Risk situation prediction method and related device for low-voltage power utilization area

Abstract

The invention provides a risk situation prediction method of a low-voltage power supply area and a related device. The method comprises the following steps: acquiring historical data of the low-voltage power supply area; grouping the historical data according to a first grouping rule to obtain a plurality of first training data groups; grouping the plurality of first training data groups according to a second grouping rule to obtain a plurality of first training subsets; preprocessing data in the plurality of first training subsets to obtain a target data set conforming to a data model input rule; dividing the target data set into a target to-be-trained data set and a target test set; constructing a to-be-trained model, and training the to-be-trained model according to the target to-be-trained data set to obtain a trained risk situation prediction model; and inputting the real-time operation data and the real-time environment data of the first time interval of the low-voltage power station area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage power station area.

Description

Technical Field

[0001] The present application belongs to the field of Internet technology, and specifically relates to a risk situation prediction method and related devices for a low-voltage power supply area. Background Art

[0002] At present, the existing technology generally adopts offline analysis for risk prediction of low-voltage power supply areas, that is, collecting data of low-voltage power supply areas, transferring the collected data to an offline platform, and analyzing the risks of low-voltage power supply areas on the offline platform.

[0003] It can be seen that the methods of the prior art are unable to dynamically monitor and predict the real-time operating status of the low-voltage power supply area. Summary of the invention

[0004] The present application provides a method and related device for predicting the risk situation of a low-voltage power supply area, in order to achieve real-time monitoring and prediction of the risk situation of a low-voltage power supply area.

[0005] In a first aspect, the present application provides a method for predicting risk situation in a low-voltage power supply area, comprising: Acquire historical data of the low-voltage power supply area; wherein the historical data includes a plurality of operation data, a plurality of environmental data and a plurality of risk data; The historical data are grouped according to a first grouping rule to obtain a plurality of first training data groups; the first grouping rule is to group the operation data, environment data and risk data at the same time into one group; each first training data group includes a plurality of input parameters and output parameters, wherein the input parameters include the operation data and the environment data, and the output parameters include the risk data; The plurality of first training data groups are grouped according to a second grouping rule to obtain a plurality of first training subsets; the second grouping rule is to divide the first training data groups in the same time interval into the same first training subset; Preprocessing the data in the plurality of first training subsets to obtain a target data set that complies with the data model input rule; Dividing the target data set into a target training data set and a target test set; Constructing a model to be trained, and training the model to be trained according to the target training data set and the target test set to obtain a trained risk situation prediction model; The real-time operation data and real-time environmental data of the first time interval of the low-voltage radio station area are input into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio station area; the second time interval is the next time interval adjacent to the first time interval.

[0006] In combination with the first aspect, in a possible embodiment, the data in the multiple first training subsets are preprocessed to obtain a target data set to be trained that conforms to the data model input rules, including: completing and normalizing the movement data of the first training data group in each first training subset to obtain multiple second training subsets; each second training subset includes multiple second training data groups after data completion and normalization; feature extraction is performed on the second training data group in each second training subset to obtain the target data set to be trained; the target data set to be trained includes multiple third training subsets, and each third training subset includes multiple third training data groups obtained after feature extraction.

[0007] In combination with the first aspect, in a possible embodiment, the movement data of the first training data group in each first training subset is completed and normalized respectively to obtain multiple second training subsets, including: determining the location where data is missing in each first training data group; completing the data at each location where data is missing by interpolation; scaling the data of each first training data group after data completion to a target range to obtain multiple second training data groups.

[0008] In combination with the first aspect, in a possible embodiment, the feature extraction of the second training data group in each second training subset to obtain the target data set to be trained includes: calculating the current fluctuation rate, voltage deviation rate, maximum average load rate and maximum distribution transformer average three-phase imbalance of each second training subset; determining the maximum ambient temperature and maximum ambient humidity of each second training subset to obtain the target data set to be trained.

[0009] In combination with the first aspect, in a possible embodiment, the constructing of the model to be trained includes: constructing the model to be trained based on a radial basis neural network; the model to be trained includes an input layer, a processing layer and an output layer; the input layer is used to receive a third training subset in the target data set to be trained; the processing layer is used to map the input third training subset to a high-dimensional space to obtain a feature label matrix; the output layer is used to process the feature label matrix to obtain the risk type of the next time interval of the low-voltage radio station area.

[0010] In combination with the first aspect, in a possible embodiment, the model to be trained is trained according to the target data set to be trained and the target test set to obtain a trained risk situation prediction model, including: inputting each third training subset in the target data set to be trained into the model to be trained in turn; the model to be trained processes the data in the input third training subset to obtain a first risk type; comparing the first risk type with the target risk type in the target test set; if accurate, completing the training of the model to be trained to obtain the risk situation prediction model; if inaccurate, optimizing the model to be trained by the Bayesian optimization method, and continuing to train the optimized model to be trained by the target data set to be trained, and repeating the above operations until the data results of the model are accurate, thereby obtaining the risk situation prediction model.

[0011] In combination with the first aspect, in a possible embodiment, the real-time operation data and real-time environmental data of the low-voltage power supply area are input into the risk situation prediction model to obtain the risk type of the next time interval of the low-voltage power supply area, including: collecting the real-time operation data and real-time environmental data of the low-voltage power supply area in the first time window; the real-time operation data include the operating voltage, current, average load rate and average three-phase imbalance of the distribution transformer; the real-time environmental data include ambient temperature and ambient humidity; preprocessing the target data to obtain the data to be processed that meets the input rules of the risk situation prediction model; using the risk situation prediction model to determine the target risk type of the low-voltage power supply area in the second time window according to the data to be processed; wherein the risk situation prediction model is obtained by training for multiple consecutive time windows based on a target training set, the time length of each time window is the same, the multiple time windows include the first time window and the second time window, and the second time window is the next time window of the first time window; outputting target warning information according to the target risk type.

[0012] In a second aspect, the present application provides a risk situation prediction device for a low-voltage power supply area, comprising: An acquisition unit, used for acquiring historical data of a low-voltage power supply area; wherein the historical data includes a plurality of operation data, a plurality of environmental data and a plurality of risk data; a grouping unit, configured to group the historical data according to a first grouping rule to obtain a plurality of first training data groups; the first grouping rule is to group the operation data, environment data and risk data at the same time into one group; each first training data group includes a plurality of input parameters and output parameters, wherein the input parameters include the operation data and the environment data, and the output parameters include the risk data; The grouping unit is further used to group the plurality of first training data groups according to a second grouping rule to obtain a plurality of first training subsets; the second grouping rule is to divide the first training data groups in the same time interval into the same first training subset; A preprocessing unit, configured to preprocess the data in the plurality of first training subsets to obtain a target data set that meets the data model input rule; and divide the target data set into a target to-be-trained data set and a target test set; A model building unit, used to build a model to be trained, and train the model to be trained according to the target training data set and the target test set to obtain a trained risk situation prediction model; A prediction unit is used to input the real-time operation data and real-time environmental data of the first time interval of the low-voltage radio station area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio station area; the second time interval is the next time interval adjacent to the first time interval.

[0013] In a third aspect, the present application provides an electronic device comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program comprises instructions for executing the steps of any one of the first and second aspects of the present application.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps described in any one of the first and second aspects of the present application.

[0015] In a fifth aspect, the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in any one of the first and second aspects of the present application. The computer program product may be a software installation package.

[0016] It can be seen that in the present application, by acquiring the historical data of the low-voltage radio area; then grouping the historical data according to the first grouping rule to obtain multiple first training data groups; then grouping the multiple first training data groups according to the second grouping rule to obtain multiple first training subsets; then preprocessing the data in the multiple first training subsets to obtain a target data set that meets the data model input rules; dividing the target data set into a target data set to be trained and a target test set; then constructing a model to be trained, training the model to be trained according to the target data set to be trained, and obtaining a trained risk situation prediction model; finally, inputting the real-time operation data and real-time environmental data of the first time interval of the low-voltage radio area into the risk situation prediction model, and obtaining the risk type of the second time interval of the low-voltage radio area. In this way, the real-time operation data and real-time environmental data of the low-voltage radio area are collected in real time, input into the trained risk situation prediction model, and the risk type of the next time interval of the low-voltage radio area is predicted, thereby realizing real-time monitoring and prediction of the risk situation of the low-voltage radio area. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art 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 drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a structural diagram of the first risk situation prediction system provided in the embodiment of the present application; Figure 2 It is a structural diagram of a second risk situation prediction system provided in an embodiment of the present application; Figure 3 It is a flow chart of a risk situation prediction method for a low-voltage power supply area provided in an embodiment of the present application; Figure 4 It is a schematic diagram of a risk situation prediction model and training process provided in an embodiment of the present application; Figure 5 It is a structural schematic diagram of a risk situation prediction device for a low-voltage power supply area provided in an embodiment of the present application; Figure 6 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, systems, products or devices.

[0021] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0022] The following is an introduction to the relevant terms involved in this application.

[0023] The RBF model, or the Radial Basis Function model, is a commonly used neural network model. The RBF network is a feedforward neural network consisting of an input layer, a hidden layer, and an output layer. The basic idea is to achieve nonlinear modeling of the input data by mapping the input data to a high-dimensional feature space and then performing linear combinations in this space. The neurons in the hidden layer use radial basis functions as activation functions. Common radial basis functions include Gaussian functions and polynomial functions. These functions are centered on a point in the input space and decrease monotonically as the distance from the center increases, thereby forming a localized response area in the input space.

[0024] The low-voltage user area refers to the power supply area from the secondary side of the distribution transformer to the user's meter box. It includes distribution transformers, low-voltage distribution cabinets, metering devices, low-voltage lines, and user terminal equipment. It is the last link of power supply and directly provides electricity to various users.

[0025] At present, the existing technology generally adopts offline analysis in the risk prediction of low-voltage power supply areas, that is, collecting data of low-voltage power supply areas, transferring the collected data to an offline platform, and analyzing the risks of low-voltage power supply areas on the offline platform. It can be seen that the existing technology method cannot dynamically monitor and predict the real-time operating status of low-voltage power supply areas.

[0026] To solve the above problems, an embodiment of the present application provides a risk situation prediction method for a low-voltage power supply area. The risk situation prediction method for a low-voltage power supply area can be applied to the scenario of risk prediction for a low-voltage power supply area. It can be done by acquiring historical data of the low-voltage power supply area; then grouping the historical data according to a first grouping rule to obtain multiple first training data groups; then grouping the multiple first training data groups according to a second grouping rule to obtain multiple first training subsets; then preprocessing the data in the multiple first training subsets to obtain a target data set that meets the data model input rules; dividing the target data set into a target data set to be trained and a target test set; then constructing a model to be trained, and training the model to be trained according to the target data set to obtain a trained risk situation prediction model; finally, inputting the real-time operation data and real-time environmental data of the first time interval of the low-voltage power supply area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage power supply area. In this way, the real-time operation data and real-time environmental data of the low-voltage power supply area are collected in real time, input into the trained risk situation prediction model, and the risk type of the low-voltage power supply area in the next time interval is predicted, thus realizing the real-time monitoring and prediction of the risk situation of the low-voltage power supply area. This solution can be applied to a variety of scenarios, including but not limited to the application scenarios mentioned above.

[0027] The following introduces the system architecture involved in the embodiments of the present application.

[0028] See also Figure 1The present application provides a risk situation prediction system 20, which includes a data acquisition module 21, a data processing module 22, a model training and optimization module 23, a model building module 24 and a risk situation prediction module 25. The data acquisition module 21 is used to obtain historical data of the low-voltage power supply area, and the data preprocessing module is used to group the historical data according to a first grouping rule to obtain multiple first training data groups; the first grouping rule refers to dividing the operating data, environmental data and risk data at the same time into a group; each first training data group includes multiple input parameters and output parameters, wherein the input parameters include the operating data and the environmental data, and the output parameters include the risk data; the multiple first training data groups are grouped according to a second grouping rule to obtain multiple first training subsets; the second grouping rule refers to dividing the first training data groups in the same time interval into the same first training training subset; the data processing module 22 is used to preprocess the data in the multiple first training subsets to obtain a target data set that meets the data model input rules; the target data set is divided into a target data set to be trained and a target test set; the model construction module 24 is used to construct a model to be trained, and the model training and optimization module 23 is used to train the model to be trained according to the target data set to be trained and the target test set to obtain a trained risk situation prediction model; the risk situation prediction module 25 is used to input the real-time operation data and real-time environmental data of the first time interval of the low-voltage radio station area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio station area.

[0029] For further information, please also refer to Figure 2 The data processing module 22 includes a data preprocessing module 221 and a feature extraction module 222, wherein the data preprocessing module 221 is used to complete and normalize the movement data of the first training data group in each first training subset, respectively, to obtain multiple second training subsets; each second training subset includes multiple second training data groups after data completion and normalization; the feature extraction module 222 is used to extract features from the second training data group in each second training subset, to obtain the target data set to be trained; the target data set to be trained includes multiple third training subsets, and each third training subset includes multiple third training data groups obtained after feature extraction.

[0030] In this way, the real-time operation data and real-time environmental data of the low-voltage power supply area are collected in real time and input into the trained risk situation prediction model to predict the risk type of the low-voltage power supply area in the next time period, thus realizing real-time monitoring and prediction of the risk situation of the low-voltage power supply area.

[0031] The specific methods are introduced in detail below.

[0032] See also Figure 3 and Figure 4 The present application also provides a risk situation prediction method for a low voltage power supply area, which can be applied to Figure 3 The risk situation prediction system shown in the figure, the risk situation prediction method of the low voltage radio area includes: Step S201, obtaining historical data of the low-voltage radio area.

[0033] The historical data includes a plurality of operation data, a plurality of environment data and a plurality of risk data.

[0034] Specifically, in this embodiment, the operation data, environmental data and risk data of the low-voltage power consumption area can be monitored in real time, and the monitored operation data, environmental data and risk data can be stored. The low-voltage power consumption monitoring system can store historical data of the low-voltage power consumption area for a certain period of time.

[0035] The historical operation data, environmental data and risk data of the low-voltage power consumption area are collected from the low-voltage power consumption monitoring system. The operation data includes the operating voltage U 0 (t), operating current I 0 (t), average load factor of distribution transformer P 0 (t), average three-phase imbalance degree of distribution transformer S 0 (t); Environmental data include the ambient temperature T of the low voltage radio area 0 (t) and ambient humidity W 0 (t); the risk data includes information data on whether a fault occurs at the corresponding moment, that is, the risk type label b(t) of the low-voltage power supply area at moment t. When b(t) is 0, it indicates normal; when b(t) is 1, it indicates overload; when b(t) is 2, it indicates voltage over limit; when b(t) is 3, it indicates short circuit; and when b(t) is 4, it indicates equipment failure.

[0036]

[0037] Step S202: group the historical data according to a first grouping rule to obtain a plurality of first training data groups.

[0038] Among them, the first grouping rule refers to dividing the operating data, environmental data and risk data at the same time into a group; each first training data group includes multiple input parameters and output parameters, the input parameters include the operating data and the environmental data, and the output parameters include the risk data.

[0039] Specifically, when storing historical data, a timestamp is associated with the historical data to identify the time when the data was generated. The operation data, environmental data, and risk data are grouped based on the timestamp to obtain a first training data group corresponding to each moment. Each first training data group is a training data, and the operation data and environmental data in each first training data group are used as input parameters during training, and the risk data is used as an output parameter during training, and the output parameter is used to verify the training result.

[0040] Step S203: group the multiple first training data groups according to a second grouping rule to obtain multiple first training subsets.

[0041] The second grouping rule refers to dividing the first training data group in the same time interval into the same first training subset; Specifically, taking the time length as the scale, the total length of the historical data collected for model training is T, and the time length T is divided into K windows (1, 2, ..., k, ..., K), and the length of each window is recorded as n. In other words, the time length T is divided into multiple time intervals, and any time scale in each time interval corresponds to a first training data group. Each time interval includes multiple first training data groups to form a first training subset. Each first training subset can train the model to be trained once.

[0042] Step S204: pre-process the data in the multiple first training subsets to obtain a target training data set that meets the data model input rules.

[0043] Specifically, before model training, the training data needs to be preprocessed so that the training data can conform to the data input format of the model to be trained. At the same time, the collected data is cleaned, denoised and normalized to improve data quality.

[0044] In a possible embodiment, the data in the multiple first training subsets are preprocessed to obtain a target data set to be trained that meets the data model input rules, including: performing data completion and normalization processing on the first training data group in each first training subset to obtain multiple second training subsets; each second training subset includes multiple second training data groups after data completion and normalization processing; performing feature extraction on the second training data group in each second training subset to obtain the target data set to be trained; the target data set to be trained includes multiple third training subsets, and each third training subset includes multiple third training data groups obtained after feature extraction.

[0045] In a specific implementation, due to various reasons (such as data loss during transmission, or collection failure due to a fault, etc.), the collected historical data may be missing. For example, a certain operation data is missing in a certain first training data group, but the environmental data and risk data are collected normally. In this embodiment, the missing data in each first training data group is supplemented by data completion, so that the data volume of each first training data group is consistent.

[0046] In addition, since the values ​​of different data of each similar type of data are different, this embodiment normalizes the data for data consistency, thereby improving data quality and improving the accuracy of model training.

[0047] In a possible embodiment, the data of the first training data group in each first training subset is respectively completed and normalized to obtain multiple second training subsets, including: determining the location where data is missing in each first training data group; completing the data at each location where data is missing by interpolation; and scaling the data of each first training data group after data completion to a target range to obtain multiple second training data groups.

[0048] Specifically, in this embodiment, in the case of missing data, an interpolation method is used to complete the missing data. Taking a first training data set as an example, for the first training data set, the positions where the data is missing are detected, and then interpolation is performed on those positions to complete the data. For example, if it is detected that the voltage data at time t in the first training data set is missing, the interpolation formula is used. Calculate the operating voltage U at this moment 0 (t), operating current I 0 (t), average load factor of distribution transformer P 0 (t), average three-phase imbalance degree of distribution transformer S 0 (t), ambient temperature T 0 (t) and ambient humidity W 0 (t), added to the first training group. For example, the operating voltage The interpolation formula is as follows:

[0049] in, represents the operating voltage at the moment before time t, The monitoring time interval may be 1 minute, 5 minutes, 10 minutes, etc., and is not limited here.

[0050] Based on operating voltage The interpolation formula is used to obtain the operating voltage at the moment before time t. , the operating voltage at the moment before time t Substituting into the operating voltage interpolation formula, the operating voltage at time t can be obtained. .

[0051] It is understandable that if the operating current I 0 (t), average load factor of distribution transformer P 0 (t), average three-phase unbalance degree of distribution transformer S 0 (t), ambient temperature T 0 (t) or ambient humidity W 0 If there are missing values ​​in the data of (t), the same processing can be done to calculate the corresponding missing values.

[0052] Furthermore, in this embodiment, the first training data group after data completion is normalized by data scaling. Specifically, each data in the first training data group is scaled to a target range, which is preferably between 0 and 1. In this range, the model can process data more easily. The data scaling formula is: .

[0053] Using the operating voltage in the first training data set For example, the operating voltage The data scaling formula is as follows:

[0054] in, is the normalized (i.e. data scaled) voltage data, Indicates taking the minimum value among the voltage data in the first training data set, It means taking the maximum value among the voltage data in the first training data group, and calculating the voltage within the target range through the voltage data scaling formula.

[0055] Based on operating voltage The data scaling formula is used to obtain the operating voltage at time t. , the minimum value of voltage data in the first training data set , and the minimum value of the voltage data in the first training data set ; Then substitute the acquired data into the operating voltage interpolation formula to obtain the normalized (i.e. data scaled) voltage data .

[0056] It is understandable that for the operating current I 0 (t), average load factor of distribution transformer P 0 (t), average three-phase unbalance degree of distribution transformer S 0 (t), ambient temperature T 0 (t) or ambient humidity W0 (t) data scaling processing, and voltage After normalization, we get current I1(t), average load rate of distribution transformer P1(t), average three-phase imbalance degree of distribution transformer S1(t), ambient temperature T1(t), and ambient humidity W1(t). Normalization process and voltage The same, no further elaboration here.

[0057] In a possible embodiment, the feature extraction of the training data group in each second training subset to obtain the target data set to be trained includes: calculating the current fluctuation rate, voltage deviation rate, maximum average load rate and maximum distribution transformer average three-phase imbalance of each second training subset; determining the maximum ambient temperature and maximum ambient humidity of each second training subset to obtain the target data set to be trained.

[0058] Specifically, there are multiple first training data sets in each time interval, that is, there are multiple operating voltages , multiple operating currents I 0 (t), average load factor of multiple distribution transformers P 0 (t), average three-phase imbalance degree S of multiple distribution transformers 0 (t), multiple ambient temperatures T 0 (t) and multiple ambient humidity W 0 (t).

[0059] Since it is a risk prediction for a time interval, these data need to be processed to obtain optimized data before the corresponding model to be trained can be trained.

[0060] The calculation process for each data is as follows: (1) Calculate the corresponding voltage deviation rate based on multiple operating voltages The voltage deviation rate is used to reflect the stability of the degree of deviation between the voltage and the rated value. The calculation formula is as follows:

[0061] in, is the standard deviation of the voltage, reflecting the degree of voltage dispersion; is the average value of the voltage, reflecting the overall level of the voltage.

[0062]

[0063] is the current data after missing value filling and normalization, and n is the window length of the collected historical data.

[0064] (2) Calculate the corresponding current fluctuation rate based on multiple operating currents The current fluctuation rate is used to reflect the stability of the current. The calculation formula is as follows:

[0065] in, is the standard deviation of the current value, reflecting the degree of dispersion of the current; is the average value of the current, reflecting the overall level of the current.

[0066]

[0067] is the current data after missing value filling and normalization, and n is the length of a time interval of the collected historical data.

[0068] (3) Based on the average load factor P of multiple distribution transformers 0 (t) Calculate the corresponding maximum average load rate Maximum average load factor The calculation formula is as follows:

[0069] in, The average load rate in the kth time interval takes the maximum value.

[0070] (4) Based on the average load factor P of multiple distribution transformers 0 (t) Calculate the corresponding maximum average load rate S K The maximum average load factor S K The calculation formula is as follows:

[0071] The maximum value of the average three-phase unbalance of the distribution transformer in the kth time interval.

[0072] (5) Based on the average load factor P of multiple distribution transformers 0 (t) Calculate the corresponding maximum average load rate T K The maximum average load rate T K The calculation formula is as follows:

[0073] Take the maximum value of the ambient temperature in the kth time interval.

[0074] (6) Based on the average load factor P of multiple distribution transformers0 (t) Calculate the corresponding maximum average load rate W K The maximum average load rate W K The calculation formula is as follows:

[0075] Take the maximum value of the ambient humidity for the kth time interval.

[0076] It can be seen that in this embodiment, by performing feature extraction on the second training subset corresponding to each time interval, data conforming to the corresponding format of the model to be trained is obtained.

[0077] Step S205: Divide the target data set into a target training data set and a target test set.

[0078] In a specific implementation, this embodiment divides the processed target data set into a target training data set and a target test set, which can be divided in a ratio of 7:3 or 8:2. However, in some cases, other ratios can be selected as needed, which is not limited here.

[0079] In the target test set, the determined current fluctuation rate, voltage deviation rate, maximum average load rate, maximum average three-phase imbalance of distribution transformer, maximum ambient temperature and maximum ambient humidity are input into the training model, as well as the corresponding risk labels. Generate a test set, which is the following feature label matrix X:

[0080] Among them, the first 6 columns are feature columns, and the 7th column is the label column; the first 6 columns are the extracted features, and the last column b K Indicates the risk of low voltage radio area in the next time window, which is a label column. So far, the construction of the target test set is completed.

[0081] Step S206: construct a model to be trained, and train the model to be trained according to the target data set to be trained to obtain a trained risk situation prediction model.

[0082] Specifically, Figure 4 As shown, the construction of the model to be trained includes: constructing the model to be trained based on a radial basis function (RBF) neural network; the model to be trained includes an input layer, a processing layer and an output layer; the input layer is used to receive a third training subset in the target data set to be trained; the processing layer is used to map the input third training subset to a high-dimensional space to obtain a feature label matrix; the output layer is used to process the feature label matrix to obtain the risk type of the next time interval of the low-voltage radio station area.

[0083] In the specific implementation, Figure 4 As shown, first, a model to be trained is constructed. Aiming at the characteristics of risk prediction of low-voltage power supply area, the traditional SVM algorithm is improved, kernel functions are introduced, and penalty parameters are adjusted to improve the prediction accuracy and generalization ability of the model. The model to be trained is constructed based on a radial basis function (RBF) neural network, wherein the RBF network is a three-layer structure, including an input layer, a processing layer (i.e., a hidden layer), and an output layer. In this embodiment, the input layer includes 6 input quantities (current fluctuation rate, voltage deviation rate, maximum average load rate, maximum average three-phase imbalance of distribution transformer, maximum ambient temperature, maximum ambient humidity), and the output layer includes 1 output quantity (risk type of low-voltage power supply area: 0, 1, 2, 3, 4). The input layer receives the input quantity data after preprocessing and feature extraction, the processing layer uses a radial basis function to map the input to a high-dimensional space, and the output layer uses a linear function to map the output of the processing layer to the target space, i.e., the output quantity. The radial basis function is used as the activation function of the neurons in the processing layer, and the present invention adopts a Gaussian function. is a radial basis function. Therefore, the parameters of the RBF neural network include three: basis function center, width and weight (i.e., the linear weight from the hidden layer to the output layer). ).

[0084] After obtaining the target training data set and the target test set, first set the basis function center of the RBF neural network. Specifically, determine the center position of the basis function in the input space. A random selection method can be used to randomly select several sample points from the target training data set as the basis function center; or the training data can be clustered using algorithms such as K-means clustering, and the cluster center is used as the basis function center. The number and distribution of basis function centers will affect the approximation ability and generalization performance of the network. Generally speaking, the more centers there are, the higher the approximation accuracy of the network. Then determine the width of the basis function of the RBF neural network. The width of the basis function determines its scope and shape. The width can be determined based on the distance between the basis function centers, for example, using the average value of the distance between all centers multiplied by an empirical coefficient as the width. When the width is large, the basis function is smooth and insensitive to local changes in the input data, but it may lead to reduced approximation accuracy; when the width is small, the basis function is local and can more accurately model the local features of the input data, but it may make the network more sensitive to noise. Then determine the weights of the RBF neural network. Specifically, weights are used to combine the outputs of the basis functions to obtain the final output of the network. The initial weights can be set randomly, usually in a small range, such as [-0.1, 0.1] or [-1, 1]. They can also be initialized based on some prior knowledge or experience. For example, if you have a certain understanding of the problem, you can set the initial weights according to the expected function characteristics.

[0085] Specifically, the processing process of the RBF neural network is as follows: (1) Output of basis function.

[0086] Calculate the distance: For each input sample x, calculate its distance to each basis function center c j The distance is usually the Euclidean distance .

[0087] Calculate the output through the basis function: Substitute the calculated distance into the selected radial basis function to obtain the output of each basis function for the input sample x. For example, using the Gaussian basis function , where σ j is the width of the jth basis function. In this way, for each input sample, a set of basis function outputs will be obtained, and these outputs constitute a vector that reflects the characteristics of the input sample under the action of different basis functions.

[0088] (2) Train the network.

[0089] Define the loss function: Choose a suitable loss function to measure the difference between the network output and the target output. The commonly used loss function is the mean square error (MSE), that is, , where y i is the target output, is the actual output of the network, and n is the number of training samples. The smaller the value of the loss function, the better the approximation effect of the network.

[0090] Select an optimization algorithm: Use an optimization algorithm to adjust the weights of the RBF neural network to minimize the loss function. Common optimization algorithms include gradient descent, stochastic gradient descent, conjugate gradient method, Levenberg-Marquardt algorithm, etc. Taking the gradient descent method as an example, it calculates the gradient of the loss function with respect to the weight, and then updates the weight in the opposite direction of the gradient, that is, ,in is the weight at the kth iteration, and α is the learning rate, which controls the step size of the weight update.

[0091] Iteratively update weights: Input the samples in the training set into the network one by one, and continuously update the weights according to the weight update amount calculated by the optimization algorithm. This process will be repeated many times until the value of the loss function converges to a smaller value, or the preset maximum number of iterations is reached. In each iteration, the network will calculate the network output based on the current weights and basis function outputs, compare it with the target output, and then adjust the weights according to the feedback of the loss function, so that the network gradually learns the correct input-output mapping relationship.

[0092] (3) Test evaluation and adjustment.

[0093] Evaluate network performance: Use the test set to test the trained RBF network, input the input samples in the test set into the network, get the network output, and compare it with the target output of the test set. Calculate some evaluation indicators, such as mean square error (MSE), mean absolute error (MAE), coefficient of determination (R2), etc., to measure the approximation performance of the network. These indicators can help us understand the performance of the network on unknown data and evaluate whether the network is overfitting or underfitting.

[0094] Adjust network parameters: According to the test results, adjust and optimize the network parameters. If the network approximation effect is not ideal, such as a large mean square error, you can consider increasing the number of basis functions or adjusting the center and width of the basis functions, or adjusting parameters such as the learning rate during training, and then retrain and test until a satisfactory approximation effect is obtained.

[0095] Visualization results: Visualize and compare the network's approximation results with the target function, draw the target function curve and the network approximation curve, and visually observe the network's approximation of the function. Through visualization, you can more clearly see which areas the network approximates well and which areas have large errors, so that targeted improvements can be made.

[0096] In a possible embodiment, the model to be trained is trained according to the target data set to be trained and the target test set to obtain a trained risk situation prediction model, including: inputting each third training subset in the target data set to be trained into the model to be trained in turn; processing the data in the input third training subset by the model to be trained to obtain a first risk type; comparing the first risk type with the target risk type in the target test set; if accurate, completing the training of the model to be trained to obtain the risk situation prediction model; if inaccurate, optimizing the model to be trained by the Bayesian optimization method, continuing to train the optimized model to be trained by the target data set to be trained, and repeating the above operations until the data results of the model are accurate, thereby obtaining the risk situation prediction model.

[0097] Specifically, the second historical data is re-collected to generate a training set, the data in the training set is input into the model to be trained, the model to be trained generates a corresponding risk type based on the data input in the training set, and the risk type is compared with the corresponding risk type in the test set. If they are consistent, it is determined that the model training is completed. If they are inconsistent, it means that there is an error in the model. The model parameters are adjusted and the data in the training set is re-trained until the model output results are accurate. Then it is determined that the model training is completed and the risk situation prediction model is obtained.

[0098] It can be seen that in this embodiment, the training results are verified through the test set to ensure that the model training is completed and improve the model training accuracy.

[0099] Step S207, input the real-time operation data and real-time environmental data of the first time interval of the low-voltage radio station area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio station area.

[0100] Specifically, the real-time operation data and real-time environmental data of the low-voltage radio area are input into the risk situation prediction model to obtain the risk type of the next time interval of the low-voltage radio area, including: Collect real-time operation data and real-time environmental data of the low-voltage power supply area in the first time window; the real-time operation data include the operating voltage, current, average load rate and average three-phase imbalance of the distribution transformer; the real-time environmental data include ambient temperature and ambient humidity; pre-process the target data to obtain the data to be processed that meets the input rules of the risk situation prediction model; use the risk situation prediction model to determine the target risk type of the low-voltage power supply area in the second time window according to the data to be processed; wherein the risk situation prediction model is obtained by training for multiple consecutive time windows based on a target training set, the time length of each time window is the same, and the multiple time windows include the first time window and the second time window, and the second time window is the next time window of the first time window; output target warning information according to the target risk type.

[0101] In a specific implementation, after obtaining a trained risk situation prediction model, the risk situation prediction model can be used to predict the risk of low-voltage radio stations.

[0102] Specifically, the real-time operation data and real-time environmental data of the low-voltage radio station area in the first time window are obtained. The time interval corresponding to the risk prediction moment of the low-voltage radio station area in the first time window is the real-time operation data and the real-time environmental data are the relevant data generated in real time during the time interval.

[0103] After obtaining the real-time operation data and real-time environment data, the real-time operation data and real-time environment data are input into the risk situation prediction model. The risk prediction model will generate the risk type of the next time window (i.e., the second time window) of the first time window based on the real-time operation data and real-time environment data. Finally, the system outputs the target warning information according to the target risk type. The user can perform corresponding processing according to the risk type, or the system can perform emergency processing.

[0104] It can be seen that in this embodiment, the real-time operation data and real-time environmental data of the low-voltage power supply area are collected in real time and input into the trained risk situation prediction model to predict the risk type of the low-voltage power supply area in the next time period, thereby realizing real-time monitoring and prediction of the risk situation of the low-voltage power supply area.

[0105] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that in order to realize the above functions, the mobile electronic device includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiment provided in this article, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0106] The embodiment of the present application can divide the electronic device into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0107] See also Figure 5 The present application also provides a risk situation prediction device 30 for a low-voltage power supply area, comprising: An acquisition unit 31 is used to acquire historical data of a low-voltage power supply area; wherein the historical data includes a plurality of operation data, a plurality of environmental data and a plurality of risk data; A grouping unit 32 is used to group the historical data according to a first grouping rule to obtain a plurality of first training data groups; the first grouping rule is to group the operation data, environment data and risk data at the same time into one group; each first training data group includes a plurality of input parameters and output parameters, wherein the input parameters include the operation data and the environment data, and the output parameters include the risk data; The grouping unit 32 is further configured to group the plurality of first training data groups according to a second grouping rule to obtain a plurality of first training subsets; the second grouping rule is to divide the first training data groups in the same time interval into one first training subset; A preprocessing unit 33, configured to preprocess the data in the plurality of first training subsets to obtain a target training data set that complies with a data model input rule; A model building unit 34 is used to build a model to be trained, and train the model to be trained according to the target training data set to obtain a trained risk situation prediction model; The prediction unit 35 is used to input the real-time operation data and real-time environmental data of the first time interval of the low-voltage radio station area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio station area.

[0108] It can be seen that in the present application, by acquiring the historical data of the low-voltage radio area; then grouping the historical data according to the first grouping rule to obtain multiple first training data groups; then grouping the multiple first training data groups according to the second grouping rule to obtain multiple first training subsets; then preprocessing the data in the multiple first training subsets to obtain a target data set to be trained that meets the data model input rules; then constructing a model to be trained, training the model to be trained according to the target data set to be trained to obtain a trained risk situation prediction model; finally, inputting the real-time operation data and real-time environmental data of the first time interval of the low-voltage radio area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio area. In this way, the real-time operation data and real-time environmental data of the low-voltage radio area are collected in real time, input into the trained risk situation prediction model, and the risk type of the next time interval of the low-voltage radio area is predicted, thereby realizing real-time monitoring and prediction of the risk situation of the low-voltage radio area.

[0109] In a possible embodiment, the data in the multiple first training subsets are preprocessed to obtain a target data set to be trained that meets the data model input rules, and the preprocessing unit 33 is specifically used to: complete and normalize the movement data of the first training data group in each first training subset to obtain multiple second training subsets; each second training subset includes multiple second training data groups after data completion and normalization; feature extraction is performed on the training data group in each second training subset to obtain the target data set to be trained; the target data set to be trained includes multiple third training subsets, and each third training subset includes multiple third training data groups obtained after feature extraction.

[0110] In a possible embodiment, the movement data of the first training data group in each first training subset is completed and normalized respectively to obtain multiple second training subsets, and the preprocessing unit 33 is specifically used to: determine the location where data is missing in each first training data group; perform data completion by interpolation at each data missing location; and scale the data of each first training data group after data completion to a target range to obtain multiple second training data groups.

[0111] In a possible embodiment, the training data group in each second training subset is subjected to feature extraction to obtain aspects of the target data set to be trained, and the preprocessing unit 33 is specifically used to: calculate the current fluctuation rate, voltage deviation rate, maximum average load rate and maximum distribution transformer average three-phase imbalance of each second training subset; determine the maximum ambient temperature and maximum ambient humidity of each second training subset to obtain the target data set to be trained.

[0112] In a possible embodiment, in terms of constructing the model to be trained, the model construction unit 34 is specifically used to: construct the model to be trained based on a radial basis neural network; the model to be trained includes an input layer, a processing layer and an output layer; the input layer is used to receive a third training subset in the target data set to be trained; the processing layer is used to map the input third training subset to a high-dimensional space to obtain a feature label matrix; the output layer is used to process the feature label matrix to obtain the risk type of the next time interval of the low-voltage radio station area.

[0113] In a possible embodiment, in terms of training the model to be trained according to the target data set to be trained to obtain a trained risk situation prediction model, the model construction unit 34 is specifically used to: input each third training subset in the target data set to be trained into the model to be trained in turn; the model to be trained processes the data in the input third training subset to obtain a first risk type; compare the first risk type with the target risk type in the third training subset; if accurate, complete the training of the model to be trained to obtain the risk situation prediction model; if inaccurate, optimize the model to be trained by the Bayesian optimization method, continue to train the optimized model to be trained by the target data set to be trained, and repeat the above operations until the data results of the model are accurate, thereby obtaining the risk situation prediction model.

[0114] In a possible embodiment, the real-time operation data and real-time environmental data of the low-voltage power supply area are input into the risk situation prediction model to obtain the risk type of the next time interval of the low-voltage power supply area. The prediction unit 35 is specifically used to: collect the real-time operation data and real-time environmental data of the low-voltage power supply area in the first time window; the real-time operation data include the operating voltage, current, average load rate and average three-phase imbalance of the distribution transformer; the real-time environmental data include ambient temperature and ambient humidity; pre-process the target data to obtain the data to be processed that meets the input rules of the risk situation prediction model; use the risk situation prediction model to determine the target risk type of the low-voltage power supply area in the second time window according to the data to be processed; wherein the risk situation prediction model is obtained by training for multiple consecutive time windows based on the target training set, the time length of each time window is the same, the multiple time windows include the first time window and the second time window, and the second time window is the next time window of the first time window; output target warning information according to the target risk type.

[0115] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0116] The present application also provides an electronic device 10, such as Figure 6As shown, it includes at least one processor 11; a display screen 12; and a memory 13, and may also include a communications interface 15 and a bus 14. The processor 11, the display screen 12, the memory 13 and the communications interface 15 can communicate with each other through the bus 14. The display screen 12 is configured to display a preset user guide interface in the initial setting mode. The communications interface 15 can transmit information. The processor 11 can call the logic instructions in the memory 13 to execute the method in the above embodiment.

[0117] Optionally, the electronic device 10 may be a mobile electronic device, or an electronic device or other device, which is not limited here.

[0118] In addition, the logic instructions in the memory 13 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0119] The memory 13, as a computer-readable storage medium, can be configured to store software programs, computer executable programs, such as program instructions or modules corresponding to the methods in the embodiments of the present disclosure. The processor 11 executes functional applications and data processing by running the software programs, instructions or modules stored in the memory 13, that is, implementing the methods in the above embodiments.

[0120] The memory 13 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the electronic device 10, etc. In addition, the memory 13 may include a high-speed random access memory and may also include a non-volatile memory. For example, a variety of media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, may also be a transient storage medium.

[0121] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps of any method described in the above method embodiments, and the above computer includes an electronic device.

[0122] The embodiment of the present application also provides a computer program product, the computer program product includes a non-transitory computer-readable storage medium storing a computer program, the computer program is operable to cause a computer to execute some or all of the steps of any method described in the method embodiment. The computer program product may be a software installation package, and the computer includes an electronic device.

[0123] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0124] In the several embodiments provided in the present application, it should be understood that the disclosed methods, devices and systems can be implemented in other ways. For example, the device embodiments described above are merely schematic; for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation; for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0126] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0127] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform some steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a volatile memory or a non-volatile memory. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM) and direct RAM bus RAM (DR RAM). Various media that can store program code are available.

[0128] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present invention, and can make various changes and modifications, including the combination of the above-mentioned different functions and implementation steps, including software and hardware implementation methods, all of which are within the scope of protection of the present invention.

Claims

1. A method for predicting risk situation in a low-voltage power station area, characterized in that: include: Acquire historical data of the low-voltage power supply area; wherein the historical data includes a plurality of operation data, a plurality of environmental data and a plurality of risk data; The historical data are grouped according to a first grouping rule to obtain a plurality of first training data groups; the first grouping rule is to group the operation data, environment data and risk data at the same time into one group; each first training data group includes a plurality of input parameters and output parameters, wherein the input parameters include the operation data and the environment data, and the output parameters include the risk data; The plurality of first training data groups are grouped according to a second grouping rule to obtain a plurality of first training subsets; the second grouping rule is to divide the first training data groups in the same time interval into the same first training subset; Preprocessing the data in the plurality of first training subsets to obtain a target data set that complies with the data model input rule; Dividing the target data set into a target training data set and a target test set; Constructing a model to be trained, and training the model to be trained according to the target training data set and the target test set to obtain a trained risk situation prediction model; The real-time operation data and real-time environmental data of the first time interval of the low-voltage radio station area are input into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio station area; the second time interval is the next time interval adjacent to the first time interval.

2. The method according to claim 1, characterized in that Preprocessing the data in the plurality of first training subsets to obtain a target data set that complies with the data model input rule includes: The movement data of the first training data group in each first training subset is supplemented and normalized respectively to obtain a plurality of second training subsets; each second training subset includes a plurality of second training data groups after data supplementation and normalization; Feature extraction is performed on the second training data group in each second training subset to obtain the target data set to be trained; the target data set to be trained includes multiple third training subsets, and each third training subset includes multiple third training data groups obtained after feature extraction.

3. The method according to claim 2, characterized in that The method of completing and normalizing the movement data of the first training data group in each first training subset to obtain multiple second training subsets includes: Determining locations where data is missing in each first training data set; Data is completed by interpolation at each location where data is missing; After the data is completed, the data of each first training data group is scaled to a target range to obtain a plurality of second training data groups.

4. The method according to claim 2, characterized in that: The extracting features of the second training data set in each second training subset to obtain the target data set to be trained comprises: Calculating the current fluctuation rate, voltage deviation rate, maximum average load rate and maximum distribution transformer average three-phase imbalance of each second training subset; The maximum ambient temperature and the maximum ambient humidity of each second training subset are determined to obtain the target data set to be trained.

5. The method according to any one of claims 1 to 4, characterized in that: The step of constructing a model to be trained includes: Constructing the model to be trained based on a radial basis neural network; The model to be trained includes an input layer, a processing layer and an output layer; the input layer is used to receive the third training subset in the target data set to be trained; The processing layer is used to map the input third training subset to a high-dimensional space to obtain a feature label matrix; the output layer is used to process the feature label matrix to obtain the risk type of the next time interval of the low-voltage radio station area.

6. The method according to any one of claims 1 to 4, characterized in that: The step of training the model to be trained according to the target data set to be trained and the target test set to obtain a trained risk situation prediction model includes: Inputting each third training subset in the target data set to be trained into the model to be trained in sequence; The model to be trained processes the input data in the third training subset to obtain a first risk type; The first risk type is compared with the target risk type in the target test set; if they are accurate, the training of the model to be trained is completed to obtain the risk situation prediction model; if they are inaccurate, the model to be trained is optimized by Bayesian optimization method, and the optimized model to be trained is continued to be trained by the target data set to be trained, and the above operation is repeated until the data result of the model is accurate, thereby obtaining the risk situation prediction model.

7. The method according to any one of claims 1 to 4, characterized in that: The step of inputting the real-time operation data and the real-time environment data of the first time interval of the low-voltage radio area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio area includes: Collecting real-time operation data and real-time environmental data of the low-voltage power supply area in the first time window; the real-time operation data includes the operation voltage, current, average load rate and average three-phase imbalance of the distribution transformer; the real-time environmental data includes the ambient temperature and ambient humidity; Preprocessing the target data to obtain data to be processed that meets the input rules of the risk situation prediction model; Determine the target risk type of the low-voltage radio area in the second time window according to the data to be processed using a risk situation prediction model; wherein the risk situation prediction model is obtained by training for a plurality of consecutive time windows based on a target training set, the time length of each time window is consistent, the plurality of time windows include the first time window and the second time window, and the second time window is the next time window of the first time window; Output target warning information according to the target risk type.

8. A risk situation prediction device for a low voltage radio area, characterized in that: include: An acquisition unit, used for acquiring historical data of a low-voltage power supply area; wherein the historical data includes a plurality of operation data, a plurality of environmental data and a plurality of risk data; a grouping unit, configured to group the historical data according to a first grouping rule to obtain a plurality of first training data groups; the first grouping rule is to group the operation data, environment data and risk data at the same time into one group; each first training data group includes a plurality of input parameters and output parameters, wherein the input parameters include the operation data and the environment data, and the output parameters include the risk data; The grouping unit is further used to group the plurality of first training data groups according to a second grouping rule to obtain a plurality of first training subsets; the second grouping rule is to divide the first training data groups in the same time interval into the same first training subset; A preprocessing unit, configured to preprocess the data in the plurality of first training subsets to obtain a target data set that meets the data model input rule; and to divide the target data set into a target to-be-trained data set and a target test set; A model building unit, used to build a model to be trained, and train the model to be trained according to the target training data set and the target test set to obtain a trained risk situation prediction model; A prediction unit is used to input the real-time operation data and real-time environmental data of the first time interval of the low-voltage radio station area into the risk situation prediction model to obtain the risk type of the second time interval of the low-voltage radio station area; the second time interval is the next time interval adjacent to the first time interval.

9. An electronic device, characterized in that: The method comprises a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for executing the steps in the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program for electronic data exchange is stored, wherein the computer program enables a computer to execute instructions of the steps in the method according to any one of claims 1 to 7.

Images