CN105740205B - A kind of switching power converters parameter recorder filtering method and its device - Google Patents

Abstract

The invention provides a switching power supply converter parameter recorder filtering method and its device, belonging to the field of switching power supplies, including an input terminal sampling module, a switching power supply converter, a temperature sensor module, an output terminal sampling module, a processor module, a non-volatile memory module, independent power supply module and USB interface module; the output terminals of the input terminal sampling module and the output terminal sampling module are connected with the processor module; the switching power converter is connected with the processor module through the temperature sensor module; the output terminal of the processor module Connect with the non-volatile memory module; connect the output end of the independent power supply module with the non-volatile memory module for power supply; connect the output end of the non-volatile memory module with the USB interface module; Item parameters are collected and stored in non-volatile memory to realize data analysis after the switching power converter fails, and to find the cause of the failure easily.

Description

Filtering method and device for parameter recorder of switching power supply converter

technical field

The invention relates to the field of switching power supplies, in particular to a filtering method and device for a parameter recorder of a switching power supply converter.

Background technique

With the development of modern power electronics technology, switching power supply technology is also constantly improving. At present, switching power supply is widely used in various electronic equipment due to its small size, light weight and high efficiency. It is an indispensable power supply device for the rapid development of today's electronic information industry. However, when the switching tube in the switching power supply works in the switching state, it will generate spike interference and harmonic signals. Although it has been rectified and filtered, the ripple and noise in the output voltage and output current are still relatively large. These noises will have a great impact on the later data processing, thereby affecting the performance of the switching power supply control system, resulting in the deterioration of the power quality of the power supply device. How to filter out these noises while ensuring data accuracy is very important.

The switching power converter parameter recorder is mainly used for real-time monitoring of the parameters of the switching power converter, and the accuracy of the data is even more important for it. The mathematical model of the traditional filtering algorithm is complicated, the CPU has a large amount of calculation, which is not good for fast sampling, and the historical data cannot be corrected, thus affecting the accuracy of the data recorded by the entire switching power converter parameter recorder. To sum up, in order to make the data recorded by the switching power converter parameter recorder more valuable for research, an accurate, fast, efficient and targeted filtering algorithm is urgently needed.

Contents of the invention

What the present invention needs to solve is the problem that the existing filtering algorithm cannot correct historical data, and provides a filtering method and device for a parameter recorder of a switching power supply converter.

The present invention solves the above problems through the following technical solutions:

A method for filtering a switching power converter parameter recorder, comprising the steps of:

Step 1, A/D sampling is performed on the signal to be tested and the noise signal at a period T;

Step 2, output the sampled data through adaptive cancellation formula e ₁ , e ₂ ... e _n ;

Step 3, carry out fast Fourier transform (FFT) to the output signal at every time T ' to determine the value of the adaptive coefficient K in the adaptive cancellation formula;

Step 4, the signal e _n-1 and e _n output by step 2 predict the signal e' _n+1 of the next point according to the value of K determined in step 3;

Step 5, performing a difference operation on the e _n+1 output by the collection operation and the predicted e'_n+1;

Step 6, when the absolute value of the difference in step 5 is less than or equal to the preset maximum allowable error value Δe _max , take e _n+1 as the effective value of the signal, and return to step 2;

Step 7, when the absolute value of the difference in step 5 is greater than the preset maximum allowable error value Δe _max , shill e _n+1 = e'_n+1;

Step 8, the signal e" _n+1 of the intermediate point is predicted by the signals e _n and e _n+2 output in step 2;

Step 9, performing a difference operation on e _n+1 in step 7 and the predicted e"_n+1;

Step 10, when the absolute value of the difference in step 9 is less than or equal to the preset maximum allowable error value Δe _max , take e _n+1 = e' _n+1 as the effective value of the signal, and return to step 2;

Step 11, when the absolute value of the difference in step 9 is greater than the preset maximum allowable error value Δe _max , shill e _n+1 = (e' _n+1 +e” _n+1 )/2, and return to step 2.

The adaptive cancellation formula in step 2 is

e(n)=d(n)-k*N ₁ (n)

In the formula, d(n) represents the sampling value of the signal to be measured, N ₁ (n) represents the sampling value of the noise signal, k is the adaptive coefficient, the initial value is set to 1, and e(n) is the operation output signal.

The process of determining the value of the adaptive coefficient K in the adaptive cancellation formula in step 3 is:

Step 3.1, performing fast Fourier transform (FFT) on the output signal with a cycle T', converting the original time domain signal into a frequency domain signal;

Step 3.2, calculate the ratio of the signal with a frequency greater than 20HZ to the total signal in the frequency domain signal, when the ratio is greater than 1/20 for the first time, the adaptive coefficient K increases by 0.1, otherwise, the value of the adaptive coefficient K remains unchanged;

Step 3.3, when the ratio of the signal with a frequency greater than 20HZ to the total signal in the next frequency domain signal is smaller than the previous time, continue to increase the value of the adaptive coefficient K until the ratio of the signal with a frequency greater than 20HZ to the total signal in the frequency domain signal The minimum value appears; otherwise, reduce the value of the adaptive coefficient K until the ratio of the signal with a frequency greater than 20HZ to the total signal in the frequency domain signal appears the minimum value.

The difference calculation formula in step 5 is

Δe=|e′ _n+1 -e _n+1 |

Δe <= Δe _max

In the formula, Δe represents the absolute value of the difference between the acquisition operation output signal e _n+1 and the predicted output signal e' _n+1 , Δe _max is the maximum allowable error value, and the value range is 0.01≤Δe _max ≤1.

A switching power converter parameter recorder device, comprising an input sampling module, a switching power converter, a temperature sensor module, an output sampling module, a processor module, a non-volatile memory module, an independent power supply module and a USB interface module; input The output ends of the end sampling module and the output end sampling module are connected to the processor module; the switching power supply converter is connected to the processor module through the temperature sensor module; the output end of the processor module is connected to the non-volatile memory module; the independent power supply module outputs The terminal is connected to the non-volatile memory module for power supply; the output terminal of the non-volatile memory module is connected to the USB interface module.

In the above solution, preferably, the input terminal sampling module includes a first A/D converter, a first differential amplifier circuit, a first noise sampling circuit, a first voltage sensor, a capacitor C1 and a resistor R1, and one end of the capacitor C1 is connected to a voltage The input terminal is connected to the resistance R1, the first noise sampling circuit and one end of the first voltage sensor at the same time, and the other end of the capacitor C1 is connected to the first noise sampling circuit, the first voltage sensor, the other end of the voltage input terminal and grounded; The other end of the resistor R1 is connected to the input end of the switching power converter, and this end is simultaneously connected to the other end of the first differential amplifier circuit; the output end of the first differential amplifier circuit, the first noise sampling circuit and the first voltage sensor are simultaneously connected to the The input end of the first A/D converter is connected; the output end of the first A/D converter is connected with the input end of the processor module.

In the above scheme, it is preferred that the output terminal sampling module includes a 2nd A/D converter, a 2nd differential amplifier circuit, a 2nd noise sampling circuit, a 2nd voltage sensor, a capacitor C2 and a resistor R2, and one end of the capacitor C2 Connect the voltage output end, this end is connected with resistance R2, one end of the second voltage sensor of the second noise sampling circuit at the same time; the other end of the capacitor C2 is connected with the second noise sampling circuit and the second voltage sensor, the other end of the voltage output end and Grounding; the other end of the resistor R2 is connected to the input end of the switching power converter, and this end is simultaneously connected to the other input end of the second differential amplifier circuit; the output of the second differential amplifier circuit, the second noise sampling circuit and the second voltage sensor The terminal is connected with the input terminal of the second A/D converter at the same time; the output terminal of the second A/D converter is connected with the processor module.

In order to prevent the impact of high voltage on the non-volatile memory module, in the above solution, it is preferred that the processor module and the non-volatile memory module are connected in an optocoupler-isolated communication manner.

In order to better collect the temperature of each device in the switching power converter, in the above scheme, it is preferable that the temperature sensor module includes at least two temperature sensors, and the temperature sensor is installed on the inductor, MOS tube, transformer and rectifier diode of the switching power converter nearby.

Advantage and effect of the present invention are:

1. The present invention adopts a bidirectional linear prediction method, which can effectively filter out the influence of switching noise and obtain more accurate sampling data;

2. Using linear forward and backward prediction methods, its mathematical model is simple, can reduce the amount of calculation of the CPU, is more conducive to rapid sampling and is more compatible with microprocessors.

3. Since the historical data is corrected by means of backward prediction, the accuracy of the sampled data is further improved, which makes the present invention more advantageous when used in a switching power supply parameter recorder.

Description of drawings

Fig. 1 is a filtering flow chart of the present invention.

Fig. 2 is an overall device structure diagram of the present invention.

Detailed ways

The present invention will be further described below in conjunction with embodiment.

A switching power converter parameter recorder filtering method, as shown in Figure 1, comprises the following steps:

Step 1. The processor module performs A/D sampling on the signal to be tested and the noise signal at a period T to obtain n sampled values of the signal to be tested and n sampled values of the noise signal, which are respectively set as d ₁ , d ₂ , d ₃ ,… ,d _n and N ₁ ,N ₂ ,N ₃ ,…,N _n ;

Step 2, analyze according to the existing adaptive cancellation system, the signal to be tested:

d(t)=s(t)+N ₀ (t)

Among them, s(t) is a useful signal, N ₀ (t) is a noise signal, which is the noise to be offset, and s(t) is not correlated with N ₀ (t).

The noise signal N ₁ (t) in the signal to be tested can be collected from the noise sampling circuit of the switching power converter parameter recorder. Since the noise signal is mainly attenuated during the collection process, and other parameters remain basically unchanged, in fact, the collected noise signal N ₁ (t) is basically in a linear relationship with the original noise signal N ₀ (t).

which is

y(t)=k*N ₁ (t)

Among them, k is an adaptive coefficient, the initial value is set to k=1, and the general value range is 1≤k≤100, the value of which will directly affect the cancellation effect of the system, if the value is too small, the cancellation effect is not obvious , if the value is too large, it will introduce more noise. y(t) is the actual response determined by N ₁ (t) via the adaptive coefficient k.

System output signal:

e(t)=d(t)-y(t)=s(t)+N ₀ (t)-y(t)

In order to improve the noise cancellation effect, y(t) should be the best estimate of N ₀ (t) in d(t):

Therefore, the output signal of the adaptive cancellation system:

That is, when N ₁ (t) is linearly related to N ₀ (t) and k takes a certain value,

e(t)=s(t)

At this time, the system output signal is the best cancellation situation.

According to the initial value of k=1, useful signals e ₁ , e ₂ ... e _n can be output within time T', T' is set according to the frequency collected by oneself, and the present invention sets T' equal to 1024 points of AD sampling time.

Step 3, after outputting the useful signals e ₁ , e ₂ ... e _n in step 2, the processor module performs fast Fourier transform (FFT) on the output signal of the adaptive cancellation system, and converts the useful signals e ₁ , e ₂ The time domain signal of e _n is converted into an easy-to-analyze frequency domain signal (signal spectrum), so as to determine the content of the signal whose frequency is greater than 20HZ in the output signal of the adaptive cancellation system, that is, the power frequency harmonic, The content of interference signals such as switching power supply pulse width modulation ripple and thermal noise generated by amplifier devices, when processing the relevant data of the switching power converter parameter recorder, the signal not greater than 20HZ can be considered as a useful signal. When increasing the value of the adaptive coefficient k, the signal content greater than 20HZ decreases, then continue to increase the value of the adaptive coefficient k until the ratio of the signal with a frequency greater than 20HZ to the total signal in the frequency domain signal appears the minimum value; otherwise , reduce the value of the adaptive coefficient k until the ratio of the signal with a frequency greater than 20HZ to the total signal in the frequency domain signal has the minimum value, and finally determine the value of the adaptive coefficient k. When k changes for the first time, return to step 2 and output e ₂ , go to the next step. The value range of the adaptive coefficient k is 1≤k≤100, and its value is set to be 1 during initialization, that is, the collected noise signal N ₁ (t) is equal to the original noise signal N ₀ (t) by default.

Step 4, the signal e' _n+1 of the next point is predicted from the signals e _n-1 and e _n output in step 2; due to the extreme sampling time, it can be regarded as a linear change

It can be predicted forward that the next collection data of e _n-1 is

e' _n+1 ＝2e _n -e _n-1

Step 5, when the next data e _n+1 of e _n is collected, the prediction error value Δe is calculated, namely

Δe=|e′ _n+1 -e _n+1 |

Let Δe _max be the maximum allowable error value. This value can be determined according to actual needs, and the value range is 0.01≤Δe _max ≤1. When dealing with the relevant data of the switching power converter parameter recorder, usually Δe _max = 0.1, which can meet the requirements of data accuracy.

Step 6, when in step 5

Δe <= Δe _max

Then e _n+1 does not need to be changed, and e _n+1 is the original collection value as the effective output value.

Step 7, when step 5

Δe＞Δe _max

Shilling e _n+1 = e' _n+1 .

Step 8, the signal e" _n+1 of the intermediate point is predicted by the signals e _n and e _n+2 output in step 2;

It can be predicted backward that the previous collection data of e _n+2 is

e” _n+1 ＝(e _n+2 +e _n )/2

Step 9, performing a difference operation on e _n+1 in step 7 and the predicted e"_n+1;

Step 10, when the absolute value of the difference in step 9 is less than or equal to the preset maximum allowable error value Δe _max , take e _n+1 = e' _n+1 as the effective value of the signal, and return to step 2;

Step 11, when the absolute value of the difference in step 9 is greater than the preset maximum allowable error value Δe _max , shill e _n+1 = (e' _n+1 +e” _n+1 )/2, and return to step 2.

A switching power converter parameter recorder device for realizing the above method, as shown in Figure 2, includes an input sampling module, a switching power converter, a temperature sensor module, an output sampling module, a processor module, and a nonvolatile memory module , an independent power supply module and a USB interface module; the output terminals of the input terminal sampling module and the output terminal sampling module are connected to the processor module; the switching power converter is connected to the processor module through the temperature sensor module; the output terminal of the processor module is connected to the non- The volatile memory module is connected; the output end of the independent power supply module is connected to the non-volatile memory module for power supply; the output end of the non-volatile memory module is connected to the USB interface module.

The input terminal sampling module includes a first A/D converter, a first differential amplifier circuit, a first noise sampling circuit, a first voltage sensor, a capacitor C1 and a resistor R1, one end of the capacitor C1 is connected to the voltage input end, and this end is simultaneously connected to the One end of the resistor R1, the first noise sampling circuit, and the first voltage sensor, the other end of the capacitor C1 is connected to the first noise sampling circuit, the first voltage sensor, and the other end of the voltage input end and grounded; the other end of the resistor R1 is connected to the switch The input end of the power converter, which end is connected to the other end of the first differential amplifier circuit at the same time; the output end of the first differential amplifier circuit, the first noise sampling circuit and the first voltage sensor are simultaneously connected with the first A/D converter The input terminal is connected; the output terminal of the first A/D converter is connected with the input terminal of the processor module, as shown in FIG. 2 .

The output terminal sampling module includes a 2nd A/D converter, a 2nd differential amplifier circuit, a 2nd noise sampling circuit, a 2nd voltage sensor, a capacitor C2 and a resistor R2, one end of the capacitor C2 is connected to the voltage output terminal, and this end is simultaneously connected to the Resistor R2, one end of the second noise sampling circuit and the second voltage sensor; the other end of the capacitor C2 is connected to the second noise sampling circuit and the second voltage sensor, and the other end of the voltage output end is grounded; the other end of the resistor R2 is connected to The input terminal of the switching power supply converter, which is connected to the other input terminal of the second differential amplifier circuit at the same time; the output terminal of the second differential amplifier circuit, the second noise sampling circuit and the second voltage sensor are simultaneously connected with the second A/D conversion The input end of the converter is connected; the output end of the second A/D converter is connected with the processor module, as shown in FIG. 2 .

The processor module and the non-volatile memory module are connected in an optocoupler isolation communication mode, and the optocoupler isolation is to use an optocoupler for isolation. The structure of the optocoupler is equivalent to packaging the light emitting diode and the phototransistor together. The optocoupler isolation circuit makes there is no electrical direct connection between the isolated two parts of the circuit, mainly to prevent interference caused by electrical connections.

The temperature sensor module includes at least two temperature sensors. The temperature sensor is installed near the inductor, MOS tube, transformer and rectifier diode of the switching power converter, which can better collect the temperature of each device.

The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to the embodiments. Those skilled in the art can also make various equivalent modifications or replacements without violating the spirit of the present invention. , these equivalent modifications or replacements are included within the scope of the present application.

Claims (9)

1. a switching power converter parameter recorder filtering method is characterized in that, comprises the steps: Step 1, A/D sampling is performed on the signal to be tested and the noise signal at a period T; Step 2, output the sampled data through adaptive cancellation formula e ₁ , e ₂ ... e _n ; Step 3, carry out fast Fourier transform (FFT) to the output signal at every time T ' to determine the value of the adaptive coefficient K in the adaptive cancellation formula; Step 4, the signal e _n-1 and e _n output by step 2 predict the signal e' _n+1 of the next point according to the value of K determined in step 3; Step 5, performing differential calculation on the e _n+1 output by the collection operation and the predicted e′ _n+1 ; Step 6, when the absolute value of the difference in step 5 is less than or equal to the preset maximum allowable error value Δe _max , take e _n+1 as the effective value of the signal, and return to step 2; Step 7, when the absolute value of the difference in step 5 is greater than the preset maximum allowable error value Δe _max , shilling e _n+1 =e'_n+1; Step 8, the signal e " _n+1 of intermediate point is predicted by the signal e _n and e _n+2 of step 2 output; Step 9, performing difference calculation with e _n+1 in step 7 and prediction e″ _n+1 ; Step 10, when the absolute value of the difference in step 9 is less than or equal to the preset maximum allowable error value Δe _max , take e _n+1 = e' _n+1 as the effective value of the signal, and return to step 2; Step 11, when the absolute value of the difference in step 9 is greater than the preset maximum allowable error value Δe _max , set e _n+1 =(e′ _n+1 +e″ _n+1 )/2, and return to step 2. 2. based on claim 1, it is characterized in that the adaptive cancellation formula in step 2 is e(n)=d(n)-k*N ₁ (n) In the formula, d(n) represents the sampling value of the signal to be measured, N ₁ (n) represents the sampling value of the noise signal, k is the adaptive coefficient, the initial value is set to 1, and e(n) is the operation output signal. 3. based on a kind of switching power converter parameter recorder filtering method according to claim 1, it is characterized in that, the process of determining the value of adaptive coefficient K in the self-adaptive cancellation formula in the step 3 is: Step 3.1, the output signal is subjected to fast Fourier transform (FFT), and the original time domain signal is converted into a frequency domain signal; Step 3.2, calculate the ratio of the signal with a frequency greater than 20HZ to the total signal in the frequency domain signal, when the ratio is greater than 1/20 for the first time, the adaptive coefficient K increases by 0.1, otherwise, the value of the adaptive coefficient K remains unchanged; Step 3.3, when the ratio of the signal with a frequency greater than 20 Hz to the total signal in the next frequency domain signal is smaller than the previous time, the adaptive coefficient K is increased by 0.1 until the ratio of the signal with a frequency greater than 20 Hz to the total signal in the frequency domain signal has a minimum value; Conversely, reduce the value of the adaptive coefficient K until the ratio of the signal with a frequency greater than 20 Hz to the total signal in the frequency domain signal appears the minimum value. 4. based on a kind of switching power converter parameter recorder filtering method according to claim 1, it is characterized in that, the differential calculation formula in the step 5 is Δe=|e′ _n+1 -e _n+1 | Δe <= Δe _max In the formula, Δe represents the absolute value of the difference between the acquisition operation output signal e _n+1 and the predicted output signal e' _n+1 , Δe _max is the maximum allowable error value, and the value range is 0.01≤Δe _max ≤1. 5. A kind of switching power converter parameter recorder device designed based on a kind of switching power converter parameter recorder filtering method described in claim 1, is characterized in that: comprise input terminal sampling module, switching power converter, temperature sensor module, output sampling module, processor module, non-volatile memory module, independent power supply module and USB interface module; The output terminals of the input terminal sampling module and the output terminal sampling module are connected to the processor module; the switching power converter is connected to the processor module through the temperature sensor module; the output terminal of the processor module is connected to the non-volatile memory module; the independent power supply module The output end is connected to the non-volatile memory module for power supply; the output end of the non-volatile memory module is connected to the USB interface module. 6. The parameter recorder device of a switching power converter according to claim 5, wherein the input sampling module includes a first A/D converter, a first differential amplifier circuit, a first noise sampling circuit, A first voltage sensor, a capacitor C1 and an inductor R1, one end of the capacitor C1 is connected to the voltage input end, and this end is connected to the resistor R1, the first noise sampling circuit and one end of the first voltage sensor at the same time, and the other end of the capacitor C1 is connected to The first noise sampling circuit, the first voltage sensor, and the other end of the voltage input end are grounded; the other end of the resistor R1 is connected to the input end of the switching power converter, and this end is connected to the other end of the first differential amplifier circuit; the first The output terminals of the differential amplifier circuit, the first noise sampling circuit and the first voltage sensor are simultaneously connected to the input terminal of the first A/D converter; the output terminal of the first A/D converter is connected to the input terminal of the processor module. 7. A kind of switch power converter parameter recorder device according to claim 5, is characterized in that: described output terminal sampling module comprises the 2nd A/D converter, the 2nd differential amplification circuit, the 2nd noise sampling circuit, The second voltage sensor, capacitor C2 and resistor R2, one end of the capacitor C2 is connected to the voltage output terminal, and this end is connected to the resistor R2, the second noise sampling circuit and one end of the second voltage sensor; the other end of the capacitor C2 Connect the second noise sampling circuit and the second voltage sensor, the other end of the voltage output end and ground; the other end of the resistor R2 is connected to the input end of the switching power converter, and this end is connected to the other input end of the second differential amplifier circuit at the same time; The output ends of the second differential amplifier circuit, the second noise sampling circuit and the second voltage sensor are simultaneously connected to the input end of the second A/D converter; the output end of the second A/D converter is connected to the processor module. 8 . The parameter recorder device of a switching power converter according to claim 5 , wherein the processor module and the non-volatile memory module are connected in an optocoupler-isolated communication manner. 9 . 9. A switching power converter parameter recorder device according to claim 5, characterized in that: the temperature sensor module includes at least two temperature sensors, and the temperature sensors are installed on the inductor, MOS tube, near transformers and rectifier diodes.