Power Monitoring System Operation and Maintenance Risk Analysis Method and System

Abstract

The present invention relates to the technical field of power system security, and particularly to a method and system for analyzing operation and maintenance risks of a power monitoring system. By judging the data quality in the data set, different inertia weights are adopted for data sets with different data qualities. Among them, for the data set with good data quality in the data set, a relatively large inertia weight is set in the early stage of iteration, and a relatively small inertia weight is set in the later stage of iteration. The change of the inertia weight is not drastic, thereby improving the particle optimization speed. For the data set with poor data quality in the data set, a completely dynamic weight is adopted, that is, the dynamic weight changes at any time with the change of the number of iterations. In this way, the acceleration change of the particle is also relatively frequent, so that the particle moves more fully during the iteration process, improving the particle optimization accuracy; thereby improving the efficiency of model training and taking into account the economy and reliability of the operation and maintenance risk analysis of the power monitoring system.

Description

Technical Field

[0001] The present invention relates to the technical field of power system security, and particularly to a method and system for analyzing operation and maintenance risks of a power monitoring system. Background Art

[0002] To a certain extent, the production and operation quality and efficiency of power enterprises determine the operating benefits of power grid enterprises and reflect the management level of power grid enterprises. The operation and maintenance work of power monitoring systems is extremely pressured and tasks are heavy, and the operation and maintenance quality and efficiency of power monitoring systems are severely restricted. Facing the complex general environment, it has become an inevitable choice to fully improve the work quality and efficiency of power grid enterprises.

[0003] In the prior art, there are solutions to improve the parameter optimization efficiency by improving the particle swarm algorithm. Generally, the inertia weight is optimized. For some data sets with uneven data quality, the same inertia weight is used to control the optimization speed of particles for data sets of different qualities, which will slow down the convergence speed of model training.

[0004] Therefore, there is an urgent need in the prior art for a technical solution for analyzing operation and maintenance risks of a power monitoring system that can improve model training efficiency. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a method for analyzing operation and maintenance risks of a power monitoring system, which is used to analyze the operation and maintenance risks of icing on transmission lines, and specifically includes:

[0006] S1: Obtain data for analyzing the operation and maintenance risks of historical icing on transmission lines;

[0007] S2: Perform preprocessing operations on the data;

[0008] S3: Divide the data preprocessed in S2 into several data sets;

[0009] S4: Establish a prediction model for the icing thickness of transmission lines;

[0010] S5: Train the prediction model for the icing thickness of transmission lines with the data sets obtained in S3; S5 is specifically:

[0011] S5.1: Improve the particle swarm algorithm;

[0012] Improving the particle swarm algorithm specifically means: when setting the inertia weight of the particle swarm algorithm, different inertia weight formulas are used for optimization for data sets of different data qualities;

[0013] Using different inertia weight formulas for optimization for data sets of different data qualities specifically means:

[0014] Define the dataset with data that has undergone data outlier handling and missing value handling as the first dataset. Among them, data outlier handling or missing value handling has occurred in the first dataset, otherwise it is the second dataset;

[0015] Among them, for the first dataset, the inertia weight formula of the improved particle swarm algorithm is:

[0016] ;

[0017] Among them, w(t) is the inertia weight at the t-th iteration, w max is the maximum value of the inertia weight, w min is the minimum value of the inertia weight, t is the current iteration number, and T is the total number of iterations;

[0018] For the second dataset, the inertia weight formula of the improved particle swarm algorithm is:

[0019] ;

[0020] S5.2: Use the improved particle swarm algorithm to determine the hyperparameters of the transmission line icing thickness prediction model;

[0021] S5.3: Train the transmission line icing thickness prediction model determined by the hyperparameters with the dataset obtained in S3;

[0022] S6: Obtain the data for transmission line operation and maintenance risk analysis in the week before the prediction date;

[0023] S7: Input the data for transmission line operation and maintenance risk analysis in the week before the prediction date obtained in S6 into the transmission line icing thickness prediction model trained in S5, and input the icing thickness prediction result;

[0024] S8: Obtain the transmission line operation and maintenance risk analysis result according to the icing thickness prediction result.

[0025] Preferably, in S1, the data obtained for historical transmission line icing operation and maintenance risk analysis includes: environmental temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current magnitude data, transmission line voltage level data, transmission line current data, conductor material data of the wire, and wire diameter data in the area where the transmission line is located.

[0026] Preferably, obtain the environmental temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current magnitude data, transmission line voltage level data, transmission line current data, conductor material data of the wire, and wire diameter data in the area where the transmission line is located in the past ten years for transmission line operation and maintenance risk analysis.

[0027] Preferably, in S2, the preprocessing operations on the numerical data in the data include outlier processing and missing value processing; the preprocessing operations on the text data in the data include data numericalization.

[0028] Preferably, the outlier processing is specifically as follows: If the deviation of the sampling point value from the average value of all data within the preset time period of the data type of the sampling point value exceeds three times the standard deviation, then the sampling point value is determined as an outlier; then the outlier is deleted.

[0029] The missing value processing is specifically as follows: The mean filling method is used to process the missing value; specifically, the average value of all data within the preset time period of the data type of the missing value is calculated and filled into the missing value position.

[0030] The data numericalization is as follows: For the wire made of copper-aluminum alloy, it is assigned a value of 1, and for the wire made of aluminum, it is assigned a value of 0.

[0031] Preferably, in S3, the data preprocessed by S2 every week is used as a data set for data set division; wherein, each data set includes the environmental temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current magnitude data, transmission line voltage level data, transmission line current data, wire conductor material data, and wire diameter data of the area where the transmission line is located at different times corresponding to the week of the data set.

[0032] Preferably, in S4, the transmission line operation and maintenance risk analysis and prediction model is a long short-term memory network model.

[0033] Preferably, the long short-term memory network model includes a forgetting gate, an input gate, and an output gate.

[0034] Preferably, in S5.2, the hyperparameters of the transmission line ice thickness prediction model are: the number of neurons in the LSTM network, the learning rate, and the number of training times.

[0035] According to another aspect of the present invention, there is provided a power monitoring system operation and maintenance risk analysis system, which uses the above-mentioned power monitoring system operation and maintenance risk analysis method to analyze the operation and maintenance risk of transmission line icing. The system includes:

[0036] A historical data acquisition module for acquiring data on the historical operation and maintenance risk analysis of transmission line icing.

[0037] A preprocessing module for performing preprocessing operations on the data.

[0038] A dataset division module, configured to divide the data processed by the preprocessing module into several datasets;

[0039] A transmission line icing thickness prediction model establishment module, configured to establish a transmission line icing thickness prediction model;

[0040] A transmission line icing thickness prediction model training module, configured to train the transmission line icing thickness prediction model with the datasets obtained by the dataset division module;

[0041] A current data acquisition module, configured to acquire data for transmission line operation and maintenance risk analysis in the week before the prediction date;

[0042] An icing thickness prediction module, configured to input the data for transmission line operation and maintenance risk analysis in the week before the prediction date acquired by the current data acquisition module into the transmission line icing thickness prediction model trained by the transmission line icing thickness prediction model training module, and input the icing thickness prediction result;

[0043] A transmission line operation and maintenance risk analysis result acquisition module, configured to obtain a transmission line operation and maintenance risk analysis result according to the icing thickness prediction result.

[0044] The embodiments of the present invention have the following technical effects:

[0045] By judging the data quality in the dataset, different inertia weights are adopted for datasets with different data qualities. Among them, for the dataset with good data quality in the dataset, a relatively large inertia weight is set in the early stage of iteration, and a relatively small inertia weight is set in the later stage of iteration. The change of the inertia weight is not drastic, thereby improving the particle optimization speed. For the dataset with poor data quality in the dataset, a completely dynamic weight is adopted, that is, as the number of iterations changes, the dynamic weight changes at any time, so that the acceleration change of the particle is also more frequent, so that the particle moves more fully during the iteration process, improving the particle optimization accuracy; thereby improving the efficiency of model training and taking into account the economy and reliability of the operation and maintenance risk analysis of the power monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to these drawings without creative efforts.

[0047] Figure 1 is a flowchart of a method for analyzing the operation and maintenance risk of a power monitoring system provided by an embodiment of the present invention;

[0048] Figure 2 It is a flowchart for training a prediction model for the icing thickness of a transmission line provided by an embodiment of the present invention. Detailed implementation manners

[0049] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope protected by the present invention.

[0050] For a power system, icing on the lines in the power system is a common risk in the operation and maintenance of the power system. In this embodiment, an operation and maintenance risk analysis is carried out for the icing of the power system lines. For icing early warning of the lines, different operation and maintenance measures are taken to effectively improve the operation and maintenance quality of the transmission lines and ensure the safe operation of the transmission lines during winter.

[0051] Appendix Figure 1 It is a flowchart of a method for analyzing the operation and maintenance risks of a power monitoring system provided by an embodiment of the present invention. The method for analyzing the operation and maintenance risks of a power monitoring system is used to analyze the operation and maintenance risks of the icing of a transmission line, and specifically includes:

[0052] S1: Obtain data for historical operation and maintenance risk analysis of the icing of a transmission line;

[0053] In this step, first analyze the influencing factors of the icing of the transmission line; generally, there are three major influencing factors for the icing of the line: meteorology, environment, and transmission line. Among them, each influencing factor contains multiple small influencing factors. For example, for the meteorological influencing factor, it includes environmental temperature, air humidity, snowfall, wind speed, etc. For the environmental influencing factor, it includes the altitude of the transmission line, etc. For the transmission line influencing factor, it includes the voltage level of the transmission line, the current of the transmission line, the material of the wire conductor, the diameter of the wire, etc.;

[0054] Therefore, based on the above analysis of the influencing factors of the icing of the transmission line, the data obtained for historical operation and maintenance risk analysis of the icing of the transmission line includes: environmental temperature data, air humidity data, snowfall data, wind speed data, altitude data of the transmission line, data on the magnitude of the line current, voltage level data of the transmission line, current data of the transmission line, data on the material of the wire conductor, data on the diameter of the wire, etc.;

[0055] In this step, environmental temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current magnitude data, transmission line voltage level data, transmission line current data, conductor material data of the wire, and wire diameter data in the area where the transmission line is located in the past ten years are obtained for the operation and maintenance risk analysis of the transmission line.

[0056] S2: Perform preprocessing operations on the data;

[0057] In this step, the preprocessing operations on the numerical data in the data include outlier processing, missing value processing, etc.;

[0058] Since a large amount of data needs to be collected in this embodiment, covering various aspects such as meteorology, environment, and transmission lines, there will generally be some abnormal data. And this embodiment is actually a data-driven risk analysis, with relatively high requirements for the accuracy of the analyzed data. Therefore, professional processing of abnormal data is required;

[0059] Among them, the specific outlier processing is as follows: If the deviation of the sampled point value from the average value of all data within the preset time period of the data type of the sampled point value exceeds three times the standard deviation, then the sampled point value is determined as an outlier; then the outlier is deleted;

[0060] Among them, the specific missing value processing is as follows: The mean filling method is used to process the missing value; specifically, the average value of all data within the preset time period of the data type of the missing value is calculated and filled into the missing value position;

[0061] In this step, the preprocessing operations on the text data in the data include data numericalization;

[0062] Among them, as can be seen from step S1, the data for risk analysis also includes text data. For example, the conductor material data of the wire. In order for the risk identification model to better utilize this type of data, this type of data needs to be converted into numerical data;

[0063] Among them, generally, the wire materials are divided into two types: copper-aluminum alloy wire and aluminum wire. The copper-aluminum alloy wire is a mixture of copper and aluminum, with the advantages of both copper and aluminum wires. It not only has good electrical conductivity but also a relatively reasonable price. The copper-aluminum alloy wire is usually used for medium-voltage and high-voltage lines and has a wide application range. The aluminum wire is widely used in the low-voltage power transmission industry due to its relatively low price. The aluminum wire is a lightweight wire, which is convenient for building long-span wire laying channels; therefore, the wire material data is numericalized as follows: For the copper-aluminum alloy wire, it is assigned a value of 1, and for the aluminum wire, it is assigned a value of 0.

[0064] S3: Divide the data preprocessed in S2 into several data sets;

[0065] Among them, in step S1, environmental temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current magnitude data, transmission line voltage level data, transmission line current data, conductor material data of the wire, and wire diameter data of the area where the transmission line is located in the past ten years are obtained; in this embodiment, by using these data to learn and train the transmission line icing thickness prediction model, an accurate line icing thickness prediction model is obtained.

[0066] Specifically, in this step, the data preprocessed by the above S2 per week is used as a data set for data set division; among them, each data set includes environmental temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current magnitude data, transmission line voltage level data, transmission line current data, conductor material data of the wire, and wire diameter data of the area where the transmission line is located at different times corresponding to the week of the data set.

[0067] Among them, each data set further includes the average value of the icing thickness of the transmission line corresponding to the week.

[0068] S4: Establish a transmission line icing thickness prediction model;

[0069] In this embodiment, the transmission line operation and maintenance risk analysis and prediction model is a long short-term memory network model. The long short-term memory network model (LSTM) is a special recurrent neural network that can be used to solve the problems of gradient disappearance and gradient explosion that may occur when traditional recurrent neural networks process long-time series data. LSTM introduces a gating mechanism, including a forget gate, an input gate, and an output gate; the gate structure is the core component of LSTM. The cell state can be regarded as a channel for information transmission, allowing information to be continuously transmitted; the gate structure is used to judge whether to retain the passed information, and it learns whether to retain or forget the information through continuous training. The input gate determines the amount of important information in the current input to be retained in the cell state, and then updates the cell state; the forget gate determines the information that should be forgotten or retained; the output gate is used to determine the new hidden state and pass the new cell state c and the new hidden state to the next LSTM network unit. The information transmission method within the LSTM network neuron can autonomously select which information needs to be retained and which information needs to be forgotten, so as to better capture the long-term dependence relationship in the long sequence data.

[0070] S5: Train the transmission line icing thickness prediction model with the data set obtained in S3;

[0071] Among them, as shown in the appendix Figure 2 It shows that S5 is specifically:

[0072] S5.1: Improve the particle swarm optimization algorithm;

[0073] In the prior art, the particle swarm optimization algorithm is generally used to determine the hyperparameters of the LSTM. Scientists found in the study of biological group interactions that in a biological group, individuals can solve some problems through their own set of interaction methods. For example, ants in an ant colony communicate information through their antennae, and they can cooperate with clear division of labor among themselves; flocks of birds always migrate according to certain rules, etc. These interactions between individuals and populations in biological groups inspired scientists, and they tried to simulate this process through computers to help solve problems that are difficult to solve in reality. This gave rise to the "particle swarm optimization algorithm". The principle of the particle swarm optimization algorithm is to regard each particle as a solution and search in the entire set space until the particle swarm finds the optimal solution or reaches the iteration limit. Among them, the velocity and position update formulas of particle i are:

[0074] ;

[0075] In the formula, v i is the current velocity of particle i, w is the inertia weight, c1 and c2 are learning factors, pbest[i] is the best point experienced by the i-th example, gbest[i] is the best point experienced by the entire particle population, rand() is a random number, and x i is the position where the i-th example is located; it can be seen from the above formula that the inertia weight is an important influencing factor for the change of particle velocity.

[0076] In the prior art, there are solutions to improve the parameter optimization efficiency by improving the particle swarm optimization algorithm, and generally the inertia weight is optimized; the optimization of the inertia weight is mainly the non-linear change of the inertia weight, and the mathematical expression is:

[0077] ;

[0078] In the formula, w(t) is the inertia weight at the t-th iteration, w max is the maximum value of the inertia weight, w min is the minimum value of the inertia weight, t is the current iteration number, and Maxt is the maximum iteration number; however, the above solutions improve the optimization efficiency of the particle population to a certain extent. However, for some data sets with uneven data quality, the same inertia weight is used to control the optimization speed of particles for data sets of different qualities, which will slow down the convergence speed of model training.

[0079] Therefore, in this embodiment, the particle swarm optimization algorithm is improved. When setting the inertia weight of the particle swarm optimization algorithm, different inertia weight formulas are used for optimization for data sets of different data qualities;

[0080] Specifically, different inertia weight formulas are used for optimization of datasets with different data qualities, specifically as follows:

[0081] A dataset in which the data in the dataset has been processed for data outliers and missing values is defined as the first dataset. Among them, there has been data outlier processing or missing value processing in the first dataset. On the one hand, it indicates that the data in this section may have inaccurate acquisition during collection, so the data quality is relatively low; otherwise, it is the second dataset;

[0082] Among them, for the first dataset, the inertia weight formula of the improved particle swarm algorithm is:

[0083] ;

[0084] Among them, w(t) is the inertia weight at the t-th iteration, w max is the maximum value of the inertia weight, w min is the minimum value of the inertia weight, t is the current iteration number, and T is the total number of iterations;

[0085] For the second dataset, the inertia weight formula of the improved particle swarm algorithm is:

[0086] ;

[0087] By judging the data quality in the dataset, different inertia weights are adopted for datasets with different data qualities. Among them, for the dataset with good data quality in the dataset, a relatively large inertia weight is set in the early stage of iteration, and a relatively small inertia weight is set in the later stage of iteration. The change of the inertia weight is not drastic, thereby improving the particle optimization speed. For the dataset with relatively poor data quality in the dataset, a completely dynamic weight is adopted, that is, as the iteration number changes, the dynamic weight changes at any time. In this way, the acceleration change of the particle is also relatively frequent, so that the particle moves more fully during the iteration process, thereby improving the particle optimization accuracy.

[0088] S5.2: Use the improved particle swarm algorithm to determine the hyperparameters of the transmission line icing thickness prediction model;

[0089] Among them, the hyperparameters of the transmission line icing thickness prediction model are: the number of neurons in the LSTM network, the learning rate, and the number of training times;

[0090] In this embodiment, first, an expert gives a relatively broad search range based on experience, and then the number of neurons, learning rate, and number of training times are optimized and solved within the search range. Among them, the search range of the number of neurons is 50 - 250, the search range of the learning rate is 0.05 - 0.35, and the search range of the number of training times is 250 - 1000; then the improved particle swarm algorithm is used to optimize and determine the hyperparameters.

[0091] S5.3: Train the transmission line icing thickness prediction model with the dataset obtained in S3 for the determined hyperparameters;

[0092] In this step, the loss function is used to determine whether the training of the transmission line icing thickness prediction model is completed.

[0093] S6: Obtain the data for the operation and maintenance risk analysis of the transmission line in the week before the prediction date;

[0094] Among them, the current data for the operation and maintenance risk analysis of the transmission line includes: environmental temperature data, air humidity data, snowfall data, wind speed data, altitude data of the transmission line, line current magnitude data, transmission line voltage level data, transmission line current data, conductor material data of the wire, and wire diameter data at different times in the week before the prediction date.

[0095] S7: Input the data for the operation and maintenance risk analysis of the transmission line in the week before the prediction date obtained in S6 into the transmission line icing thickness prediction model trained in S5, and input the icing thickness prediction result;

[0096] Among them, the output of the transmission line icing thickness prediction model is the icing thickness prediction result value.

[0097] S8: Obtain the operation and maintenance risk analysis result of the transmission line according to the icing thickness prediction result;

[0098] Among them, in this step, when the predicted value of the transmission line icing thickness is greater than 2 cm, the operation and maintenance risk of the transmission line is high, and operation and maintenance measures such as increasing the inspection frequency and deicing the line need to be taken.

[0099] Embodiment 2. This embodiment provides a power monitoring system operation and maintenance risk analysis system. The system uses the power monitoring system operation and maintenance risk analysis method described in Embodiment 1 to analyze the operation and maintenance risk of transmission line icing. The system includes:

[0100] A historical data acquisition module, used to acquire the data for the operation and maintenance risk analysis of historical transmission line icing;

[0101] A preprocessing module, used to perform preprocessing operations on the data;

[0102] A dataset division module, configured to divide the data processed by the preprocessing module into several datasets;

[0103] A transmission line icing thickness prediction model establishment module, configured to establish a transmission line icing thickness prediction model;

[0104] A transmission line icing thickness prediction model training module, configured to train the transmission line icing thickness prediction model with the datasets obtained by the dataset division module;

[0105] A current data acquisition module, configured to acquire data for transmission line operation and maintenance risk analysis in the week before the prediction date;

[0106] An icing thickness prediction module, configured to input the data for transmission line operation and maintenance risk analysis in the week before the prediction date acquired by the current data acquisition module into the transmission line icing thickness prediction model trained by the transmission line icing thickness prediction model training module, and input the icing thickness prediction result;

[0107] A transmission line operation and maintenance risk analysis result acquisition module, configured to obtain a transmission line operation and maintenance risk analysis result according to the icing thickness prediction result.

[0108] Embodiment 3. This embodiment discloses an electronic device, which includes one or more processors and a memory.

[0109] The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0110] The memory may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may run the program instructions to implement a method for analyzing the operation and maintenance risks of a power monitoring system according to any embodiment of the present application as described above and / or other desired functions. Various contents such as initial external parameters and thresholds may also be stored in the computer-readable storage medium.

[0111] In one example, the electronic device may further include: an input device and an output device, and these components are interconnected through a bus system and / or other forms of connection mechanisms (not shown). The input device may include, for example, a keyboard, a mouse, and the like. The output device may output various information to the outside, including warning prompt information, braking force, etc. The output device may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0112] In addition, according to specific application scenarios, the electronic device may further include any other appropriate components.

[0113] In addition to the above methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, and when the computer program instructions are run by a processor, the processor is caused to execute all or each step of a method for analyzing operation and maintenance risks of a power monitoring system provided by any embodiment of the present application.

[0114] The computer program product may be written in any combination of one or more programming languages for programming code to perform the operations of the embodiments of the present application. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, executed as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0115] In addition, an embodiment of the present application may also be a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are run by a processor, the processor is caused to execute the steps of a method for analyzing operation and maintenance risks of a power monitoring system provided by any embodiment of the present application.

[0116] The computer-readable storage medium may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing the operation and maintenance risk of power monitoring system, which is used to analyze the operation and maintenance risk of ice accumulation on power transmission lines, and is characterized in that: include: S1: Obtain historical data for transmission line ice accumulation operation and maintenance risk analysis; In said S1, the data acquired for the historical transmission line ice accumulation operation and maintenance risk analysis include: ambient temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current data, transmission line voltage level data, transmission line current data, wire conductor material data, and wire diameter data of the area where the transmission line is located. The ambient temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current data, transmission line voltage level data, transmission line current data, wire conductor material data, and wire diameter data of the area where the transmission line is located in the past ten years are acquired for transmission line operation and maintenance risk analysis; S2: preprocessing the data; in S2, preprocessing the numerical data in the data includes outlier processing and missing value processing; preprocessing the text data in the data includes data digitization, and processing the outlier specifically includes: if the deviation between the sampling point value and the average value of all data within a preset time period of the data type of the sampling point value exceeds three times the standard deviation, the sampling point value is identified as an outlier; and then the outlier is deleted; The specific processing of missing values ​​is: using the mean filling method to process the missing values; specifically, calculating the average value of all data within a preset time period of the data type of the missing value and filling it into the missing value position; The data is digitized as follows: for copper-aluminum alloy wires, a value of 1 is assigned to them, and for aluminum wires, a value of 0 is assigned to them; S3: the data preprocessed by S2 is divided into several data sets; in S3, the data preprocessed by S2 each week is used as a data set for data set division; wherein each data set includes the ambient temperature data, air humidity data, snowfall data, wind speed data, transmission line altitude data, line current size data, transmission line voltage level data, transmission line current data, wire conductor material data, and wire diameter data of the area where the power transmission line is located at different times of the week corresponding to the data set; S4: Establish a prediction model for ice thickness of transmission lines; S5: training the transmission line ice thickness prediction model using the data set obtained in S3; S5 specifically comprises: S5.1: Improve the particle swarm algorithm; The improvement of the particle swarm algorithm is as follows: when setting the inertia weight of the particle swarm algorithm, different inertia weight formulas are used for optimization for data sets with different data qualities; For data sets with different data qualities, different inertia weight formulas are used for optimization: A data set that has data processed with outliers or missing values ​​in the data set is defined as a first data set, wherein the first data set has data processed with outliers or missing values, otherwise it is a second data set; Among them, for the first data set, the inertia weight formula of the improved particle swarm algorithm is: ； Among them, w(t) is the inertia weight of t iterations, w max is the maximum value of the inertia weight, w min is the minimum value of the inertia weight, t is the current iteration number, and T is the total iteration number; For the second data set, the inertia weight formula of the improved particle swarm algorithm is: ； S5.2: Determine the hyperparameters of the transmission line ice thickness prediction model using an improved particle swarm algorithm; S5.3: training the transmission line ice thickness prediction model determined by the hyperparameters using the data set obtained in S3; S6: Obtain data for transmission line operation and maintenance risk analysis one week before the forecast date; S7: input the data used for transmission line operation and maintenance risk analysis one week before the prediction date obtained in S6 into the transmission line ice thickness prediction model trained by S5, and input the ice thickness prediction result; S8: Obtaining a transmission line operation and maintenance risk analysis result based on the ice thickness prediction result.

2. The method for analyzing operation and maintenance risks of a power monitoring system according to claim 1, characterized in that: In S4, the transmission line operation and maintenance risk analysis and prediction model is a long short-term memory network model.

3. The method for analyzing operation and maintenance risks of a power monitoring system according to claim 2, characterized in that: The long short-term memory network model includes a forget gate, an input gate and an output gate.

4. The method for analyzing operation and maintenance risks of a power monitoring system according to claim 1, characterized in that: In S5.2, the hyperparameters of the transmission line ice thickness prediction model are: the number of neurons in the LSTM network, the learning rate and the number of training times.

5. The power monitoring system operation and maintenance risk analysis system is characterized by: The system adopts the power monitoring system operation and maintenance risk analysis method according to any one of claims 1 to 4, which is used to analyze the operation and maintenance risk of ice accumulation on the transmission line. The system includes: A historical data acquisition module, used to obtain data on historical transmission line ice accumulation operation and maintenance risk analysis; A preprocessing module, used for performing preprocessing operations on the data; A data set division module, used for dividing the data processed by the preprocessing module into several data sets; A transmission line ice thickness prediction model establishment module is used to establish a transmission line ice thickness prediction model; A transmission line ice thickness prediction model training module, used to train the transmission line ice thickness prediction model using the data set obtained by the data set division module; The current data acquisition module is used to obtain data for transmission line operation and maintenance risk analysis one week before the forecast date; An icing thickness prediction module is used to input the data used for transmission line operation and maintenance risk analysis one week before the prediction date acquired by the current data acquisition module into the transmission line icing thickness prediction model trained by the transmission line icing thickness prediction model training module, and input the icing thickness prediction result; The transmission line operation and maintenance risk analysis result acquisition module is used to obtain the transmission line operation and maintenance risk analysis result according to the ice thickness prediction result.

Images