Three-remote test method and equipment for secondary equipment of transformer substation and storage medium

Abstract

The invention discloses a three-remote test method and device for secondary equipment of a transformer substation and a storage medium. The method comprises the following steps: transmitting a three-remote test signal to the secondary equipment in the transformer substation; acquiring a response signal output by the secondary equipment for the three-remote test signal; inputting the three-remote test signal into an expectation response semantic simulator for processing to obtain an expectation response feature; inputting the response signal into a response signal feature extractor for processing to obtain an actual response feature; calculating the distance between the expected response feature and the actual response feature; and detecting the operation state of the secondary equipment according to the distance. According to the embodiment, the three-remote test is performed on the secondary equipment in the transformer substation by comparing the difference between the actual response and the expected response, the operation state of the secondary equipment in the transformer substation is detected, the dependence on technicians can be effectively reduced, the test convenience is greatly improved, the consumed time is short, the coverage rate is high, and the test efficiency is improved. And the test consistency and reliability can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid testing, and in particular to a three-remote testing method, equipment and storage medium for secondary equipment of a transformer substation. Background Art

[0002] As one of the components of the power grid, the secondary equipment in the substation is mainly responsible for monitoring, protecting and controlling the primary equipment to ensure the stable operation of the power system.

[0003] In order to maintain the normal operation of secondary equipment, technical personnel are currently mainly relied on to test the secondary equipment, send and receive signals one by one according to the test requirements, and manually record the test results.

[0004] This testing method is cumbersome and time-consuming, especially in a system composed of complex secondary equipment. The test will take hours or even days. During the testing process, due to the limitations of the technicians' experience and knowledge, the test coverage is low and is easily affected by human subjective factors, resulting in poor test consistency and reliability. Summary of the invention

[0005] In view of this, the present invention provides a three-remote testing method, equipment and storage medium for substation secondary equipment, so as to improve the consistency and reliability of testing of secondary equipment in the substation.

[0006] A first aspect of the present invention provides a three-remote control test method for a substation secondary device, comprising:

[0007] Transmit three-telemetry test signals to secondary equipment in the substation;

[0008] Collecting a response signal output by the secondary device in response to the three-remote test signal;

[0009] Inputting the three-remote test signals into an expected response semantic simulator for processing to obtain expected response features;

[0010] Inputting the response signal into a response signal feature extractor for processing to obtain an actual response feature;

[0011] Calculating the distance between the expected response feature and the actual response feature;

[0012] The operating state of the secondary device is detected according to the distance.

[0013] A second aspect of the present invention provides a three-remote test device for a substation secondary device, comprising:

[0014] Three-remote test signal transmission module, used to transmit three-remote test signals to secondary equipment in the substation;

[0015] A response signal acquisition module, used for acquiring the response signal output by the secondary device to the three-remote test signal;

[0016] An expected response feature generation module is used to input the three-remote test signals into an expected response semantic simulator for processing to obtain expected response features;

[0017] An actual response feature generating module, used for inputting the response signal into a response signal feature extractor for processing to obtain an actual response feature;

[0018] A distance calculation module, used to calculate the distance between the expected response feature and the actual response feature;

[0019] The operating status detection module is used to detect the operating status of the secondary device according to the distance.

[0020] A third aspect of the present invention provides an electronic device, the electronic device comprising:

[0021] at least one processor; and

[0022] a memory communicatively connected to the at least one processor; wherein,

[0023] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the three-remote testing method for substation secondary equipment as described in the first aspect above.

[0024] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the three-remote control test method for substation secondary equipment as described in the first aspect above.

[0025] A fifth aspect of the present invention provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the three-remote testing method of substation secondary equipment as described in the first aspect above.

[0026] In this embodiment, the three-remote test signal is transmitted to the secondary equipment in the substation; the response signal output by the secondary equipment for the three-remote test signal is collected; the three-remote test signal is input into the expected response semantic simulator for processing to obtain the expected response feature; the response signal is input into the response signal feature extractor for processing to obtain the actual response feature; the distance between the expected response feature and the actual response feature is calculated; and the operating status of the secondary equipment is detected based on the distance. This embodiment provides an automated testing framework for the three-remote test, which performs the three-remote test on the secondary equipment in the substation by comparing the difference between the actual response and the expected response, and detects the operating status of the secondary equipment in the substation, which can effectively reduce the dependence on technical personnel, greatly improve the simplicity of the test, shorten the time consumption, and increase the coverage rate, which can effectively improve the consistency and reliability of the test.

[0027] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 It is a flow chart of a three-remote testing method for substation secondary equipment provided in Embodiment 1 of the present invention.

[0030] Figure 2 It is a structural schematic diagram of a three-remote test device for substation secondary equipment provided in Embodiment 2 of the present invention.

[0031] Figure 3 It is a structural schematic diagram of an electronic device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can cover sequential implementations other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] Embodiment 1

[0035] See also Figure 1 , shows a flowchart of a three-remote test method for a substation secondary device provided by the first embodiment of the present invention, the method can be executed by a three-remote test device for a substation secondary device, the three-remote test device for a substation secondary device can be implemented in the form of hardware and / or software, and the three-remote test device for a substation secondary device can be configured in an electronic device of a test platform. Figure 1 As shown, the method includes:

[0036] Step 101: Transmit three-remote test signals to secondary equipment in the substation.

[0037] Among the secondary equipment in substations, the three remote technologies (i.e., remote signaling, remote measurement, and remote control) are one of the technologies for realizing remote monitoring and control.

[0038] Among them, telesignaling refers to the detection of remote status signals, such as the position (on / off) of the circuit breaker, the status of the protection device, etc. The test platform can simulate the changes in the status signals and verify whether the secondary equipment can accurately reflect the changes in the status signals.

[0039] Telemetry involves the transmission of remote measurement data, such as current and voltage values. The test platform can generate simulated measurement values ​​to check whether the secondary equipment can correctly receive and process the measurement values.

[0040] Remote control refers to the operation of controlling equipment from a distance, such as remotely controlling the opening and closing of circuit breakers. During testing, the test platform can send control commands to confirm that the secondary equipment can perform the corresponding operations.

[0041] In this embodiment, the simulation terminal (i.e., the simulation signal generator) is connected to the secondary equipment to be tested. The test platform can configure the simulation terminal and the secondary equipment regularly or irregularly according to the function type to be tested (remote control, telemetry, and telesignaling). It simulates various three-remote test signals that may be generated by the primary equipment during the actual operation of the substation without interfering with the actual operation of the power grid, such as voltage signals, current signals, etc., and transmits the three-remote test signals to the secondary equipment to test the secondary equipment and verify whether the secondary equipment can correctly respond to the three-remote test signals.

[0042] Step 102: Collect the response signal output by the secondary device to the three-remote test signal.

[0043] In this embodiment, the secondary device is equipped with a data collector. While transmitting the three-remote test signals, the data collector is started to collect the response signals output by the secondary device to the three-remote test signals, thereby verifying whether the secondary device can correctly respond to the simulated three-remote test signals (telesignaling, telemetering, and remote control).

[0044] During the acquisition process, ensure that the acquisition time window is sufficient to cover the entire test cycle so as to completely capture the response signal.

[0045] Step 103: input the three remote test signals into the expected response semantic simulator for processing to obtain expected response features.

[0046] In this embodiment, an expected response semantic simulator can be constructed and trained in advance based on deep learning, and the three-remote test signals can be input into the expected response semantic simulator. The response semantic simulator can anticipate the characteristics of the response signal generated when the secondary device correctly captures and responds to the three-remote test signals, which are recorded as expected response characteristics.

[0047] In one embodiment of the present invention, step 103 may include the following steps:

[0048] Step 1031: Load the expected response semantic simulator.

[0049] In this embodiment, the expected response semantic simulator is constructed based on the seq2seq (Sequence to Sequence) architecture in deep learning.

[0050] Among them, the expected response semantic simulator includes an encoder Encoder with a one-dimensional (1D) convolutional neural network (CNN) structure and a decoder Decoder with a recurrent neural network (RNN) structure.

[0051] During the testing process, the expected response semantic simulator that has been trained in advance can be loaded into the memory for execution.

[0052] Step 1032: Input the three remote test signals into the encoder for encoding to obtain a sequence of local implicit correlation features.

[0053] The test platform inputs the simulated three-remote test signals into the encoder in the expected response semantic simulator to encode them and obtain a sequence of local implicit association features.

[0054] Step 1033: Perform feature optimization aggregation processing based on cluster contribution analysis on the sequence of local implicit association features to obtain global semantically significant aggregation features.

[0055] Since each local implicit correlation feature in the sequence of local implicit correlation features contains local correlation feature information about the three remote test signals, these local correlation feature information have correlation semantics and global aggregation representation information. However, not all local implicit correlation features are equally important to the final expected response features, and some local implicit correlation features may contain redundant information or noise.

[0056] Based on this, the sequence of local implicit correlation features can be subjected to feature optimization aggregation processing based on cluster contribution analysis to obtain global semantically significant aggregation features.

[0057] Among them, the feature optimization aggregation processing based on cluster contribution analysis aims to use the simulated field analysis theory to perform field-based explicit modeling of the cluster contribution of each local implicit correlation feature, improve the accuracy of feature aggregation, and thus improve the performance and generalization ability of the expected response semantic simulator.

[0058] In one embodiment of the present invention, step 1033 may further include the following steps:

[0059] Step 10331: Perform self-supervised clustering on the sequence of local implicit correlation features to obtain category features.

[0060] In this embodiment, the sequence of local implicit correlation features can be clustered by self-supervision to obtain category features. The so-called self-supervision clustering does not rely on pre-labeled training samples, but clusters by automatically learning the intrinsic structure and pattern of the sequence of local implicit correlation features.

[0061] Exemplarily, the mean of each local implicit correlation feature in the sequence of local implicit correlation features may be calculated to obtain the category feature.

[0062] Step 10332: According to the category features, the sequence of local implicit correlation features is modulated and aggregated based on the feature clustering contribution field distribution to obtain global semantically significant aggregate features.

[0063] In this embodiment, on the basis of the category features, the sequence of local implicit association features can be modulated and aggregated based on the feature clustering contribution field distribution to obtain global semantically significant aggregate features.

[0064] In one embodiment of the present invention, step 10332 may further include the following steps:

[0065] Step 103321: Calculate the implicit clustering contribution factor of each local implicit association feature in the sequence of local implicit association features relative to the category feature to obtain a sequence of implicit clustering contribution factors.

[0066] In this embodiment, the implicit clustering contribution factor of each local implicit association feature in the sequence of local implicit association features relative to the category feature is calculated to obtain a sequence of implicit clustering contribution factors.

[0067] Exemplarily, the absolute value of the ratio between the local implicit association feature and the category feature by position is taken to obtain the cluster interaction association feature; the logarithmic function value of the feature value at each position in the cluster interaction association feature with a base of 2 is calculated to obtain the cluster interaction value; the feature value at each position in the cluster interaction value is used as the weighting coefficient, and the feature value at each position in the local implicit association feature is weighted and summed to obtain the cluster contribution factor; the exponential function value of the cluster contribution factor with a natural constant as the base is calculated to obtain the implicit cluster contribution factor.

[0068] Step 103322: vectorize the sequence of implicit clustering contribution factors to obtain a clustering contribution field distribution vector.

[0069] In this embodiment, each implicit clustering contribution factor in the sequence of implicit clustering contribution factors is vectorized and arranged to obtain a clustering contribution field distribution vector.

[0070] Step 103323: Input the cluster contribution field distribution vector into the cluster contribution field explicit modeling module based on the self-attention mechanism for processing to obtain the cluster contribution field modulation weight.

[0071] In this embodiment, the cluster contribution field distribution vector can be input into a cluster contribution field explicit modeling module based on a self-attention mechanism (Self-Attention) for processing to obtain the cluster contribution field modulation weight.

[0072] Exemplarily, the matrix multiplications between the cluster contribution field distribution vector and the query weight matrix (Q matrix), key weight matrix (K matrix) and value weight matrix (V matrix) in the self-attention mechanism are calculated respectively to obtain the cluster contribution field distribution query vector (Q matrix), the cluster contribution field distribution key vector (K matrix) and the cluster contribution field distribution value vector (V matrix);

[0073] The product of the cluster contribution field distribution query vector and the transposed vector of the cluster contribution field distribution key vector is divided by the square root of the length of the cluster contribution field distribution key vector by position to obtain the cluster contribution field semantic interaction matrix.

[0074] Use functions such as Softmax to normalize the cluster contribution field semantic interaction matrix.

[0075] If the normalization process is completed, the product between the cluster contribution field distribution value vector and the cluster contribution field semantic interaction matrix is ​​calculated to obtain the cluster contribution field modulation weight.

[0076] Step 103324: Use each feature value in the cluster contribution field modulation weight to perform weighted summation of the sequence of local implicit correlation features by position to obtain a global semantically significant aggregation feature.

[0077] In this embodiment, each feature value in the cluster contribution field modulation weight can be used as a weight, and the product between it and the sequence of local implicit association features is calculated by position, and the products are summed to obtain the global semantically significant aggregation feature.

[0078] In this embodiment, the process of aggregating the sequence of local implicit association features into global semantically significant aggregate features can be expressed as follows:

[0079] X={x1,x2,...,x k ,...,x n}

[0080]

[0081] v d ={d1; d2; ...; d j ;...;d n}

[0082]

[0083] v q =W q v d

[0084] v k =W k v d

[0085] v v =W b v d

[0086]

[0087] Among them, X is the sequence of local implicit correlation features, x1, x2, x k ,x n are the first, second, kth and nth local implicit correlation features in the sequence of local implicit correlation features, respectively, x i is the ith local implicit correlation feature in the sequence of local implicit correlation features, N is the number of local implicit correlation features in the sequence of local implicit correlation features, x c is the category feature, x j is the jth local implicit correlation feature in the sequence of local implicit correlation features, is the feature value of the kth position in the jth local implicit correlation feature, x ck is the feature value of the kth position in the category feature, log represents the logarithmic function value with base 2, exp represents the natural exponential function value, and d j is the implicit clustering contribution factor corresponding to the jth local implicit correlation feature, v d is the cluster contribution field distribution vector obtained by quantizing multiple implicit cluster contribution factors, W q , W k and W b are the query weight matrix, key weight matrix and value weight matrix respectively, v q 、v k and v v are cluster contribution field distribution query vector, cluster contribution field distribution key vector and cluster contribution field distribution value vector respectively, L is the length of cluster contribution field distribution key vector, Softmax(·) is the Softmax function, is vector multiplication, v w The cluster contribution field modulation weight, x f It is a global semantically significant aggregation feature.

[0088] In this embodiment, firstly, a self-supervised learning method is used to perform cluster analysis on the sequence of local implicit correlation features, and the local correlation feature patterns and structures of the simulated three-remote test signals in the set are automatically discovered to generate category features.

[0089] In this process, self-supervised clustering can reveal the intrinsic structure of the local correlation features of each simulated three-remote test signal, which helps the expected response semantic simulator to learn meaningful feature representations without explicit labels, enhances the expected response semantic simulator's understanding of the correlation between the various local features of the simulated three-remote test signals, and helps to reveal the intrinsic connection between each feature vector.

[0090] Furthermore, the implicit clustering contribution factor between the local implicit association features and the category features is calculated to reflect the contribution of each local implicit association feature to its cluster.

[0091] Accordingly, the calculation of the implicit clustering contribution factor helps to identify and quantify the role of each local implicit correlation feature in the clustering process, providing a basis for subsequent feature selection and weight allocation, so that the model can pay more attention to the features that contribute significantly to clustering.

[0092] Next, the calculated sequence of implicit clustering contribution factors is converted into a clustering contribution field distribution vector, that is, the sequence of implicit clustering contribution factors is integrated into the vector field by using data integration to construct the clustering contribution field.

[0093] Then, the cluster contribution field distribution vector is explicitly modeled using the self-attention mechanism to generate the cluster contribution field modulation weight. That is, the cluster contribution degree of each implicit cluster contribution factor in the cluster contribution field is explicitly modeled based on the field autocorrelation global analysis, which can strengthen the focus on the feature vectors with large cluster contributions, while suppressing unimportant features, improving the quality of feature representation and the accuracy of the subsequent global semantic representation of the simulated three-remote test signal and the semantic representation of the expected response signal.

[0094] Finally, the clustering contribution field modulation weights are used to perform weighted fusion on the sequence of local implicit correlation features to generate global semantically significant aggregation features. The global semantically significant aggregation features provide a comprehensive feature representation, which not only contains the local correlation feature information of the original simulated three-remote test signal, but also integrates the knowledge learned in the clustering process, so that the expected response semantic simulator can more accurately identify and analyze signals when processing complex tasks, thereby improving the overall performance and reliability of the expected response semantic simulator.

[0095] Step 1034: input the global semantically significant aggregated features into a decoder for decoding to obtain expected response features.

[0096] In this embodiment, the global semantically significant aggregated features are input into a decoder of an expected response semantic simulator to decode the global semantically significant aggregated features, and the response behavior of the secondary device to be tested in a real operating environment is predicted, thereby obtaining the expected response features.

[0097] When the decoder structure is RNN, RNN can capture the global semantic long-distance correlation features and relationships of the simulated three-remote test signals in the global semantic significant aggregation features after feature aggregation representation, thereby decoding and generating the expected response signal semantic simulation to obtain the expected response features.

[0098] Step 104: Input the response signal into a response signal feature extractor for processing to obtain actual response features.

[0099] In this embodiment, the response signal is input into a response signal feature extractor, and features are extracted from the response signal and recorded as actual response features.

[0100] When the structure of the encoder of the expected response semantic simulator is a one-dimensional (1D) convolutional neural network (CNN), the structure of the response signal feature extractor is a one-dimensional (1D) convolutional neural network (CNN). The encoder of the expected response semantic simulator and the response signal feature extractor have the same or similar structure, which helps to extract features at the same level.

[0101] Furthermore, the convolutional neural network can identify local patterns in the response signal through convolution operations, which can reduce the amount of data and thus abstract the key features of the response signal. At the same time, building a response signal feature extractor based on the convolutional neural network can enhance the robustness of the response signal feature extractor to noise and changes. Even if the input response signal is subject to some interference or slight changes, the actual response features extracted can still reflect the main features of the response signal, which can facilitate subsequent comparative analysis.

[0102] The response signal feature extractor includes a convolution layer, a pooling layer, and a fully connected layer. In the convolution layer, the response signal can be convolved based on a one-dimensional (1D) convolution kernel to obtain a convolution feature, wherein the one-dimensional convolution kernel slides on the response signal to extract local implicit correlation features in the response signal, and the convolution kernel can capture patterns in the response signal, which may be waveforms of a specific frequency, periodic patterns, or other meaningful features. In the pooling layer, the convolution feature vector is pooled to obtain a pooled feature vector, wherein the pooling process is used to reduce the spatial dimension of the data while retaining the most important feature information. The pooling operation helps the model focus on the most significant features and reduce the risk of overfitting. The pooled feature vector is nonlinearly activated to obtain an activated feature vector. In the fully connected layer, the activated feature vector is mapped to the actual response feature.

[0103] Step 105: Calculate the distance between the expected response feature and the actual response feature.

[0104] In this embodiment, the distance between the expected response feature and the actual response feature, such as the Hamming distance, may be calculated to quantify the difference between the expected response feature and the actual response feature.

[0105] Among them, the Hamming distance can be used to measure the number of differences between two binary strings of equal length. Here, it is used to represent the difference between two semantic encoding feature vectors. By calculating the Hamming distance, the subtle changes between the expected response features and the actual response features can be captured more accurately, thereby improving the reliability of the test results.

[0106] Step 106: Detect the operating status of the secondary device based on the distance.

[0107] In this embodiment, the difference between the actual response and the expected response can be compared, thereby achieving a more reliable and intelligent three-remote automatic test of the substation secondary equipment to verify the operating status of the substation secondary equipment, that is, whether it operates in the expected manner and whether there are any abnormalities.

[0108] When the secondary equipment can correctly respond to the three-remote test signals, it can be ensured that it can work normally in actual operation.

[0109] By comparing the difference between the actual response signal and the expected response signal, it is possible to promptly detect whether there is any abnormality in the equipment, so that measures can be taken to correct or prevent it.

[0110] In a specific implementation, the distance may be compared with a preset threshold.

[0111] If the distance is greater than or equal to a preset threshold, it means that the difference between the expected response characteristic and the actual response characteristic is large, and the operating state of the secondary device is determined to be abnormal.

[0112] If the distance is smaller than a preset threshold, it means that the difference between the expected response characteristic and the actual response characteristic is small, and it is determined that the operating state of the secondary device is normal.

[0113] In this embodiment, the three-remote test signal is transmitted to the secondary equipment in the substation; the response signal output by the secondary equipment for the three-remote test signal is collected; the three-remote test signal is input into the expected response semantic simulator for processing to obtain the expected response feature; the response signal is input into the response signal feature extractor for processing to obtain the actual response feature; the distance between the expected response feature and the actual response feature is calculated; and the operating status of the secondary equipment is detected based on the distance. This embodiment provides an automated testing framework for the three-remote test, which performs the three-remote test on the secondary equipment in the substation by comparing the difference between the actual response and the expected response, and detects the operating status of the secondary equipment in the substation, which can effectively reduce the dependence on technical personnel, greatly improve the simplicity of the test, shorten the time consumption, and increase the coverage rate, which can effectively improve the consistency and reliability of the test.

[0114] Embodiment 2

[0115] See also Figure 2 , shows a schematic diagram of the structure of a three-remote test device for a substation secondary device provided by the second embodiment of the present invention. Figure 2 As shown, the device comprises:

[0116] Three-remote test signal transmission module 201, used to transmit three-remote test signals to secondary equipment in the substation;

[0117] A response signal acquisition module 202, used for acquiring a response signal output by the secondary device in response to the three-remote test signal;

[0118] An expected response feature generation module 203 is used to input the three-remote test signals into an expected response semantic simulator for processing to obtain expected response features;

[0119] An actual response feature generating module 204 is used to input the response signal into a response signal feature extractor for processing to obtain an actual response feature;

[0120] A distance calculation module 205 is used to calculate the distance between the expected response feature and the actual response feature;

[0121] The operating status detection module 206 is configured to detect the operating status of the secondary device according to the distance.

[0122] In one embodiment of the present invention, the expected response feature generation module 203 includes:

[0123] An expected response semantics simulator loading module is used to load the expected response semantics simulator; wherein the expected response semantics simulator includes an encoder with a convolutional neural network structure and a decoder with a recurrent neural network structure; the response signal feature extractor has a convolutional neural network structure;

[0124] An encoding module, used for inputting the three-remote test signal into the encoder for encoding to obtain a sequence of local implicit correlation features;

[0125] A feature aggregation module, used to perform feature optimization aggregation processing based on cluster contribution analysis on the sequence of local implicit association features to obtain global semantically significant aggregation features;

[0126] A decoding module is used to input the global semantically significant aggregation feature into the decoder for decoding to obtain the expected response feature.

[0127] In one embodiment of the present invention, the feature aggregation module includes:

[0128] A self-supervised clustering module, used for performing self-supervised clustering on the sequence of local implicit association features to obtain category features;

[0129] The modulation and aggregation module is used to modulate and aggregate the sequence of the local implicit association features based on the feature clustering contribution field distribution according to the category features to obtain global semantically significant aggregation features.

[0130] In one embodiment of the present invention, the self-supervised clustering module is further used for:

[0131] The mean of the sequence of the local implicit association features is calculated to obtain the category features.

[0132] In one embodiment of the present invention, the modulation aggregation module includes:

[0133] An implicit clustering contribution factor calculation module, used to calculate the implicit clustering contribution factor of each local implicit association feature in the sequence of local implicit association features relative to the category feature, to obtain a sequence of implicit clustering contribution factors;

[0134] A cluster contribution field distribution vector calculation module is used to vectorize the sequence of the implicit cluster contribution factors to obtain a cluster contribution field distribution vector;

[0135] A cluster contribution field modulation weight calculation module is used to input the cluster contribution field distribution vector into a cluster contribution field explicit modeling module based on a self-attention mechanism for processing to obtain a cluster contribution field modulation weight;

[0136] The global semantically significant aggregate feature generation module is used to use each feature value in the cluster contribution field modulation weight to sum the sequence of local implicit association features by position weight to obtain the global semantically significant aggregate feature.

[0137] In one embodiment of the present invention, the implicit clustering contribution factor calculation module is further used for:

[0138] Taking the absolute value of the ratio between the local implicit association feature and the category feature according to the position to obtain the clustering interactive association feature;

[0139] Calculate the logarithmic function value of the feature value at each position in the cluster interaction correlation feature with base 2 to obtain the cluster interaction value;

[0140] Taking each position feature value in the cluster interaction value as a weighting coefficient, weighted summing each position feature value in the local implicit association feature to obtain a cluster contribution factor;

[0141] An exponential function value of the cluster contribution factor with a natural constant as the base is calculated to obtain an implicit cluster contribution factor.

[0142] In one embodiment of the present invention, the cluster contribution field modulation weight calculation module is further used for:

[0143] Calculating matrix multiplications between the cluster contribution field distribution vector and the query weight matrix, the key weight matrix and the value weight matrix respectively to obtain a cluster contribution field distribution query vector, a cluster contribution field distribution key vector and a cluster contribution field distribution value vector;

[0144] Divide the product of the cluster contribution field distribution query vector and the transposed vector of the cluster contribution field distribution key vector by the square root of the length of the cluster contribution field distribution key vector by position to obtain a cluster contribution field semantic interaction matrix;

[0145] Normalizing the cluster contribution field semantic interaction matrix;

[0146] If the normalization process is completed, the product of the cluster contribution field distribution value vector and the cluster contribution field semantic interaction matrix is ​​calculated to obtain the cluster contribution field modulation weight.

[0147] In one embodiment of the present invention, the operating status detection module 206 includes:

[0148] an abnormality determination module, configured to determine that the operating state of the secondary device is abnormal if the distance is greater than or equal to a preset threshold;

[0149] The normal determination module is used to determine that the operating state of the secondary device is normal if the distance is less than a preset threshold.

[0150] The three-remote test device for substation secondary equipment provided in the embodiment of the present invention can execute the three-remote test method for substation secondary equipment provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the three-remote test method for substation secondary equipment.

[0151] Embodiment 3

[0152] See also Figure 3 , shows a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0153] like Figure 3As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0154] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0155] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as the three-remote testing method for substation secondary equipment.

[0156] In some embodiments, the three-remote test method of the substation secondary equipment can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the three-remote test method of the substation secondary equipment described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the three-remote test method of the substation secondary equipment in any other appropriate manner (for example, by means of firmware).

[0157] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0158] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0159] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0160] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0161] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0162] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0163] Embodiment 4

[0164] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the three-remote control test method for a substation secondary device provided in any embodiment of the present invention is implemented.

[0165] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0166] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0167] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A three-remote test method for substation secondary equipment, characterized in that: include: Transmit three-telemetry test signals to secondary equipment in the substation; Collecting a response signal output by the secondary device in response to the three-remote test signal; Inputting the three-remote test signals into an expected response semantic simulator for processing to obtain expected response features; Inputting the response signal into a response signal feature extractor for processing to obtain an actual response feature; Calculating the distance between the expected response feature and the actual response feature; The operating state of the secondary device is detected according to the distance.

2. The method according to claim 1, characterized in that The inputting of the three-remote test signal into the expected response semantic simulator for processing to obtain the expected response feature includes: Loading an expected response semantic simulator; wherein the expected response semantic simulator includes an encoder having a convolutional neural network structure and a decoder having a recurrent neural network structure; the response signal feature extractor has a convolutional neural network structure; Inputting the three-remote test signal into the encoder for encoding to obtain a sequence of local implicit correlation features; Performing feature optimization aggregation processing based on cluster contribution analysis on the sequence of local implicit association features to obtain global semantically significant aggregation features; The global semantically significant aggregated features are input into the decoder for decoding to obtain expected response features.

3. According to the method of claim 2, the step of performing feature optimization aggregation processing based on cluster contribution analysis on the sequence of local implicit association features to obtain global semantically significant aggregate features comprises: Performing self-supervised clustering on the sequence of local implicit association features to obtain category features; According to the category features, the sequence of the local implicit association features is modulated and aggregated based on the feature clustering contribution field distribution to obtain a global semantically significant aggregation feature.

4. The method according to claim 3, characterized in that The self-supervised clustering of the sequence of local implicit association features to obtain category features includes: The mean of the sequence of the local implicit association features is calculated to obtain the category features.

5. The method according to claim 3, characterized in that: The modulating and aggregating the sequence of the local implicit association features based on the feature clustering contribution field distribution according to the category features to obtain the global semantically significant aggregation features includes: Calculating the implicit clustering contribution factor of each local implicit association feature in the sequence of local implicit association features relative to the category feature to obtain a sequence of implicit clustering contribution factors; Vectorizing and arranging the sequence of the implicit clustering contribution factors to obtain a clustering contribution field distribution vector; Inputting the cluster contribution field distribution vector into the cluster contribution field explicit modeling module based on the self-attention mechanism for processing to obtain the cluster contribution field modulation weight; The sequence of the local implicit association features is weighted and summed by position using each feature value in the cluster contribution field modulation weight to obtain a global semantically significant aggregation feature.

6. The method according to claim 5, characterized in that The step of calculating the implicit clustering contribution factor of each local implicit association feature in the sequence of local implicit association features relative to the category feature to obtain a sequence of implicit clustering contribution factors includes: Taking the absolute value of the ratio between the local implicit association feature and the category feature according to the position to obtain the clustering interactive association feature; Calculate the logarithmic function value of the feature value at each position in the cluster interaction correlation feature with base 2 to obtain the cluster interaction value; Taking each position feature value in the cluster interaction value as a weighting coefficient, weighted summing each position feature value in the local implicit association feature to obtain a cluster contribution factor; An exponential function value of the cluster contribution factor with a natural constant as the base is calculated to obtain an implicit cluster contribution factor.

7. The method according to claim 5, characterized in that The cluster contribution field distribution vector is input into the cluster contribution field explicit modeling module based on the self-attention mechanism for processing to obtain the cluster contribution field modulation weight, including: Calculating matrix multiplications between the cluster contribution field distribution vector and the query weight matrix, the key weight matrix and the value weight matrix respectively to obtain a cluster contribution field distribution query vector, a cluster contribution field distribution key vector and a cluster contribution field distribution value vector; Divide the product of the cluster contribution field distribution query vector and the transposed vector of the cluster contribution field distribution key vector by the square root of the length of the cluster contribution field distribution key vector by position to obtain a cluster contribution field semantic interaction matrix; Normalizing the cluster contribution field semantic interaction matrix; If the normalization process is completed, the product of the cluster contribution field distribution value vector and the cluster contribution field semantic interaction matrix is ​​calculated to obtain the cluster contribution field modulation weight.

8. The method according to any one of claims 1 to 7, characterized in that The detecting the operating state of the secondary device according to the distance includes: If the distance is greater than or equal to a preset threshold, determining that the operating state of the secondary device is abnormal; If the distance is less than a preset threshold, it is determined that the operating state of the secondary device is normal.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the three-remote testing method for substation secondary equipment according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the three-remote control test method for a substation secondary device according to any one of claims 1 to 8 is implemented.

Images