Voltage transformer secondary circuit voltage drop monitoring method and device, computer equipment, storage medium and computer program product

Abstract

The invention relates to a voltage transformer secondary circuit voltage drop monitoring method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring associated data of a to-be-monitored voltage transformer secondary circuit through a wireless network; preprocessing the associated data to obtain preprocessed associated data; initial data features corresponding to the preprocessed associated data are obtained through the trained pressure drop monitoring model; respectively determining a first data feature and a second data feature corresponding to the initial data feature; performing fusion processing on the first data feature and the second data feature to obtain a fusion data feature corresponding to the preprocessed associated data until the obtained fusion data feature is an Nth fusion data feature corresponding to the preprocessed associated data, and according to the Nth fusion data feature, performing fusion processing on the first data feature and the second data feature; and determining a target voltage drop monitoring result of the secondary circuit of the voltage transformer. By adopting the method, the monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer can be improved.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for monitoring voltage drop in a secondary circuit of a voltage transformer. Background Art

[0002] In substations, in order to ensure stable operation of the power system, it is crucial to accurately monitor the voltage drop in the secondary circuit of the voltage transformer.

[0003] In traditional technology, in the process of monitoring the voltage drop in the secondary circuit of the voltage transformer, manual monitoring using a portable secondary voltage drop test instrument is generally adopted; however, this method is prone to monitoring errors due to operating errors, which in turn causes low monitoring accuracy of the voltage drop in the secondary circuit of the voltage transformer. Summary of the invention

[0004] Based on this, it is necessary to provide a voltage transformer secondary circuit voltage drop monitoring method, device, computer equipment, computer readable storage medium and computer program product that can improve the monitoring accuracy of the voltage transformer secondary circuit voltage drop in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for monitoring a voltage drop in a secondary circuit of a voltage transformer, comprising:

[0006] Acquire, via a wireless network, associated data of a secondary circuit of a voltage transformer to be monitored; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer;

[0007] Preprocessing the associated data to obtain preprocessed associated data;

[0008] Performing feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data;

[0009] Respectively determine a first data feature and a second data feature corresponding to the initial data feature;

[0010] Performing fusion processing on the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data;

[0011] The fused data feature is used as a new initial data feature, and the process jumps to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data; N is a positive integer greater than or equal to 2;

[0012] A target voltage drop monitoring result of the voltage transformer secondary circuit is determined according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

[0013] In one embodiment, determining the target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature includes:

[0014] Respectively determining a third data feature, a fourth data feature, and a fifth data feature corresponding to the Nth fused data feature;

[0015] Performing multiple fusion processing on the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature to obtain a target data feature corresponding to the secondary circuit of the voltage transformer;

[0016] According to the target data characteristics, a target voltage drop monitoring result of the voltage transformer secondary circuit is determined.

[0017] In one embodiment, the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature are subjected to multiple fusion processing to obtain the target data feature corresponding to the secondary circuit of the voltage transformer, including:

[0018] The third data feature and the fourth data feature are merged to obtain a sixth data feature corresponding to the secondary circuit of the voltage transformer;

[0019] The fifth data feature and the sixth data feature are merged to obtain a seventh data feature corresponding to the secondary circuit of the voltage transformer;

[0020] The Nth fused data feature and the seventh data feature are fused to obtain the target data feature.

[0021] In one embodiment, determining the target voltage drop monitoring result of the voltage transformer secondary circuit according to the target data feature includes:

[0022] According to the target data characteristics, a plurality of preset voltage drop monitoring results of the secondary circuit of the voltage transformer and a predicted probability of each preset voltage drop monitoring result are obtained;

[0023] From each of the preset pressure drop monitoring results, the preset pressure drop monitoring result with the maximum predicted probability is screened out as the target pressure drop monitoring result.

[0024] In one of the embodiments, after determining the target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature, the method further includes:

[0025] When the target voltage drop monitoring result indicates that an abnormal voltage drop exists in the secondary circuit of the voltage transformer, extracting key associated data from the associated data;

[0026] Determining the abnormal result corresponding to the secondary circuit of the voltage transformer according to the key associated data;

[0027] Generate alarm information corresponding to the secondary circuit of the voltage transformer according to the abnormal result;

[0028] The alarm information is sent to a target terminal associated with the secondary circuit of the voltage transformer.

[0029] In one embodiment, the trained voltage drop monitoring model is trained in the following manner:

[0030] Acquire sample associated data of a sample voltage transformer secondary circuit; the sample associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the sample voltage transformer secondary circuit;

[0031] Preprocessing the sample-related data to obtain preprocessed sample-related data;

[0032] Performing feature extraction processing on the preprocessed sample associated data through the pressure drop monitoring model to be trained to obtain initial sample data features corresponding to the preprocessed sample associated data;

[0033] Respectively determine a first sample data feature and a second sample data feature corresponding to the initial sample data feature;

[0034] Performing fusion processing on the first sample data feature and the second sample data feature to obtain a fused sample data feature corresponding to the preprocessed sample associated data;

[0035] The fused sample data feature is used as a new initial sample data feature, and the process jumps to the step of respectively determining the first sample data feature and the second sample data feature corresponding to the initial sample data feature, until the obtained fused sample data feature is the Nth fused sample data feature corresponding to the preprocessed sample associated data; N is a positive integer greater than or equal to 2;

[0036] Determining a predicted voltage drop monitoring result of the secondary circuit of the sample voltage transformer according to the Nth fused sample data feature;

[0037] The actual voltage drop monitoring result of the secondary circuit of the sample voltage transformer is obtained, and according to the difference between the predicted voltage drop monitoring result and the actual voltage drop monitoring result, the voltage drop monitoring model to be trained is iteratively trained to obtain the trained voltage drop monitoring model.

[0038] In a second aspect, the present application also provides a voltage transformer secondary circuit voltage drop monitoring device, comprising:

[0039] A data acquisition module, used to acquire the associated data of the secondary circuit of the voltage transformer to be monitored through a wireless network; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer;

[0040] A data processing module, used for preprocessing the associated data to obtain preprocessed associated data;

[0041] A feature extraction module, used to perform feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data;

[0042] A feature determination module, used to respectively determine a first data feature and a second data feature corresponding to the initial data feature;

[0043] A feature fusion module, used for fusing the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data;

[0044] A feature processing module, used for taking the fused data feature as a new initial data feature, and jumping to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data; N is a positive integer greater than or equal to 2;

[0045] A result determination module is used to determine a target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

[0046] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0047] Acquire, via a wireless network, associated data of a secondary circuit of a voltage transformer to be monitored; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer;

[0048] Preprocessing the associated data to obtain preprocessed associated data;

[0049] Performing feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data;

[0050] Respectively determine a first data feature and a second data feature corresponding to the initial data feature;

[0051] Performing fusion processing on the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data;

[0052] The fused data feature is used as a new initial data feature, and the process jumps to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data; N is a positive integer greater than or equal to 2;

[0053] A target voltage drop monitoring result of the voltage transformer secondary circuit is determined according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

[0054] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0055] Acquire, via a wireless network, associated data of a secondary circuit of a voltage transformer to be monitored; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer;

[0056] Preprocessing the associated data to obtain preprocessed associated data;

[0057] Performing feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data;

[0058] Respectively determine a first data feature and a second data feature corresponding to the initial data feature;

[0059] Performing fusion processing on the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data;

[0060] The fused data feature is used as a new initial data feature, and the process jumps to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data; N is a positive integer greater than or equal to 2;

[0061] A target voltage drop monitoring result of the voltage transformer secondary circuit is determined according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

[0062] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0063] Acquire, via a wireless network, associated data of a secondary circuit of a voltage transformer to be monitored; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer;

[0064] Preprocessing the associated data to obtain preprocessed associated data;

[0065] Performing feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data;

[0066] Respectively determine a first data feature and a second data feature corresponding to the initial data feature;

[0067] Performing fusion processing on the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data;

[0068] The fused data feature is used as a new initial data feature, and the process jumps to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data; N is a positive integer greater than or equal to 2;

[0069] A target voltage drop monitoring result of the voltage transformer secondary circuit is determined according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

[0070] The above-mentioned voltage transformer secondary circuit voltage drop monitoring method, device, computer equipment, storage medium and computer program product first obtain the power data, temperature data, vibration data, noise data, humidity data and weather data of the voltage transformer secondary circuit to be monitored through a wireless network as associated data, and pre-process the associated data to obtain pre-processed associated data, and then use the trained voltage drop monitoring model to perform feature extraction processing on the pre-processed associated data to obtain initial data features corresponding to the pre-processed associated data, then respectively determine the first data feature and the second data feature corresponding to the initial data feature, and fuse the first data feature and the second data feature to obtain the fused data feature corresponding to the pre-processed associated data, then use the fused data feature as a new initial data feature, and jump to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the pre-processed associated data, and finally, determine the target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature. In this way, in the process of monitoring the voltage drop of the secondary circuit of the voltage transformer, various types of data of the secondary circuit of the voltage transformer are collected as related data through the wireless network, and through a series of processing processes such as preprocessing, model processing and multiple feature processing, the target voltage drop monitoring result of the secondary circuit of the voltage transformer can be accurately obtained, thereby improving the determination accuracy of the target voltage drop monitoring result of the secondary circuit of the voltage transformer, which is beneficial to improving the monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer; moreover, the entire process does not require manual monitoring using a portable secondary voltage drop test instrument, avoiding the method of manual monitoring using a portable secondary voltage drop test instrument, which is prone to monitoring errors due to operating errors, thereby resulting in low monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer, further improving the monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0072] Figure 1 Schematic diagram of a flow chart of a method for monitoring voltage drop in a secondary circuit of a voltage transformer in one embodiment;

[0073] Figure 2 A schematic diagram of the structure of the Xception-efficient multi-scale attention model in one embodiment;

[0074] Figure 3 Schematic diagram of the structure of an efficient multi-scale attention module in one embodiment;

[0075] Figure 4 A schematic diagram of a flow chart of a method for monitoring voltage drop in a secondary circuit of a voltage transformer in another embodiment;

[0076] Figure 5 It is a schematic diagram of a flow chart of an intelligent voltage transformer secondary circuit voltage drop online monitoring method in one embodiment;

[0077] Figure 6 is a structural block diagram of a voltage transformer secondary circuit voltage drop monitoring device in one embodiment;

[0078] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0080] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0081] In an exemplary embodiment, Figure 1 As shown, a method for monitoring the voltage drop of the secondary circuit of a voltage transformer is provided. This embodiment uses the method applied to a server as an example for illustration; it can be understood that the method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be but is not limited to various personal computers, laptops, smart phones and tablets; the server can be implemented with an independent server or a server cluster composed of multiple servers. In this embodiment, the method includes the following steps:

[0082] Step S101, obtaining associated data of a voltage transformer secondary circuit to be monitored through a wireless network; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the voltage transformer secondary circuit.

[0083] The wireless network refers to the Wireless LAN Authentication and Privacy Infrastructure (WAPI) wireless network.

[0084] The voltage transformer secondary circuit is used to represent the electrical circuit constructed around the secondary side of the voltage transformer.

[0085] The associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer.

[0086] Among them, the power data refers to data related to the electrical characteristics of the voltage transformer secondary circuit, including the voltage amplitude on the voltage transformer side, the voltage amplitude on the instrument side and the angle difference of the voltage transformer secondary circuit.

[0087] Among them, the temperature data refers to the temperature values ​​of each key component in the secondary circuit of the voltage transformer (such as the secondary winding of the voltage transformer, connecting wires, fuses, etc.).

[0088] Among them, vibration data refers to the vibration amplitude, frequency and other related data generated by the equipment and components in the secondary circuit of the voltage transformer (such as the voltage transformer body).

[0089] The noise data refers to the sound intensity, frequency and other related data in the surrounding environment of the voltage transformer secondary circuit.

[0090] The humidity data refers to the air humidity of the environment where the secondary circuit of the voltage transformer is located.

[0091] Among them, weather data refers to the meteorological conditions of the environment in which the secondary circuit of the voltage transformer is located, such as temperature, humidity, rainfall, wind speed, wind direction, air pressure and other comprehensive data.

[0092] Exemplarily, in response to a monitoring instruction for a secondary circuit of a voltage transformer, the server obtains power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer to be monitored through a wireless network; then, the server combines and processes the power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer to be monitored in a preset combination manner to obtain associated data of the secondary circuit of the voltage transformer to be monitored.

[0093] Step S102: preprocess the associated data to obtain preprocessed associated data.

[0094] Among them, preprocessing includes data cleaning, data expansion and data normalization.

[0095] Among them, data cleaning refers to deleting incomplete and erroneous data.

[0096] Among them, data expansion refers to increasing the amount of data at the same time point by copying, limiting the amplitude and scaling the original data proportionally.

[0097] The formula for data normalization is as follows:

[0098] , formula (1)

[0099] in, , and They are respectively the relevant data of the secondary circuit of the voltage transformer (such as the temperature data at each moment), the maximum value of the relevant data of the secondary circuit of the voltage transformer (such as the maximum value of the temperature data at each moment) and the minimum value of the relevant data of the secondary circuit of the voltage transformer (such as the minimum value of the temperature data at each moment); These are the normalized data of the secondary circuit of the voltage transformer.

[0100] The preprocessed associated data refers to the associated data after preprocessing.

[0101] Exemplarily, the server uses data processing technology to perform preprocessing such as data cleaning, data expansion, and data normalization on the associated data to obtain preprocessed associated data as the preprocessed associated data.

[0102] Step S103 , performing feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data.

[0103] Among them, the voltage drop monitoring model refers to a network model that can use the associated data of the secondary circuit of the voltage transformer to obtain the target voltage drop monitoring result of the secondary circuit of the voltage transformer, such as Xception (extreme neural network architecture)-efficient multi-scale attention model.

[0104] The initial data features refer to the data features corresponding to the associated data after preprocessing.

[0105] Exemplarily, the server inputs the preprocessed associated data into a trained voltage drop monitoring model, performs feature extraction on the preprocessed associated data through the trained voltage drop monitoring model, and obtains data features corresponding to the preprocessed associated data as initial data features.

[0106] For example, see Figure 2,The server inputs the preprocessed associated data into the Xception-efficient multi-scale attention model, and obtains the initial data features through two groups of 3*3 convolutional layers and activation layers in the Xception-efficient multi-scale attention model.

[0107] Step S104, respectively determining the first data feature and the second data feature corresponding to the initial data feature.

[0108] The first data feature refers to a data feature obtained by performing a 3*3 separable convolution process on the initial data feature.

[0109] The second data feature refers to a data feature obtained by performing 1*1 convolution processing on the initial data feature.

[0110] Exemplarily, the server performs feature extraction processing on the initial data features respectively to obtain first data features and second data features corresponding to the initial data features.

[0111] For example, see Figure 2 , the server performs 3*3 separable convolution processing on the initial data features through the Xception-efficient multi-scale attention model to obtain the first data features corresponding to the initial data features, and performs 1*1 convolution processing on the initial data features to obtain the second data features corresponding to the initial data features.

[0112] Step S105: fusing the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data.

[0113] The fusion processing may refer to weighted summation processing.

[0114] The fused data feature refers to a data feature obtained by fusing the first data feature and the second data feature.

[0115] Exemplarily, the server obtains a first weight corresponding to the first data feature and a second weight corresponding to the second data feature; then, the server fuses the first data feature and the second data feature according to the first weight and the second weight to obtain a fused data feature corresponding to the preprocessed associated data.

[0116] For example, see Figure 2 ,The server performs weighted sum processing on the first data feature and the second data feature through the Xception-efficient multi-scale attention model to obtain the fused data feature corresponding to the preprocessed associated data.

[0117] Step S106, taking the fused data feature as a new initial data feature, and jumping to the steps of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the associated data after preprocessing; N is a positive integer greater than or equal to 2.

[0118] Wherein, N is a positive integer greater than or equal to 2. In actual scenarios, N is 8.

[0119] Exemplarily, the server judges the obtained fused data feature and obtains a judgment result; when the judgment result indicates that the obtained fused data feature is not the Nth fused data feature, the server uses the fused data feature as a new initial data feature, and jumps to the steps of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the associated data after preprocessing.

[0120] For example, see Figure 2 The server judges the obtained fused data feature through the Xception-efficient multi-scale attention model to obtain a judgment result; when the judgment result indicates that the obtained fused data feature is not the eighth fused data feature, the server uses the fused data feature as a new initial data feature, and jumps to the steps of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the eighth fused data feature corresponding to the associated data after preprocessing.

[0121] Step S107, determining a target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

[0122] Among them, the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the secondary circuit of the voltage transformer.

[0123] The abnormal voltage drop is used to indicate that the current voltage drop value of the secondary circuit of the voltage transformer is greater than the preset voltage drop value. It should be noted that the preset voltage drop value depends on the situation.

[0124] Exemplarily, the server determines, based on the Nth fused data feature, a voltage drop monitoring result corresponding to the Nth fused data feature as a target voltage drop monitoring result of the secondary circuit of the voltage transformer.

[0125] In the above-mentioned voltage transformer secondary circuit voltage drop monitoring method, firstly, the power data, temperature data, vibration data, noise data, humidity data and weather data of the voltage transformer secondary circuit to be monitored are obtained through a wireless network as associated data, and the associated data are preprocessed to obtain preprocessed associated data, and then the preprocessed associated data are subjected to feature extraction processing through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data, and then, the first data feature and the second data feature corresponding to the initial data feature are respectively determined, and the first data feature and the second data feature are fused to obtain fused data features corresponding to the preprocessed associated data, and then, the fused data feature is used as a new initial data feature, and the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature is jumped to, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data, and finally, the target voltage drop monitoring result of the voltage transformer secondary circuit is determined according to the Nth fused data feature. In this way, in the process of monitoring the voltage drop of the secondary circuit of the voltage transformer, various types of data of the secondary circuit of the voltage transformer are collected as related data through the wireless network, and through a series of processing processes such as preprocessing, model processing and multiple feature processing, the target voltage drop monitoring result of the secondary circuit of the voltage transformer can be accurately obtained, thereby improving the determination accuracy of the target voltage drop monitoring result of the secondary circuit of the voltage transformer, which is beneficial to improving the monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer; moreover, the entire process does not require manual monitoring using a portable secondary voltage drop test instrument, avoiding the method of manual monitoring using a portable secondary voltage drop test instrument, which is prone to monitoring errors due to operating errors, thereby resulting in low monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer, further improving the monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer.

[0126] In an exemplary embodiment, the above step S107, based on the Nth fused data feature, determines the target voltage drop monitoring result of the voltage transformer secondary circuit, specifically includes the following contents: respectively determining the third data feature, the fourth data feature and the fifth data feature corresponding to the Nth fused data feature; performing multiple fusion processing on the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature to obtain the target data feature corresponding to the voltage transformer secondary circuit; and determining the target voltage drop monitoring result of the voltage transformer secondary circuit based on the target data feature.

[0127] Among them, the third data feature refers to the data feature obtained by performing one-dimensional horizontal average pooling processing on the Nth fused data feature.

[0128] Among them, the fourth data feature refers to the data feature obtained by performing one-dimensional longitudinal average pooling processing on the Nth fused data feature.

[0129] Among them, the fifth data feature refers to the data feature obtained by performing 3*3 convolution processing on the Nth fused data feature.

[0130] The target data feature refers to a data feature obtained by performing multiple fusion processes on the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature.

[0131] Exemplarily, the server performs feature extraction processing on the Nth fused data feature respectively to obtain the third data feature, the fourth data feature and the fifth data feature corresponding to the Nth fused data feature; then, the server performs multiple fusion processing on the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature to obtain the target data feature corresponding to the secondary circuit of the voltage transformer; then, the server determines the voltage drop monitoring result corresponding to the target data feature based on the target data feature as the target voltage drop monitoring result of the secondary circuit of the voltage transformer.

[0132] For example, see Figure 3 The server uses the Xception-efficient multi-scale attention model to perform one-dimensional horizontal average pooling on the Nth fused data feature to obtain the third data feature corresponding to the Nth fused data feature, performs one-dimensional vertical average pooling on the Nth fused data feature to obtain the fourth data feature corresponding to the Nth fused data feature, and performs 3*3 convolution on the Nth fused data feature to obtain the fifth data feature corresponding to the Nth fused data feature.

[0133] In this embodiment, by respectively determining the third data feature, the fourth data feature and the fifth data feature, and fusing them with the Nth fused data feature again, it is possible to mine and integrate information related to the secondary circuit of the voltage transformer from more dimensions, and thus obtain more accurate target data features, which is beneficial to improving the accuracy of determining the target voltage drop monitoring results of the secondary circuit of the voltage transformer.

[0134] In an exemplary embodiment, the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature are fused multiple times to obtain the target data feature corresponding to the secondary circuit of the voltage transformer, which specifically includes the following contents: the third data feature and the fourth data feature are fused to obtain the sixth data feature corresponding to the secondary circuit of the voltage transformer; the fifth data feature and the sixth data feature are fused to obtain the seventh data feature corresponding to the secondary circuit of the voltage transformer; the Nth fused data feature and the seventh data feature are fused to obtain the target data feature.

[0135] The sixth data feature refers to a data feature obtained by fusing the third data feature and the fourth data feature.

[0136] The seventh data feature refers to a data feature obtained by fusing the fifth data feature and the sixth data feature.

[0137] Exemplarily, the server fuses the third data feature and the fourth data feature according to the weight corresponding to the third data feature and the weight corresponding to the fourth data feature to obtain the sixth data feature corresponding to the secondary circuit of the voltage transformer; then, the server fuses the fifth data feature and the sixth data feature according to the weight corresponding to the fifth data feature and the weight corresponding to the sixth data feature to obtain the seventh data feature corresponding to the secondary circuit of the voltage transformer; then, the server fuses the Nth fused data feature and the seventh data feature according to the weight corresponding to the Nth fused data feature and the weight corresponding to the seventh data feature to obtain the target data feature.

[0138] For example, see Figure 3 , the server uses the Xception-efficient multi-scale attention model to perform 1*1 convolution processing, activation processing and re-weighting processing on the third data feature and the fourth data feature, respectively, to obtain the sixth data feature corresponding to the secondary circuit of the voltage transformer; then, the server performs average pooling processing, activation processing and matrix multiplication processing on the fifth data feature and the sixth data feature, respectively, to obtain the seventh data feature corresponding to the secondary circuit of the voltage transformer; then, the server activates the seventh data feature, and re-weights the activated seventh data feature and the Nth fused data feature to obtain the target data feature.

[0139] In this embodiment, by integrating the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature, a more comprehensive target data feature can be obtained, thereby improving the accuracy of determining the target data feature, which is beneficial to provide an accurate data basis for subsequent data processing.

[0140] In an exemplary embodiment, a target voltage drop monitoring result of a voltage transformer secondary circuit is determined based on target data characteristics, specifically including the following contents: based on the target data characteristics, a plurality of preset voltage drop monitoring results of the voltage transformer secondary circuit and a predicted probability of each preset voltage drop monitoring result are obtained; from each preset voltage drop monitoring result, the preset voltage drop monitoring result with the largest predicted probability is screened out as the target voltage drop monitoring result.

[0141] The preset voltage drop monitoring result refers to a preset voltage drop monitoring result, including the presence of abnormal voltage drop in the secondary circuit of the voltage transformer and the absence of abnormal voltage drop in the secondary circuit of the voltage transformer. It should be noted that the preset voltage drop monitoring result depends on the situation.

[0142] The predicted probability is used to indicate the possibility that the voltage drop monitoring model determines that the preset voltage drop monitoring result is correct.

[0143] Exemplarily, the server obtains multiple preset voltage drop monitoring results of the voltage transformer secondary circuit and the predicted probability of each preset voltage drop monitoring result through the trained voltage drop monitoring model according to the target data characteristics; then, the server filters out the preset voltage drop monitoring result with the largest predicted probability from each preset voltage drop monitoring result, and uses the preset voltage drop monitoring result as the target voltage drop monitoring result.

[0144] For example, see Figure 3 The server performs global average pooling processing on the target data features through the Xception-efficient multi-scale attention model to obtain the target data features after pooling processing; then, the server obtains multiple preset voltage drop monitoring results of the secondary circuit of the voltage transformer and the predicted probability of each preset voltage drop monitoring result through the fully connected layer and activation layer in the Xception-efficient multi-scale attention model according to the target data features after pooling processing, and screens out the preset voltage drop monitoring result with the largest predicted probability from each preset voltage drop monitoring result, and uses the preset voltage drop monitoring result as the target voltage drop monitoring result.

[0145] In this embodiment, by screening out the preset voltage drop monitoring result with the largest predicted probability from each preset voltage drop monitoring result as the target voltage drop monitoring result, it is equivalent to determining the voltage drop monitoring result that best matches the secondary circuit of the voltage transformer from multiple preset voltage drop monitoring results, which is beneficial to improving the accuracy of determining the target voltage drop monitoring result.

[0146] In an exemplary embodiment, the above step S107, after determining the target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature, specifically includes the following contents: when the target voltage drop monitoring result indicates that there is an abnormal voltage drop in the voltage transformer secondary circuit, extracting key associated data in the associated data; determining the abnormal result corresponding to the voltage transformer secondary circuit according to the key associated data; generating alarm information corresponding to the voltage transformer secondary circuit according to the abnormal result; and sending the alarm information to the target terminal associated with the voltage transformer secondary circuit.

[0147] Among them, key related data refers to related data whose importance is greater than the preset importance.

[0148] Among them, the abnormal result is used to indicate the reason corresponding to the abnormal voltage drop in the secondary circuit of the voltage transformer, such as poor contact at the connection point, damage or aging of the wire, etc.

[0149] The alarm information is used to indicate the information corresponding to a fault in the secondary circuit of the voltage transformer.

[0150] The target terminal refers to the terminal associated with the secondary circuit of the voltage transformer.

[0151] Exemplarily, when the target voltage drop monitoring result indicates that there is an abnormal voltage drop in the secondary circuit of the voltage transformer, the server inputs the associated data into the importance prediction model to obtain the importance of each associated data; then, the server screens out the associated data whose importance is greater than the preset importance from each associated data as the key associated data; then, the server determines the abnormal result corresponding to the key associated data based on the key associated data, as the abnormal result corresponding to the secondary circuit of the voltage transformer; then, the server generates alarm information corresponding to the abnormal result based on the abnormal result, as the alarm information corresponding to the secondary circuit of the voltage transformer; then, the server determines the terminal associated with the secondary circuit of the voltage transformer as the target terminal, and establishes a network path between the server and the target terminal; finally, the server sends the alarm information to the target terminal through the network path.

[0152] In this embodiment, when the target voltage drop monitoring result shows that there is an abnormal voltage drop, by extracting key related data from the related data, it is possible to focus on important information that is closely related to the abnormal situation, thereby making the judgment of the abnormal situation more accurate; moreover, the alarm information corresponding to the abnormal result is sent to the target terminal associated with the secondary circuit of the voltage transformer, ensuring that the alarm can be conveyed to relevant personnel in a timely manner, avoiding further deterioration of the abnormal situation, and is conducive to ensuring the safe and stable operation of the power system.

[0153] In an exemplary embodiment, the voltage drop monitoring method for the secondary circuit of a voltage transformer provided in the present application also includes a training step of a trained voltage drop monitoring model, which specifically includes the following contents: obtaining sample associated data of a sample voltage transformer secondary circuit; the sample associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the sample voltage transformer secondary circuit; preprocessing the sample associated data to obtain preprocessed sample associated data; performing feature extraction processing on the preprocessed sample associated data through the voltage drop monitoring model to be trained to obtain initial sample data features corresponding to the preprocessed sample associated data; respectively determining first sample data features and second sample data features corresponding to the initial sample data features; and extracting the first sample data features and the second sample data features from the first sample data features and the second sample data features. The features are fused to obtain fused sample data features corresponding to the preprocessed sample associated data; the fused sample data features are used as new initial sample data features, and the steps of respectively determining the first sample data features and the second sample data features corresponding to the initial sample data features are jumped to, until the obtained fused sample data features are the Nth fused sample data features corresponding to the preprocessed sample associated data; N is a positive integer greater than or equal to 2; according to the Nth fused sample data features, the predicted voltage drop monitoring result of the secondary circuit of the sample voltage transformer is determined; the actual voltage drop monitoring result of the secondary circuit of the sample voltage transformer is obtained, and according to the difference between the predicted voltage drop monitoring result and the actual voltage drop monitoring result, the voltage drop monitoring model to be trained is iteratively trained to obtain a trained voltage drop monitoring model.

[0154] The sample voltage transformer secondary circuit refers to a voltage transformer secondary circuit used for training a voltage drop monitoring model.

[0155] The sample associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the sample voltage transformer.

[0156] The preprocessed sample-related data refers to the sample-related data after preprocessing.

[0157] The initial sample data features refer to the data features corresponding to the sample associated data after preprocessing.

[0158] The first sample data feature refers to a data feature obtained by performing a 3*3 separable convolution process on the initial sample data feature.

[0159] The second sample data feature refers to a data feature obtained by performing 1*1 convolution processing on the initial sample data feature.

[0160] The fused sample data feature refers to a data feature obtained by fusing the first sample data feature and the second sample data feature.

[0161] Among them, the predicted voltage drop monitoring result refers to the predicted value corresponding to the voltage drop monitoring result of the secondary circuit of the sample voltage transformer.

[0162] Among them, the actual voltage drop monitoring result refers to the actual value corresponding to the voltage drop monitoring result of the secondary circuit of the sample voltage transformer.

[0163] Exemplarily, in response to a model training instruction for a voltage drop monitoring model to be trained, the server obtains power data, temperature data, vibration data, noise data, humidity data and weather data of a sample voltage transformer secondary circuit from a database, all of which are used as sample associated data of the sample voltage transformer secondary circuit; then, the server uses data processing technology to perform data cleaning, data expansion and data normalization and other preprocessing on the sample associated data to obtain preprocessed sample associated data as preprocessed sample associated data; then, the server inputs the preprocessed sample associated data into the voltage drop monitoring model to be trained, and performs feature extraction processing on the preprocessed sample associated data through the voltage drop monitoring model to be trained to obtain data features corresponding to the preprocessed sample associated data as initial sample data features; then, the server performs feature extraction processing on the initial sample data features respectively to obtain first sample data features and second sample data features corresponding to the initial sample data features; then, the server performs fusion processing on the first sample data features and the second sample data features to obtain fused sample data features corresponding to the preprocessed sample associated data; then, the server judges the obtained fused sample data features , and obtain a judgment result; when the judgment result indicates that the obtained fused sample data feature is not the Nth fused sample data feature, the server uses the fused sample data feature as a new initial sample data feature, and jumps to the steps of respectively determining the first sample data feature and the second sample data feature corresponding to the initial sample data feature, until the obtained fused sample data feature is the Nth fused sample data feature corresponding to the preprocessed sample associated data; then, the server determines the voltage drop monitoring result corresponding to the Nth fused sample data feature according to the Nth fused sample data feature, as the predicted voltage drop monitoring result of the secondary circuit of the sample voltage transformer; then, the server obtains the actual voltage drop monitoring result of the secondary circuit of the sample voltage transformer from the database, and obtains the loss value according to the difference between the predicted voltage drop monitoring result and the actual voltage drop monitoring result; then, the server adjusts the model parameters of the voltage drop monitoring model to be trained according to the loss value; then, the server retrains the voltage drop monitoring model after the model parameters are adjusted, until the loss value obtained by the trained voltage drop monitoring model is less than the loss value threshold, then stops the training, and uses the trained voltage drop monitoring model as the trained voltage drop monitoring model.

[0164] In this embodiment, by pre-training the voltage drop monitoring model, it is convenient to predict the target voltage drop monitoring result of the secondary circuit of the voltage transformer to be monitored after obtaining the relevant data of the secondary circuit of the voltage transformer to be monitored in practical applications; moreover, the voltage drop monitoring model receives new data in each round of iteration, and performs internal improvements and optimizations on the model, so as to facilitate more effective predictions, which is beneficial to improving the prediction accuracy of the voltage drop monitoring model.

[0165] In an exemplary embodiment, Figure 4 As shown, another voltage transformer secondary circuit voltage drop monitoring method is provided, which is described by taking the application of the method to a server as an example, and includes the following steps:

[0166] Step S401, obtaining associated data of a voltage transformer secondary circuit to be monitored through a wireless network; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the voltage transformer secondary circuit.

[0167] Step S402: pre-process the associated data to obtain pre-processed associated data.

[0168] Step S403: performing feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data.

[0169] Step S404: determine the first data feature and the second data feature corresponding to the initial data feature respectively.

[0170] Step S405: fusing the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data.

[0171] Step S406, taking the fused data feature as a new initial data feature, and jumping to the steps of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the associated data after preprocessing; N is a positive integer greater than or equal to 2.

[0172] Step S407, respectively determine the third data feature, the fourth data feature and the fifth data feature corresponding to the Nth fused data feature.

[0173] Step S408: fusing the third data feature and the fourth data feature to obtain a sixth data feature corresponding to the secondary circuit of the voltage transformer.

[0174] Step S409: fusing the fifth data feature and the sixth data feature to obtain a seventh data feature corresponding to the secondary circuit of the voltage transformer.

[0175] Step S410, fusing the Nth fused data feature and the seventh data feature to obtain a target data feature.

[0176] Step S411, obtaining a plurality of preset voltage drop monitoring results of the secondary circuit of the voltage transformer and a predicted probability of each preset voltage drop monitoring result according to the target data characteristics.

[0177] Step S412, from each preset voltage drop monitoring result, select the preset voltage drop monitoring result with the largest predicted probability as the target voltage drop monitoring result; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the secondary circuit of the voltage transformer.

[0178] In the above-mentioned voltage transformer secondary circuit voltage drop monitoring method, in the process of monitoring the voltage drop of the voltage transformer secondary circuit, various types of data of the voltage transformer secondary circuit are collected as related data through a wireless network, and through a series of processing processes such as preprocessing, model processing and multiple feature processing, the target voltage drop monitoring result of the voltage transformer secondary circuit can be accurately obtained, thereby improving the determination accuracy of the target voltage drop monitoring result of the voltage transformer secondary circuit, which is beneficial to improving the monitoring accuracy of the voltage transformer secondary circuit voltage drop; moreover, the entire process does not require manual monitoring using a portable secondary voltage drop test instrument, avoiding the method of manual monitoring using a portable secondary voltage drop test instrument, which is prone to monitoring errors due to operating errors, thereby resulting in low monitoring accuracy of the voltage transformer secondary circuit voltage drop, further improving the monitoring accuracy of the voltage transformer secondary circuit voltage drop.

[0179] In an exemplary embodiment, in order to more clearly illustrate the voltage transformer secondary circuit voltage drop monitoring method provided in the embodiment of the present application, the voltage transformer secondary circuit voltage drop monitoring method is specifically described below with a specific embodiment. In one embodiment, Figure 5As shown, the present application also provides an intelligent online monitoring method for the voltage drop of the secondary circuit of the voltage transformer. In the process of monitoring the voltage drop of the secondary circuit of the voltage transformer, the relevant data of the secondary circuit of the voltage transformer is first collected, and then the collected data is transmitted back to the background monitoring system through the WAPI wireless network. The collected data is preprocessed by data cleaning, data expansion and data normalization. Then, the initial parameters of the network are set, and the training data set is input into the preset Xception-efficient multi-scale attention model for training. The weight of the model loss function is updated using the random gradient descent method with momentum. When the number of iterations reaches the threshold, the model is evaluated and the training is completed if it meets the expected model performance. Finally, the newly collected data related to the secondary circuit of the voltage transformer is preprocessed and then input into the trained deep learning model to obtain the latest online monitoring results of the voltage drop of the secondary circuit of the voltage transformer and analyze the cause of the abnormal voltage drop based on the collected data and the obtained voltage drop results. Specifically including the following contents:

[0180] Step 1: Collect relevant data of the secondary circuit of the voltage transformer.

[0181] Among them, the specific contents of the relevant data of the voltage transformer secondary circuit are: the voltage amplitude on the voltage transformer side of the voltage transformer secondary circuit, the voltage amplitude on the instrument side, the angle difference value, the temperature data of the equipment, the vibration data, the surrounding noise data, the ambient humidity data, the real-time power flow data of the power grid, the weather data and the label data value. The label data value is specifically the calculated secondary circuit voltage drop value.

[0182] Step 2: The collected data is transmitted back to the substation backend monitoring system using the WAPI wireless network.

[0183] Step 3: Preprocess the collected data in the background monitoring system, including data cleaning, data expansion and data normalization.

[0184] Among them, data cleaning specifically means: deleting incomplete and erroneous data.

[0185] Data expansion specifically refers to increasing the amount of data at the same time point by duplicating the original data, limiting the amplitude, and scaling it proportionally.

[0186] The formula for data normalization is:

[0187] , formula (1)

[0188] in, , and They are respectively the relevant data of the secondary circuit of the voltage transformer, the maximum value of the relevant data of the secondary circuit of the voltage transformer and the minimum value of the relevant data of the secondary circuit of the voltage transformer; These are the normalized data of the secondary circuit of the voltage transformer.

[0189] Step 4: Build a deep learning model and train it on a large amount of preprocessed data.

[0190] Among them, the specific content of the deep learning model is: using Xception as the backbone feature extraction network, and introducing efficient multi-scale attention into Xception to form the Xception-efficient multi-scale attention model to improve the performance of the network.

[0191] Among them, the loss function of the model The calculation formula is:

[0192] , formula (2)

[0193] Wherein, M represents the number of relevant data samples of the secondary circuit of the voltage transformer; log() is a logarithmic function with base 10; is the label of the relevant data i of the secondary circuit of the voltage transformer; N is the number of weights; represents the L2 regularization coefficient, which is added to control the overfitting of the weights; j represents the number of weight sequences; is the weight to be updated.

[0194] Among them, the calculation formula of the stochastic gradient descent method with momentum is:

[0195] , formula (3)

[0196] Among them, the weight that needs to be updated for iteration t is :

[0197] , formula (4)

[0198] Among them, t, , and They are the number of iterations, the weights to be updated after iteration t, the momentum after iteration t, and the momentum after iteration t-1; is the learning rate; For L The gradient of is the momentum coefficient; The weights that need to be updated for iteration t-1.

[0199] Figure 2 It is a structural diagram of the Xception-efficient multi-scale attention model in this embodiment. The specific content of the Xception-efficient multi-scale attention model is:

[0200] Input data;

[0201] Connect to the convolution of 32 3×3 convolution kernels, ReLU function, convolution of 32 3×3 convolution kernels and ReLU function processing in sequence;

[0202] Then access 2 branches:

[0203] Branch 1 is connected to 128 separable convolutions of 3×3 convolution kernels, ReLU function, 128 separable convolutions of 3×3 convolution kernels and 3×3 maximum pooling processing in sequence;

[0204] Branch 2 is connected to the convolution processing of 1×1 convolution kernel;

[0205] Then merge branch 1 and branch 2;

[0206] Then connect to 2 branches:

[0207] Branch 1 is connected to the ReLU function, the separable convolution of 256 3×3 convolution kernels, the ReLU function, the separable convolution of 256 3×3 convolution kernels and the 3×3 maximum pooling process in sequence;

[0208] Branch 2 is connected to the convolution processing of 1×1 convolution kernel;

[0209] Then merge branch 1 and branch 2;

[0210] Then connect to 2 branches:

[0211] Branch 1 is connected to the ReLU function, the separable convolution of 728 3×3 convolution kernels, the ReLU function, the separable convolution of 728 3×3 convolution kernels and the 3×3 maximum pooling process in sequence;

[0212] Branch 2 is connected to the convolution processing of 1×1 convolution kernel;

[0213] Then merge branch 1 and branch 2;

[0214] Then connect to 2 branches:

[0215] Branch 1 is connected in sequence to the ReLU function, the separable convolution of 728 3×3 convolution kernels, the ReLU function, the separable convolution of 728 3×3 convolution kernels, the ReLU function, and the separable convolution of 728 3×3 convolution kernels;

[0216] Branch 2 is connected to the convolution of the 1×1 convolution kernel and performs 8 processing;

[0217] Then merge branch 1 and branch 2;

[0218] Then connect the ReLU function, separable convolution with 256 3×3 convolution kernels, ReLU function, separable convolution with 256 3×3 convolution kernels, 3×3 maximum pooling processing, ReLU function, separable convolution with 256 3×3 convolution kernels, ReLU function, separable convolution with 256 3×3 convolution kernels, efficient multi-scale attention module, global average pooling, full connection and Softmax function.

[0219] Finally, output the results.

[0220] Figure 3 : is a structural diagram of the efficient multi-scale attention module in this embodiment. The specific contents of the efficient multi-scale attention module are:

[0221] Input features;

[0222] Then connect a group convolution, and then connect 2 branches:

[0223] Branch 1 outputs directly;

[0224] Branch 2 is connected to 2 branches:

[0225] Branch 2 first accesses the convolution of the 3×3 convolution kernel;

[0226] Branch 1 is connected to 2 branches:

[0227] Branch 1 is connected to 1D horizontal average pooling;

[0228] Branch 2 is connected to 1D longitudinal average pooling;

[0229] Then merge branch 1 and branch 2, connect to the convolution of 1×1 convolution kernel, and then connect to 2 branches:

[0230] Branch 1 is connected to the Sigmoid function;

[0231] Branch 2 is connected to the Sigmoid function;

[0232] Then merge branch 1 and branch 2, and successively connect reweighting, group normalization, average pooling, and Softmax function;

[0233] Then, the output of the Softmax function of branch 1 and the output of the convolution of branch 2 with the 3×3 convolution kernel are merged and connected to the Matmul function of branch 1.

[0234] Branch 2 is then connected to average pooling, Softmax function, and Matmul function, and the output of the Softmax function of branch 1 is merged and connected to the Matmul function of branch 2;

[0235] Then merge branch 1 and branch 2 and connect to the Sigmoid function;

[0236] Then merge branch 1 and branch 2 and re-weight them;

[0237] Output features.

[0238] In the above embodiment, in the process of monitoring the voltage drop of the secondary circuit of the voltage transformer, various types of data of the secondary circuit of the voltage transformer are collected as related data through the wireless network, and through a series of processing processes such as preprocessing, model processing and multiple feature processing, the target voltage drop monitoring result of the secondary circuit of the voltage transformer can be accurately obtained, thereby improving the determination accuracy of the target voltage drop monitoring result of the secondary circuit of the voltage transformer, which is beneficial to improving the monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer; moreover, the entire process does not need to be monitored manually using a portable secondary voltage drop test instrument, avoiding the method of using a portable secondary voltage drop test instrument for manual monitoring, which is prone to monitoring errors due to operating errors, thereby resulting in low monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer, further improving the monitoring accuracy of the voltage drop of the secondary circuit of the voltage transformer. At the same time, WAPI wireless network data transmission is used instead of traditional wired network data transmission. Since there is no need for complex wired wiring, the use of existing WAPI wireless network data transmission can reduce the wiring cost of the monitoring system and save the cost of equipment installation and maintenance; WAPI wireless network data transmission has high transmission speed and stability, eliminates the limitations of traditional wired connections, improves the flexibility and convenience of the monitoring system, and makes the installation and wiring of the monitoring device simpler and faster; this method can simulate various abnormal conditions in the secondary circuit of the voltage transformer for recording, and construct a deep learning network model for training, which can truly realize more accurate real-time detection of the voltage drop in the secondary circuit of the voltage transformer, find out the cause of the abnormal voltage drop in a more timely manner and issue an early warning, and solve the existing monitoring The device has limited data processing capabilities and cannot achieve intelligent analysis and learning of data. This method uses Xception as the backbone feature extraction network and introduces an efficient multi-scale attention module. Xception uses separable convolution instead of traditional convolution operations to learn features more effectively and achieve better performance at the same computational complexity. The efficient multi-scale attention module focuses on retaining the information of each channel and reducing computational overhead, groups the channel dimensions into multiple sub-features, and evenly distributes spatial semantic features in each feature group. Due to its flexible and lightweight characteristics, the efficient multi-scale attention module can be easily used in Xception, and without changing the network depth, the efficient multi-scale attention module is superior to several attention mechanisms in recent years.

[0239] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0240] Based on the same inventive concept, the embodiment of the present application also provides a voltage transformer secondary circuit voltage drop monitoring device for implementing the voltage transformer secondary circuit voltage drop monitoring method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more voltage transformer secondary circuit voltage drop monitoring device embodiments provided below can refer to the limitations of the voltage transformer secondary circuit voltage drop monitoring method above, and will not be repeated here.

[0241] In an exemplary embodiment, Figure 6 As shown, a voltage transformer secondary circuit voltage drop monitoring device is provided, including: a data acquisition module 601, a data processing module 602, a feature extraction module 603, a feature determination module 604, a feature fusion module 605, a feature processing module 606 and a result determination module 607, wherein:

[0242] The data acquisition module 601 is used to acquire the associated data of the secondary circuit of the voltage transformer to be monitored through the wireless network; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer.

[0243] The data processing module 602 is used to pre-process the associated data to obtain pre-processed associated data.

[0244] The feature extraction module 603 is used to perform feature extraction processing on the pre-processed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the pre-processed associated data.

[0245] The feature determination module 604 is used to respectively determine the first data feature and the second data feature corresponding to the initial data feature.

[0246] The feature fusion module 605 is used to fuse the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data.

[0247] The feature processing module 606 is used to use the fused data feature as a new initial data feature and jump to the steps of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the associated data after preprocessing; N is a positive integer greater than or equal to 2.

[0248] The result determination module 607 is used to determine the target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

[0249] In an exemplary embodiment, the result determination module 607 is also used to respectively determine the third data feature, the fourth data feature and the fifth data feature corresponding to the Nth fused data feature; perform multiple fusion processes on the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature to obtain the target data feature corresponding to the secondary circuit of the voltage transformer; and determine the target voltage drop monitoring result of the secondary circuit of the voltage transformer based on the target data feature.

[0250] In an exemplary embodiment, the result determination module 607 is also used to fuse the third data feature and the fourth data feature to obtain the sixth data feature corresponding to the secondary circuit of the voltage transformer; fuse the fifth data feature and the sixth data feature to obtain the seventh data feature corresponding to the secondary circuit of the voltage transformer; and fuse the Nth fused data feature and the seventh data feature to obtain the target data feature.

[0251] In an exemplary embodiment, the result determination module 607 is also used to obtain multiple preset voltage drop monitoring results of the voltage transformer secondary circuit and the predicted probability of each preset voltage drop monitoring result based on the target data characteristics; from each preset voltage drop monitoring result, the preset voltage drop monitoring result with the largest predicted probability is screened out as the target voltage drop monitoring result.

[0252] In an exemplary embodiment, the voltage transformer secondary circuit voltage drop monitoring device also includes an information sending module, which is used to extract key related data in the related data when the target voltage drop monitoring result indicates that there is an abnormal voltage drop in the voltage transformer secondary circuit; determine the abnormal result corresponding to the voltage transformer secondary circuit based on the key related data; generate alarm information corresponding to the voltage transformer secondary circuit based on the abnormal result; and send the alarm information to the target terminal associated with the voltage transformer secondary circuit.

[0253] In an exemplary embodiment, the voltage transformer secondary circuit voltage drop monitoring device also includes a signal model training module, which is used to obtain sample associated data of the sample voltage transformer secondary circuit; the sample associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the sample voltage transformer secondary circuit; the sample associated data is preprocessed to obtain preprocessed sample associated data; through the voltage drop monitoring model to be trained, feature extraction processing is performed on the preprocessed sample associated data to obtain initial sample data features corresponding to the preprocessed sample associated data; first sample data features and second sample data features corresponding to the initial sample data features are respectively determined; the first sample data features and the second sample data features are fused to obtain to the fused sample data feature corresponding to the preprocessed sample associated data; taking the fused sample data feature as the new initial sample data feature, and jumping to the steps of respectively determining the first sample data feature and the second sample data feature corresponding to the initial sample data feature, until the obtained fused sample data feature is the Nth fused sample data feature corresponding to the preprocessed sample associated data; N is a positive integer greater than or equal to 2; according to the Nth fused sample data feature, determining the predicted voltage drop monitoring result of the secondary circuit of the sample voltage transformer; obtaining the actual voltage drop monitoring result of the secondary circuit of the sample voltage transformer, and according to the difference between the predicted voltage drop monitoring result and the actual voltage drop monitoring result, iteratively training the voltage drop monitoring model to be trained to obtain a trained voltage drop monitoring model.

[0254] Each module in the voltage transformer secondary circuit voltage drop monitoring device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0255] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 7As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store power data, temperature data, etc. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a voltage transformer secondary circuit voltage drop monitoring method is implemented.

[0256] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0257] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0258] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0259] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0260] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0261] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0262] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for monitoring voltage drop in a secondary circuit of a voltage transformer, characterized in that: The method comprises: Acquire, via a wireless network, associated data of a secondary circuit of a voltage transformer to be monitored; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer; Preprocessing the associated data to obtain preprocessed associated data; Performing feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data; Respectively determine a first data feature and a second data feature corresponding to the initial data feature; Performing fusion processing on the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data; The fused data feature is used as a new initial data feature, and the process jumps to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data; N is a positive integer greater than or equal to 2; A target voltage drop monitoring result of the voltage transformer secondary circuit is determined according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

2. The method according to claim 1, characterized in that Determining the target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature includes: Respectively determining a third data feature, a fourth data feature, and a fifth data feature corresponding to the Nth fused data feature; Performing multiple fusion processing on the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature to obtain a target data feature corresponding to the secondary circuit of the voltage transformer; According to the target data characteristics, a target voltage drop monitoring result of the voltage transformer secondary circuit is determined.

3. The method according to claim 2, characterized in that The performing multiple fusion processing on the Nth fused data feature, the third data feature, the fourth data feature and the fifth data feature to obtain the target data feature corresponding to the secondary circuit of the voltage transformer includes: The third data feature and the fourth data feature are merged to obtain a sixth data feature corresponding to the secondary circuit of the voltage transformer; The fifth data feature and the sixth data feature are merged to obtain a seventh data feature corresponding to the secondary circuit of the voltage transformer; The Nth fused data feature and the seventh data feature are fused to obtain the target data feature.

4. The method according to claim 2, characterized in that: Determining the target voltage drop monitoring result of the voltage transformer secondary circuit according to the target data feature includes: According to the target data characteristics, a plurality of preset voltage drop monitoring results of the secondary circuit of the voltage transformer and a predicted probability of each preset voltage drop monitoring result are obtained; From each of the preset pressure drop monitoring results, the preset pressure drop monitoring result with the maximum predicted probability is screened out as the target pressure drop monitoring result.

5. The method according to claim 1, characterized in that After determining the target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature, the method further includes: When the target voltage drop monitoring result indicates that an abnormal voltage drop exists in the secondary circuit of the voltage transformer, extracting key associated data from the associated data; Determining the abnormal result corresponding to the secondary circuit of the voltage transformer according to the key associated data; Generate alarm information corresponding to the secondary circuit of the voltage transformer according to the abnormal result; The alarm information is sent to a target terminal associated with the secondary circuit of the voltage transformer.

6. The method according to any one of claims 1 to 5, characterized in that: The trained pressure drop monitoring model is obtained by training in the following manner: Acquire sample associated data of a sample voltage transformer secondary circuit; the sample associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the sample voltage transformer secondary circuit; Preprocessing the sample-related data to obtain preprocessed sample-related data; Performing feature extraction processing on the preprocessed sample associated data through the pressure drop monitoring model to be trained to obtain initial sample data features corresponding to the preprocessed sample associated data; Respectively determine a first sample data feature and a second sample data feature corresponding to the initial sample data feature; Performing fusion processing on the first sample data feature and the second sample data feature to obtain a fused sample data feature corresponding to the preprocessed sample associated data; The fused sample data feature is used as a new initial sample data feature, and the process jumps to the step of respectively determining the first sample data feature and the second sample data feature corresponding to the initial sample data feature, until the obtained fused sample data feature is the Nth fused sample data feature corresponding to the preprocessed sample associated data; N is a positive integer greater than or equal to 2; Determining a predicted voltage drop monitoring result of the secondary circuit of the sample voltage transformer according to the Nth fused sample data feature; The actual voltage drop monitoring result of the secondary circuit of the sample voltage transformer is obtained, and according to the difference between the predicted voltage drop monitoring result and the actual voltage drop monitoring result, the voltage drop monitoring model to be trained is iteratively trained to obtain the trained voltage drop monitoring model.

7. A voltage transformer secondary circuit voltage drop monitoring device, characterized in that: The device comprises: A data acquisition module, used to acquire the associated data of the secondary circuit of the voltage transformer to be monitored through a wireless network; the associated data at least includes power data, temperature data, vibration data, noise data, humidity data and weather data of the secondary circuit of the voltage transformer; A data processing module, used for preprocessing the associated data to obtain preprocessed associated data; A feature extraction module, used to perform feature extraction processing on the preprocessed associated data through the trained voltage drop monitoring model to obtain initial data features corresponding to the preprocessed associated data; A feature determination module, used to respectively determine a first data feature and a second data feature corresponding to the initial data feature; A feature fusion module, used for fusing the first data feature and the second data feature to obtain a fused data feature corresponding to the pre-processed associated data; A feature processing module, used for taking the fused data feature as a new initial data feature, and jumping to the step of respectively determining the first data feature and the second data feature corresponding to the initial data feature, until the obtained fused data feature is the Nth fused data feature corresponding to the preprocessed associated data; N is a positive integer greater than or equal to 2; A result determination module is used to determine a target voltage drop monitoring result of the voltage transformer secondary circuit according to the Nth fused data feature; the target voltage drop monitoring result is used to indicate whether there is an abnormal voltage drop in the voltage transformer secondary circuit.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images