A method of enhancing a pipeline pressure leak signal

Abstract

The application belongs to the technical field of pipeline leakage detection, and particularly relates to a pipeline pressure leakage signal enhancement method, which comprises the following steps: obtaining upstream and downstream pipeline pressure signals; performing trend detection and trend elimination on the upstream and downstream pipeline pressure signals to obtain the upstream and downstream pipeline pressure signals after trend elimination; performing multi-scale characterization on the prominence degree of interval sub-signal amplitude in the upstream and downstream pipeline pressure signals after trend elimination to obtain the prominence degree characterization function of each interval sub-signal in the upstream and downstream signals and the optimal filtering parameter; and multiplying the upstream and downstream pipeline pressure signals filtered by the optimal filtering parameter with the prominence degree characterization function of each interval sub-signal in the upstream and downstream signals to obtain the enhanced upstream and downstream pipeline pressure signals. The method solves the problem of weak leakage signal enhancement and ensures the optimization of the leakage signal enhancement effect.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of pipeline leakage detection, and particularly relates to a pipeline pressure leakage signal enhancement method. BACKGROUND

[0002] At present, for pipeline leakage signal enhancement, mainly there are methods based on wavelet (packet) and EMD or VMD, and the main objects of processing are acoustic emission signals and piezoelectric acceleration sensor output signals.

[0003] Among them, the signal enhancement method based on wavelet (packet) mainly includes: wavelet denoising method based on soft and hard threshold, entropy-based optimal wavelet basis selection, optimal signal frequency band (decomposition scale) selection based on wavelet packet entropy, etc.

[0004] The signal enhancement method based on EMD or VMD mainly includes the following: 1) According to the extreme value selection of the cross-correlation curve of one original reference signal and another IMF obtained by EMD decomposition, the signal denoising method of the effective component (IMF) of the signal. 2) Adopting genetic iterative algorithm to adaptively optimize VMD parameters, and adopting singular value kurtosis difference spectrum to adaptively optimize SVD reconstruction order; then adopting parameter-optimized VMD to decompose the leakage signal, and adopting kurtosis analysis method to screen and reconstruct the decomposed modal components; finally, adopting order-optimized SVD to perform nonlinear filtering on the reconstructed signal, and finally improving the signal-to-noise ratio of the micro-leakage signal. 3) Adopting VMD to decompose the signal, and then adopting the adaptive denoising method according to the fuzzy density function.

[0005] At present, the existing methods have the following problems: 1) The above methods are mainly aimed at the output signal of the piezoelectric sensor, and the corresponding leakage signal is of a burst type, while the pipeline leakage pressure signal often appears as a slowly varying leakage signal due to the long propagation distance, so the signal characteristics are completely different. 2) In the leakage monitoring based on pressure signals, there often exist trend signals due to frequent changes in pipeline transportation process, and once the weak leakage occurs in the pressure rising or falling stage, the leakage signal may be completely submerged by the trend signal. 3) The above methods all assume that the leakage signal appears in the quasi-Gaussian distributed background noise signal, and the assumption is that there is only one abnormal sub-signal in a frame of monitoring signal. However, in the pipeline leakage monitoring based on pressure signals, there may be multiple abnormal sub-signals (including interference signals and leakage signals) in a frame of monitoring signal, and in this case, because which abnormal sub-signal is the leakage signal is unknown, the signal effective frequency band range corresponding to each abnormal signal may be different, and how to determine the object of signal enhancement becomes a difficult problem. 4) Most of the above signal enhancement methods do not take into account the characteristics of the pair of leakage signals, which may cause the objects of upstream and downstream signal enhancement not to be the pair of associated signals, and it is difficult to achieve the expected effect. SUMMARY

[0006] To at least partially overcome the problems existing in the related art, the present application provides a pipeline pressure leakage signal enhancement method, device and system, which helps to enhance the pipeline pressure leakage signal.

[0007] To achieve the above object, the present application adopts the following technical solutions:

[0008] The present application provides a pipeline pressure leakage signal enhancement method, which comprises:

[0009] obtaining an upstream pipeline pressure signal and a downstream pipeline pressure signal;

[0010] trend detection and trend elimination are performed on the upstream pipeline pressure signal and the downstream pipeline pressure signal to obtain a trend-eliminated pipeline upstream and downstream pressure signal SP(ch, k), wherein ch is the number of channels, corresponding to the pipeline upstream and downstream, ch = 1-2; k = 1-N, N is the data frame length;

[0011] The prominence of the interval sub-signal amplitude in the trend-eliminated pipeline upstream and downstream pressure signal SP(ch, k) is characterized by multi-scale to obtain a prominence characterization function of each interval sub-signal in the upstream and downstream pressure signal, denoted as FlagSum;

[0012] According to the prominence characterization function FlagSum of each interval sub-signal in the upstream and downstream signal, the position of the most prominent interval sub-signal in the upstream and downstream signal is determined;

[0013] determine whether the most prominent interval sub-signal in the upstream and downstream signal is a correlation signal pair;

[0014] If the most prominent interval sub-signal is a correlation signal pair, the best filtering parameter is determined;

[0015] Wavelet denoising and high-pass filtering are performed on the trend-eliminated pipeline upstream and downstream pressure signal SP(ch, k) using the best filtering parameter to obtain a filtered pipeline upstream and downstream pressure signal P(ch, k), wherein ch is the number of channels, ch = 1-2, corresponding to the pipeline upstream and downstream; k = 1-N, N is the data frame length;

[0016] The filtered pipeline upstream and downstream pressure signal P(ch, k) obtained is multiplied by the prominence characterization function FlagSum of each interval sub-signal in the upstream and downstream signal to obtain an enhanced pipeline upstream and downstream pressure abnormal signal.

[0017] Further, the determination of whether the most prominent interval sub-signal in the upstream and downstream signal is a correlation signal pair further comprises:

[0018] if the most prominent interval sub-signal is not a pair of associated signals, judging the negative extreme value in the prominence degree representation function of the upstream pipeline pressure signal and the downstream pipeline pressure signal;

[0019] taking the most prominent negative polarity interval sub-signal in the signal with larger negative extreme value as a reference, searching for an associated signal in the other signal;

[0020] determining the optimal filtering parameter according to the associated signal searched in the other signal.

[0021] Further, the prominence degree of the interval sub-signal amplitude in the pipeline upstream and downstream pressure signals after the trend is eliminated is characterized in multiple scales to obtain the prominence degree representation function of each interval sub-signal in the upstream and downstream signals, including:

[0022] selecting a wavelet base, a wavelet filtering scale adjustment range, an initial cutoff frequency of high-pass filtering, and a cutoff frequency adjustment (increase) step, filtering the pipeline upstream and downstream pressure signals after the trend is eliminated to obtain filtered signals;

[0023] dividing the filtered upstream and downstream first signals into positive and negative intervals to obtain the interval number NC of the signals, the starting position SSt(j) and the ending position SEnd(j) of each interval sub-signal, the peak value Peak(j) of each interval sub-signal and its position PeakPos(j), and j is the interval serial number, j ∈ [1, NC];

[0024] based on the hierarchical clustering method, the interval sub-signal peak values of the signals are characterized in multiple scales to obtain the prominence degree representation function FlagSum of each interval sub-signal in the signals.

[0025] Further, the interval sub-signal peak values of the signals are characterized in multiple scales using the hierarchical clustering method to obtain the prominence degree representation function FlagSum of each interval sub-signal in the signals, including:

[0026] the single connection distance threshold adjustment range of the hierarchical clustering is set as Set_Dist(l)-SetDist(M), the step is step_D, and Set_Dist(1)<Set_Dist(2)<…<SetDist(M-1)<SetDist(M), a total of M groups;

[0027] taking the absolute value of the peak value sequence of the interval sub-signals in the signals, and arranging them in descending order from large to small to obtain the descendingly arranged interval sub-signal peak value sequence Sort_Peak(j) with positive amplitude, where j ∈ [1, NC], NC is the total number of positive and negative intervals of the current signal, and

[0028] Sort_Peak(1)>Sort_Peak(2)>…>Sort_Peak(NC);

[0029] mapping relationship between interval sequence numbers of interval sub-signals before and after sorting;

[0030] differential calculation is performed on the descendingly arranged interval sub-signal peak value sequence with positive amplitude to obtain a differential sequence d_SortPeak(j) of the descendingly arranged interval sub-signal peak value sequence with positive amplitude, j∈[1, NC-1], NC being the total number of positive and negative intervals of the current signal;

[0031] According to the current single connection distance threshold Set_Dist(m), m∈[1, M], M being the number of groups of the single connection distance threshold, the prominent interval sub-signals in the interval sub-signals of the current signal are determined to obtain a prominence degree representation FlagS(m, j) of the interval sub-signals of the current signal under the current single connection distance threshold condition, wherein m corresponds to the sequence number of the single connection distance threshold, and j corresponds to the interval sequence number before sorting.

[0032] Further, it further comprises:

[0033] The single connection distance threshold Set_Dist(m), m∈[1, M], is adjusted step by step according to the step distance step_D to obtain the prominence degree representation FlagS(m, j) of the interval sub-signals of the current signal under different single connection distance threshold conditions, and the multi-scale representation FlagZ and the multi-scale representation function FlagSum of the prominence degree of the interval sub-signals of the current signal under the current wavelet base and wavelet filter scale, the current high-pass filter cutoff frequency condition, and in the single connection distance threshold adjustment range Set_Dist(l)-SetDist(M) are obtained:

[0034]

[0035] wherein, l∈[1, L], L being the product of the wavelet filter scale adjustment times and the high-pass filter cutoff frequency adjustment times.

[0036] Further, it further comprises:

[0037] The negative maximum value of the interval sub-signals in FlagZ is searched, and the negative maximum value of the interval sub-signals in FlagZ is stored in the array maxFlag(l);

[0038] The start and end positions of the interval sub-signals are stored in arrays SSt(l), SEnd(l) respectively, and the wavelet filter scale and high-pass filter cutoff frequency are stored in arrays Ws_Tab(l) and FC_Tab(l).

[0039] Further comprising:

[0040] The cutoff frequency of the high-pass filter is adjusted according to the adjustment step of the cutoff frequency of the high-pass filter, the multiscale representation FlagZ of the prominence of the interval sub-signals of the signal within the single connection distance threshold range under the current wavelet basis and wavelet filter scale is recalculated, and the values of the multiscale representation function FlagSum(l, k) and maxFlag(l), SSt(l), SEnd(l), Ws_Tab(l), FC_Tab(l) under the current filter parameter condition are obtained, wherein l∈[1, L] and k∈[1, N].

[0041] Further comprising:

[0042] The wavelet filter scale is adjusted, the multiscale representation FlagZ of the prominence of the interval sub-signals of the signal within the single connection distance threshold range under the condition of different high-pass filter cutoff frequencies is recalculated, and the values of the multiscale representation function FlagSum(l, k) and maxFlag(l), SSt(l), SEnd(l), Ws_Tab(l), FC_Tab(l) sequence under the current filter parameter condition are obtained, wherein l∈[1, L] and k∈[1, N];

[0043] Finally, the multiscale representation function FlagSum of the signal is obtained, which is marked as FlagSum_1(k), k∈[1, N].

[0044] Further, the judgment of whether the most prominent interval sub-signals in the upstream and downstream signals are a pair of associated signals comprises:

[0045] The negative extreme value and its corresponding interval in the prominence representation function corresponding to the pair of upstream and downstream pressure signals are searched, and the prominence representation function signal within the corresponding interval range is intercepted for cross-correlation calculation;

[0046] If the time delay value obtained by the correlation calculation is within the preset time delay interval, the negative extreme value interval sub-signals in the upstream and downstream pressure signals within the interval range are associated signals;

[0047] If the time delay value obtained by the correlation calculation is outside the preset time delay interval, the absolute values of the most prominent negative polarity interval sub-signals in the prominence degree representation functions of the upstream and downstream are compared, the interval sub-signal with the larger absolute value is taken as a reference, and the negative polarity interval sub-signal with the largest absolute amplitude in the preset time delay interval in the prominence degree representation function of the other channel is searched as a correlation signal.

[0048] The values of the FlagZ, FlagSum(l, k), maxFlag(l), SSt(l), SEnd(l), Ws_Tab(l), and FC_Tab(l) sequences corresponding to the interval sub-signals are recalculated.

[0049] Further, the best filtering parameter is determined according to the correlation signal searched in the other channel, and the best filtering parameter includes:

[0050] The negative extreme value in the negative maximum value array maxFlag(l) is searched.

[0051] The best filtering parameter of the corresponding signal is retrieved from the arrays Ws_Tab(l) and FC_Tab(l) corresponding to the signal according to the serial number corresponding to the negative extreme value.

[0052] The above technical solutions are adopted in the present application, and the present application at least has the following beneficial effects:

[0053] The application provides a pipeline pressure leakage signal enhancement method, which comprises the following steps: acquiring an upstream pipeline pressure signal and a downstream pipeline pressure signal; performing trend detection and trend elimination on the upstream pipeline pressure signal and the downstream pipeline pressure signal to obtain a pipeline upstream and downstream pressure signal SP(ch, k) without trend, wherein ch is the number of channels, corresponding to the pipeline upstream and downstream, ch=1-2; k=1-N, N is the data frame length; performing multi-scale characterization on the prominence of the interval sub-signal amplitude in the pipeline upstream and downstream pressure signal SP(ch, k) without trend to obtain a prominence characterization function of each interval sub-signal in the upstream and downstream pressure signal, denoted as FlagSum; determining the position of the most prominent interval sub-signal in the upstream and downstream signal according to the prominence characterization function FlagSum of each interval sub-signal in the upstream and downstream signal; judging whether the most prominent interval sub-signal in the upstream and downstream signal is a correlation signal pair; if the most prominent interval sub-signal is a correlation signal pair, determining the optimal filtering parameter; performing wavelet denoising and high-pass filtering on the pipeline upstream and downstream pressure signal SP(ch, k) without trend by using the optimal filtering parameter to obtain a filtered pipeline upstream and downstream pressure signal P(ch, k), wherein ch is the number of channels, ch=1-2, corresponding to the pipeline upstream and downstream; k=1-N, N is the data frame length; performing multiplication operation on the filtered pipeline upstream and downstream pressure signal P(ch, k) and the prominence characterization function FlagSum of each interval sub-signal in the upstream and downstream signal to obtain an enhanced pipeline upstream and downstream pressure abnormal signal. The signal enhancement problem of the leakage signal appearing in the trend pressure signal is solved, the problem that the evaluation target is difficult to determine when there are multiple abnormal sub-signals in a frame of signal is overcome, the characteristics that the leakage signal appears in pairs in two signals are considered in the signal enhancement, and the optimization of the upstream and downstream leakage signal enhancement effect is ensured.

[0054] It should be understood that the general description above and the following detailed description are only exemplary and explanatory, and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0056] Figure 1 It is a flow chart of a pipeline pressure leakage signal enhancement method according to an exemplary embodiment;

[0057] Figure 2 It is a schematic diagram of a pipeline original pressure signal according to an exemplary embodiment;

[0058] Figure 3 is a schematic diagram of a pressure signal after trend elimination according to an exemplary embodiment;

[0059] Figure 4 is a schematic diagram of a signal after db9 wavelet denoising of the upstream pressure signal according to an exemplary embodiment;

[0060] Figure 5 is a schematic diagram of a signal after filtering by a first-order RC high-pass filter with a discrete frequency of 200 Hz according to an exemplary embodiment;

[0061] Figure 6 is a schematic diagram of a FlagPZ obtained from the upstream signal according to an exemplary embodiment;

[0062] Figure 7 is a schematic diagram of a FlagSum obtained from the upstream signal according to an exemplary embodiment;

[0063] Figure 8 is a schematic diagram of a FlagSum of the upstream and downstream according to an exemplary embodiment;

[0064] Figure 9 is a schematic diagram of a result AP according to an exemplary embodiment;

[0065] Figure 10 is a schematic diagram of a final signal enhancement result according to an exemplary embodiment. DETAILED DESCRIPTION

[0066] In order to make the purposes, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be described in detail below. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0067] Please refer to Figure 1 , Figure 1 is a flowchart of a pipeline pressure leakage signal enhancement method according to an exemplary embodiment, as shown in Figure 1 , the pipeline pressure leakage signal enhancement method comprises the following steps:

[0068] Step S101, obtaining an upstream pipeline pressure signal and a downstream pipeline pressure signal;

[0069] As shown in Figure 2 , Figure 2 is an original pressure signal obtained in an embodiment of the present application.

[0070] Step S102, trend detection and trend elimination are performed on the upstream pipeline pressure signal and the downstream pipeline pressure signal to obtain the trend-eliminated pipeline upstream and downstream pressure signal SP(ch, k), where ch is the number of channels corresponding to the pipeline upstream and downstream, ch = 1-2; k = 1-N, N is the data frame length;

[0071] As shown in Figure 3 , the trend-eliminated pressure signal is the pressure signal after trend elimination in an embodiment of the present application. Figure 3

[0072] Specifically, trend detection and trend elimination are performed on the pipeline upstream and downstream pressure signal to obtain the trend-eliminated pipeline upstream and downstream pressure signal SP(ch, k), where ch is the number of channels corresponding to the pipeline upstream and downstream pressure signal, ch = 1-2; k = 1-N, N is the data frame length. Trend elimination can be achieved by using a fitting-based method, a decomposition method, and a high-pass filtering method. The fitting-based method, the decomposition method, and the high-pass filtering method all belong to the prior art, and the present application is an improvement thereof.

[0073] Step S103, the prominence of the interval sub-signal amplitude in the trend-eliminated pipeline upstream and downstream pressure signal SP(ch, k) is characterized by multi-scale to obtain the prominence characterization function of each interval sub-signal in the upstream and downstream pressure signal, denoted as FlagSum.

[0074] Step S104, according to the prominence characterization function FlagSum of each interval sub-signal in the upstream and downstream signal, the position of the most prominent interval sub-signal in the upstream and downstream signal is determined.

[0075] Step S105, it is judged whether the most prominent interval sub-signal in the upstream and downstream signal is a correlation signal pair.

[0076] Specifically, the interval sub-signal corresponding to the position of the most prominent negative polarity interval sub-signal in the upstream and downstream is intercepted, and it is judged by correlation calculation whether the interval sub-signal corresponding to the position of the most prominent negative polarity interval sub-signal in the upstream and downstream signal is a correlation signal pair.

[0077] The negative extreme value and its corresponding interval in the prominence characterization function FlagSum corresponding to the upstream and downstream signal pair are searched, and the interval sub-signal in the corresponding FlagSum is intercepted for cross-correlation calculation. If the actual calculated time delay delay >=-DELAY and delay <=DELAY (DELAY is the maximum time delay of the leakage signal in the pipeline in reverse transmission or forward transmission, which can be calculated according to the pipeline length and the sound speed in the medium), the intercepted signal is a correlation signal.

[0078] ​Step S106, if the most prominent interval sub-signal is a correlation signal pair, determine the optimal filter parameter;

[0079] Specifically, according to the maxFlag sequence corresponding to the upstream and downstream pressure signals, search for the sequence number where the maximum value of maxFlag is located, and according to the sequence number corresponding to the maximum value, index to the corresponding wavelet filter scale and high-pass filter cutoff frequency from the Ws_Tab and FC_Tab arrays corresponding to the upstream and downstream pressure signals, as the optimal filter parameters of the upstream and downstream pressure signals.

[0080] Specifically, according to the sequence number corresponding to the maximum value, index to the corresponding wavelet filter scale and high-pass filter cutoff frequency from the Ws_Tab and FC_Tab arrays corresponding to the upstream and downstream pressure signals, and perform wavelet denoising and high-pass filtering on the original trend-removed pressure signal SP(ch, k) to obtain a signal P(ch, k), ch = 1-2, k = 1-N.

[0081] Step S107, use the optimal filter parameters to perform wavelet denoising and high-pass filtering on the trend-removed upstream and downstream pipeline pressure signals SP(ch, k) to obtain filtered upstream and downstream pipeline pressure signals P(ch, k), where ch is the channel number, ch = 1-2, corresponding to the upstream and downstream of the pipeline; k = 1-N, N is the data frame length.

[0082] Step S108, multiply the filtered upstream and downstream pipeline pressure signals P(ch, k) by the prominence degree representation function FlagSum of each interval sub-signal in the upstream and downstream signals to obtain enhanced upstream and downstream pipeline pressure anomaly signals.

[0083] According to the formula AP = P * FlagSum, AP is the signal obtained after signal enhancement according to the above.

[0084] It can be understood that the application provides a pipeline pressure leakage signal enhancement method, which comprises the following steps: acquiring an upstream pipeline pressure signal and a downstream pipeline pressure signal; performing trend detection and trend elimination on the upstream pipeline pressure signal and the downstream pipeline pressure signal to obtain a pipeline upstream and downstream pressure signal SP(ch, k) without trend, wherein ch is the number of channels, corresponding to the pipeline upstream and downstream, ch = 1-2; k = 1-N, N is the data frame length; performing multi-scale characterization on the prominence of the interval sub-signal amplitude in the pipeline upstream and downstream pressure signal SP(ch, k) without trend to obtain a prominence characterization function of each interval sub-signal in the upstream and downstream pressure signal, denoted as FlagSum; determining the position of the most prominent interval sub-signal in the upstream and downstream signal according to the prominence characterization function FlagSum of each interval sub-signal in the upstream and downstream signal; judging whether the most prominent interval sub-signal in the upstream and downstream signal is a correlation signal pair; if the most prominent interval sub-signal is a correlation signal pair, determining the best filtering parameter; performing wavelet denoising and high-pass filtering on the pipeline upstream and downstream pressure signal SP(ch, k) without trend by using the best filtering parameter to obtain a filtered pipeline upstream and downstream pressure signal P(ch, k), wherein ch is the number of channels, ch = 1-2, corresponding to the pipeline upstream and downstream; k = 1-N, N is the data frame length; performing multiplication operation on the filtered pipeline upstream and downstream pressure signal P(ch, k) and the prominence characterization function FlagSum of each interval sub-signal in the upstream and downstream signal to obtain an enhanced pipeline upstream and downstream pressure abnormal signal. The signal enhancement problem of the leakage signal appearing in the trend pressure signal is solved, and the problem of difficult to determine the evaluation target when there are multiple abnormal sub-signals in a frame of signal is overcome; the characteristics of the leakage signal appearing in pairs in two signals are considered in signal enhancement, and the optimization of the upstream and downstream leakage signal enhancement effect is ensured.

[0085] As a further improvement of the above method, in one embodiment, the judging whether the most prominent interval sub-signal in the upstream and downstream signal is a correlation signal pair further comprises:

[0086] If the most prominent interval sub-signal is not a correlation signal pair, judging the negative extreme value in the prominence characterization function of the upstream pipeline pressure signal and the downstream pipeline pressure signal;

[0087] Taking the most prominent negative polarity interval sub-signal in the signal with a larger negative extreme value as a reference, searching for a correlation signal in the other signal;

[0088] Determining the best filtering parameter according to the correlation signal searched in the other signal.

[0089] Specifically, if the most prominent interval sub-signal is not a correlation signal pair, the absolute values of the most prominent negative polarity interval sub-signal in the two prominent degree representation functions FlagSum signals are compared, and the interval sub-signal of the signal in the signal with the larger absolute value is selected as the reference signal. Within the cross-correlation time delay ± DELAY range, the corresponding correlation interval sub-signal in the other signal is determined, and the correlation interval sub-signal is taken as the target. The parameters (FlagZ(ch,j), SSt(ch,j), SEnd(ch,j), Ws_Tab(ch,j) and FC_Tab(ch,j)) of the correlation interval sub-signal are reacquired by the method in step S103.

[0090] In some embodiments, the optimal filtering parameters are determined according to the correlation signal searched in the other signal, including:

[0091] Searching for a negative extreme value in the negative extreme value array maxFlag(l);

[0092] According to the serial number corresponding to the negative extreme value, the optimal filtering parameters of the corresponding signal are retrieved from the arrays Ws_Tab(l) and FC_Tab(l) corresponding to the signal.

[0093] Wherein, the corresponding different wavelet filtering scale and high-pass filter cutoff frequency are obtained from the multi-scale representation FlagZ of the prominence degree of the signal interval sub-signal.

[0094] In some embodiments, the prominence degree of the interval sub-signal amplitude in the pipeline upstream and downstream pressure signals after trend elimination is multi-scale represented, and the prominence degree representation function of each interval sub-signal in the upstream and downstream signals is obtained, including:

[0095] Selecting a wavelet basis, a wavelet filtering scale adjustment range, a high-pass filtering initial cutoff frequency, and a cutoff frequency adjustment (increase) step, filtering the pipeline upstream and downstream pressure signals after trend elimination to obtain filtered signals;

[0096] Dividing the filtered upstream and downstream first signals into positive and negative intervals to obtain the interval number NC of the signals, the starting position SSt(j) and the ending position SEnd(j) of each interval sub-signal, the peak value Peak(j) and its position PeakPos(j) of each interval sub-signal, and j is the interval serial number, j ∈ [1, NC];

[0097] Based on the hierarchical clustering method, the prominence degree representation function FlagSum of each interval sub-signal in the signal is obtained by multi-scale representing the peak values of the interval sub-signals of the signal.

[0098] In one embodiment, the selected Db9 wavelet base is used, the wavelet filter scale variation range [Scale_S, Scale_E] is [5-10], a first-order RC high-pass filter is used, the time constant Tc=4s, the cutoff frequency of the high-pass filter is adjusted by adjusting the discretization frequency, the adjustment range is [100Hz-2000Hz], the step is 100Hz, the signal amplitude characteristic single connection threshold range [SetDist(1), SetDist(M)] is [0.01-0.5] and the step_D=0.01, and the total iteration number M=50. The product of the number of wavelet filter scale variation and the number of high-pass filter cutoff frequency variation is L=126.

[0099] As shown in Figure 4 , it is the signal of the upstream pressure signal after denoising by the db9 wavelet with the scale 5 in one embodiment of the application.

[0100] As shown in Figure 5 , it is the signal after filtering by the first-order RC high-pass filter with the discretization frequency of 200Hz in one embodiment of the application.

[0101] Under the conditions that the db9 wavelet filter scale is 5, the high-pass filter is the first-order RC high-pass filter with the time constant Tc=4s and the discretization frequency of 200Hz, the signal amplitude highlight characteristic single connection threshold range [SetDist(1), SetDist(M)] is [0.01-0.5] and the step_D=0.01, and the total iteration number M=50, the multi-scale representation FlagPZ of the interval sub-signal of the upstream signal is as shown in Figure 6 , and the highlight characteristic function FlagSum of the interval sub-signal of the upstream signal is as shown in Figure 7 . Under the conditions that the wavelet filter scale variation range is [5-10] and the first-order RC high-pass filter with the time constant Tc=4s is used, the discretization frequency adjustment range of the high-pass filter is [100Hz-2000Hz], and the highlight characteristic function FlagSum of the interval sub-signal of the upstream and downstream signals is as shown in Figure 8 .

[0102] In some embodiments, the multi-scale representation of the peak value of the interval sub-signal of the signal is obtained by using the hierarchical clustering method, and the highlight characteristic function FlagSum of each interval sub-signal in the signal is obtained, including:

[0103] The single connection distance threshold adjustment range of the hierarchical clustering is set as Set_Dist(l)-SetDist(M), the step is step_D, and Set_Dist(1)<Set_Dist(2)<…<SetDist(M-1)<SetDist(M), a total of M groups.

[0104] Taking absolute value of the peak sequence of the interval sub-signal in the signal, and arranging in descending order from large to small, a positive amplitude interval sub-signal peak sequence Sort_Peak(j) in descending order is obtained, wherein j∈[1,NC], NC is the total number of positive and negative intervals of the current signal, and

[0105] Sort_Peak(1)>Sort_Peak(2)>…>Sort_Peak(NC);

[0106] Recording the mapping relationship of the interval sequence numbers of the interval sub-signals before and after sorting;

[0107] Difference calculation is performed on the positive amplitude interval sub-signal peak sequence Sort_Peak(j) in descending order to obtain a difference sequence d_SortPeak(j) of the positive amplitude interval sub-signal peak sequence in descending order, j∈[1,NC-1], NC is the total number of positive and negative intervals of the current signal;

[0108] According to the current single connection distance threshold Set_Dist(m), m∈[1,M], M is the number of the single connection distance threshold groups, the prominent interval sub-signals in the interval sub-signals of the current signal are determined to obtain the prominence degree representation FlagS(m,j) of the interval sub-signals of the current signal under the current single connection distance threshold condition, wherein m corresponds to the sequence number of the single connection distance threshold, and j corresponds to the interval sequence number before sorting.

[0109] As a further improvement of the above method, in one embodiment, further comprising: adjusting the single connection distance threshold Set_Dist(m), m∈[1,M] step by step according to the step distance step_D to obtain the prominence degree representation FlagS(m,j) of the interval sub-signals of the current signal under different single connection distance threshold conditions, and to obtain the multi-scale representation FlagZ and the multi-scale representation function FlagSum of the prominence degree of the interval sub-signals of the current signal under the current wavelet base and wavelet filter scale, the current high-pass filter cutoff frequency condition, and in the single connection distance threshold adjustment range Set_Dist(l)-SetDist(M):

[0110]

[0111] Wherein, l∈[1,L], L is the product of the wavelet filter scale adjustment times and the high-pass filter cutoff frequency adjustment times.

[0112] As a further improvement of the above method, in one embodiment, judging whether the interval sub-signal corresponding to the position of the most prominent negative polarity interval sub-signal in the upstream and downstream signals is a relevant signal pair comprises:

[0113] Searching for the negative maximum value of the interval sub-signal in FlagZ and storing the negative maximum value of the interval sub-signal in FlagZ into an array maxFlag(l);

[0114] Storing the start and end positions of the interval sub-signal into arrays SSt(l), SEnd(l) respectively, and storing the wavelet filter scale and high-pass filter cutoff frequency into arrays Ws_Tab(l) and FC_Tab(l).

[0115] Searching for the negative maximum value and its corresponding interval in the prominence degree representation function FlagSum of the interval sub-signal corresponding to the upstream and downstream pressure signals, and performing cross-correlation calculation on the interval sub-signal in the corresponding interval in the prominence degree representation function FlagSum;

[0116] If the calculated time delay value is within the preset time delay interval, the interval sub-signal corresponding to the position of the most prominent negative polarity interval sub-signal is taken as the relevant signal.

[0117] Specifically, in one embodiment, the negative maximum values and their intervals in FlagSum are searched, the negative maximum values are -5331 and -2414 respectively, and the interval ranges are [19801, 28799] and [28926, 31810] respectively. Cross-correlation calculation is performed on the interval sub-signal taken from the prominence degree representation function FlagSum, and the time delay is -2485, which is within the maximum time delay [-4500, 4500], indicating that the taken interval sub-signal is relevant. The negative maximum values searched from the maxFlag array corresponding to the upstream and downstream signals are -50 and -32 respectively, and the best wavelet filter scale corresponding to the upstream and downstream signals is 9 and 9 respectively, and the discretization frequency is 500 Hz and 300 Hz respectively. The above parameters are used to denoise the upstream and downstream pressure signals after trend elimination, and the result is shown in Fig. 2. Figure 9

[0118] The product of P and FlagSum is calculated according to the formula AP=P*FlagSum, and the final signal enhancement result is shown in Fig. 3. Figure 10

[0119] ​​In some embodiments, the method further comprises adjusting the cutoff frequency of the high-pass filter according to the adjustment step size of the cutoff frequency of the high-pass filter, recalculating the multi-scale representation FlagZ of the prominence of the interval sub-signals of the signal within the single connection distance threshold range under the current wavelet basis and wavelet filter scale condition, and obtaining the values of the multi-scale representation function FlagSum(l, k) and maxFlag(l), SSt(l), SEnd(l), Ws_Tab(l), and FC_Tab(l) under the current filter parameter condition, where l∈[1, L] and k∈[1, N].

[0120] In some embodiments, the method further comprises adjusting the wavelet filter scale, recalculating the multi-scale representation FlagZ of the prominence of the interval sub-signals of the signal within the single connection distance threshold range under the condition of different cutoff frequencies of the high-pass filter, and obtaining the values of the multi-scale representation function FlagSum(l, k) and maxFlag(l), SSt(l), SEnd(l), Ws_Tab(l), and FC_Tab(l) sequence under the current filter parameter condition, where l∈[1, L] and k∈[1, N].

[0121] Finally, the multi-scale representation function FlagSum of the signal is obtained, which is denoted as FlagSum_1(k), where k∈[1, N].

[0122] As a further improvement of the above method, in one embodiment, the judgment of whether the most prominent interval sub-signals in the upstream and downstream signals are a pair of associated signals comprises:

[0123] The negative extreme value and its corresponding interval in the prominence representation function corresponding to the pair of upstream and downstream pressure signals are searched, and the signals of the prominence representation function within the corresponding interval range are intercepted for cross-correlation calculation;

[0124] If the time delay value obtained by the correlation calculation is within the preset time delay interval, then the negative polarity interval sub-signals in the upstream and downstream pressure signals within the interval range are associated signals;

[0125] If the time delay value obtained by the correlation calculation is outside the preset time delay interval, then the absolute values of the most prominent negative polarity interval sub-signals in the prominence representation functions of the upstream and downstream signals are compared, and the interval sub-signal with the larger absolute value is taken as a reference. Within the preset time delay interval in the other prominence representation function, the negative polarity interval sub-signal with the largest absolute amplitude is searched as an associated signal;

[0126] According to the above method, the values of FlagZ, FlagSum(l, k), maxFlag(l), SSt(l), SEnd(l), Ws_Tab(l), and FC_Tab(l) sequence of the corresponding interval sub-signals are recalculated.

[0127] In some embodiments, according to the maxFlag sequence corresponding to the upstream and downstream pressure signals, the sequence number where the maximum value of the maxFlag is located is searched, and according to the sequence number corresponding to the maximum value, the corresponding wavelet filtering scale and high-pass filter cutoff frequency are indexed from the Ws_Tab and FC_Tab arrays corresponding to the upstream and downstream pressure signals, as the optimal filtering parameters of the upstream and downstream pressure signals.

[0128] In some embodiments, according to the optimal filtering parameters corresponding to the upstream and downstream pressure signals, the upstream and downstream pressure signals SP(ch, k) after trend elimination are wavelet denoised and high-pass filtered, and the upstream and downstream pressure signals are actively signal enhanced using the multi-scale representation function of the pressure signals, including:

[0129] According to the optimal filtering parameters corresponding to the upstream and downstream pressure signals, the upstream and downstream pressure signals SP(ch, k) after trend elimination are wavelet denoised and high-pass filtered, and the filtered upstream and downstream pressure signals P(ch, k) are obtained, where ch is the channel number corresponding to the upstream and downstream of the pipeline, ch = 1-2; k = 1-N, N is the data frame length.

[0130] In some embodiments, the abnormal interval sub-signals in the pressure signals P(ch, k) filtered by the optimal filtering parameters are actively signal enhanced, and the non-abnormal interval sub-signals are actively signal suppressed using the multi-scale representation function of the upstream and downstream pressure signals.

[0131] AP(1, k) = P(1, k) * FlagSum_1(k),

[0132] AP(2, k) = P(2, k) * FlagSum_2(k),

[0133] Where k = 1-N, AP(1, k) and AP(2, k) are the enhanced pressure signals, where k = 1-N.

[0134] It can be understood that the same or similar parts in the above embodiments can be mutually referred to, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0135] It should be noted that in the description of the present application, the terms "first", "second" and the like are only for descriptive purposes, and cannot be understood as indicating or implying relative importance. In addition, in the description of the present application, unless otherwise specified, the meaning of "a plurality of" or "a plurality" is at least two.

[0136] It should be understood that when an element is referred to as being "on" or "connected to" another element, it can be directly on or connected to the other element or intervening elements can be present. In addition, the term "connected" as used herein can include wirelessly connected. Also, the term "on" as used herein can include "directly on" and "indirectly on" when used in the context of interlayers.

[0137] Any process or method described in flow chart form or otherwise described herein can be understood as a representation of executable instructions, code, or a module, segment, or portion of code for execution, including one or more steps for accomplishing a particular logic function or process, and the preferred embodiments of the present application encompass additional implementations that can not be described in detail in the description of the preferred embodiments, including implementations that can not perform the steps in the order shown or discussed, including implementations that perform functions in substantially simultaneous fashion, or in the reverse order of the steps, as will be understood by those of ordinary skill in the art to which the embodiments of the present application pertain.

[0138] It should be understood that portions of the present application can be realized with a hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be realized as software or firmware to be executed by a suitable instruction-executing system, stored in a storage medium. For example, if realized with hardware, as in another embodiment, it can be realized with any one or a combination of the following technologies known in the art: discrete logic circuit having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and the like.

[0139] Those of ordinary skill in the art can understand that all or part of the steps carried out by the above-mentioned embodiments can be completed by programs instructing relevant hardware, and the programs can be stored in a computer-readable storage medium, and when executed, include one or a combination of the steps of the method embodiments.

[0140] In addition, each functional unit in each embodiment of the present application can be integrated in one processing module, or each unit can be physically present separately, or two or more units can be integrated in one module. The above-mentioned integrated module can be realized in the form of hardware or in the form of a software functional module. When the integrated module is realized in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0141] The above-mentioned storage medium can be a read-only memory, a magnetic disk or an optical disk, etc.

[0142] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are contained in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily mean the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0143] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary, and cannot be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above-described embodiments within the scope of the present application.

Claims

1. A method of enhancing a pipeline pressure leak signal, the method comprising: The method comprises: acquiring upstream pipeline pressure signals and downstream pipeline pressure signals; trend detection and trend elimination are performed on the upstream pipe pressure signal and the downstream pipe pressure signal to obtain the trend-eliminated upstream and downstream pipe pressure signals SP (ch, k), wherein ch is the number of channels, corresponding to the upstream and downstream of the pipe, ch = 1-2; k = 1-N, N is the length of the data frame; The upstream and downstream pressure signals of the pipeline are trend-removed SP The prominence degree of the interval sub-signal in (ch, k) is multi-scale characterized to obtain the prominence degree characterization function of each interval sub-signal in the upstream and downstream pressure signals, denoted as FlagSum ; According to the prominence degree function of each interval sub-signal in the upstream and downstream pressure signals FlagSum , the position of the most prominent interval sub-signal in the upstream and downstream signals is determined; determining whether the most prominent interval sub-signal in the upstream and downstream signals is a relevant signal pair; if the most prominent interval sub-signal is a relevant signal pair, determining the optimal filtering parameter; wavelet denoising and high-pass filtering are performed on the pipeline upstream and downstream pressure signals subjected to the trend elimination using the optimal filtering parameters to obtain filtered pipeline upstream and downstream pressure signals SP ( ch , k ) wavelet denoising and high-pass filtering are performed on the pipeline upstream and downstream pressure signals subjected to the trend elimination using the optimal filtering parameters to obtain filtered pipeline upstream and downstream pressure signals P ( ch , k ) wavelet denoising and high-pass filtering are performed on the pipeline upstream and downstream pressure signals subjected to the trend elimination using the optimal filtering parameters to obtain filtered pipeline upstream and downstream pressure signals ch is the number of channels, ch =1-2, corresponding to the pipeline upstream and downstream; k =1-N, N is the data frame length; The filtered upstream and downstream pressure signals of the pipeline P ( ch , k The prominence representation function of each interval sub-signal in the upstream and downstream signals. FlagSum By performing a multiplication operation, the enhanced upstream and downstream pressure anomaly signals of the pipeline are obtained.

2. The method of claim 1, wherein, The determination of whether the most prominent interval sub-signal in the upstream and downstream signals is a relevant signal pair further comprises: if the most prominent interval sub-signal is not a relevant signal pair, determining the negative extreme value in the prominence degree representation function of the upstream and downstream pipeline pressure signals; taking the most prominent negative polarity interval sub-signal in the signal with a larger negative extreme value as a reference, and searching for a relevant signal in the other signal; determining the optimal filtering parameter according to the relevant signal searched in the other signal.

3. The method of claim 2, wherein, The prominence degree of the interval sub-signal amplitude in the pipeline upstream and downstream pressure signals after the trend is eliminated is represented in multiple scales to obtain a prominence degree representation function of each interval sub-signal in the upstream and downstream signals, comprising: selecting a wavelet base, a wavelet filter scale adjustment range, an initial cutoff frequency of a high-pass filter, and a cutoff frequency adjustment step, filtering the pipeline upstream and downstream pressure signals after the trend is eliminated to obtain filtered signals; The filtered upstream and downstream first path signals are divided into positive and negative intervals to obtain the interval number NC of the signals, the starting position and ending position of each interval sub-signal, the peak value and position of each interval sub-signal, and the interval serial number SSt ( j ) and ending position SEnd ( j ) of each interval sub-signal, Peak ( j ) and position PeakPos ( j ) of each interval sub-signal, j is the interval serial number, j ∈[1,NC] Based on the hierarchical clustering method, the interval sub-signal peak of the signal is characterized in multiple scales, and the prominence degree characterization function of each interval sub-signal in the signal is obtained FlagSum .

4. The method of claim 3, wherein, The hierarchical clustering method is used to perform multi-scale characterization on the interval sub-signal peak values of the signal, so as to obtain a prominence degree characterization function of each interval sub-signal in the signal FlagSum comprising: The single-linkage distance threshold adjustment range of the hierarchical clustering is set as Set_Dist (l)- SetDist (M), the step distance is step_D , and Set_Dist (1)< Set_Dist (2)<…< SetDist (M-1)< SetDist (M), a total of M groups; taking absolute value of the peak sequence of the interval sub-signal in the signal and arranging in descending order from large to small to obtain a descendingly arranged positive interval sub-signal peak sequence Sort_Peak j , wherein j ∈[1,NC], NC is the total number of positive and negative intervals of the current signal, and​ Sort_Peak (1) Sort_Peak (2) Sort_Peak (NC); mapping the interval serial numbers of the interval sub-signals before and after sorting; a sequence of the positive amplitude interval sub-signal peak values in descending order Sort_Peak j performing a difference calculation to obtain a difference sequence of the sequence of the positive amplitude interval sub-signal peak values in descending order d_SortPeak j j , wherein NC is a total number of positive and negative intervals of the current signal.​​​ According to the current single connection distance threshold Set_Dist ( m ), m ∈[1,M],M is the number of the single connection distance threshold group, determine the current signal interval sub-signal in the prominent interval sub-signal, get the current signal interval sub-signal prominence degree representation under the current single connection distance threshold condition FlagS ( m , j ), wherein m corresponding to the serial number of single connection distance threshold, j corresponding to the interval serial number before sorting.

5. The method of claim 4, wherein, Further comprising: According to step distance step_D Adjusting single connection distance threshold value Set_Dist ( m ), m ∈[1,M] to obtain the prominence degree representation of the interval sub-signal of the current signal under different single connection distance threshold values FlagS ( m , j ) to obtain the multi-scale representation of the prominence degree of the interval sub-signal of the current signal within the single connection distance threshold value adjustment range Set_Dist (l) SetDist (M) under the current wavelet base and wavelet filter scale, the current high-pass filter cutoff frequency FlagZ and the multi-scale representation function FlagSum : wherein l ∈ [1, L], L is the product of the wavelet filter scale adjustment number and the high-pass filter cutoff frequency adjustment number.

6. The method of claim 5, wherein, Further comprising: search FlagZ The negative maxima of the sub-signal in the interval, and the FlagZ The negative maxima of the sub-signals in the interval are stored in the array max. Flag ( l )middle; The start and end positions of the interval sub-signals are stored in arrays SSt , l , SEnd , l , while the wavelet filter scale and high-pass filter cutoff frequency are stored in arrays Ws_Tab , l , and FC_Tab , l .

7. The method of claim 6, wherein, Further comprising: adjusting the cut-off frequency of the high-pass filter according to a step size of the adjustment of the cut-off frequency of the high-pass filter, re-computing the multi-scale representation of the prominence of the interval sub-signals of the signal in the single-connection distance threshold range under the current wavelet basis and wavelet filter scale conditions FlagZ , obtaining the multi-scale representation function under the current filter parameter conditions FlagSum ( l , k ) and max Flag ( l )、 SSt ( l )、 SEnd ( l )、 Ws_Tab ( l )、 FC_Tab ( l ) values, wherein l ∈[1,L], k ∈[1,N] 8. The method of claim 7, wherein, Further comprising: adjusting said wavelet filtering scale, re-computing the multiscale representation of the prominence of the interval sub-signals of said signal in the range of said single connectivity distance threshold at different high-pass filter cutoff frequencies FlagZ , obtaining said multiscale representation function at current filtering parameters FlagSum ( l , k ) and max Flag ( l ), SSt ( l ), SEnd ( l ), Ws_ Tab ( l ), FC_Tab ( l ) sequence, where l ∈ [1, L], k ∈ [1, N] ; a multiscale representation function of the signal is finally obtained FlagSum , denoted by FlagSum_1 ( k ), k ∈ [1, N].

9. The method of claim 2, wherein, The determination of whether the most prominent interval sub-signal in the upstream and downstream signals is a relevant signal pair comprises: searching for a negative extreme value and its corresponding interval in the prominence degree representation function corresponding to the upstream and downstream pressure signals, and performing cross-correlation calculation on the prominence degree representation function signals in the corresponding interval range; if the time delay value obtained by the correlation calculation is within a preset time delay interval, the negative extreme value interval sub-signal in the upstream and downstream pressure signals in the interval range is a relevant signal; If the time delay value obtained by the correlation calculation is outside the preset time delay interval, the absolute values of the most prominent negative polarity interval sub-signals in the prominence degree representation functions of the upstream and downstream are compared, the interval sub-signal with the larger absolute value is taken as a reference, and the negative polarity interval sub-signal with the largest absolute amplitude in the preset time delay interval in the other prominence degree representation function is searched as a correlation signal; recomputing the values of the FlagZ, FlagSum ( l , k )、max Flag ( l )、 SSt ( l )、 SEnd ( l )、 Ws_ Tab ( l )、 FC_Tab ( l ) sequences of the corresponding interval sub-signals.

10. The method of claim 2, wherein, The correlation signal searched in the other signal is used to determine the optimal filtering parameter, including: searching for negative maximums in the array max Flag l negative values in the array max​ According to the serial number corresponding to the negative value, the best filtering parameter of the corresponding signal is searched from the array corresponding to the signal Ws_Tab ( l )、 FC_Tab ( l )

Images