CN104343440A - Method and system for detecting mud pressure pulse signal - Google Patents

Abstract

The invention discloses a method and system for detecting a mud pressure pulse signal. The method includes the following steps: S1, receiving a signal containing mud pressure pulse coding information transmitted from a mud medium; S2, receiving the received signal Perform low-pass filtering; S3, perform average algorithm processing on the signal after low-pass filtering; S4, perform Chebyshev low-pass filtering on the signal processed by the average algorithm; S5, amplify the signal after Chebyshev low-pass filtering processing; S6, performing base subtraction processing on the amplified data; S7, demodulating the downhole measurement data from the received signal according to the encoding protocol. In particular, the present invention can maximize the proportion of pulses with the same width as the effective pulse shape after amplifying the signal. From this point of view, the amplitude of the effective pulse and other interference pulses is separated Therefore, it is beneficial to detection and decoding, and the accuracy of decoding is improved.

Description

Detection method and system of mud pressure pulse signal

technical field

The invention relates to the measurement-while-drilling technology in exploration and development of geological resources, in particular to a detection method and system for mud pressure pulse signals.

Background technique

At present, inclinometers while drilling (MWD) and logging while drilling (LWD) are widely used in the field of geological resource exploration and development drilling. They are characterized by the use of wireless systems for measurement, replacing traditional wired The long cable of the measurement system is easy to use, which realizes the real-time measurement of downhole measurement data and improves the efficiency of geological exploration and drilling.

Instruments such as MWD and LWD mostly use mud (drilling fluid) as the medium to transmit downhole measurement data to the ground. The downhole measurement data is first adjusted to a certain sequence of pressure pulse signals according to the set coding rules, the mud pressure is changed through the mud pulser to generate a series of pressure pulses in the mud fluid, and a standpipe pressure sensor is installed on the mud pipeline on the ground Measure the pressure change of the mud, and then complete the signal detection and identification through hardware and software, and demodulate the downhole measurement data information through certain calculation algorithms and decoding rules.

Since the pressure signal will attenuate during mud transmission, and will be affected by various factors such as mud pumps and echoes, a lot of interference will be superimposed on the mud pulse signal, especially in the case of severe noise interference, it is impossible to detect effective The signal seriously affects the normal work of the well site and limits the use of such measurement-while-drilling instruments. Therefore, detecting the weak signal submerged in the strong background noise signal is the key technology for identifying the data transmitted by the coding sequence in the mud channel system.

The patent application with application number 200410005525.0 discloses "a method and device for receiving and detecting mud pressure pulse signal", which uses different methods to solve the noise and interference problems, but the mud pressure pulse signal processed by this method still exists Serious decoding error problem, so it needs to be improved urgently.

Contents of the invention

The purpose of the present invention is to provide a detection method and system of mud pressure pulse signal, which can effectively solve the problems existing in the prior art, especially the decoding error still existing in the mud pressure pulse signal after denoising and de-interference processing serious problem.

In order to solve the above technical problems, the present invention adopts the following technical scheme: a detection method of mud pressure pulse signal, comprising the following steps:

S1, receiving the signal containing mud pressure pulse coding information transmitted from the mud medium;

S2, performing low-pass filtering on the received signal;

S3, performing averaging algorithm processing on the low-pass filtered signal;

S4, performing Chebyshev low-pass filtering on the signal processed by the averaging algorithm;

S5, performing amplification processing on the signal after Chebyshev low-pass filtering;

S6, performing base reduction processing on the enlarged data;

S7. Demodulate the downhole measurement data from the received signal according to the encoding protocol.

Preferably, in step S1, the time slot location encoding method is used to encode and modulate the downhole measurement data into a signal containing mud pressure pulse encoding information; thus, only one pulse is an effective pulse within the set time, and interference can be effectively removed impact of pulses.

More preferably, when time slot positioning coding is used, the synchronization header is 4 equally spaced pulses, and the status code is added after the synchronization header to determine the following decoding sequence, thereby ensuring accurate synchronization while improving decoding efficiency.

Preferably, in step S2, a fourth-order hardware low-pass filter is performed on the received signal.

In step S3 of the present invention, the averaging algorithm processing of the low-pass filtered signal includes: continuously collecting 32 signals, removing the maximum value and the minimum value and performing averaging processing.

Preferably, performing Chebyshev low-pass filtering described in step S4 comprises: select ChebyshevII type filter (stop band etc. ripple), design eight different cutoff frequencies (0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4Hz) filter for filtering, which can be selected according to different situations of noise in the data, and the received signal is convoluted with the designed filter to filter out the noise higher than the set cut-off frequency, so that To achieve the most effective noise reduction effect. After a large number of experimental verifications, the present invention designs the cut-off frequencies of the Chebyshev low-pass filter to be 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4Hz, thereby making the filtering effect of the Chebyshev low-pass filter more obvious.

More preferably, in step S4, when performing Chebyshev low-pass filtering, the passband side frequency is 2Hz, the maximum passband attenuation is less than 0.1dB, the stopband side frequency is 3Hz, and the stopband minimum attenuation is greater than 60dB, so that it can be more effective to filter out high-frequency noise.

Preferably, the amplifying processing of the Chebyshev low-pass filtered signal described in step S5 includes: designing a waveform template similar to the effective mud pressure pulse shape; performing the Chebyshev low-pass filtered signal with the waveform template The arithmetic processing makes the pressure pulse similar to the waveform template amplified. The principle of amplification processing is to design a waveform template that is most similar to the effective pulse shape. This template is dot-multiplied (that is, arithmetic processing) with the filtered waveform. After the dot multiplication, pulses of different shapes will be amplified. The amplification range of similar pulses is the largest, thus achieving the effect of amplifying effective pulses and improving the pulse recognition rate.

In step S6 of the present invention, the third-order Runge-Kutta algorithm is used to subtract the base value of the amplified data, so that the base value of the pulse waveform can be pulled to near 0, which is beneficial to use a threshold value to determine synchronization The effective pulse of the head, and only when the base value is consistent, can the effective pulse be found by searching the maximum energy according to the slot during decoding (because the base value corresponds to the pump pressure, the pump pressure may fluctuate, you need to use when decoding to find the synchronous head) to the threshold, only when the base value is adjusted to 0, can the threshold be used to find the sync head), in addition, the algorithm is used to suppress the error, and the accuracy is high.

Realize the detection system of the mud pressure pulse signal of aforementioned method, comprise: riser pressure sensor, data acquisition processor and computer, data acquisition processor is connected with standpipe pressure sensor and computer respectively; Described riser pressure sensor is used for Measure the mud pressure at the standpipe, and convert the change of standpipe pressure into an electrical signal; the data acquisition processor is used to receive the signal from the standpipe pressure sensor, and perform low-pass filtering and signal averaging algorithm processing on the signal; computer , used to receive the data sent by the data acquisition processor, and sequentially perform Chebyshev low-pass filter processing, amplification processing and base value processing on the data; then demodulate the downhole measurement data according to the coding protocol of the time slot positioning coding method come out.

After a lot of testing and research, the inventor found that: because the low-pass filter filters out only high-frequency signals, the actual mud pressure pulse signal not only has high-frequency interference caused by pumps, but also has low-frequency interference caused by many other factors. If this low-frequency interference is not filtered out, it will easily be mistaken for an effective pulse by the decoding system, thus causing decoding errors.

Compared with the prior art, the present invention performs low-pass filtering, average algorithm processing, Chebyshev low-pass filtering, amplification processing and base value processing on the mud pressure pulse signal, so that the noise of the mud pump and the drilling process Other random clutter noises generated in the filter are filtered out, and weak effective signals are extracted; especially after the present invention amplifies the signal, the proportion of pulses consistent with the effective pulse shape width can be maximized. From this perspective In other words, the amplitudes of effective pulses and other forms of interference pulses are separated, which is beneficial to detection and decoding, and improves the accuracy of decoding; in addition, the present invention adopts the time slot positioning coding method to code and modulate the downhole measurement data to include The signal of the mud pressure pulse encoding information, thus further realizing the function of noise removal from the encoding method (because the time slot method is to define that there is only one effective pulse in a period of time or in a specific number of slots, so according to the method with the largest energy , can lock the effective pulse, and other interference pulses during this period will be ignored; the method with the largest energy, that is, according to the encoding protocol, each pulse will occupy two slot widths, so in the algorithm, by every two A slot width is used to carry out an integral operation on the data processed by correlative amplification and subtraction of the base value, that is, to add the values of all data points in every two slots, and the one with the largest result value is regarded as an effective pulse), which further improves the decoding efficiency. Accuracy. In addition, experimental research shows that: the third-order hardware low-pass filter cannot effectively filter out high-frequency interference, while the fourth-order hardware low-pass filter can meet the system filtering requirements, so the present invention performs fourth-order hardware low-pass to the received signal Filtering saves resources. In addition, the present invention adjusts the parameters of related amplifiers, including width and shape, wherein, the width is consistent with the pulse width, and the shape is divided into top hat and triangle, which can be determined according to the shape of the original pulse, so that it can more effectively filter In addition to various random clutter noise. The present invention also collects 32 signals continuously according to the signal sampling frequency, the data transmission cycle and the test experiment results when the data acquisition processor communicates with the decoding computer data, and performs average processing after removing the maximum value and the minimum value, so that the system performance can be satisfied On the basis of this, the effects of noise removal and interference removal are better. At last, when time slot positioning coding is adopted in the present invention, the synchronization header is 4 equally spaced pulses, and the status code is added after the synchronization header to determine the following decoding sequence, thereby ensuring accurate synchronization while improving decoding efficiency; if the synchronization header When it is 3 pulses, it is easy to mistake the interference pulse as a valid pulse, causing a synchronization error; when the synchronization head is 5 pulses, the synchronization time is increased and the decoding efficiency is reduced; and if the synchronization head is 4 pulses, it can be effective Finding the synchronization not only improves the decoding efficiency but also ensures accurate synchronization; in addition, the next pulse after synchronization is the status code, which parameters of the next data sequence can be determined through the status code, and then the energy maximum can be passed according to the encoding protocol. method to find a valid pulse and decode it.

Description of drawings

Fig. 1 is the work flowchart of a kind of embodiment of the present invention;

Fig. 2 is a schematic diagram of the system structure of an embodiment of the present invention;

Figure 3 is a schematic diagram of the waveform effect after the original data waveform is filtered by a Chebyshev low-pass filter with a cutoff frequency of 1;

Fig. 4 is a schematic diagram of the waveform effect after the original data waveform is filtered by a Chebyshev low-pass filter with a cutoff frequency of 0.5 or 1;

Fig. 5 is the schematic diagram that finds synchronous head and state code when software decoding;

Figure 6 is an effect diagram of software decoding by time slot;

FIG. 7 is a comparison diagram of effects before and after the enlargement process.

Reference signs: 1-standpipe pressure sensor, 2-data acquisition processor, 3-computer.

The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

Detailed ways

Embodiments of the present invention: a detection method of mud pressure pulse signal, as shown in Figure 1, comprises the following steps:

S1, using the time slot positioning coding method to code and modulate the downhole measurement data into a signal containing mud pressure pulse code information; receiving the signal containing mud pressure pulse code information transmitted from the mud medium;

S2, using the low-pass filter MAX292 to perform fourth-order hardware low-pass filtering on the received signal;

S3, performing average algorithm processing on the low-pass filtered signal; the average algorithm processing includes: continuously collecting 32 signals, removing the maximum value and the minimum value and performing average processing;

S4, performing Chebyshev low-pass filtering on the signal processed by the average algorithm; wherein, the passband side frequency is 2Hz, the maximum passband attenuation is less than 0.1dB, the stopband side frequency is 3Hz, and the stopband minimum attenuation is greater than 60dB; Select the ChebyshevII type filter during Bishev low-pass filtering, design the filter of eight kinds of different cut-off frequencies to filter, filter out the noise higher than the set cut-off frequency; Described eight kinds of different cut-off frequencies can be 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4Hz;

S5, amplifying the signal after Chebyshev low-pass filtering; said amplifying processing includes: designing a waveform template similar to the shape of the effective mud pressure pulse; Operation processing, which makes the pressure pulse similar to the waveform template amplified;

S6, using the third-order Runge-Kutta algorithm to perform base value reduction processing on the enlarged data;

S7, demodulate the downhole measurement data from the received signal according to the encoding protocol, in which the synchronization header is 4 equally spaced pulses, and the status code is added after the synchronization header to determine the subsequent decoding sequence; the effect diagram of decoding according to the time slot As shown in Figure 6; the schematic diagram of finding the synchronization header and status code during decoding is shown in Figure 5.

Realize the detection system of the mud pressure pulse signal of said method, as shown in Figure 2, comprise: standpipe pressure sensor 1, data acquisition processor 2 and computer 3, data acquisition processor 2 is connected with standpipe pressure sensor 1 and computer 3 respectively Connect; the standpipe pressure sensor 1 is used to measure the mud pressure at the standpipe, and the change of the standpipe pressure is converted into an electrical signal; the data acquisition processor 2 is used to receive the signal from the standpipe pressure sensor 1, And the signal is processed by low-pass filtering and signal averaging algorithm; the computer 3 is used to receive the data sent by the data acquisition processor 2, and sequentially perform Chebyshev low-pass filtering processing, amplification processing and base subtraction processing on the data ; Then demodulate the downhole measurement data according to the encoding protocol of the time slot positioning encoding method.

The inventor has carried out experimental research to the Chebyshev low-pass filter effect in the above-mentioned embodiment, as shown in Figure 3 (Fig. Schematic diagram, in which, the upper one is the original data waveform, and the lower one is the schematic diagram of the waveform effect filtered by the Chebyshev low-pass filter with a cutoff frequency of 1), and the results show that: using the Chebyshev low-pass filter with a cutoff frequency of 1 After filtering, high-frequency noise can be effectively filtered out. Especially as shown in Figure 4 (Figure 4 is a schematic diagram of the waveform effect after the original data waveform is filtered by a Chebyshev low-pass filter with a cut-off frequency of 0.5 or 1, wherein the upper one is a Chebyshev low-pass filter with a cut-off frequency of 1 The schematic diagram of the waveform effect after filtering by the pass filter, the middle one is the original data waveform, and the lower one is the schematic diagram of the waveform effect after filtering by the Chebyshev low-pass filter with a cutoff frequency of 0.5), using the Chebyshev low-pass filter with a cutoff frequency of 0.5Hz The effect of low-pass filter filtering is better; In addition, the inventor has also carried out research to the effect of amplification processing, as shown in Figure 7 (Fig. Schematic diagram of the data waveform after low-pass filtering, average algorithm processing, and Chebyshev low-pass filtering processing of the invention; the following is the data after low-pass filtering, average algorithm processing, Chebyshev low-pass filtering, and amplification processing of the present invention Waveform schematic diagram), as can be seen from Figure 7: after the amplification process of the present invention, the secondary value of the clutter can be reduced, and the ratio of the pulse consistent with the effective pulse shape width is maximized, thereby pulling the effective pulse from other pulses. The amplitude of the interfering pulse of the shape is beneficial to the detection and decoding, and improves the accuracy of decoding.

Claims (10)

1. a detection method of mud pressure pulse signal, is characterized in that, comprises the following steps: S1, receiving the signal containing mud pressure pulse coding information transmitted from the mud medium; S2, performing low-pass filtering on the received signal; S3, performing averaging algorithm processing on the low-pass filtered signal; S4, performing Chebyshev low-pass filtering on the signal processed by the averaging algorithm; S5, performing amplification processing on the signal after Chebyshev low-pass filtering; S6, performing base reduction processing on the enlarged data; S7. Demodulate the downhole measurement data from the received signal according to the encoding protocol. 2. The detection method of mud pressure pulse signal according to claim 1, characterized in that, in step S1, the downhole measurement data is coded and modulated into a signal containing mud pressure pulse code information by using a time slot positioning coding method. 3. the detection method of mud pressure pulse signal according to claim 2, it is characterized in that, when adopting time slot positioning coding, synchronous header is 4 equally spaced pulses, adds status code behind the synchronous header and determines the decoding sequence of the back . 4. The detection method of the mud pressure pulse signal according to claim 1, characterized in that, in step S2, a fourth-order hardware low-pass filter is performed on the received signal. 5. the detection method of mud pressure pulse signal according to claim 1, is characterized in that, the signal after low-pass filtering described in step S3 is carried out average algorithm processing and comprises: continuously collect 32 signals, remove maximum value and Averaged after the minimum value. 6. according to the detection method of the arbitrary described mud pressure pulse signal of claim 1～5, it is characterized in that, described in the step S4, carry out Chebyshev low-pass filter and comprise: select ChebyshevII type filter for use, design eight different The cutoff frequency filter performs filtering to filter out the noise higher than the set cutoff frequency; the eight different cutoff frequencies are 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4Hz respectively. 7. the detection method of mud pressure pulse signal according to claim 6, it is characterized in that, in step S4, when carrying out Chebyshev low-pass filtering, passband edge frequency is 2Hz, passband maximum attenuation is less than 0.1dB, and the resistance The band side frequency is 3Hz, and the minimum attenuation of the stop band is greater than 60dB. 8. the detection method of mud pressure pulse signal according to claim 1, is characterized in that, described in the step S5 carries out amplifying processing to the signal after the Chebyshev low-pass filter comprises: design and effective mud pressure pulse shape are close The waveform template; the Chebyshev low-pass filtered signal is processed with the waveform template, so that the pressure pulse similar to the waveform template is amplified. 9 . The detection method of mud pressure pulse signal according to claim 8 , characterized in that, in step S6 , the third-order Runge-Kutta algorithm is used to perform base subtraction processing on the amplified data. 10. Realize the detection system of the mud pressure pulse signal of the described method of claim 1～9, it is characterized in that, comprise: riser pressure sensor (1), data acquisition processor (2) and computer (3), data acquisition processing Device (2) is connected with standpipe pressure sensor (1) and computer (3) respectively; Described standpipe pressure sensor (1) is used for measuring the mud pressure at standpipe place, and the change of standpipe pressure is converted into electrical Signal; data acquisition processor (2), used to receive the signal from the standpipe pressure sensor (1), and carry out low-pass filtering and signal averaging algorithm processing on the signal; computer (3), used to receive the data acquisition processor (2) Send the data, and perform Chebyshev low-pass filter processing, amplification processing and base value processing on the data in sequence; then demodulate the downhole measurement data according to the coding protocol of the time slot positioning coding method.