Secondary equipment fault monitoring method and system for power system
By using predictive models and weight update thresholds in power systems, the problem of insufficient accuracy in secondary equipment failure monitoring in the prior art is solved, and higher fault identification accuracy and less environmental interference impact are achieved.
Patent Information
- Application Number
- CN202411350453.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-26
- Publication Date
- 2025-02-07
AI Technical Summary
The existing power system secondary equipment fault monitoring methods rely on manual inspection and threshold-based automatic monitoring systems, and there are problems of insufficient identification, affected by environmental interference and equipment aging.
A secondary equipment fault monitoring method is adopted for power systems. By acquiring the monitoring data of the equipment, a preset prediction model is used to generate a predicted value, and the initial threshold is updated according to the spatial weight, time weight and failure rate weight to generate a fault alarm signal.
It improves the accuracy of fault identification, can predict and alarm faults in advance, and reduces the impact on environmental interference and equipment aging.
Smart Images
Figure CN119397432A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of secondary equipment monitoring, and in particular to a secondary equipment fault monitoring method and system for a power system. Background Art
[0002] The fiber optic current differential protection device is a type of relay protection device. It determines whether there is a fault in the transmission line by comparing the magnitude and phase of the current at both ends. If an abnormality is detected, it will quickly send a signal to activate the circuit breaker and cut off the faulty line to protect the safe and stable operation of the system.
[0003] In the power system, secondary equipment refers to equipment used to monitor, measure, control and protect primary equipment (such as transformers, transmission lines, etc.). Secondary equipment includes relay protection devices, automation systems, communication equipment, etc. However, secondary equipment failure may cause unstable operation of the power system and even cause large-scale power outages, so the demand for fault monitoring is very urgent.
[0004] Existing methods for monitoring faults in secondary equipment in power systems mainly rely on regular manual inspections and some threshold-based automatic monitoring systems. Manual inspections are costly, inefficient, and difficult to detect problems in real time; although threshold-based automatic monitoring systems can monitor in real time, they are often not accurate enough in identifying faults and are easily affected by environmental interference and equipment aging. Summary of the invention
[0005] In order to improve the accuracy of fault identification, the present invention provides a secondary equipment fault monitoring method and system for a power system.
[0006] In a first aspect, the present invention provides a secondary equipment fault monitoring method for a power system, which adopts the following technical solution:
[0007] A method for monitoring secondary equipment faults in a power system, comprising the steps of:
[0008] Acquire monitoring data of the equipment, where the monitoring data is voltage, current or temperature; generate a predicted value according to a preset prediction model, update the initial threshold used to determine whether the monitoring data is abnormal to obtain the latest threshold, and generate a fault alarm signal when the predicted value is greater than the latest threshold; wherein, the method for updating the initial threshold is: acquire historical data of monitoring data from the equipment to be analyzed and other similar equipment in the area; calculate the spatial weight, time weight and failure rate weight, and use the product of the spatial weight, time weight and failure rate weight as the comprehensive weight, and use the average of the cumulative sum of the products of the comprehensive weight and the monitoring data of other secondary equipment as the latest threshold of the equipment to be analyzed at the next moment, and repeat the update, wherein the latest threshold of the previous update is used as the initial threshold of the next update, and the monitoring data is voltage, current or temperature.
[0009] Optionally, the spatial weight is calculated as follows: respectively calculate the Euclidean distances between the device to be analyzed and other devices of the same type, perform Gaussian distribution on all Euclidean distances, and obtain the first standard deviation of the first Gaussian distribution; normalize the ratio of the square of the Euclidean distance to twice the square of the first standard deviation and use it as the spatial weight.
[0010] Optionally, the time weight is calculated as follows: calculate the absolute value of the difference between the current moment and each historical moment, perform Gaussian distribution on the absolute value of the difference, and obtain the second standard deviation of the second Gaussian distribution; normalize the ratio of the absolute value of the difference to twice the square of the second standard deviation and use it as the spatial weight.
[0011] Optionally, a cycle is set, and the ratio of the number of failures to the cycle is used as the failure rate; the difference in failure rates between adjacent cycles is calculated, and the L2 norm of the failure rate difference is calculated; the failure rate difference is Gaussian distributed to obtain the third standard deviation of the third Gaussian distribution; and the normalized result of the ratio of the L2 norm of the failure rate difference to twice the square of the third standard deviation is used as the failure rate weight.
[0012] Optionally, the prediction model setting method is: setting a data set, the data set includes: voltage, current, temperature and time; inputting the data set into a preset neural network model to perform model training on the neural network model to obtain a prediction model.
[0013] Optionally, the loss function for model training is the mean square error loss function.
[0014] In a second aspect, the present invention provides a secondary equipment fault monitoring system for a power system, which adopts the following technical solution:
[0015] A secondary equipment fault monitoring system for an electric power system comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the secondary equipment fault monitoring method for an electric power system is implemented.
[0016] The beneficial effect is that the above-mentioned method for monitoring faults of secondary equipment in the power system is used to generate a computer program, and is stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.
[0017] Compared with the prior art, the present invention has the following technical effects:
[0018] 1. By setting the spatial weight, time weight and failure rate weight, within the period window in the area, for a secondary device, the sum of the comprehensive weights of the same type of secondary devices in the area and the corresponding operating status values is calculated, and the mean of the sum is taken as the latest threshold. The predicted value of the operating status is generated through the prediction model. When the predicted value is greater than the threshold, an alarm signal is issued, which can predict and alarm the fault in advance and improve the accuracy of fault identification.
[0019] 2. By comparing the secondary devices of the same type in the same area, a standard value of the operating status (i.e. the latest threshold) is given to each secondary device, so that the threshold setting can refer to the environment and the surrounding equipment. That is, it provides an adaptive threshold setting method that is not easily affected by environmental interference and equipment aging, thereby improving the accuracy of fault identification. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.
[0021] Figure 1 This is a method flow chart of a method for monitoring secondary equipment faults in an electric power system according to an embodiment of the present invention; the equipment in the figure all refers to secondary equipment. DETAILED DESCRIPTION
[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0023] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.
[0024] The embodiment of the present invention discloses a secondary equipment fault monitoring method for an electric power system, comprising the steps of: acquiring monitoring data of the equipment, the monitoring data being voltage, current or temperature; generating a prediction value according to a preset prediction model, updating an initial threshold used to determine whether the monitoring data is abnormal to obtain a latest threshold, and generating a fault alarm signal when the prediction value is greater than the latest threshold.
[0025] Among them, refer to Figure 1 , the method to update the initial threshold is:
[0026] S1: Obtain historical data of monitoring data from the secondary equipment to be analyzed and other similar secondary equipment in the area.
[0027] In one embodiment, the monitoring data may be data of multiple dimensions such as voltage, current or temperature. The present invention takes data of one dimension of current data as an example, and the same is true for data of other dimensions. Exemplarily, the region is the entire power system area. When the secondary device to be analyzed is a current detection device, the same type of device refers to other current detection devices in the same region.
[0028] S2: Calculate the spatial weight, time weight and failure rate weight, take the product of the spatial weight, time weight and failure rate weight as the comprehensive weight, take the average of the cumulative sum of the products of the comprehensive weight and the monitoring data of other secondary equipment as the latest threshold of the secondary equipment to be analyzed at the next moment, and update repeatedly, where the latest threshold of the previous update is used as the initial threshold of the next update.
[0029] In one embodiment, the spatial weight is calculated as follows: the Euclidean distances between the secondary device to be analyzed and the secondary devices of the same type are calculated respectively, all the Euclidean distances are Gaussian distributed, and the first standard deviation of the first Gaussian distribution is obtained; the ratio of the square of the Euclidean distance to twice the square of the first standard deviation is normalized and used as the spatial weight.
[0030] The spatial weight is expressed mathematically as:
[0031]
[0032] Among them, r a represents the spatial weight, exp represents the exponential function with e as the base, |||xx i ∣∣ indicates that the secondary device x to be analyzed is different from the secondary device x of the same type i The Euclidean distance between a Indicates the first standard deviation.
[0033] For the secondary equipment to be analyzed, the operating data of the same type of secondary equipment can reflect the normal operating status of the current secondary equipment, that is, for the current detection device to be analyzed, the operating data of other current monitoring devices of the same type reflect the normal operating status of the current detection device to be analyzed. Therefore, this proximity is reflected by setting spatial weights. The closer the distance to the adjacent device, the higher the weight; conversely, the farther the distance, the lower the weight.
[0034] In one embodiment, the time weight is calculated as follows: the absolute value of the difference between the current moment and each historical moment is calculated respectively, the absolute value of the difference is Gaussian distributed, and the second standard deviation of the second Gaussian distribution is obtained; the ratio of the absolute value of the difference to twice the square of the second standard deviation is normalized and used as the spatial weight.
[0035] The mathematical expression of time weight is:
[0036]
[0037] Among them, r b represents the time weight, exp represents the exponential function with e as the base, y represents the current time, and y j represents the historical moment j, σ b represents the second standard deviation.
[0038] Newer data may better reflect the current status of secondary equipment, so recent data can be given a higher weight, that is, the smaller the time difference, the higher the time weight; conversely, the larger the time difference, the lower the time weight.
[0039] In one embodiment, the calculation method of the failure rate weight is as follows: set a period, and use the ratio of the number of failures to the period as the failure rate; calculate the difference in failure rates between adjacent periods, and calculate the L2 norm of the failure rate difference; Gaussian distribute the failure rate difference to obtain the third standard deviation of the third Gaussian distribution; and use the normalized result of the ratio of the L2 norm of the failure rate difference to the square of twice the third standard deviation as the failure rate weight. Exemplarily, the period can be one month or one week.
[0040] The mathematical expression of the failure rate weight is:
[0041]
[0042] Among them, r c represents the failure rate weight, exp represents the exponential function with e as the base, and zz n Represents the failure rate difference between adjacent cycles. Gaussian distribution is performed on all failure rate differences. σ c Denotes the third standard deviation.
[0043] The failure rate weight expresses the fluctuation of the failure rate change. The greater the fluctuation of the failure rate change, the higher the possibility that the operation of the secondary equipment itself is abnormal, so the failure rate weight of the equipment is smaller.
[0044] The product of the spatial weight, the time weight and the fault rate weight is used as the comprehensive weight, the comprehensive weight and the product of the monitoring data of other secondary devices of the same type are accumulated, and the mean of the accumulated sum is used as the latest threshold of the secondary device to be analyzed at the next moment. Exemplarily, for the current detection device to be analyzed, the respective comprehensive weights of other current monitoring devices are calculated, and the mean of the accumulated sum of the current value of each current detection device and the product of their respective comprehensive weights is used as the latest threshold.
[0045] A prediction value is generated according to the preset prediction model. When the prediction value is greater than the latest threshold, a fault alarm signal is generated. Otherwise, no alarm is generated.
[0046] Specifically, a data set is set, and the data set includes: voltage, current, temperature and time; the data set is put into a preset neural network model to train the neural network model to obtain a prediction model. The loss function of the model training is the mean square error loss function. The neural network model uses the LSTM (Long Short-Term Memory) network model. The LSTM network model is an existing technology, and the model construction and training are existing technologies, which will not be repeated here.
[0047] An embodiment of the present invention further discloses a secondary equipment fault monitoring system for a power system, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a secondary equipment fault monitoring method for a power system according to the present invention is implemented.
[0048] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.
[0049] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high bandwidth memory HBM (High Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.
[0050] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.
[0051] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A secondary equipment fault monitoring method for a power system, characterized in that: Includes steps: Acquire monitoring data of secondary equipment; the monitoring data is voltage, current or temperature; Generate a prediction value according to a preset prediction model, update the initial threshold used to determine whether the monitoring data is abnormal to obtain the latest threshold, and generate a fault alarm signal when the prediction value is greater than the latest threshold; Among them, the method for updating the initial threshold is: Obtain historical data of monitoring data from the secondary equipment to be analyzed and the same type of secondary equipment in the area; Calculate the spatial weight, time weight and failure rate weight, take the product of the spatial weight, time weight and failure rate weight as the comprehensive weight, take the average of the cumulative sum of the product of the comprehensive weight and the monitoring data as the latest threshold of the secondary equipment to be analyzed at the next moment, and update repeatedly; among which, the latest threshold of the previous update is used as the initial threshold of the next update.
2. The secondary equipment fault monitoring method for a power system according to claim 1, characterized in that: The calculation method of spatial weight is: Calculate the Euclidean distances between the secondary device to be analyzed and the devices of the same type respectively, perform Gaussian distribution on all Euclidean distances, and obtain the first standard deviation of the first Gaussian distribution; The ratio of the square of the Euclidean distance to twice the square of the first standard deviation is normalized and used as the spatial weight.
3. The secondary equipment fault monitoring method for a power system according to claim 1, characterized in that: The time weight is calculated as: Calculate the absolute value of the difference between the current moment and each historical moment respectively, perform Gaussian distribution on the absolute value of the difference, and obtain the second standard deviation of the second Gaussian distribution; The ratio of the absolute value of the difference to twice the square of the second standard deviation is normalized and used as the spatial weight.
4. The secondary equipment fault monitoring method for a power system according to claim 1, characterized in that: The calculation method of failure rate weight is: Set the cycle and take the ratio of the number of failures to the cycle as the failure rate; Calculate the failure rate difference between adjacent cycles and calculate the L2 norm of the failure rate difference; Performing Gaussian distribution on the failure rate difference to obtain a third standard deviation of the third Gaussian distribution; The normalized result of the ratio of the L2 norm of the failure rate difference to twice the square of the third standard deviation is used as the failure rate weight.
5. The method for monitoring secondary equipment faults in a power system according to claim 1, characterized in that: The preset method of the prediction model is: setting a data set, the data set includes: voltage, current, temperature and time; inputting the data set into a preset neural network model to train the neural network model to obtain a prediction model.
6. The method for monitoring secondary equipment faults in a power system according to claim 5, characterized in that: The loss function for model training is the mean square error loss function.
7. A secondary equipment fault monitoring system for a power system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the secondary equipment fault monitoring method for a power system according to any one of claims 1 to 6 is implemented.