Electric power system fault intelligent emergency processing method based on big data

Abstract

The invention relates to the technical field of power system fault processing, in particular to a power system fault intelligent emergency processing method based on big data, and the method comprises the steps: collecting various data in a power system; based on the voltage data and the temperature data of the key nodes and the vibration data of the equipment, the attention weight of the local power grid is determined; based on the frequency data of the local power grid where the key node is located and the voltage data of the key node, combining the attention weight of the local power grid to obtain a short-circuit fault index of the key node; and based on the short-circuit fault indexes and distances of all the key nodes, determining an optimal troubleshooting sequence of the key nodes by adopting an optimization algorithm, and carrying out emergency processing on the power system fault by utilizing the optimal troubleshooting sequence. The invention aims to optimize the troubleshooting sequence after the fault occurs, shorten the troubleshooting processing time, and improve the short circuit fault positioning efficiency, thereby facilitating the rapid repair of the fault.

Description

Technical Field

[0001] The present application relates to the technical field of power system fault processing, and specifically to an intelligent emergency processing method for power system faults based on big data. Background Art

[0002] With the continued growth of global energy demand and the increasing complexity of power grids, ensuring the stability and reliability of power systems faces unprecedented challenges. Exploring and implementing a big data-based intelligent emergency response method for power system faults has far-reaching significance for improving the digital transformation and intelligent management of the entire energy industry. In power systems, short-circuit faults are relatively detrimental, potentially damaging equipment or components and impacting power supply in corresponding lines or subgrids.

[0003] Existing technologies typically focus on troubleshooting and addressing the fault location after it occurs, based on the moment of fault occurrence and subsequent changes in power data. However, this approach can only determine the approximate fault area. When the short circuit affects a large area, troubleshooting is very slow. Short circuit faults exhibit certain characteristics before they occur. While these characteristics are insufficient to accurately predict a short circuit, they can focus on subgrids, lines, or critical nodes. If these pre-fault characteristics are not captured promptly and accurately, and the data is not prioritized, troubleshooting can take a long time, hindering rapid fault repair. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides an intelligent emergency processing method for power system faults based on big data to solve the existing problems.

[0005] The present application discloses a method for intelligent emergency response to power system faults based on big data, which adopts the following technical solutions: An embodiment of the present application provides a method for intelligent emergency handling of power system faults based on big data, the method comprising the following steps: Collect various data in the power system; the various data include temperature data and voltage data of each key node in the power system, vibration data of each device in the power system, and frequency data of the local power grid where the power system is located; Determine the importance of the local power grid based on the degree of distortion of the voltage data of each key node in the frequency domain signal, the temperature data of each key node, and the degree of the upward trend of the vibration data of each device; Based on the frequency data of the local power grid where the key node is located and the degree of the downward trend of the voltage data of the key node, and weighted by the importance weight of the local power grid where the key node is located, the short circuit fault index of the corresponding key node is obtained; An objective function is constructed based on the short-circuit fault index of all key nodes and the distance between key nodes. The optimal troubleshooting sequence of key nodes is determined using an optimization algorithm using the objective function, and the optimal troubleshooting sequence is used to perform emergency processing on power system faults.

[0006] Preferably, the key node is an important hub or component for ensuring the stability and safety of the power system.

[0007] Preferably, the frequency data is: In the local power grid where the power system is located, Class A power quality online monitoring devices are deployed in all substations and all major power centers to monitor the frequency of the AC power at each collection moment; The mean of all frequencies at each acquisition moment is calculated as the frequency data of the local power grid at the corresponding acquisition moment.

[0008] Preferably, the calculation formula for the importance weight of the local power grid is: ; is the weight of the local power grid, It is the sum of the distortion of the voltage data of all key nodes in the local power grid in the frequency domain signal. is the sum of the upward trend of temperature data of all key nodes in the local power grid, It is the sum of the upward trend of the vibration data of all devices in the local power grid; among them, the distortion degree of each key node is: the voltage data of the corresponding key node is converted into a spectrum diagram, the spectrum diagram is used as the input of the total harmonic distortion, and the THD value is output.

[0009] Preferably, the degree of the upward trend is specifically: The sequence of similar data in the order of collection time is used as the input of the exponential smoothing method, and the smoothed values ​​of each collection time output are combined into a smoothed sequence; The cumulative sum of the differences between all adjacent elements in the smoothed sequence is used as the degree of the upward trend of the corresponding similar data; among them, the difference between adjacent elements is the difference between the adjacent subsequent element and the previous element.

[0010] Preferably, the calculation formula of the failure index of the key node is: ; is the short-circuit fault index of the key node, is the normalized weight of the local power grid where the key node is located, 、 They are respectively the voltage drop degree of the voltage data of the key nodes and the frequency drop degree of the frequency data of the local power grid where the key nodes are located.

[0011] Preferably, the method for acquiring the voltage drop degree of the voltage data and the frequency drop degree of the frequency data is specifically as follows: The voltage drop degree and the frequency drop degree are collectively referred to as the drop degree; the sequence of the voltage data in the order of acquisition time and the sequence of the frequency data in the order of acquisition time are both used as the sequence to be analyzed; The moving average method is used to obtain the moving average of the sequence to be analyzed at each acquisition moment to form the moving average sequence to be analyzed; The slope of the fitted straight line of the moving average sequence to be analyzed is recorded as the degree of decline of the sequence to be analyzed.

[0012] Preferably, the output value of the objective function is the total investigation cost, and the calculation formula of the total investigation cost is: ; 、 are the normalized short circuit fault index and the normalized distance weight respectively, is an exponential function with a natural constant as its base.

[0013] Preferably, the determining the optimal troubleshooting sequence of key nodes by using an optimization algorithm using an objective function includes: based on the objective function, taking the distance matrix and the short-circuit fault set as inputs of an ant colony optimization algorithm, and outputting the optimal troubleshooting sequence.

[0014] Preferably, any element in the distance matrix is ​​the distance between two key nodes corresponding to its row and column; the short-circuit fault set is a set consisting of short-circuit fault indices of all key nodes.

[0015] This application has at least the following beneficial effects: This application calculates the importance weight of the local power grid based on the degree of distortion of the voltage data of each key node in the local power grid on the frequency domain signal before the short-circuit fault occurs, the temperature data of the key nodes and the rising trend change of the vibration signal of the equipment, and is used to measure the degree of appearance of the precursor characteristics of the short-circuit fault in the local power grid; after the short-circuit fault occurs, the short-circuit fault index of each key node is calculated based on the detected frequency data of the local power grid and the falling trend change of the voltage data of the key nodes, and the degree of appearance of the precursor characteristics of the short-circuit fault in the local power grid, and is used to measure the probability of a short-circuit fault occurring at each key node. In this way, not only the changes in the data after the short-circuit fault occurs are taken into account, but also the characteristics shown before the fault occurs are used for weighting, which optimizes the troubleshooting order after the fault occurs, shortens the troubleshooting time, and improves the efficiency of short-circuit fault locating, thereby facilitating the rapid repair of the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. 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 any creative work.

[0017] Figure 1 A flowchart of a method for intelligent emergency handling of power system faults based on big data provided in one embodiment of the present application; Figure 2 A flow chart of a method for analyzing the importance weight of a local power grid provided in one embodiment of the present application; Figure 3 A flow chart of a method for analyzing a short-circuit fault index of a key node provided in one embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to further illustrate the technical means and effects adopted by this application to achieve the predetermined invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation method, structure, features and effects of an intelligent emergency handling method for power system faults based on big data proposed in this application. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0020] The following describes in detail a specific solution of an intelligent emergency handling method for power system faults based on big data provided by this application in conjunction with the accompanying drawings.

[0021] An embodiment of the present application provides an intelligent emergency handling method for power system faults based on big data.

[0022] Specifically, we provide the following intelligent emergency response method for power system faults based on big data, please refer to Figure 1 , the method comprises the following steps: The first step is to collect various data in the power system; the various data include temperature data and voltage data of each key node in the power system, vibration data of each device in the power system, and frequency data of the local power grid where the power system is located.

[0023] Temperature and voltage data are collected through sensors installed at key nodes in the power system, including but not limited to substations, distribution stations, transformers, and distribution lines. Temperature signals are collected through temperature sensors, and voltage signals are collected through voltage transformers. Key nodes are nodes that are crucial to the operation and control of the power system. They are responsible for tasks such as power transmission and distribution, and are important hubs or components that ensure the stability and security of the power system.

[0024] Vibration sensors are installed on various devices in the power system to collect vibration signals from these devices, including but not limited to transformers, generators, and motors. The specific locations of the sensors at key nodes and on various devices in the power system are determined by the implementer based on actual conditions.

[0025] At a preset time interval t, temperature, voltage, and vibration data are acquired at each acquisition moment. The data collected within the acquisition period T is organized into a temperature data sequence, a voltage data sequence, and a vibration data sequence, respectively. In this embodiment, t and T are set to 1 second and 1 minute, respectively. The implementer can set these values ​​based on actual conditions.

[0026] As the power system transmits electricity to the transmission lines, Class A online power quality monitoring devices are deployed at each substation and major power center along the transmission lines. These devices monitor and collect the AC power frequency at all substations and major power centers, with a data collection period of T. All collected data is preprocessed, and invalid data or missing values ​​are corrected using a forward-filling method. This forward-filling method is well known, and the detailed process is omitted here.

[0027] In the same local power grid, the frequencies collected at all substations and all major power centers at the same time are averaged and processed in this way. The averages of the frequencies calculated at all times are used to form the frequency sequence of the current corresponding local power grid in chronological order. Using this method, the frequency sequence of each local power grid can be obtained.

[0028] At this point, various types of data in the power system can be collected through the above method, including temperature data series and voltage data series of each key node, vibration data series of each device in the power system, and frequency series of each local power grid.

[0029] The second step is to determine the importance weight of the local power grid based on the distortion degree of the voltage data of each key node in the frequency domain signal, the temperature data of each key node and the rising trend degree of the vibration data of each device.

[0030] Short-circuit failures, in addition to operational errors by power workers, are largely due to aging equipment or deteriorating insulation materials. Aging equipment or deteriorating insulation materials in power systems can lead to local overheating, manifesting as a slow, long-term temperature rise. Because power systems have certain compensation mechanisms, these temperature changes only affect the same local power grid, or even just the nearest few key nodes. Secondly, loose components in power system equipment can cause short-circuit failures. Before a short-circuit occurs, the current may be abnormally distributed, increasing the electromotive force in the local power grid and causing increased equipment vibration. Finally, insulation degradation can increase the voltage harmonic content, making the voltage waveform irregular, exhibiting sawtooth or other non-sinusoidal characteristics.

[0031] Based on the above characteristics, this application analyzes the importance weight of each local power grid. The flow chart of the local power grid importance weight analysis method is shown in the attached figure. Figure 2 This embodiment analyzes the importance weight of any local power grid, specifically: Calculate the sum of the rising trend of the temperature data of all key nodes in the local power grid; calculate the sum of the rising trend of the vibration data of all devices in the local power grid; calculate the sum of the distortion of the voltage data of all key nodes in the local power grid in the frequency domain signal; and determine the importance weight of the local power grid based on these three sums. The calculation formula for the importance weight of the local power grid is: ; is the weight of the local power grid, It is the sum of the distortion of the voltage data of all key nodes in the local power grid in the frequency domain signal. is the sum of the upward trend of temperature data of all key nodes in the local power grid, It is the sum of the upward trend of vibration data of all equipment in the local power grid.

[0032] The specific process of obtaining is as follows: In this embodiment, the degree of distortion of the voltage data at a key node in the frequency domain is measured using the total harmonic distortion (THD) value output by the THD calculation. The specific method involves converting the voltage data sequence at the key node into a frequency spectrum, using the frequency spectrum as the input for the total harmonic distortion (THD) calculation, and outputting the THD value. The THD calculation process is well-known and will not be described in detail. The algorithms used to convert the frequency spectrum include Fourier transforms, wavelet transforms, and other well-known techniques, and will not be described in detail.

[0033] THD is an important indicator of signal quality, representing the overall distortion of each harmonic component in a signal. When a signal contains the fundamental frequency and its integer multiples of harmonics, these harmonics, which may be present due to device nonlinearities or other reasons, are called harmonic distortion. A higher THD value indicates more severe signal distortion.

[0034] In other embodiments of the present application, the spectrum flatness of the frequency spectrum of the voltage data of a key node may be used as the degree of distortion of the voltage data of the corresponding key node in the frequency domain signal.

[0035] Thus, the sum of the distortion levels of the voltage data of all key nodes in the local power grid in the frequency domain signal is obtained.

[0036] The process of obtaining the degree of the upward trend is specifically as follows: For each key node, the temperature data of each key node at different times are the same type of data. For the equipment in the local power grid, the vibration data of each equipment at different times are the same type of data.

[0037] The sequence of similar data in the order of collection time is used as the input of the exponential smoothing method, and the smoothed values ​​of each collection time are output to form a smoothed sequence; the cumulative sum of the differences between all adjacent elements in the smoothed sequence is used as the degree of the upward trend of the corresponding similar data; among which, the difference between adjacent elements is the difference between the adjacent subsequent element and the previous element.

[0038] In other embodiments of the present application, the degree of upward trend may be determined by forming a sequence of similar data in the order of their collection time, calculating the slope of a straight line fitted to all elements in the sequence, and using the slope as the degree of upward trend for the corresponding similar data. The process of straight line fitting is a well-known technique, and can be performed using a least squares method, principal component analysis, or the like. These methods are well-known and will not be further described.

[0039] The specific process of obtaining is as follows: The temperature data series of each key node is used as the input of the exponential smoothing method, and the smoothing coefficient of the exponential smoothing method is preset. , the exponential smoothing method is used to calculate the smoothing value of each key node at each acquisition moment, and the smoothing values ​​are combined into a temperature smoothing sequence of each key node according to the order of acquisition moments. In this embodiment, the smoothing coefficient The value is 0.3. The exponential smoothing method is a well-known technology, and the specific process will not be described in detail.

[0040] The cumulative sum of the differences between all adjacent elements in the temperature smoothing sequence of each key node is used as the degree of the upward trend of the temperature data of the corresponding key node; wherein the difference between adjacent elements is the difference between the adjacent subsequent element and the previous element.

[0041] In other embodiments of the present application, all elements in the temperature data sequence are fitted into the slope of a straight line as the degree of the rising trend of the temperature data of the corresponding key node.

[0042] Thus, the sum of the rising trend degrees of the temperature data of all key nodes in the local power grid is obtained.

[0043] The specific process of obtaining is as follows: The vibration data sequence of each device is used as the input of the exponential smoothing method, and the smoothing coefficient of the exponential smoothing method is preset. ,The exponential smoothing method is used to calculate the smoothing value of each device at each acquisition moment, and the smoothing values ​​are combined into a vibration smoothing sequence for each device according to the order of acquisition moments.

[0044] The cumulative sum of the differences between all adjacent elements in the vibration smoothing sequence of each device is used as the degree of the upward trend of the vibration data of the corresponding device; wherein the difference between adjacent elements is the difference between the adjacent subsequent element and the previous element.

[0045] In other embodiments of the present application, all elements in the vibration data sequence are fitted into the slope of a straight line as the degree of the upward trend of the vibration data of the corresponding device.

[0046] Thus, the sum of the upward trend degrees of the vibration data of all devices in the local power grid is obtained.

[0047] It is understandable that before a short circuit occurs, the aging of equipment or the deterioration of insulation materials will cause the temperature in the local power grid to rise, and at the same time, the vibration of the equipment in the local power grid will increase, so that the temperature smoothing sequence of each key node or the vibration smoothing sequence of the equipment will show an upward trend, and the overall performance will be an increasing sequence, thus and and due to the increase of voltage harmonics, the voltage signal will deviate from the ideal sine wave, which will make the THD value larger, that is, Therefore, before a short-circuit fault occurs, the more obvious the precursor characteristics of the short-circuit fault in the local power grid are, the greater the probability of a short circuit, and the greater the corresponding weight of the local power grid.

[0048] The third step is to obtain the short circuit fault index of the corresponding key node based on the frequency data of the local power grid where the key node is located and the degree of the downward trend of the voltage data of the key node, and weighting it using the importance weight of the local power grid where the key node is located.

[0049] When a short circuit occurs, the frequency of the local grid affected by the short circuit, or any grid that is strongly coupled with the local grid where the short circuit occurs (i.e., exchanges a large amount of current), will also decrease in frequency. The more severe the impact of the short circuit, the greater the frequency drop. Simultaneously, the voltage will also decrease due to the short circuit, and the more severe the impact, the greater the voltage drop.

[0050] Based on the above characteristics, this application analyzes the short circuit fault index of each key node. The flow chart of the short circuit fault index analysis method of the key node is shown in the attached figure. Figure 3 As shown, this embodiment analyzes the short circuit fault index of any key node, specifically: Calculate the voltage drop degree of the voltage data of the key node; calculate the frequency drop degree of the frequency data of the local power grid where the key node is located; determine the short circuit fault index of the key node based on the voltage drop degree, frequency drop degree and the normalized weight of the local power grid where the key node is located. The calculation formula of the short circuit fault index of the key node is: ; is the short-circuit fault index of the key node, is the normalized weight of the local power grid where the key node is located, 、 They are respectively the voltage drop degree of the voltage data of the key nodes and the frequency drop degree of the frequency data of the local power grid where the key nodes are located.

[0051] The method for obtaining the voltage drop degree of the voltage data and the frequency drop degree of the frequency data is specifically as follows: The voltage drop degree and the frequency drop degree are collectively referred to as the drop degree; the sequence of the voltage data arranged in the order of acquisition time and the sequence of the frequency data arranged in the order of acquisition time are both used as the sequence to be analyzed; the moving average method is used to obtain the moving average of the sequence to be analyzed at each acquisition moment to form the moving average sequence to be analyzed; the slope of the fitting line of the moving average sequence to be analyzed is recorded as the drop degree of the sequence to be analyzed.

[0052] In other embodiments of the present application, the cumulative sum of all elements in the differential sequence of the moving average sequence to be analyzed can also be used as the degree of decline of the sequence to be analyzed. The method for obtaining the differential sequence is a well-known technique and will not be described in detail.

[0053] in, and The specific process of obtaining is as follows: Take the voltage data sequence of the key nodes as input, use the moving average method, and set the window size , get the moving average of each acquisition moment, and construct the voltage moving average sequence of the key nodes in chronological order. The value is 20. The moving average method is a well-known technology and the specific process will not be described in detail.

[0054] The frequency sequence of the local power grid where the key node is located is used as input, and the moving average method is used to set the window size. , get the moving average of each acquisition moment, and construct the frequency moving average sequence of the local power grid according to the time sequence.

[0055] In this embodiment, the slope of the fitted straight line of the voltage moving average sequence of the key node is used as the voltage drop degree of the voltage data of the key node; the slope of the fitted straight line of the frequency moving average sequence of the local power grid where the key node is located is used as the frequency drop degree of the frequency data of the local power grid.

[0056] In other embodiments of the present application, the cumulative sum of all elements in the differential sequence of the moving average sequence of the voltage of the key node can be used as the voltage drop degree of the voltage data of the key node. The cumulative sum of all elements in the differential sequence of the moving average sequence of the frequency of the local power grid can be used as the frequency drop degree of the frequency data of the mobile power grid.

[0057] It is understandable that when a short circuit occurs, the local power grid where the key node where the short circuit occurs is located and the local power grid with a strong coupling relationship with it will experience a frequency drop, and the greater the impact of the short circuit, the greater the degree of frequency drop, that is, is a negative number and At the same time, the voltage of the key node affected by the short circuit fault will also show a downward trend, and the greater the impact of the short circuit, the greater the degree of voltage drop, that is, is a negative number and The larger the value of and the greater the weight of the local power grid, the greater the probability of a short circuit fault occurring in the corresponding local power grid. Therefore, when a short circuit fault occurs, the larger the short circuit fault index of the key node, the more likely the corresponding key node is to be the node where the short circuit occurs, or the node that is severely affected by the short circuit. Conversely, the less the key node is affected by the short circuit, the smaller its short circuit fault index is.

[0058] The fourth step is to construct an objective function based on the short-circuit fault index of all key nodes and the distance between key nodes, use the objective function to adopt an optimization algorithm to determine the optimal troubleshooting order of key nodes, and use the optimal troubleshooting order to perform emergency treatment on power system faults.

[0059] According to the third step, the short circuit fault index of each key node can be calculated.

[0060] When a short-circuit fault occurs, the short-circuit fault index of the key node is calculated by combining the frequency data of the local power grid where the key node is located and the voltage data changes of the key node. Short-circuit troubleshooting is started from the key node with the largest short-circuit fault index. This way, the location of the fault can be quickly found in a short time.

[0061] In order to quickly determine the fault location of the power system, the present application forms a short-circuit fault set with the short-circuit fault indices of all key nodes.

[0062] At the same time, taking into account the commuting costs caused by the distance between different key nodes during the investigation, the three-dimensional coordinate information of each key node is used to obtain the distance matrix. The rows and columns of the distance matrix are the key nodes, and the number of rows and columns of the distance matrix is ​​the number of corresponding key nodes. Any element in the distance matrix is ​​the distance between the two key nodes corresponding to its row and column.

[0063] In this embodiment, Manhattan distance is used as the distance between two key nodes. In other embodiments, the Euclidean distance between two key nodes can also be used as the distance between them. The calculation of Manhattan distance is a well-known technique and will not be described in detail.

[0064] Using the distance matrix and the short-circuit fault set as input, the ant colony optimization algorithm is used. The objective function is the total troubleshooting cost, and the optimal troubleshooting sequence is output to minimize the total troubleshooting cost. The total troubleshooting cost is a function of fault severity and travel distance. A greater travel distance indicates a higher troubleshooting cost, while a greater fault severity indicates a lower troubleshooting cost.

[0065] Among them, the ant colony optimization algorithm is a well-known technology, and the specific process is not repeated here. In other embodiments of the present application, other optimization algorithms, such as greedy algorithm, particle swarm optimization algorithm, etc., can also be used to output the optimal screening order using the minimum screening cost.

[0066] In this embodiment, the number of ants in the ant colony optimization algorithm is set to 10% of the number of all key nodes in the power system, the maximum number of iterations is 150, the volatility factor is 0.3, and the pheromone factor is 2. The output of the objective function of the ant colony optimization algorithm is the total troubleshooting cost, and the calculation formula for the total troubleshooting cost is: ; 、 are respectively the normalized short circuit fault index and the normalized distance weight. In this embodiment, Take the value 0.6, The value is 0.4; the normalization adopts the maximum-minimum value normalization, which is a well-known technology and will not be described in detail. It is an exponential function with the natural constant e as its base.

[0067] It should be noted that, since the value of the short-circuit fault index may be both positive and negative, the purpose of using the exponential function is to avoid the denominator being 0.

[0068] In this way, the changing characteristics of various data before and after the short-circuit fault occurs are taken into account, allowing staff to check the key nodes one by one from front to back in the optimal troubleshooting order, thereby finding the specific location of the short-circuit fault as quickly as possible and performing corresponding fault processing, which is conducive to the rapid repair of the fault.

[0069] The various embodiments in this application are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0070] It should be noted that, unless otherwise specified and limited, terms such as "include", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, the phrase "including a ..." defines an element, does not exclude the presence of other identical elements in the article or device including the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items.

[0071] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not invented herein.

[0072] It will be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for intelligent emergency handling of power system faults based on big data, characterized in that: The method comprises the following steps: Collect various data in the power system; the various data include temperature data and voltage data of each key node in the power system, vibration data of each device in the power system, and frequency data of the local power grid where the power system is located; Determine the importance of the local power grid based on the degree of distortion of the voltage data of each key node in the frequency domain signal, the temperature data of each key node, and the degree of the upward trend of the vibration data of each device; Based on the frequency data of the local power grid where the key node is located and the degree of the downward trend of the voltage data of the key node, and weighted by the importance weight of the local power grid where the key node is located, the short circuit fault index of the corresponding key node is obtained; An objective function is constructed based on the short-circuit fault index of all key nodes and the distance between key nodes. The optimal troubleshooting sequence of key nodes is determined using an optimization algorithm using the objective function, and the optimal troubleshooting sequence is used to perform emergency processing on power system faults.

2. The intelligent emergency handling method for power system faults based on big data according to claim 1, characterized in that: The key nodes are important hubs or components that ensure the stability and safety of the power system.

3. The intelligent emergency handling method for power system faults based on big data according to claim 1, characterized in that: The frequency data are: In the local power grid where the power system is located, Class A power quality online monitoring devices are deployed in all substations and all major power centers to monitor the frequency of the AC power at each collection moment; The mean of all frequencies at each acquisition moment is calculated as the frequency data of the local power grid at the corresponding acquisition moment.

4. The intelligent emergency handling method for power system faults based on big data according to claim 1, characterized in that: The calculation formula of the importance weight of the local power grid is: ; is the weight of the local power grid, It is the sum of the distortion of the voltage data of all key nodes in the local power grid in the frequency domain signal. is the sum of the upward trend of temperature data of all key nodes in the local power grid, It is the sum of the upward trend of the vibration data of all devices in the local power grid; among them, the distortion degree of each key node is: the voltage data of the corresponding key node is converted into a spectrum diagram, the spectrum diagram is used as the input of the total harmonic distortion, and the THD value is output.

5. The intelligent emergency handling method for power system faults based on big data according to claim 4, characterized in that: The degree of the upward trend is specifically: The sequence of similar data in the order of collection time is used as the input of the exponential smoothing method, and the smoothed values ​​of each collection time output are combined into a smoothed sequence; The cumulative sum of the differences between all adjacent elements in the smoothed sequence is used as the degree of the upward trend of the corresponding similar data; among them, the difference between adjacent elements is the difference between the adjacent subsequent element and the previous element.

6. The intelligent emergency handling method for power system faults based on big data according to claim 4, characterized in that: The calculation formula of the failure index of the key node is: ; is the short-circuit fault index of the key node, is the normalized weight of the local power grid where the key node is located, 、 They are respectively the voltage drop degree of the voltage data of the key nodes and the frequency drop degree of the frequency data of the local power grid where the key nodes are located.

7. The intelligent emergency handling method for power system faults based on big data according to claim 6, characterized in that: The method for obtaining the voltage drop degree of the voltage data and the frequency drop degree of the frequency data is specifically as follows: The voltage drop degree and the frequency drop degree are collectively referred to as the drop degree; the sequence of the voltage data in the order of acquisition time and the sequence of the frequency data in the order of acquisition time are both used as the sequence to be analyzed; The moving average method is used to obtain the moving average of the sequence to be analyzed at each acquisition moment to form the moving average sequence to be analyzed; The slope of the fitted straight line of the moving average sequence to be analyzed is recorded as the degree of decline of the sequence to be analyzed.

8. The intelligent emergency handling method for power system faults based on big data according to claim 6, characterized in that: The output value of the objective function is the total investigation cost, and the calculation formula of the total investigation cost is: ; 、 are the normalized short circuit fault index and the normalized distance weight respectively, is an exponential function with a natural constant as its base.

9. The intelligent emergency handling method for power system faults based on big data according to claim 8, characterized in that: The method of using the objective function and the optimization algorithm to determine the optimal troubleshooting sequence of key nodes includes: based on the objective function, taking the distance matrix and the short-circuit fault set as inputs of the ant colony optimization algorithm, and outputting the optimal troubleshooting sequence.

10. The intelligent emergency handling method for power system faults based on big data according to claim 9, characterized in that: Any element in the distance matrix is ​​the distance between two key nodes corresponding to its row and column; the short-circuit fault set is a set consisting of short-circuit fault indices of all key nodes.

Images