake valve opens before the top dead center (during the exhaust stroke) or closes after the bottom dead center (during the compression stroke), the piston goes up with the intake valve open. Then, the combustion gas or the intake air is blown back. When the combustion gas or the intake air is blown back, the intake pulsation is disturbed by the influence of the combustion gas or the intake air, so that the learning accuracy of the learning value of the variation between cylinders may be reduced.
[0023]
As a countermeasure, the valve timing of the intake / exhaust valve has no valve overlap and the valve opening period of the intake valve is from the top dead center to the bottom dead center in the operating region where the variation learning execution condition is satisfied. It is also possible to learn the inter-cylinder variation learning value by changing so as to fall within the point. By doing so, during the execution of the inter-cylinder variation learning, it is possible to eliminate the blowback of the combustion gas and the intake air, and it is possible to prevent disturbance of the intake pulsation due to the return of the combustion gas and the intake air. The learning accuracy of the variation learning value can be improved.
[0024]
Further, when the present invention is applied to a system for controlling the intake air amount by controlling the variable intake valve mechanism to vary the lift amount of the intake valve, a great effect can be obtained. In other words, in the intake air amount control by the variable intake valve mechanism, the intake air amount variation among the cylinders tends to be large, and it is desirable to have a function of correcting the intake air amount variation between the cylinders. In addition, since it is necessary to accurately detect variations in the intake air amount between cylinders, the effect of applying the present invention is large.
[0025]
In this case, the current inter-cylinder variation learning value, the lift amount of the intake valve when the inter-cylinder variation learning value is learned, and the current intake valve lift amount are used as claimed in claim 10. It is preferable to estimate the inter-cylinder intake air amount variation rate or the cylinder-by-cylinder intake air amount according to the operation region of (1). In a system in which the intake air amount is controlled by a variable intake valve mechanism, the ratio of the variation of the actual lift amount to the target lift amount changes according to the lift amount of the intake valve, and the variation of the intake air amount between cylinders changes. The variation in the intake air amount between the cylinders changes according to the lift amount of the intake valve. Therefore, by using the inter-cylinder variation learning value, the lift amount of the intake valve when learning the inter-cylinder variation learning value, and the current lift amount of the intake valve, the cylinder corresponding to the current lift amount of the intake valve is used. It is possible to accurately estimate the inter-intake air amount variation rate or the cylinder-by-cylinder intake air amount.
[0026]
Further, the intake air amount control by the variable intake valve mechanism is inhibited and the intake air amount is controlled by the throttle valve until the learning of the inter-cylinder variation learning value is completed. After completion of the learning, the intake air amount control by the variable intake valve mechanism may be executed. In this manner, the control of the intake air amount can be ensured by the intake air amount control by the throttle valve until the learning of the learning value between cylinders is completed. It is possible to avoid the deterioration of the efficiency and the exhaust emission.
[0027]
BEST MODE FOR CARRYING OUT THE INVENTION
<< Embodiment (1) >>
Hereinafter, an embodiment (1) of the present invention will be described with reference to FIGS. First, a schematic configuration of the entire engine control system will be described with reference to FIG. For example, a four-cylinder engine 11, which is an internal combustion engine, has four cylinders of a first cylinder # 1 to a fourth cylinder # 4, and an air cleaner 13 is provided at the most upstream portion of an intake pipe 12 of the engine 11. An air flow meter 14 (detection means) for detecting the amount of intake air is provided downstream of the air cleaner 13. As the air flow meter 14, an air flow meter capable of detecting a reverse flow of intake air is used. A throttle valve 15 whose opening is adjusted by a DC motor or the like and a throttle opening sensor 16 for detecting the throttle opening are provided downstream of the air flow meter 14.
[0028]
Further, a surge tank 17 is provided downstream of the throttle valve 15, and the surge tank 17 is provided with an intake pipe pressure sensor 18 (detection means) for detecting an intake pipe pressure. The surge tank 17 is provided with an intake manifold 19 for introducing air into each cylinder of the engine 11, and a fuel injection valve 20 for injecting fuel is attached near an intake port of the intake manifold 19 of each cylinder. I have. An ignition plug 21 is attached to a cylinder head of the engine 11 for each cylinder, and the air-fuel mixture in the cylinder is ignited by spark discharge of each ignition plug 21.
[0029]
Further, the intake valve 28 and the exhaust valve 29 of the engine 11 are provided with variable valve lift mechanisms 30 and 31 for varying the valve lift amount, respectively. Further, the intake valve 28 and the exhaust valve 29 may be provided with a variable valve timing mechanism for varying the valve timing (opening / closing timing). Note that the exhaust valve 29 may be provided with only the variable valve timing mechanism without providing the variable valve lift mechanism 31.
[0030]
On the other hand, an exhaust pipe 22 of the engine 11 is provided with a catalyst 23 such as a three-way catalyst for purifying CO, HC, NOx and the like in the exhaust gas. An exhaust gas sensor 24 (air-fuel ratio sensor, oxygen sensor, etc.) for detecting / lean or the like is provided. In the cylinder block of the engine 11, a cooling water temperature sensor 25 for detecting a cooling water temperature and a crank angle sensor 26 for outputting a pulse signal every time the crankshaft of the engine 11 rotates at a constant crank angle (for example, 30 ° C. A). Installed. The crank angle and the engine speed are detected based on the output signal of the crank angle sensor 26.
[0031]
Outputs of these various sensors are input to an engine control circuit (hereinafter referred to as “ECU”) 27. The ECU 27 is mainly composed of a microcomputer, and executes various engine control programs stored in a built-in ROM (storage medium), thereby controlling the fuel injection amount of the fuel injection valve 20 according to the engine operating state. The ignition timing of the ignition plug 21 is controlled.
[0032]
Next, the configuration of the variable valve lift mechanism 30 of the intake valve 28 will be described with reference to FIG. Note that the variable valve lift mechanism 31 of the exhaust valve 29 has substantially the same configuration as the variable valve lift mechanism 30 of the intake valve 28, and a description thereof will be omitted.
[0033]
As shown in FIG. 2, a link arm 34 is provided between the camshaft 32 for driving the intake valve 28 and the rocker arm 33, and is rotated above the link arm 34 by a motor 41 such as a stepping motor. A dynamically driven control shaft 35 is provided. An eccentric cam 36 is provided on the control shaft 35 so as to be integrally rotatable. The link arm 34 swings via a support shaft (not shown) at a position eccentric with respect to the axis of the eccentric cam 36. It is movably supported. A swing cam 38 is provided at the center of the link arm 34, and a side surface of the swing cam 38 is in contact with an outer peripheral surface of a cam 37 provided on the camshaft 32. A pressing cam 39 is provided at the lower end of the link arm 34, and the lower end surface of the pressing cam 39 is in contact with the upper end surface of a roller 40 provided at the center of the rocker arm 33.
[0034]
Thus, when the cam 37 rotates by the rotation of the camshaft 32, the swing cam 38 of the link arm 34 moves right and left following the outer peripheral surface shape of the cam 37, and the link arm 34 swings left and right. . When the link arm 34 swings left and right, the pressing cam 39 moves left and right, so that the roller 40 of the rocker arm 33 moves up and down according to the lower end surface shape of the pressing cam 39, and the rocker arm 33 swings up and down. Move. The vertical movement of the rocker arm 33 causes the intake bubble 28 to move up and down.
[0035]
On the other hand, when the eccentric cam 36 rotates by the rotation of the control shaft 35, the position of the support shaft of the link arm 34 moves, and the initial contact point position between the pressing cam 39 of the link arm 34 and the roller 40 of the rocker arm 33 is changed. Change. The lower end surface of the pressing cam 39 of the link arm 34 has a base curved surface 39a formed on the left side thereof with a curvature such that the pressing amount of the rocker arm 33 is 0 (the valve lift amount of the intake valve 28 is 0). The pressing curved surface 39b is formed with a curvature such that the pressing amount of the rocker arm 33 increases toward the right from the base curved surface 39a (the valve lift amount of the intake valve 28 increases).
[0036]
In the case of the high lift mode in which the valve lift amount of the intake valve 28 is increased, the initial contact point between the pressing cam 39 of the link arm 34 and the roller 40 of the rocker arm 33 is moved rightward by the rotation of the control shaft 35. Let it. As a result, when the pressing cam 39 moves left and right due to the rotation of the cam 37, the section of the lower end surface of the pressing cam 39 that contacts the roller 40 moves to the right, and the maximum pressing amount of the rocker arm 33 increases. As a result, the maximum valve lift amount of the intake valve 28 increases, and the period during which the rocker arm 33 is pressed becomes longer, so that the valve opening period of the intake bubble 28 becomes longer.
[0037]
On the other hand, in the low lift mode in which the valve lift of the intake valve 28 is reduced, the initial contact point between the pressing cam 39 of the link arm 34 and the roller 40 of the rocker arm 33 is moved to the left by the rotation of the control shaft 35. Move to Accordingly, when the pressing cam 39 moves left and right due to the rotation of the cam 37, the section of the lower end surface of the pressing cam 39 that contacts the roller 40 moves to the left, and the maximum pressing amount of the rocker arm 33 decreases. As a result, the maximum valve lift of the intake valve 28 is reduced, and the period during which the rocker arm 33 is pressed is shortened, so that the valve opening period of the intake bubble 28 is shortened.
[0038]
In the variable valve lift mechanism 30 described above, if the control shaft 35 is rotated by the motor 41 to continuously move the initial contact point position between the pressing cam 39 of the link arm 34 and the roller 40 of the rocker arm 33, The maximum valve lift amount and the valve opening period (hereinafter simply referred to as “intake valve lift amount”) of the intake valves 28 of all cylinders (# 1 to # 4) can be continuously and collectively changed.
[0039]
The ECU 27 executes a variable valve control program (not shown) stored in the ROM to control the variable valve lift mechanism 30 of the intake valve 28 on the basis of the accelerator opening, the engine operating state, and the like, thereby controlling the intake valve. The intake air amount is controlled by continuously varying the valve lift amount of the valve. In the case of a system using both the variable valve lift mechanism 30 and the variable valve timing mechanism, both the valve lift amount and the valve timing may be continuously varied to control the intake air amount.
[0040]
Further, the ECU 27 calculates the intake air amount variation rate DEV between the cylinders based on the output of the air flow meter 14 by executing each routine for correcting the variation between cylinders described later, and a predetermined variation learning execution condition is satisfied. The inter-cylinder intake air amount variation ratio DEV calculated in the operating region where the engine is operating is learned as an inter-cylinder variation learning value GDEV. When the variation learning execution condition is not satisfied in the operating region, the inter-cylinder intake air variation ratio DEV corresponding to the current operating region is calculated using the inter-cylinder variation learning value GDEV. The intake valve lift amount VVL is corrected based on the amount variation rate DEV to correct the variation in the intake air amount between the cylinders.
[0041]
Hereinafter, processing contents of each routine for correcting variation between cylinders executed by the ECU 27 in the embodiment (1) will be described.
[0042]
[Base routine for calculating variation between cylinders]
The cylinder-to-cylinder variation calculation base routine shown in FIG. 3 is started at the A / D conversion timing of the output voltage of the air flow meter 14 (for example, at a cycle of 4 ms). When this routine is started, first, at step 101, the output voltage VAFM after the filtering process of the air flow meter 14 is read, and then the process proceeds to step 102, where the air flow meter 14 is read using the instantaneous air flow rate GAFM map of FIG. Is converted to an instantaneous air flow rate GAFM that passes through the air flow meter 14. It should be noted that the instantaneous air flow rate GAFM map in FIG. 4 is a map in the case where the air flow meter 14 can also detect a reverse flow. In the case of a reverse flow, the instantaneous air flow rate GAFM has a negative value.
[0043]
Thereafter, the routine proceeds to step 103, where the count value of the crank angle counter CCRNK is read. Since the crank angle counter CCRNK is incremented by “1” every 30 ° C. based on the output signal of the crank angle sensor 26, 24 counts of the crank angle counter CCRNK correspond to one cycle (720 ° C. A). . Note that the crank angle counter CCRNK is reset to “0” when it becomes “24”. The crank rotation position of the crank angle counter CCRNK = 0 corresponds to the compression top dead center (compression TDC) of the first cylinder # 1, and the crank rotation positions of the crank angle counter CCRNK = 6, 12, and 18 correspond to the first rotation position, respectively. The compression TDC is set to correspond to the compression TDC of the third cylinder # 3, the fourth cylinder # 4, and the second cylinder # 2.
[0044]
Then, in the next step 104, the maximum instantaneous air flow rate calculation routine shown in FIG. 6 is executed to calculate the maximum instantaneous air flow rate GAFM maximum value GAPEAK (#i) in the period corresponding to the intake stroke of each cylinder. calculate. Here, (#i) is a cylinder number, which means any of (# 1) to (# 4).
[0045]
After that, the routine proceeds to step 105, where the inter-cylinder intake air amount variation ratio calculation routine shown in FIG. 7 is executed, and the inter-cylinder intake air amount variation ratio DEV is calculated based on the maximum instantaneous air flow rate GAPAK (#i) of each cylinder. (#i) is calculated.
[0046]
Thereafter, the process proceeds to step 106, where it is determined whether the variation learning execution condition is satisfied. Here, the variation learning execution condition is to satisfy both of the following conditions (1) and (2), for example.
(1) The fuel must be cut, cranked, or decelerated (for example, when the brakes are applied or the accelerator is off).
{Circle around (2)} The engine speed NE is equal to or lower than a predetermined speed (for example, 2000 rpm) and the intake air amount GA is equal to or higher than a predetermined amount (for example, a value corresponding to the intake air amount during idling operation).
[0047]
In this case, the condition (1) is set for the following reason. At the time of fuel cut or cranking, combustion does not occur in the cylinder and is not affected by rotation fluctuation due to combustion. Therefore, the output waveform of the air flow meter 14 accurately reflects variations in the intake air amount of each cylinder. A pulsation waveform is obtained, and the cylinder-to-cylinder intake air amount variation rate DEV (#i) can be accurately calculated and learned. Also, in a system that controls the intake air amount by varying the intake valve lift amount, the intake valve lift amount decreases during deceleration operation, and the ratio of the variation of the actual lift amount to the target lift amount in each cylinder increases. In addition, the variation in the intake air amount of each cylinder becomes large, and it is possible to accurately calculate and learn the variation ratio DEV (#i) of the intake air amount between the cylinders.
[0048]
The condition (2) is set for the following reason. As the engine speed NE decreases, the time of one cycle increases, and the time (or the number of times of sampling) of the output of the air flow meter 14 can be increased. When the intake air amount GA is large, the amplitude of the intake pulsation increases, and the characteristic values of the intake pulsation (maximum value, minimum value, area, etc.) used to calculate the inter-cylinder intake air amount variation rate DEV (#i). Can be easily determined. For these reasons, in the operating region where the engine speed NE is low and the intake air amount GA is large, the state in which the inter-cylinder intake air amount variation rate DEV (#i) can be learned with high accuracy.
[0049]
If both of the conditions (1) and (2) are satisfied, the variation learning execution condition is satisfied. However, if there is a condition that does not satisfy any one of the conditions (1) and (2), the variation learning is performed. The learning execution condition is not satisfied.
[0050]
If it is determined that the variation learning execution condition is satisfied, the routine proceeds to step 107, where the cylinder-to-cylinder intake air amount variation ratio DEV (#i) calculated in the operation region where the variation learning execution condition is satisfied is set to the cylinder. It learns (stores) it as the inter-variation learning value GDEV (#i). The processing of steps 105 to 107 described above plays a role as an inter-cylinder variation learning means referred to in the claims.
[0051]
Thereafter, the routine proceeds to step 108, where the average intake valve lift amount VLL obtained when the inter-cylinder variation learning value GDEV (#i) is learned is learned as the reference intake valve lift amount GVVL. In this case, the average intake valve lift amount VLL may be calculated from, for example, the rotational position of the drive motor 41 of the variable valve lift mechanism 30 or the like.
[0052]
Thereafter, the routine proceeds to step 109, where the inter-cylinder variation learning completion flag GXVVL is set to "ON" which means the completion of inter-cylinder variation learning, and this routine ends.
[0053]
On the other hand, if it is determined in step 106 that the variation learning execution condition is not satisfied, the process proceeds to step 110 to determine whether the learning of the cylinder-to-cylinder variation learning value GDEV (#i) has been completed (inter-cylinder variation). Learning completion flag GXVVL = ON). If the inter-cylinder variation learning has not been completed, the present routine is terminated.
[0054]
On the other hand, when it is determined that the learning of the inter-cylinder variation learning value GDEV (#i) has been completed, the process proceeds to step 111, and the map of the inter-cylinder intake air amount variation rate DEV (#i) shown in FIG. 5 is used. By using the inter-cylinder variation learning value GDEV (#i) and the ratio (VVL / GVVL) of the current average intake valve lift amount VLL and the reference intake valve lift amount GVVL, the current average intake valve lift amount VLL is calculated. The corresponding intake air amount variation ratio DEV (#i) is calculated. In this case, the stored value of the cylinder-to-cylinder intake air amount variation rate DEV (#i) calculated in step 105 is rewritten to the estimated value calculated in step 111. The processing of step 111 serves as an inter-cylinder variation estimating means referred to in the claims.
[0055]
Incidentally, the cylinder-by-cylinder intake air amount can be obtained by multiplying the intake air amount variation ratio DEV (#i) between cylinders by the intake air amount GA (average air flow rate) detected by the air flow meter 14.
[0056]
In this routine described above, the inter-cylinder intake air amount variation ratio DEV (#i) is calculated in all the operation regions in step 105, but the process in step 105 is determined as "Yes" in step 106. It may be changed so that it is executed when it is executed. In this case, the inter-cylinder intake air amount variation ratio DEV (#i) is calculated only in the operation region where the variation learning execution condition is satisfied, and the calculated value is learned as the inter-cylinder variation learning value GDEV (#i). .
[0057]
[Routine for calculating the maximum instantaneous air flow rate of each cylinder]
When the instantaneous air flow maximum value calculation routine of each cylinder in FIG. 6 is started in step 104 of FIG. 3, first, in step 201, it is determined whether or not the instantaneous air flow rate GAFM detected by the air flow meter 14 is the maximum value. judge. This maximum value may be determined by comparing the current value of the instantaneous air flow rate GAFM with the previous value, for example, and determining whether or not the change direction of the instantaneous air flow rate GAFM has reversed from increasing to decreasing.
[0058]
If it is determined in step 201 that the instantaneous air flow rate GAFM is not the local maximum value, the routine is terminated without performing the subsequent processing. Thereafter, when the instantaneous air flow rate GAFM reaches the maximum value, the process proceeds to step 202, where it is determined whether or not the crank angle counter CCRNK is a period of 12 to 17 (that is, a period corresponding to the intake stroke of the first cylinder # 1). If CCRNK = 12 to 17, the routine proceeds to step 203, where the maximum value of the instantaneous air flow rate GAFM during the period corresponding to the intake stroke of the first cylinder # 1 is determined by the instantaneous air flow rate maximum of the first cylinder # 1. The value is GAPEAK (# 1).
[0059]
If "No" is determined in step 202, and if it is determined in step 204 that the crank angle counter is CCRNK = 6 to 11 (that is, the period corresponding to the intake stroke of the second cylinder # 2), the process proceeds to step 204. Proceeding to 205, the maximum value of the instantaneous air flow rate GAFM during the period corresponding to the intake stroke of the second cylinder # 2 is defined as the maximum instantaneous air flow rate value GAPEAK (# 2) of the second cylinder # 2.
[0060]
If "No" is determined in steps 202 and 204, and if it is determined in step 206 that the crank angle counter is CCRNK = 18 to 23 (that is, a period corresponding to the intake stroke of the third cylinder # 3). Then, the process proceeds to step 207, where the maximum value of the instantaneous air flow rate GAFM during the period corresponding to the intake stroke of the third cylinder # 3 is set to the maximum instantaneous air flow rate value GAPAK (# 3) of the third cylinder # 3.
[0061]
If all of the above steps 202, 204, and 206 are determined to be "No", that is, if the crank angle counter CCRNK is a period of 0 to 5 (a period corresponding to the intake stroke of the fourth cylinder # 4), the process proceeds to the step. Proceeding to 208, the maximum value of the instantaneous air flow rate GAFM during the period corresponding to the intake stroke of the fourth cylinder # 4 is set as the maximum instantaneous air flow rate value GAPEAK (# 4) of the fourth cylinder # 4.
[0062]
As shown in FIG. 12, the output (instantaneous air flow rate GAFM) of the air flow meter 14 has a pulsating waveform reflecting the intake air amount of each cylinder. Therefore, if the characteristic values (maximum value, minimum value, average value, amplitude value, area, trajectory length, etc.) of the instantaneous air flow rate GAFM detected by the air flow meter 14 for each period corresponding to the intake stroke of each cylinder are calculated, Since the characteristic value of the pulsation waveform reflecting the intake air amount of each cylinder can be calculated, the characteristic value is used to calculate the intake air amount variation ratio between the cylinders reflecting the intake air amount variation of each cylinder. can do.
[0063]
[Routine for calculating variation rate of intake air amount between cylinders]
The routine for calculating the variation rate of the intake air amount between cylinders shown in FIG. 7 (step 105 in FIG. 3) is executed for each stroke (in the case of a four-cylinder engine, every 180 ° C. A). When this routine is started, first, in step 301, the maximum instantaneous air flow rate value GAPEAK (#i) of each cylinder is read, and in the next step 302, the maximum instantaneous air flow rate value GAPEAK (#i) of each cylinder is read. The average value of the maximum instantaneous air flow rate of each cylinder is obtained by smoothing processing between cylinders.
GAPEAKSM (#i) = GAPEAKSM (# i-1) + K1 × {GAPEAK (#i) -GAPEAKSM (# i-1)}
[0064]
Here, K1 is a smoothing coefficient, and GAPEKSM (# i-1) is a smoothed value of the maximum instantaneous air flow rate of the (# i-1) th cylinder.
Thereafter, the routine proceeds to step 303, where the inter-cylinder intake air amount variation ratio DEV (#i) of each cylinder is calculated by the following equation.
[0065]
(Equation 1)

[0066]
The denominator in the above equation is the average value of the smoothed value of the maximum instantaneous air flow rate GAPEKSM (#i) of all cylinders, and K2 is used to convert the variation of the maximum value of the instantaneous air flow rate into the variation of the intake air amount. It is a correction coefficient.
[0067]
As is apparent from the above [Equation 1], the inter-cylinder intake air amount variation rate DEV (#i) of each cylinder is calculated by subtracting the smoothed value GAPEKSM (#i) of the inter-cylinder intake air amount variation rate of all cylinders. Is divided by the average value of the smoothed values of the variation rate of the intake air amount between cylinders, and multiplied by the correction coefficient K2.
[0068]
[Cylinder variation correction routine]
The cylinder-to-cylinder variation correction routine shown in FIG. 8 is executed at a predetermined cycle during engine operation. When this routine is started, first, in step 401, it is determined whether or not learning of the inter-cylinder variation learning value GDEV (#i) has been completed (whether or not the inter-cylinder variation learning completion flag GXVVL = ON). . If the inter-cylinder variation learning has not been completed, this routine ends without executing the process related to the inter-cylinder variation correction after step 402.
[0069]
On the other hand, if it is determined in step 401 that the inter-cylinder variation learning has been completed, the process related to the inter-cylinder variation correction from step 402 is executed as follows. First, at step 402, the inter-cylinder intake air amount variation ratio DEV (#i) finally determined by the inter-cylinder variation calculation base routine of FIG. 3 is read.
[0070]
Thereafter, the routine proceeds to step 403, where the basic lift correction amount FVVL1 (#i) corresponding to the cylinder-to-cylinder intake air amount variation rate DEV (#i) is calculated using the map of FIG. In the map of FIG. 9, in the region where the inter-cylinder intake air amount variation rate DEV (#i) is a positive value, the basic lift correction amount FVVL1 (#i) is a reduced value (negative value), and the inter-cylinder intake air amount variation. In a region where the rate DEV (#i) is a negative value, the basic lift correction amount FVVL1 (#i) is an increased value (plus value). That is, as the intake air amount of a certain cylinder becomes larger than the average intake air amount of all cylinders, the amount of reduction correction by the basic lift correction amount FVVL1 (#i) increases, and conversely, the intake air amount of a certain cylinder becomes , The larger the basic lift correction amount FVVL1 (#i), the larger the increase correction amount. Note that, in a predetermined region where the inter-cylinder intake air amount variation rate DEV (#i) is around 0, the basic lift correction amount FVVL1 (#i) is set to 0, and the intake valve lift amount VVL is not corrected.
[0071]
Then, in the next step 404, after reading the engine rotational speed NE and the intake air amount GA (average air flow rate) detected by the air flow meter 14, the process proceeds to step 405, in which a map of the correction coefficient FVVL2 shown in FIG. 10 is searched. Then, a correction coefficient FVVL2 (#i) corresponding to the current engine operating conditions (for example, the engine speed NE and the intake air amount GA) is calculated for each cylinder.
[0072]
In general, when the intake air amount GA decreases (the intake valve lift amount decreases), the intake valve lift amount is more likely to be affected by the correction. Therefore, the map of the correction coefficient FVVL2 in FIG. The correction coefficient FVVL2 is set to be small.
[0073]
Thereafter, the routine proceeds to step 406, where the basic lift correction amount FVVL1 (#i) of each cylinder is multiplied by the correction coefficient FVVL2 (#i) to obtain the lift correction amount FVVL (#i) of each cylinder.
FVVL (#i) = FVVL1 (#i) × FVVL2 (#i)
Thereafter, the routine proceeds to step 407, where the final target valve lift amount VVLM is obtained by adding the lift correction amount FVVL (#i) of each cylinder to the average valve lift amount VVL of all cylinders before correction.
[0074]
In this case, the period of the crank angle counter CCRNK = 12 to 17 (that is, the period corresponding to the intake stroke of the first cylinder # 1) is determined using the valve lift correction amount FVVL (# 1) of the first cylinder # 1. The target valve lift amount VVLM is calculated by the following equation.
VVLM = VVL + FVVL (# 1)
[0075]
During the period of the crank angle counter CCRNK = 6 to 11 (that is, the period corresponding to the intake stroke of the second cylinder # 2), the final target valve lift is calculated using the valve lift correction amount FVVL (# 2) of the second cylinder # 2. The quantity VVLM is calculated by the following equation.
VVLM = VVL + FVVL (# 2)
[0076]
During the period of the crank angle counter CCRNK = 18 to 23 (that is, the period corresponding to the intake stroke of the third cylinder # 3), the final target valve lift is calculated using the valve lift correction amount FVVL (# 3) of the third cylinder # 3. The quantity VVLM is calculated by the following equation.
VVLM = VVL + FVVL (# 3)
[0077]
During the period of the crank angle counter CCRNK = 0 to 5 (that is, the period corresponding to the intake stroke of the fourth cylinder # 4), the final target valve lift is calculated using the valve lift correction amount FVVL (# 4) of the fourth cylinder # 4. The quantity VVLM is calculated by the following equation.
VVLM = VVL + FVVL (# 4)
[0078]
Thereafter, the process proceeds to step 408, in which the motor 41 of the variable valve lift mechanism 30 of the intake valve 28 is driven at a high speed in accordance with the final target valve lift amount VVLM of each cylinder, which varies according to the intake stroke of each cylinder, The intake valve lift is corrected for each intake stroke of the cylinder to correct the intake air amount of each cylinder. Thereby, the variation in the intake air amount between the cylinders is corrected.
[0079]
In the inter-cylinder variation correction routine described above, the inter-cylinder variation correction is not performed before the inter-cylinder variation learning is completed. However, before the inter-cylinder variation learning is completed, the calculation is performed in step 105 of FIG. The inter-cylinder variation correction may be executed using the inter-cylinder intake air amount variation rate DEV (#i).
[0080]
When the present invention is applied to a variable valve mechanism configured to drive the intake valve 28 of each cylinder by an electromagnetic actuator, the control amount of the electromagnetic actuator of each cylinder is controlled according to the final target valve lift amount VVLM of each cylinder. The correction may correct the intake valve lift amount of each cylinder to correct the intake air amount of each cylinder.
[0081]
An execution example of the above-described embodiment (1) will be described with reference to a time chart shown in FIG. While the inter-cylinder variation correction execution condition is satisfied and the inter-cylinder variation correction execution flag is on, the inter-cylinder intake air amount of each cylinder is determined based on the output (instant air flow rate GAFM) of the air flow meter 14 for each cycle. The variation rate DEV (#i) is calculated, and the variation rate DEV (#i) calculated when the variation learning execution condition is satisfied and the variation learning execution flag is ON is determined as the inter-cylinder variation learning value. While learning as GDEV (#i), the average intake valve lift amount VLL at the time of learning the inter-cylinder variation learning value GDEV (#i) is learned as the reference intake valve lift amount GVVL.
[0082]
Then, after the learning of the inter-cylinder variation learning value GDEV (#i) is completed and the inter-cylinder variation learning completion flag GXVL is turned on, the variation learning execution condition is not satisfied and the inter-cylinder variation learning execution flag is turned off. During this period, the final inter-cylinder intake air is obtained using the inter-cylinder variation learning value GDEV (#i) and the ratio (VVL / GVVL) of the current average intake valve lift amount VLL and the reference intake valve lift amount GVVL. The amount variation rate DEV (#i) is calculated.
[0083]
According to the present embodiment (1) described above, the inter-cylinder intake air calculated in the operating region where the variation learning execution condition is satisfied and the calculation accuracy of the inter-cylinder intake air amount variation rate DEV (#i) can be ensured. Since the amount variation rate DEV (#i) is learned as the inter-cylinder variation learning value GDEV (#i), the inter-cylinder variation learning value GDEV (#i) can be learned with high accuracy. Then, in the operating region where the variation learning execution condition is not satisfied, the cylinder-to-cylinder intake air amount variation ratio DEV (#i) corresponding to the operating region at that time is used by using the accurate cylinder-to-cylinder variation learning value GDEV (#i). ), The operation area in which the output waveform of the air flow meter 14 does not become a pulsating waveform accurately reflecting the variation in the intake air amount of each cylinder, or the time (or the number of times of sampling) for sampling the output of the air flow meter 14 is sufficient. Even in an operation region where it cannot be ensured, the inter-cylinder intake air amount variation ratio DEV (#i) can be calculated relatively accurately. As a result, the detection accuracy of the inter-cylinder variation can be improved in almost all operation regions, so that the intake air amount variation between cylinders can be accurately corrected in almost all operation regions, and the torque variation between cylinders can be improved. And both variations in the air-fuel ratio can be reduced.
[0084]
<< Embodiment (2) >>
By the way, when the variation learning execution condition is satisfied, the valve operating conditions of the intake valve 28, the exhaust valve 29, and the throttle valve 15 are necessarily operating conditions suitable for calculating the cylinder-to-cylinder intake air amount variation rate DEV. Not necessarily.
[0085]
Therefore, in the embodiment (2) of the present invention shown in FIGS. 13 to 15, when the operation range in which the variation learning execution condition is satisfied is satisfied, the valve operating conditions of the intake valve 28, the exhaust valve 29, and the throttle valve 15 are learned. By changing to the valve operating condition, the inter-cylinder intake air amount variation ratio DEV is calculated, and the calculated value is learned as the inter-cylinder variation learning value GDEV.
[0086]
Hereinafter, the processing content of the cylinder-to-cylinder variation calculation base routine of FIG. 13 that executes this processing will be described. When this routine is started, the maximum value GAPEAK (#i of the instantaneous air flow rate GAFM during the period corresponding to the intake stroke of each cylinder is performed by the same processing as in steps 101 to 104 in FIG. 3 described in the embodiment (1). ), The process proceeds to step 106, and it is determined whether or not the variation learning execution condition is satisfied in the same manner as in the embodiment (1).
[0087]
As a result, if it is determined that the variation learning execution condition is satisfied, the routine proceeds to step 106a, where the throttle opening TA is changed to the throttle opening KTA at the time of learning. The throttle opening KTA at the time of this learning is set to a full opening or an opening at which the intake pipe pressure becomes close to the atmospheric pressure. As a result, the amplitude of the intake pulsation detected by the air flow meter 14 disposed on the upstream side of the throttle valve 15 can be increased, and the characteristic of the intake pulsation used for calculating the cylinder-to-cylinder intake air amount variation rate DEV (#i). Since it is easy to determine the value (for example, the maximum value), the state becomes suitable for learning the inter-cylinder variation learning value GDEV (#i).
[0088]
Thereafter, the routine proceeds to step 106b, where the intake valve lift amount VVL is changed to the intake valve lift amount KVVL at the time of learning. The intake valve lift amount KVVL at the time of this learning is set to a predetermined lift amount or less or a minimum value. As a result, the intake valve lift amount VVL decreases, and the ratio of the variation in the actual lift amount to the target lift amount in each cylinder increases. Therefore, the variation in the intake air amount in each cylinder increases, and the inter-cylinder variation learning value GDEV The state becomes suitable for the learning of (#i).
[0089]
Thereafter, the routine proceeds to step 106c, where the intake valve timing INVVT is changed to the intake valve timing KINVT for learning, and the exhaust valve timing EXVVT is changed to the exhaust valve timing KEXVT for learning. As shown in FIG. 14, the intake valve timing KINVT at the time of learning and the exhaust valve timing EXVVT at the time of learning have no valve overlap between the intake valve 28 and the exhaust valve 29 and the opening period of the intake valve 28 is at the top dead center ( It is set so as to fall between TDC) and the bottom dead center (BDC). This makes it possible to eliminate the blowback of the combustion gas and the intake air, thereby preventing the disturbance of the intake pulsation due to the return of the combustion gas and the intake air, and is suitable for learning the inter-cylinder variation learning value GDEV (#i). State.
[0090]
After the valve operating conditions of the throttle valve 15, the intake valve 28, and the exhaust valve 29 are changed to the learning valve operating conditions as described above, the process proceeds to step 106d, and the inter-cylinder operation of FIG. An intake air amount variation ratio calculation routine is executed to calculate an inter-cylinder intake air amount variation ratio DEV (#i) based on the instantaneous air flow maximum value GAPAK (#i) of each cylinder. Thereafter, the routine proceeds to step 107, in which the inter-cylinder intake air amount variation rate DEV (#i) calculated in step 106d is learned (stored) as an inter-cylinder variation learning value GDEV (#i). The other processing is the same as the processing of each step of FIG. 3 described in the embodiment (1).
[0091]
In the embodiment (2) described above, as shown in the time chart of FIG. 15, during the period in which the variation learning execution condition is satisfied and the variation learning execution flag is ON, the throttle opening TA and the intake valve lift amount are set. VVL, intake valve timing INVT, and exhaust valve timing EXVT are changed to learning valve operating conditions (KTA, KVVL, KINVT, KEXVT) suitable for learning variation among cylinders, and the intake air between cylinders calculated in that state. The amount variation rate DEV (#i) is learned as an inter-cylinder variation learning value GDEV (#i). As a result, the learning accuracy of the inter-cylinder variation learning value GDEV (#i) can be improved.
[0092]
<< Embodiment (3) >>
Next, an embodiment (3) of the present invention will be described with reference to FIGS. In the present embodiment (3), the intake air amount control by the variable intake valve control is performed until the learning of the inter-cylinder variation learning value GDEV (#i) is completed by executing the intake air amount control method switching routine shown in FIG. And the intake air amount is controlled by the throttle valve control, and after the learning of the inter-cylinder variation learning value GDEV (#i) is completed, the intake air amount control by the variable intake valve control is executed.
[0093]
The intake air amount control method switching routine of FIG. 16 that executes this process is started at a predetermined cycle (for example, a 4 ms cycle). When this routine is started, first, at step 501, it is determined whether or not learning of the inter-cylinder variation learning value GDEV (#i) has been completed (whether or not the inter-cylinder variation learning completion flag GXVVL = ON). . If the inter-cylinder variation learning has not been completed, the routine proceeds to step 502, where the intake air amount control by the variable intake valve control is prohibited, and the intake air amount control by the throttle valve control is executed. In the intake air amount control by the throttle valve control, the intake air amount is controlled by varying the throttle opening TA based on the accelerator opening, the engine operating state, and the like. In this case, the intake valve lift amount VVL is fixed to a predetermined value, for example, a maximum value (for example, 10 mm).
[0094]
Thereafter, when it is determined in step 501 that the learning of the inter-cylinder variation learning value GDEV (#i) has been completed, the process proceeds to step 503, where the intake air amount control by the variable intake valve control is executed. During the intake air amount control by the variable intake valve control, the intake air amount variation between cylinders is based on the cylinder-to-cylinder intake air amount variation rate DEV (#i) estimated using the cylinder-to-cylinder variation learning value GDEV (#i). And the intake valve amount is controlled by varying the intake valve lift amount VVL based on the accelerator opening, the engine operating state, and the like. In this case, the throttle opening TA is controlled to maintain the intake pipe pressure PM at a predetermined negative pressure (for example, -5 kPa) in order to reduce intake pulsation noise and the like.
[0095]
In the above-described embodiment (3), as shown in the time chart of FIG. 17, before the learning of the inter-cylinder variation learning value GDEV (#i) is completed (when the inter-cylinder variation learning completion flag GXVVL is OFF). Prohibits intake air amount control by variable intake valve control and controls intake air amount by throttle valve control. In this way, until the learning of the inter-cylinder variation learning value GDEV (#i) is completed, the control accuracy of the intake air amount can be ensured by the intake air amount control by the throttle valve control. Deterioration of drivability and deterioration of exhaust emission due to unlearning of the value GDEV (#i) can be avoided.
[0096]
Then, after the learning of the inter-cylinder variation learning value GDEV (#i) is completed and the inter-cylinder variation learning completion flag GXVL is turned on, the intake air amount control by the variable intake valve control is started. In this manner, after the cylinder-to-cylinder variation learning value GDEV (#i) is used to accurately estimate the cylinder-to-cylinder intake air amount variation rate DEV (#i) in almost the entire operation range, the estimated cylinder The intake air amount control by the variable intake valve control can be executed while accurately correcting the intake air amount variation between cylinders using the inter-intake air amount variation ratio DEV (#i), and the intake by the variable intake valve control. The air amount control can be executed with high accuracy.
[0097]
<< Embodiment (4) >>
In the above-described embodiments (1) to (3), the inter-cylinder intake air amount variation ratio DEV (#i) is calculated using the instantaneous air flow rate GAFM detected by the air flow meter 14, but FIGS. In the embodiment (4) of the present invention, the variation ratio DEV (#i) of the intake air amount between cylinders is calculated using the instantaneous intake pipe pressure PMAP detected by the intake pipe pressure sensor 18 (detection means). I have.
[0098]
During the operation of the engine, as shown in FIG. 19, the output of the intake pipe pressure sensor 18 (the instantaneous intake pipe pressure PMAP) pulsates in synchronization with the pulsation of the output of the air flow meter 14 (the instantaneous air flow rate GAFM). When the output (instantaneous air flow rate GAFM) becomes maximum, the output of the intake pipe pressure sensor 18 (instantaneous intake pipe pressure PMAP) becomes minimum.
[0099]
Therefore, in the present embodiment (4), the instantaneous intake pipe pressure minimum value routine for each cylinder shown in FIG. 18 is executed to execute the instantaneous intake pipe pressure minimum value PMPEAK (#i ) Is calculated, and the inter-cylinder intake air amount variation rate DEV (#i) is calculated using the instantaneous intake pipe pressure minimum value PMPEAK (#i).
[0100]
The routine for calculating the minimum value of the instantaneous intake pipe pressure of each cylinder shown in FIG. 18 is started at the A / D conversion timing of the output of the intake pipe pressure sensor 18 (for example, at a cycle of 4 ms). When this routine is started, first, in step 601, it is determined whether or not the instantaneous intake pipe pressure PMAP detected by the intake pipe pressure sensor 18 is a minimum value. This minimum value may be determined, for example, by comparing the current value of the instantaneous intake pipe pressure PMAP with the previous value and determining whether or not the change direction of the instantaneous intake pipe pressure PMAP has reversed from decreasing to increasing.
[0101]
When it is determined in step 601 that the instantaneous intake pipe pressure PMAP is the minimum value, the process proceeds to step 602, where the crank angle counter CCRNK = 12 to 17 (that is, corresponding to the intake stroke of the first cylinder # 1). Period), and if CCRNK = 12 to 17, the routine proceeds to step 603, where the minimum value of the instantaneous intake pipe pressure PMAP in the period corresponding to the intake stroke of the first cylinder # 1 is set to the first value. It is assumed that the instantaneous intake pipe pressure minimum value PMPEAK (# 1) of the cylinder # 1.
[0102]
If "No" is determined in step 602, and if it is determined in step 604 that the crank angle counter CCRNK is a period of 6 to 11 (that is, a period corresponding to the intake stroke of the second cylinder # 2), the process proceeds to step 604. Proceeding to 605, the minimum value of the instantaneous intake pipe pressure PMAP in the period corresponding to the intake stroke of the second cylinder # 2 is set as the instantaneous intake pipe pressure minimum value PMPEAK (# 2) of the second cylinder # 2.
[0103]
If "No" is determined in steps 602 and 604, and if it is determined in step 606 that the crank angle counter CCRNK is a period of 18 to 23 (that is, a period corresponding to the intake stroke of the third cylinder # 3). Then, the process proceeds to step 607, where the minimum value of the instantaneous intake pipe pressure PMAP in the period corresponding to the intake stroke of the third cylinder # 3 is set as the instantaneous intake pipe pressure minimum value PMPEAK (# 3) of the third cylinder # 3.
[0104]
If all of the above steps 602, 604 and 606 are determined to be "No", that is, if the crank angle counter CCRNK is a period of 0 to 5 (a period corresponding to the intake stroke of the fourth cylinder # 4), the step Proceeding to 608, the minimum value of the instantaneous intake pipe pressure PMAP in the period corresponding to the intake stroke of the fourth cylinder # 4 is set as the instantaneous intake pipe pressure minimum value PMPEAK (# 4) of the fourth cylinder # 4.
[0105]
Using the instantaneous intake pipe pressure minimum value PMPEAK (#i) of each cylinder obtained in this manner, the same as the method of calculating the variation rate of the inter-cylinder intake air amount described in the embodiment (1) (see FIG. 7). The variation rate DEV (#i) of the inter-cylinder intake air amount of each cylinder is calculated by the method.
Also in the embodiment (4) described above, the variation rate DEV (#i) of the inter-cylinder intake air amount variation can be accurately calculated.
[0106]
<< Embodiment (5) >>
In the embodiment (5) of the present invention shown in FIG. 20, the in-cylinder pressure of each cylinder is detected by focusing on the characteristic that the in-cylinder pressure increases as the amount of intake air drawn into the cylinder increases. The maximum value of the in-cylinder pressure during the period before and after the compression TDC of the cylinder is obtained. In order to realize this, in the present embodiment (5), an in-cylinder pressure sensor (detection means) for detecting an in-cylinder pressure is provided for each cylinder. The in-cylinder pressure sensor may be one incorporated in the ignition plug 21 or may be provided separately from the ignition plug 21.
[0107]
In the present embodiment (7), the in-cylinder pressure local maximum value calculation routine of each cylinder shown in FIG. 20 is executed at the A / D conversion timing (for example, a cycle of 4 ms) of the in-cylinder pressure sensor. When this routine is started, first, in step 701, it is determined whether or not the in-cylinder pressure CPS detected by the in-cylinder pressure sensor is a local maximum value. This maximum value may be determined, for example, by comparing the current value of the in-cylinder pressure CPS with the previous value and determining whether or not the change direction of the in-cylinder pressure CPS has been reversed from increasing to decreasing.
[0108]
When it is determined in step 701 that the in-cylinder pressure CPS is the maximum value, the process proceeds to step 702, where the crank angle counter CCRNK is 22 to 3 (that is, 90 ° C. before and after the compression TDC of the first cylinder # 1). If CCRNK = 22-3, the routine proceeds to step 703, where the maximum value of the in-cylinder pressure CPS in a period within 90 ° C. before and after the compression TDC of the first cylinder # 1. Is the maximum in-cylinder pressure value CPSPEAK (# 1) of the first cylinder # 1.
[0109]
The determination in step 702 is “No”, and the determination in step 704 is that the crank angle counter is CCRNK = 16 to 21 (that is, a period within 90 ° C. before and after the compression TDC of the second cylinder # 2). If so, the process proceeds to step 705, where the maximum value of the in-cylinder pressure CPS during a period within 90 ° C. before and after the compression TDC of the second cylinder # 2 is set to the maximum in-cylinder pressure value CPSPEAK (# 2) of the second cylinder # 2. .
[0110]
It is determined "No" in the above steps 702 and 704, and in step 706, the crank angle counter CCRNK is a period of 4 to 9 (that is, a period within 90 ° C. before and after the compression TDC of the third cylinder # 3). If it is determined, the process proceeds to step 707, where the maximum value of the in-cylinder pressure CPS during a period within 90 ° C. before and after the compression TDC of the third cylinder # 3 is set to the maximum value CPSPEAK (# 3) of the third cylinder # 3. And
[0111]
If all of the above steps 702, 704, and 706 are determined to be "No", that is, the period of the crank angle counter CCRNK = 10 to 15 (the period within 90 ° C. before and after the compression TDC of the fourth cylinder # 4). In this case, the process proceeds to step 708, where the maximum value of the in-cylinder pressure CPS during a period within 90 ° C. before and after the compression TDC of the fourth cylinder # 4 is set to the maximum in-cylinder pressure value CPSPEAK (# 4) of the fourth cylinder # 4. .
[0112]
Using the in-cylinder pressure maximum value CPSPEAK (#i) of each cylinder obtained in this way, the same method as the method for calculating the variation rate of the intake air amount between the cylinders described in the embodiment (1) (see FIG. 7) is used. The inter-cylinder intake air amount variation ratio DEV (#i) of each cylinder is calculated.
[0113]
Also in the present embodiment (5) described above, it is possible to accurately calculate the inter-cylinder intake air amount variation rate DEV (#i).
Note that the calculation method of the cylinder-to-cylinder variation rate may be changed as appropriate. In short, the maximum value, the minimum value, the average value, the amplitude value, the area, the intake air amount, the intake pipe pressure, and the in-cylinder pressure for each predetermined period. The inter-cylinder variation value may be calculated based on at least one of the trajectory lengths and the like.
[0114]
Further, in each of the above embodiments (1) to (5), the present invention is applied to the system for controlling the intake air amount by the variable intake valve control. May be applied. In a system in which only the intake air amount is controlled by the throttle valve control, the amplitude of the intake pulsation increases in a high load region in which the throttle valve is fully opened or greatly opened, and the characteristic value of the intake pulsation (maximum value, minimum value, area, etc.) ) Can be easily determined, it is preferable to learn the inter-cylinder variation learning value in this high load region.
[0115]
The scope of the present invention is not limited to a four-cylinder engine, and the present invention may be applied to a multi-cylinder engine having five or more cylinders or three or less cylinders.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of an entire engine control system according to an embodiment (1) of the present invention.
FIG. 2 is a front view of a variable valve lift mechanism.
FIG. 3 is a flowchart showing the flow of processing of a base routine for calculating variation between cylinders according to the embodiment (1).
FIG. 4 is a diagram conceptually showing a map for converting an output voltage VAFM of an air flow meter into an instantaneous air flow rate GAFM.
FIG. 5 is a diagram conceptually showing a map of an inter-cylinder intake air amount variation rate DEV.
FIG. 6 is a flowchart showing a processing flow of an instantaneous air flow local maximum value calculation routine of each cylinder according to the embodiment (1).
FIG. 7 is a flowchart showing a processing flow of a routine for calculating a variation rate of intake air amount between cylinders according to the embodiment (1).
FIG. 8 is a flowchart showing a processing flow of an inter-cylinder variation correction routine according to the embodiment (1).
FIG. 9 is a diagram conceptually showing a map of a basic lift correction amount FVVL1.
FIG. 10 is a diagram conceptually showing a map of a correction coefficient FVVL2.
FIG. 11 is a time chart showing an execution example of the embodiment (1).
FIG. 12 is a time chart showing the behavior of the output of the air flow meter.
FIG. 13 is a flowchart showing a part of a processing flow of a cylinder variation calculation base routine according to the embodiment (2).
FIG. 14 is a valve characteristic diagram for explaining an intake valve timing advance position during learning and an exhaust valve timing advance position during learning.
FIG. 15 is a time chart showing an execution example of the embodiment (2).
FIG. 16 is a flowchart showing a processing flow of an intake air amount control method switching routine according to the embodiment (3).
FIG. 17 is a time chart showing an execution example of the embodiment (3).
FIG. 18 is a flowchart showing a flow of a routine for calculating a minimum value of an instantaneous intake pipe pressure of each cylinder according to the embodiment (4).
FIG. 19 is a time chart for explaining the correlation between the output of the air flow meter and the output of the intake pipe pressure sensor.
FIG. 20 is a flowchart showing a processing flow of an in-cylinder pressure maximum value calculation routine of each cylinder according to the embodiment (5).
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 11 ... Engine (internal combustion engine), 12 ... Intake pipe, 14 ... Air flow meter (detection means), 15 ... Throttle valve, 18 ... Intake pipe pressure sensor (detection means), 20 ... Fuel injection valve, 21 ... Spark plug, 22 ... exhaust pipe, 24 ... exhaust gas sensor, 26 ... crank angle sensor, 27 ... ECU (cylinder variation learning means, cylinder variation estimation means), 28 ... intake valve, 29 ... exhaust valve, 30, 31 ... variable valve lift mechanism .

Claims (12)

Detecting means for detecting at least one of an intake air amount flowing through an intake pipe of an internal combustion engine having a plurality of cylinders, an intake pipe pressure, and an in-cylinder pressure;
In an operating region where a predetermined variation learning execution condition is satisfied, an intake air amount variation rate between cylinders or an intake air amount for each cylinder is calculated based on an output of the detection means, and the calculated value is used as an inter-cylinder variation learning value. Inter-cylinder variation learning means for learning as
In the operating range where the variation learning execution condition is not satisfied, the cylinder-to-cylinder variation estimating the cylinder-by-cylinder intake air amount variation rate or the cylinder-by-cylinder intake air amount using the inter-cylinder variation learning value according to the operating region at that time. An inter-cylinder variation detecting apparatus for an internal combustion engine, comprising: an estimating unit. The apparatus according to claim 1, wherein the variation learning execution condition is satisfied at the time of fuel cut and / or at the time of cranking. 3. The apparatus according to claim 1, wherein the variation learning execution condition is satisfied during a deceleration operation. 4. The inter-cylinder engine according to any one of claims 1 to 3, wherein the variation learning execution condition is satisfied when the engine rotation speed is equal to or lower than a predetermined rotation speed and the intake air amount is equal to or higher than a predetermined amount. Variation detection device. The inter-cylinder variation learning means changes at least one of the intake valve, the exhaust valve, and the throttle valve operating condition to a predetermined learning-time valve operating condition during an operating region in which the variation learning execution condition is satisfied. The inter-cylinder variation detection apparatus for an internal combustion engine according to any one of claims 1 to 4, wherein the inter-cylinder variation learning value is learned by learning. The inter-cylinder variation learning means is configured to change the opening of the throttle valve to a fully open position or an opening at which the intake pipe pressure is close to the atmospheric pressure and to change the inter-cylinder variation during an operating region where the variation learning execution condition is satisfied. The apparatus according to claim 5, wherein the learning value is learned. The inter-cylinder variation learning means learns the inter-cylinder variation learning value by changing a lift amount of the intake valve to a predetermined lift amount or less or a minimum value in an operation region where the variation learning execution condition is satisfied. The inter-cylinder variation detection apparatus for an internal combustion engine according to claim 5 or 6, wherein: The cylinder-to-cylinder variation learning means may be configured such that when the variation learning execution condition is satisfied, the valve timing of the intake / exhaust valve has no valve overlap and the valve opening period of the intake valve is from top dead center to bottom dead center. The inter-cylinder variation detection apparatus according to any one of claims 5 to 7, wherein the inter-cylinder variation learning value is learned by changing the learning value to fall within the range. 9. An inter-cylinder engine according to claim 1, further comprising means for controlling an intake air amount by controlling a variable intake valve mechanism to vary a lift amount of the intake valve. Variation detection device. The cylinder-to-cylinder variation estimating means is a cylinder-to-cylinder variation learning value learned by the cylinder-to-cylinder variation learning means, a lift amount of the intake valve when learning the cylinder-to-cylinder variation learning value, and a lift amount of the current intake valve. The cylinder-to-cylinder variation detection apparatus for an internal combustion engine according to claim 9, wherein the cylinder-to-cylinder intake air variation rate or the cylinder-by-cylinder intake air amount is estimated based on the current intake valve lift amount using: Until the learning of the inter-cylinder variation learning value by the inter-cylinder variation learning means is completed, the intake air amount control by the variable intake valve mechanism is prohibited, and the intake air amount is controlled by a throttle valve. 11. The inter-cylinder variation detection apparatus for an internal combustion engine according to claim 9, further comprising means for executing intake air amount control by the variable intake valve mechanism after the completion of the control. Means for controlling the intake valve amount by controlling the variable intake valve mechanism to vary the lift amount of the intake valve;
Detecting means for detecting at least one of an intake air amount flowing through an intake pipe of an internal combustion engine having a plurality of cylinders, an intake pipe pressure, and an in-cylinder pressure;
An inter-cylinder variation learning unit that calculates an intake air amount variation ratio between cylinders or an intake air amount for each cylinder based on an output of the detection unit and learns the calculated value as an inter-cylinder variation learning value;
Until the learning of the cylinder-to-cylinder variation learning value by the cylinder-to-cylinder variation learning means is completed, the intake air amount control by the variable intake valve mechanism is prohibited, and the intake air amount is controlled by the throttle valve. Means for performing intake air amount control by the variable intake valve mechanism after the completion of the operation.

Images