A method and device for monitoring high-frequency switching power supply for substation relay protection

Abstract

The present application discloses a method and device for monitoring a high-frequency switching power supply for relay protection in a substation. The method includes obtaining a first audio signal of the high-frequency switching power supply based on an ultrasonic microphone; collecting a first sequence of data corresponding to the first audio signal, and calculating a second audio signal based on the first sequence of data; extracting an upper envelope and a lower envelope from the second audio signal, respectively, and calculating an eigenvalue signal; inputting the eigenvalue signal into a trained recurrent neural network to obtain a prediction feature; performing a two-norm calculation on the prediction feature and the sample feature, and determining early warning information of the high-frequency switching power supply based on the calculation result. By collecting the acoustic wave signal of the high-frequency switching power supply and performing acoustic wave monitoring to monitor the high-frequency switching power supply in real time, not only can the monitoring of the switching power supply for relay protection in the substation be realized throughout its entire life cycle, but also the early warning of relay protection failure can be greatly advanced, thereby leaving a sufficient maintenance window for maintenance.

Description

Technical Field

[0001] The present application belongs to the field of signal monitoring technology, and in particular relates to a method and device for monitoring a high-frequency switching power supply for relay protection in a substation. Background Art

[0002] Relay protection in substations generally refers to when the insulation of transmission lines or power distribution equipment is damaged, the load is short-circuited, resulting in a decrease in impedance and a corresponding increase in current. Relay protection disconnects (cuts off) the short-circuited transmission lines or power distribution equipment. If not handled in a timely manner, it can easily cause equipment damage, fire, threaten personal safety, and even disrupt the stable operation of the power supply system.

[0003] The high-frequency switching power supply is a key component that drives the relay protection device to ensure that it can drive the switch mechanism to cut off when a fault occurs. However, if the high-frequency switching power supply has a slight abnormality in normal times and the circuit cannot detect it, it may not be able to drive the switch mechanism when a fault occurs, which will cause significant losses. Therefore, it is very important to monitor the high-frequency switching power supply.

[0004] Currently, ultra-high frequency (UHF) technology is primarily used to monitor high-frequency switching power supplies. This involves using analyzers to collect, analyze, and monitor the UHF electromagnetic field generated by partial discharges when a high-frequency switching power supply experiences a severe anomaly. However, in practice, partial discharges in switching power supplies often indicate a serious fault, leaving a very short maintenance window. This hinders the coordinated maintenance planning of substations and creates the risk of untimely repairs. Summary of the Invention

[0005] This application aims to solve the technical problems mentioned above, such as when partial discharge occurs in the switching power supply, it indicates a serious fault and its maintenance window is very short, which is not conducive to the overall planning of the substation maintenance plan, and there is a risk of not repairing the fault in time. A high-frequency switching power supply monitoring method and device for substation relay protection are proposed. The technical solution is as follows:

[0006] In a first aspect, an embodiment of the present application provides a method for monitoring a high-frequency switching power supply for relay protection in a substation, comprising:

[0007] Acquire a first audio signal of the high-frequency switching power supply based on an ultrasonic microphone;

[0008] collecting a first sequence of data corresponding to the first audio signal using at least two microphones, and calculating a second audio signal based on the first sequence of data; wherein the data volume of the second audio signal is less than the data volume of the first audio signal;

[0009] extracting an upper envelope and a lower envelope from the second audio signal, respectively, and calculating an eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope;

[0010] When it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint conditions, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction features;

[0011] The second norm of the prediction features and sample features is calculated, and the early warning information of the high-frequency switching power supply is determined based on the calculation results.

[0012] In an optional solution of the first aspect, calculating the second audio signal according to the first sequence data includes:

[0013] Calculating the average value of each subsequence data in the first sequence data to obtain the second sequence data;

[0014] Calculate the covariance matrix based on the first sequence data and the second sequence data;

[0015] Determining at least two eigenvalues ​​of the covariance matrix and constructing a singular value matrix based on the at least two eigenvalues;

[0016] A second audio signal is calculated based on the covariance matrix, the singular value matrix, and the preset weight ratios.

[0017] In another optional solution of the first aspect, extracting the upper envelope and the lower envelope from the second audio signal respectively includes:

[0018] Constructing a bandpass signal based on the second audio signal and a preset signal expression, and filtering the bandpass signal expression to obtain a filtered signal;

[0019] Construct a composite signal based on the bandpass signal and the filtered signal;

[0020] Substituting the composite signal into a preset upper envelope expression to obtain the upper envelope; and

[0021] Substitute the composite signal into the preset lower envelope expression to obtain the lower envelope.

[0022] In another optional solution of the first aspect, calculating the eigenvalue signal according to the second audio signal, the upper envelope, and the lower envelope includes:

[0023] Calculate the mean of the upper envelope and the lower envelope to obtain the average envelope;

[0024] A difference calculation is performed on the second audio signal and the average envelope to obtain a characteristic value signal.

[0025] In yet another optional solution of the first aspect, after extracting the upper envelope and the lower envelope from the second audio signal, and calculating the eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope, when it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition, inputting the eigenvalue signal into the trained recurrent neural network, and before obtaining the prediction feature, further comprising:

[0026] Determine whether the difference between the number of extreme value points and the number of zero-crossing points in the characteristic curve corresponding to the characteristic value signal is within a preset number range;

[0027] When it is detected that the difference value is within a preset number interval, determining whether the average value between the envelope formed by all the maximum points and the envelope formed by all the minimum points at any time in the characteristic curve corresponding to the characteristic value signal is zero;

[0028] When it is detected that the average value is zero, it is determined that the characteristic curve corresponding to the characteristic value signal satisfies the constraint condition.

[0029] In another optional solution of the first aspect, before inputting the characteristic value signal into the trained recurrent neural network to obtain the prediction feature, the method further includes:

[0030] performing a difference calculation on the second audio signal and the eigenvalue signal, and determining whether the difference calculation result is a monotonic function;

[0031] Input the eigenvalue signal into the trained recurrent neural network to obtain the prediction features, including:

[0032] When it is determined that the difference calculation result is a monotonic function, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction feature.

[0033] In another optional solution of the first aspect, determining the early warning information of the high-frequency switching power supply according to the calculation result includes:

[0034] When it is detected that the calculation result is within a preset threshold range, early warning information of the high-frequency switching power supply is generated according to damage information corresponding to the preset threshold range.

[0035] In a second aspect, an embodiment of the present application provides a high-frequency switching power supply monitoring device for substation relay protection, comprising:

[0036] An audio acquisition module, configured to acquire a first audio signal of the high-frequency switching power supply based on an ultrasonic microphone;

[0037] a first processing module configured to collect a first sequence of data corresponding to the first audio signal using at least two microphones, and calculate a second audio signal based on the first sequence of data; wherein the data volume of the second audio signal is smaller than the data volume of the first audio signal;

[0038] a second processing module, configured to extract an upper envelope and a lower envelope from the second audio signal, and calculate an eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope;

[0039] A model prediction module is used to input the eigenvalue signal into a trained recurrent neural network to obtain a prediction feature when it is determined that the characteristic curve corresponding to the eigenvalue signal meets the constraint conditions;

[0040] The early warning module is used to calculate the binary norm of the prediction features and sample features, and determine the early warning information of the high-frequency switching power supply based on the calculation results.

[0041] In an optional solution of the second aspect, the first processing module includes:

[0042] Calculating the average value of each subsequence data in the first sequence data to obtain the second sequence data;

[0043] Calculate the covariance matrix based on the first sequence data and the second sequence data;

[0044] Determining at least two eigenvalues ​​of the covariance matrix and constructing a singular value matrix based on the at least two eigenvalues;

[0045] A second audio signal is calculated based on the covariance matrix, the singular value matrix, and the preset weight ratios.

[0046] In another optional solution of the second aspect, the second processing module includes:

[0047] Constructing a bandpass signal based on the second audio signal and a preset signal expression, and filtering the bandpass signal expression to obtain a filtered signal;

[0048] Construct a composite signal based on the bandpass signal and the filtered signal;

[0049] Substituting the composite signal into a preset upper envelope expression to obtain the upper envelope; and

[0050] Substitute the composite signal into the preset lower envelope expression to obtain the lower envelope.

[0051] In another optional solution of the second aspect, the second processing module further includes:

[0052] Calculate the mean of the upper envelope and the lower envelope to obtain the average envelope;

[0053] A difference calculation is performed on the second audio signal and the average envelope to obtain a characteristic value signal.

[0054] In yet another optional solution of the second aspect, the apparatus further comprises:

[0055] After extracting the upper envelope and the lower envelope from the second audio signal, and calculating the eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope, when it is determined that the characteristic curve corresponding to the eigenvalue signal meets the constraint conditions, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction feature.

[0056] Determine whether the difference between the number of extreme value points and the number of zero-crossing points in the characteristic curve corresponding to the characteristic value signal is within a preset number range;

[0057] When it is detected that the difference value is within a preset number interval, determining whether the average value between the envelope formed by all the maximum points and the envelope formed by all the minimum points at any time in the characteristic curve corresponding to the characteristic value signal is zero;

[0058] When it is detected that the average value is zero, it is determined that the characteristic curve corresponding to the characteristic value signal satisfies the constraint condition.

[0059] In yet another optional solution of the second aspect, the apparatus further comprises:

[0060] Before inputting the eigenvalue signal into the trained recurrent neural network to obtain the prediction feature,

[0061] performing a difference calculation on the second audio signal and the eigenvalue signal, and determining whether the difference calculation result is a monotonic function;

[0062] Input the eigenvalue signal into the trained recurrent neural network to obtain the prediction features, including:

[0063] When it is determined that the difference calculation result is a monotonic function, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction feature.

[0064] In another optional solution of the second aspect, the early warning module includes:

[0065] When it is detected that the calculation result is within a preset threshold range, early warning information of the high-frequency switching power supply is generated according to damage information corresponding to the preset threshold range.

[0066] In a third aspect, an embodiment of the present application further provides a high-frequency switching power supply monitoring device for substation relay protection, comprising a processor and a memory;

[0067] The processor is connected to the memory;

[0068] a memory for storing executable program code;

[0069] The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the high-frequency switching power supply monitoring method for substation relay protection provided by the first aspect of the embodiment of the present application or any implementation method of the first aspect.

[0070] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the high-frequency switching power supply monitoring method for substation relay protection provided by the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect can be implemented.

[0071] In an embodiment of the present application, during the monitoring of a high-frequency switching power supply, a first audio signal of the high-frequency switching power supply can be obtained based on an ultrasonic microphone; a first sequence of data corresponding to the first audio signal can be collected based on at least two microphones, and a second audio signal can be calculated based on the first sequence of data; an upper envelope and a lower envelope can be extracted from the second audio signal, and an eigenvalue signal can be calculated based on the second audio signal, the upper envelope, and the lower envelope; when it is determined that the characteristic curve corresponding to the eigenvalue signal meets the constraint conditions, the eigenvalue signal is input into a trained recurrent neural network to obtain a predicted feature; a two-norm calculation is performed on the predicted feature and the sample feature, and early warning information of the high-frequency switching power supply is determined based on the calculation result. By collecting the acoustic wave signal of the high-frequency switching power supply and performing acoustic wave monitoring to monitor the high-frequency switching power supply in real time, not only can the monitoring of the switching power supply of the substation relay protection be realized throughout its entire life cycle, but also the early warning of the relay protection failure can be greatly advanced, thereby leaving a sufficient maintenance window for maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0073] Figure 1 An overall flow chart of a high-frequency switching power supply monitoring method for substation relay protection provided in an embodiment of the present application;

[0074] Figure 2 A schematic structural diagram of a high-frequency switching power supply monitoring device for substation relay protection provided in an embodiment of the present application;

[0075] Figure 3 A schematic structural diagram of another high-frequency switching power supply monitoring device for substation relay protection provided in an embodiment of the present application. DETAILED DESCRIPTION

[0076] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0077] In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application. Different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments described. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, even though the embodiment may not be clearly described in the following text.

[0078] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the function and arrangement of the elements described without departing from the scope of the present application. Various examples may appropriately omit, replace, or add various processes or components. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted, or combined. In addition, features described in some examples may be combined in other examples.

[0079] See also Figure 1 , Figure 1 The figure shows an overall flow chart of a high-frequency switching power supply monitoring method for substation relay protection provided by an embodiment of the present application.

[0080] like Figure 1 As shown, the high-frequency switching power supply monitoring method for substation relay protection may include at least the following steps:

[0081] Step 102: Acquire a first audio signal of a high-frequency switching power supply based on an ultrasonic microphone.

[0082] In an embodiment of the present application, a method for monitoring a high-frequency switching power supply for relay protection in a substation can be applied to a control terminal. The control terminal can collect audio signals emitted by the high-frequency switching power supply based on an ultrasonic microphone disposed near the high-frequency switching power supply. An ultrasonic acquisition card then acquires the audio signals collected by the ultrasonic microphone and performs signal processing on the audio signals. It is understood that when the result of the signal processing indicates that the high-frequency switching power supply is at risk, the control terminal can also immediately send an early warning message to the operation and maintenance room for inspection by staff. The distance between the ultrasonic microphone and the high-frequency switching power supply can be a preset distance to ensure the accuracy of the audio signals collected by the ultrasonic microphone.

[0083] Specifically, during the process of monitoring the high-frequency switching power supply, the control terminal may control the ultrasonic collector to obtain the first audio signal emitted by the high-frequency switching power supply when determining that the high-frequency switching power supply is in a working state, and feed the first audio signal back to the control terminal through the ultrasonic acquisition card.

[0084] Step 104: Collect a first sequence of data corresponding to the first audio signal using at least two microphones, and calculate a second audio signal based on the first sequence of data.

[0085] Specifically, after acquiring the first audio signal, the control terminal may, but is not limited to, control at least two microphones to collect the first audio signal. Each microphone may collect sequence data corresponding to the first audio signal, and the set of sequence data collected by each microphone may be used as the first sequence data, wherein the first subsequence data in the first sequence data may be, but is not limited to, the sequence data collected by the first microphone, the second subsequence data in the first sequence data may be, but is not limited to, the sequence data collected by the second microphone, and the nth subsequence data in the first sequence data may be, but is not limited to, the sequence data collected by the nth microphone. It is understandable that the acquisition time of each microphone remains consistent so that the sequence data collected by each microphone remains consistent in length, thereby improving the consistency and validity of the data. Of course, for a segment of audio signal, the control terminal may also divide the audio signal into multiple segments according to the acquisition time. Each microphone may respectively collect sequence data corresponding to each segment of audio signal, thereby obtaining multiple first sequence data.

[0086] Furthermore, after obtaining the first sequence data, the control terminal may calculate an average value for each subsequence data in the first sequence data. The average value may be calculated by dividing each subsequence data by the sum of all subsequence data, and the set of results of the average value calculation for each subsequence data may be used as the second sequence data.

[0087] Here, the first sequence data can be expressed as For example, the method for calculating the average value of the nth subsequence data in the first sequence data may be, but is not limited to, as follows:

[0088]

[0089] Combining the above formula, the second sequence data can be expressed as .

[0090] Furthermore, after obtaining the second sequence data through average calculation, the control terminal may calculate a covariance matrix based on the first sequence data and the second sequence data. In the process of calculating the covariance matrix, the first sequence data and the second sequence data may be combined to obtain normalized sequence data, but is not limited to the above. The normalized sequence data may be expressed as follows:

[0091]

[0092] In the above formula, Can correspond to normalized sequence data, Can correspond to the first sequence data, It can correspond to the second sequence data.

[0093] Next, the control terminal may calculate a covariance matrix based on the normalized sequence data. The covariance matrix may be, but is not limited to, expressed as follows:

[0094]

[0095] In the above formula, It can correspond to the covariance matrix, and n can correspond to the number of subsequence data in the first sequence data. Can correspond to normalized sequence data, Can be understood as The transposed matrix of .

[0096] Furthermore, after calculating the covariance matrix, the control terminal may determine a plurality of eigenvalues ​​corresponding to the covariance matrix by combining the covariance matrix and the eigenvalue expression, wherein the covariance matrix and the eigenvalue expression may be, but is not limited to, expressed as follows:

[0097]

[0098] In the above formula, It can be corresponded to the covariance matrix, and I can be understood as the unit matrix. It can be understood as an eigenvalue, and the eigenvalue can have multiple solutions (that is, the eigenvalue can be expressed as ).

[0099] It is understandable that after obtaining multiple eigenvalues ​​corresponding to the covariance matrix, a corresponding eigenvector matrix can be obtained according to each eigenvalue, wherein the expression corresponding to the eigenvector matrix and the eigenvalue can be, but is not limited to, expressed as follows:

[0100]

[0101] Here, V can be understood as the eigenvalue The corresponding eigenvector matrix, and for different eigenvector matrices , which can be understood as the corresponding eigenvalue The basic system of solutions to the homogeneous equations in the above formula.

[0102] It should be noted that in order to achieve dimensionality reduction processing of the audio signal, a new covariance matrix of different matrix dimensions can also be constructed based on the covariance matrix mentioned above. The matrix dimension of the new covariance matrix can be, but is not limited to, expressed as n*n. The matrix dimension of the covariance matrix mentioned above can be, but is not limited to, expressed as m*m, where m is greater than n. The new covariance matrix can be, but is not limited to, expressed as the following formula:

[0103]

[0104] Then, referring to the above-mentioned expressions of the covariance matrix and the eigenvalues, and the expressions corresponding to the eigenvector matrix and the eigenvalues, the multiple eigenvalues ​​of the new covariance matrix and the corresponding eigenvector matrix can be obtained respectively. It should be noted that the multiple eigenvalues ​​of the covariance matrix mentioned above are consistent with the multiple eigenvalues ​​of the new covariance matrix constructed here, and the eigenvector matrix corresponding to the multiple eigenvalues ​​mentioned above is different in dimension from the eigenvector matrix corresponding to the multiple eigenvalues ​​here.

[0105] Furthermore, after determining the multiple eigenvalues ​​of the covariance matrix mentioned above, the control terminal may also substitute the multiple eigenvalues ​​into a preset singular value matrix expression to obtain a singular value matrix, wherein the preset singular value matrix expression may be, but is not limited to, expressed as follows:

[0106]

[0107] In the above formula, It can be corresponded to a singular value matrix,

[0108] Furthermore, after obtaining the singular value matrix, the control terminal may construct a new covariance matrix based on the singular value matrix, the covariance matrix mentioned above, and the preset weight ratio to obtain the second audio signal. This may be, but is not limited to, first substituting the singular value matrix, the eigenvector matrix corresponding to the covariance matrix mentioned above, and the eigenvector matrix corresponding to the constructed new covariance matrix into a preset signal component distribution expression, which is as follows:

[0109]

[0110] In the above formula, A can be corresponded to the signal component distribution, It can be corresponded to a singular value matrix, It can be corresponded to constructing the i-th eigenvector matrix corresponding to the new covariance matrix, It can correspond to the i-th eigenvector matrix corresponding to the covariance matrix mentioned above.

[0111] Then, after calculating the signal component distribution, the control terminal can calculate a second audio signal based on the signal component distribution and the preset weight ratio, wherein the data volume of the second audio signal is smaller than the data volume of the first audio signal, thereby reflecting the dimensionality reduction processing effect of the audio signal.

[0112] It is understandable that the preset weight ratio mentioned above may correspond to 90%, and is not limited thereto.

[0113] Step 106 : extract the upper envelope and the lower envelope from the second audio signal respectively, and calculate an eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope.

[0114] Specifically, after obtaining the second audio signal, the control terminal may substitute the second audio signal into a preset signal expression to construct a bandpass signal, which may be, but is not limited to, expressed as follows:

[0115]

[0116] In the above formula, Can be used as a bandpass signal. corresponds to the second audio signal, It can be corresponded to a bandpass frequency constant, and t can be corresponded to a time variable.

[0117] Next, the control terminal may filter the bandpass signal to obtain a filtered signal, which may be, but is not limited to, represented as follows:

[0118]

[0119] Furthermore, after obtaining the bandpass signal and the filtered signal respectively, the control terminal may substitute the bandpass signal and the filtered signal into a composite signal expression to obtain a composite signal, which may be, but is not limited to, expressed as follows:

[0120]

[0121] In the above formula, Can correspond to composite signal, Can be used as a bandpass signal. It can be corresponded to a filtered signal, and J can be understood as an imaginary unit.

[0122] Furthermore, the control terminal may substitute the composite signal into a preset upper envelope expression to obtain an upper envelope, wherein the upper envelope may be, but is not limited to, expressed as follows:

[0123]

[0124] The control terminal may also substitute the composite signal into a preset lower envelope expression to obtain a lower envelope, wherein the lower envelope may be, but is not limited to, expressed as follows:

[0125]

[0126] Furthermore, after obtaining the upper envelope and the lower envelope, the control terminal may perform mean calculation on the upper envelope and the lower envelope to obtain an average envelope, the calculation expression of which may be, but is not limited to, the following:

[0127]

[0128] After obtaining the average of the upper envelope and the lower envelope, the control terminal may perform a difference calculation on the average of the upper envelope and the lower envelope according to the second audio signal mentioned above, thereby obtaining a characteristic value signal, the calculation expression of which may be, but is not limited to, the following:

[0129]

[0130] As an optional embodiment of the present application, after extracting the upper envelope and the lower envelope from the second audio signal, and calculating the eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope, when it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition, the eigenvalue signal is input into the trained recurrent neural network, and before obtaining the prediction feature, the method further includes:

[0131] Determine whether the difference between the number of extreme value points and the number of zero-crossing points in the characteristic curve corresponding to the characteristic value signal is within a preset number range;

[0132] When it is detected that the difference value is within a preset number interval, determining whether the average value between the envelope formed by all the maximum points and the envelope formed by all the minimum points at any time in the characteristic curve corresponding to the characteristic value signal is zero;

[0133] When it is detected that the average value is zero, it is determined that the characteristic curve corresponding to the characteristic value signal satisfies the constraint condition.

[0134] Among them, the preset number interval here can be 0 to 1, that is, when the difference between the number of extreme points and the number of zero-crossing points is detected to be 0 or 1, it can be then determined whether the average value between the envelope formed by all maximum points at any time and the envelope formed by all minimum points in the characteristic curve corresponding to the eigenvalue signal is zero, and when the average value continues to be detected to be zero, it can be determined that the characteristic curve corresponding to the eigenvalue signal meets the constraint conditions.

[0135] It can be understood that when it is determined that the characteristic curve corresponding to the eigenvalue signal does not meet the constraint conditions, the upper envelope and the lower envelope can be extracted from the eigenvalue signal respectively, and a new eigenvalue signal can be calculated based on the eigenvalue signal, the upper envelope and the lower envelope, and then the new eigenvalue signal can be judged whether it meets the constraint conditions. The process of calculating the new eigenvalue signal and the process of judging whether the constraint conditions are met can be referred to the above embodiment and will not be repeated here.

[0136] It should be noted that after determining that the characteristic curve corresponding to the eigenvalue signal meets the constraint conditions, the eigenvalue signal can be stored, and then the upper envelope and the lower envelope can be extracted from the above-mentioned average envelope respectively. According to the average envelope, the upper envelope and the lower envelope, the new eigenvalue signal can be calculated, and then it can be judged whether the new eigenvalue signal meets the constraint conditions. If it does, the new eigenvalue signal can be stored, and the process is repeated until the difference between the new eigenvalue signal and the corresponding signal used to extract the upper envelope and the lower envelope is a monotonic function, and all the stored eigenvalue signals can be used as the final eigenvalue signal.

[0137] Step 108: When it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction feature.

[0138] Specifically, when it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition, the control terminal can input the eigenvalue signal into the trained recurrent neural network to predict the next characteristic signal. For example, it may specifically include n eigenvalues ​​such as average deviation, skewness, kurtosis, and spectral entropy ratio. After the eigenvalue signal is input into the trained recurrent neural network, it can be but not limited to predicting the data of n eigenvalues ​​such as average deviation, skewness, kurtosis, and spectral entropy ratio in the next 1 day.

[0139] It is understandable that the recurrent neural network mentioned in the embodiments of the present application can be but is not limited to a common RNN neural network, which will not be described in detail here.

[0140] Step 110: perform a binary norm calculation on the prediction features and the sample features, and determine the early warning information of the high-frequency switching power supply according to the calculation results.

[0141] Specifically, after obtaining the predicted features, the control terminal can perform a two-norm calculation based on the predicted features and known sample features corresponding to normal switching power supplies, and determine whether the calculation result is within a preset threshold range. It is understood that when the calculation result is detected to be within the preset threshold range, corresponding damage information can be obtained based on the threshold range, and early warning information about the high-frequency switching power supply can be generated based on this damage information to prompt personnel to immediately perform maintenance based on the damage information. The preset threshold range can include at least two threshold ranges, and different threshold ranges can correspond to damage information about the switching power supply, facilitating rapid maintenance.

[0142] See also Figure 2 , Figure 2 A schematic structural diagram of a high-frequency switching power supply monitoring device for substation relay protection provided by an embodiment of the present application is shown.

[0143] like Figure 2 As shown, the high-frequency switching power supply monitoring device for substation relay protection may include at least an audio acquisition module 201, a first processing module 202, a second processing module 203, a model prediction module 204, and an early warning module 205, wherein:

[0144] An audio acquisition module 201 is configured to acquire a first audio signal of a high-frequency switching power supply based on an ultrasonic microphone;

[0145] A first processing module 202 is configured to collect a first sequence of data corresponding to the first audio signal using at least two microphones, and calculate a second audio signal based on the first sequence of data; wherein the data volume of the second audio signal is less than the data volume of the first audio signal;

[0146] The second processing module 203 is configured to extract an upper envelope and a lower envelope from the second audio signal, and calculate an eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope;

[0147] The model prediction module 204 is configured to input the eigenvalue signal into a trained recurrent neural network to obtain a prediction feature when it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition;

[0148] The early warning module 205 is used to perform a bi-norm calculation on the prediction features and the sample features, and determine early warning information of the high-frequency switching power supply according to the calculation results.

[0149] In some possible embodiments, the first processing module includes:

[0150] Calculating the average value of each subsequence data in the first sequence data to obtain the second sequence data;

[0151] Calculate the covariance matrix based on the first sequence data and the second sequence data;

[0152] Determining at least two eigenvalues ​​of the covariance matrix and constructing a singular value matrix based on the at least two eigenvalues;

[0153] A second audio signal is calculated based on the covariance matrix, the singular value matrix, and the preset weight ratios.

[0154] In some possible embodiments, the second processing module includes:

[0155] Constructing a bandpass signal based on the second audio signal and a preset signal expression, and filtering the bandpass signal expression to obtain a filtered signal;

[0156] Construct a composite signal based on the bandpass signal and the filtered signal;

[0157] Substituting the composite signal into a preset upper envelope expression to obtain the upper envelope; and

[0158] Substitute the composite signal into the preset lower envelope expression to obtain the lower envelope.

[0159] In some possible embodiments, the second processing module further includes:

[0160] Calculate the mean of the upper envelope and the lower envelope to obtain the average envelope;

[0161] A difference calculation is performed on the second audio signal and the average envelope to obtain a characteristic value signal.

[0162] In some possible embodiments, the device further includes:

[0163] After extracting the upper envelope and the lower envelope from the second audio signal, and calculating the eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope, when it is determined that the characteristic curve corresponding to the eigenvalue signal meets the constraint conditions, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction feature.

[0164] Determine whether the difference between the number of extreme value points and the number of zero-crossing points in the characteristic curve corresponding to the characteristic value signal is within a preset number range;

[0165] When it is detected that the difference value is within a preset number interval, determining whether the average value between the envelope formed by all the maximum points and the envelope formed by all the minimum points at any time in the characteristic curve corresponding to the characteristic value signal is zero;

[0166] When it is detected that the average value is zero, it is determined that the characteristic curve corresponding to the characteristic value signal satisfies the constraint condition.

[0167] In some possible embodiments, the device further includes:

[0168] Before inputting the eigenvalue signal into the trained recurrent neural network to obtain the prediction feature,

[0169] performing a difference calculation on the second audio signal and the eigenvalue signal, and determining whether the difference calculation result is a monotonic function;

[0170] Input the eigenvalue signal into the trained recurrent neural network to obtain the prediction features, including:

[0171] When it is determined that the difference calculation result is a monotonic function, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction feature.

[0172] In some possible embodiments, the early warning module includes:

[0173] When it is detected that the calculation result is within a preset threshold range, early warning information of the high-frequency switching power supply is generated according to damage information corresponding to the preset threshold range.

[0174] Those skilled in the art will clearly understand that the technical solutions of the embodiments of the present application can be implemented with the help of software and / or hardware. "Unit" and "module" in this specification refer to software and / or hardware that can independently perform or cooperate with other components to perform specific functions, where the hardware can be, for example, a field-programmable gate array (FPGA) or an integrated circuit (IC).

[0175] See also Figure 3 , Figure 3 A structural schematic diagram of another high-frequency switching power supply monitoring device for substation relay protection provided by an embodiment of the present application is shown.

[0176] like Figure 3 As shown, the high-frequency switching power supply monitoring device 300 for substation relay protection may include at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 and at least one communication bus 302 .

[0177] The communication bus 302 may be used to implement connection and communication among the above components.

[0178] The user interface 303 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.

[0179] The network interface 304 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, and the like.

[0180] The processor 301 may include one or more processing cores. Using various interfaces and circuits, the processor 301 connects to various components within the high-frequency switching power supply monitoring device for substation relay protection 300. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and accessing data stored in the memory 305, the processor 301 executes various functions and processes data within the high-frequency switching power supply monitoring device for routing substation relay protection 300. Optionally, the processor 301 may be implemented in hardware using at least one of a DSP, FPGA, and PLA. The processor 301 may integrate one or a combination of a CPU, a GPU, and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing content displayed on the display; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 301 but implemented as a separate chip.

[0181] Among them, the memory 305 may include RAM and ROM. Optionally, the memory 305 includes a non-transitory computer-readable medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be optionally at least one storage device located away from the aforementioned processor 301. As Figure 3 As shown, the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a high-frequency switching power supply monitoring application for substation relay protection.

[0182] Specifically, the processor 301 may be configured to call a high-frequency switching power supply monitoring application for substation relay protection stored in the memory 305 and specifically perform the following operations:

[0183] Acquire a first audio signal of the high-frequency switching power supply based on an ultrasonic microphone;

[0184] collecting a first sequence of data corresponding to the first audio signal using at least two microphones, and calculating a second audio signal based on the first sequence of data; wherein the data volume of the second audio signal is less than the data volume of the first audio signal;

[0185] extracting an upper envelope and a lower envelope from the second audio signal, respectively, and calculating an eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope;

[0186] When it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint conditions, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction features;

[0187] The second norm of the prediction features and sample features is calculated, and the early warning information of the high-frequency switching power supply is determined based on the calculation results.

[0188] In some possible embodiments, calculating the second audio signal according to the first sequence data includes:

[0189] Calculating the average value of each subsequence data in the first sequence data to obtain the second sequence data;

[0190] Calculate the covariance matrix based on the first sequence data and the second sequence data;

[0191] Determining at least two eigenvalues ​​of the covariance matrix and constructing a singular value matrix based on the at least two eigenvalues;

[0192] A second audio signal is calculated based on the covariance matrix, the singular value matrix, and the preset weight ratios.

[0193] In some possible embodiments, extracting the upper envelope and the lower envelope from the second audio signal respectively includes:

[0194] Constructing a bandpass signal based on the second audio signal and a preset signal expression, and filtering the bandpass signal expression to obtain a filtered signal;

[0195] Construct a composite signal based on the bandpass signal and the filtered signal;

[0196] Substituting the composite signal into a preset upper envelope expression to obtain the upper envelope; and

[0197] Substitute the composite signal into the preset lower envelope expression to obtain the lower envelope.

[0198] In some possible embodiments, calculating the eigenvalue signal according to the second audio signal, the upper envelope, and the lower envelope includes:

[0199] Calculate the mean of the upper envelope and the lower envelope to obtain the average envelope;

[0200] A difference calculation is performed on the second audio signal and the average envelope to obtain a characteristic value signal.

[0201] In some possible embodiments, after extracting the upper envelope and the lower envelope from the second audio signal, and calculating the eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope, when it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition, the eigenvalue signal is input into the trained recurrent neural network, and before obtaining the prediction feature, the method further includes:

[0202] Determine whether the difference between the number of extreme value points and the number of zero-crossing points in the characteristic curve corresponding to the characteristic value signal is within a preset number range;

[0203] When it is detected that the difference value is within a preset number interval, determining whether the average value between the envelope formed by all the maximum points and the envelope formed by all the minimum points at any time in the characteristic curve corresponding to the characteristic value signal is zero;

[0204] When it is detected that the average value is zero, it is determined that the characteristic curve corresponding to the characteristic value signal satisfies the constraint condition.

[0205] In some possible embodiments, before inputting the feature value signal into the trained recurrent neural network to obtain the prediction feature, the method further includes:

[0206] performing a difference calculation on the second audio signal and the eigenvalue signal, and determining whether the difference calculation result is a monotonic function;

[0207] Input the eigenvalue signal into the trained recurrent neural network to obtain the prediction features, including:

[0208] When it is determined that the difference calculation result is a monotonic function, the eigenvalue signal is input into the trained recurrent neural network to obtain the prediction feature.

[0209] In some possible embodiments, determining the warning information of the high-frequency switching power supply according to the calculation result includes:

[0210] When it is detected that the calculation result is within a preset threshold range, early warning information of the high-frequency switching power supply is generated according to damage information corresponding to the preset threshold range.

[0211] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a microdrive, a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0212] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0213] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0214] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interface, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0215] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0216] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0217] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk, or optical disk, etc., various media that can store program code.

[0218] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be performed by instructing related hardware through a program, and the program may be stored in a computer-readable memory, which may include a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0219] The above are merely exemplary embodiments of the present disclosure and are not intended to limit the scope of the present disclosure. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure herein, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not recorded in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for monitoring a high-frequency switching power supply for relay protection in a substation, characterized in that: include: Acquire a first audio signal of the high-frequency switching power supply based on an ultrasonic microphone; collecting first sequence data corresponding to the first audio signal using at least two microphones, and calculating a second audio signal based on the first sequence data; wherein the data volume of the second audio signal is less than the data volume of the first audio signal, and the first sequence data includes the sequence data collected by each of the microphones; extracting an upper envelope and a lower envelope from the second audio signal, respectively, and calculating an eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope; When it is determined that the characteristic curve corresponding to the characteristic value signal satisfies the constraint condition, the characteristic value signal is input into the trained recurrent neural network to obtain a prediction feature; Performing a two-norm calculation on the prediction feature and the sample feature, and determining the early warning information of the high-frequency switching power supply according to the calculation result; The step of calculating the second audio signal based on the first sequence data includes: Calculating an average value of each subsequence data in the first sequence data to obtain second sequence data; Calculating a covariance matrix based on the first sequence data and the second sequence data; Determining at least two eigenvalues ​​of the covariance matrix, and constructing a singular value matrix based on the at least two eigenvalues; A second audio signal is calculated according to the covariance matrix, the singular value matrix, and the preset weight ratio.

2. The method according to claim 1, characterized in that The extracting the upper envelope and the lower envelope from the second audio signal respectively includes: constructing a bandpass signal based on the second audio signal and a preset signal expression, and filtering the bandpass signal expression to obtain a filtered signal; constructing a composite signal based on the bandpass signal and the filtered signal; Substituting the composite signal into a preset upper envelope expression to obtain an upper envelope; and Substituting the composite signal into a preset lower envelope expression, a lower envelope is obtained.

3. The method according to claim 2, characterized in that The calculating of the eigenvalue signal according to the second audio signal, the upper envelope, and the lower envelope includes: Calculating the mean of the upper envelope and the lower envelope to obtain an average envelope; A difference calculation is performed on the second audio signal and the average envelope to obtain a eigenvalue signal.

4. The method according to claim 1, wherein After extracting the upper envelope and the lower envelope from the second audio signal, respectively, and calculating the eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope, when it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition, inputting the eigenvalue signal into the trained recurrent neural network to obtain the prediction feature, the method further includes: Determining whether the difference between the number of extreme value points and the number of zero-crossing points in the characteristic curve corresponding to the characteristic value signal is within a preset number range; When it is detected that the difference is within the preset number interval, determining whether an average value between an envelope formed by all maximum points and an envelope formed by all minimum points at any time in a characteristic curve corresponding to the characteristic value signal is zero; When it is detected that the average value is zero, it is determined that the characteristic curve corresponding to the characteristic value signal meets the constraint condition.

5. The method according to claim 4, characterized in that Before inputting the characteristic value signal into the trained recurrent neural network to obtain the prediction feature, the method further includes: performing a difference calculation on the second audio signal and the eigenvalue signal, and determining whether the difference calculation result is a monotonic function; Inputting the characteristic value signal into a trained recurrent neural network to obtain prediction features includes: When it is determined that the difference calculation result is a monotonic function, the characteristic value signal is input into a trained recurrent neural network to obtain a prediction feature.

6. The method according to claim 1, characterized in that Determining the warning information of the high-frequency switching power supply according to the calculation result includes: When it is detected that the calculation result is within a preset threshold range, early warning information of the high-frequency switching power supply is generated according to damage information corresponding to the preset threshold range.

7. A high-frequency switching power supply monitoring device for substation relay protection, characterized in that: include: An audio acquisition module, configured to acquire a first audio signal of the high-frequency switching power supply based on an ultrasonic microphone; a first processing module, configured to collect first sequence data corresponding to the first audio signal using at least two microphones, and calculate a second audio signal based on the first sequence data; wherein the data volume of the second audio signal is smaller than the data volume of the first audio signal, and the first sequence data includes the sequence data collected by each of the microphones; a second processing module, configured to extract an upper envelope and a lower envelope from the second audio signal, and calculate an eigenvalue signal based on the second audio signal, the upper envelope, and the lower envelope; A model prediction module is used to input the eigenvalue signal into a trained recurrent neural network to obtain a prediction feature when it is determined that the characteristic curve corresponding to the eigenvalue signal satisfies the constraint condition; an early warning module, configured to perform a two-norm calculation on the prediction feature and the sample feature, and determine early warning information of the high-frequency switching power supply according to the calculation result; The step of calculating the second audio signal based on the first sequence data includes: Calculating an average value of each subsequence data in the first sequence data to obtain second sequence data; Calculating a covariance matrix based on the first sequence data and the second sequence data; Determining at least two eigenvalues ​​of the covariance matrix, and constructing a singular value matrix based on the at least two eigenvalues; A second audio signal is calculated according to the covariance matrix, the singular value matrix, and the preset weight ratio.

8. A high-frequency switching power supply monitoring device for substation relay protection, characterized in that: including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the steps of the method according to any one of claims 1 to 6.

Images