Fault monitoring and early warning method, system and equipment suitable for 10KV switch cabinet and medium

Abstract

The invention discloses a 10KV switch cabinet fault monitoring and early warning method, system, equipment and medium, belongs to the field of switch cabinets, and aims to construct a four-dimensional monitoring system of a bus chamber, a cable chamber, a functional unit chamber and a relay instrument chamber in order to solve the problems that existing monitoring depends on manpower, early warning lags and the like. The temperature data is processed in a partitioned mode through a spectral clustering algorithm, and singular point interference is eliminated in combination with a beta distribution dynamic threshold algorithm; a multi-source data fusion fault diagnosis model is built based on an artificial bee colony algorithm, feature vectors are built with temperature as a main part and electrical quantity as an auxiliary part, and temperature curves of all compartments are predicted; the system is provided with a visual interface, data is transmitted through radio frequency, early warning of faults is achieved through intelligent analysis of a server side, the requirement for unattended operation is met, and the operation safety, reliability and economical efficiency of the switch cabinet are improved.

Description

Technical Field

[0001] This invention relates to the field of switchgear technology, and in particular to methods, systems, equipment and media for fault monitoring and early warning of 10KV switchgear. Background Technology

[0002] Switchgear is one of the important devices in a power system that plays a protective and control role. It plays a key role in the switching operation of power equipment and lines, and can quickly disconnect faults when power equipment or lines fail, ensuring the normal operation of non-faulty sections and the safety of equipment and maintenance personnel.

[0003] Currently, the monitoring of switchgear in my country mainly relies on manual inspection, which has problems such as limited detection methods and low efficiency. It is difficult to provide timely warnings of potential faults and prevent the escalation of accidents. Moreover, the detection of switchgear mostly adopts relatively simple detection methods, and the basis for diagnosing faults and abnormalities is a single parameter anomaly. There is currently a lack of a method that integrates multiple parameters for comprehensive evaluation.

[0004] Existing fault diagnosis methods mostly use static diagnostic values, which cannot dynamically change their values ​​based on actual conditions such as operating status and environmental conditions. This greatly limits the accuracy and flexibility of the diagnostic methods. On the other hand, traditional switchgear fault early warning systems have significant time delays and are not intelligent enough to meet the requirements of intelligent early warning. Summary of the Invention

[0005] The main objective of this invention is to provide a method, system, device, and medium for fault monitoring and early warning of 10KV switchgear, aiming to solve existing technical problems.

[0006] To achieve the above objectives, the present invention provides a method for fault monitoring and early warning of 10KV switchgear, comprising the following steps; S1: Temperature acquisition devices are installed in the busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment of the 10KV switchgear to collect temperature data of each compartment in real time. S2: Perform partitioning processing on the collected temperature data and eliminate singular interference in multi-source monitoring data; S3: Using preprocessed multi-source monitoring data as input, construct a multi-source input feature vector with temperature as the main component and electrical quantity as the auxiliary component, build a multi-source data fusion fault diagnosis model, generate predicted temperature curves for each compartment, perform fault diagnosis and temperature prediction, and generate and output fault early warning information.

[0007] Furthermore, the temperature acquisition methods for the busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment are specifically as follows: The busbar compartment will have wireless temperature sensors installed at the busbar crimp points to measure the temperature. The cable compartment uses a current transformer to measure the outgoing current, and together with the bus voltage, obtains the active power, reactive power, and power factor. A wireless temperature sensor is installed at the cable joint to measure the temperature. The functional unit room uses a current transformer to measure the opening and closing coil current of the bus circuit breaker and the breaking current of the circuit breaker, and installs a wireless temperature sensor at the moving contact of the circuit breaker to measure the temperature. The relay instrument compartment uses non-contact infrared sensing temperature measurement.

[0008] Furthermore, the steps for partitioning the collected temperature data include: S211: Abstract the temperature network of the switchgear into a weighted undirected connected graph, construct a Laplace matrix, and take the second and third smallest eigenvalues ​​of the matrix to construct a two-dimensional space; S212: Randomly select k nodes from n nodes as initial cluster centers, where the value of k is the same as the number of switch cabinet compartments; S213: Calculate the distance from each node to each cluster center, and divide all nodes according to the minimum distance; S214: Recalculate the cluster centers; S215: Calculate the standard measure function. If the preset conditions are met, stop the calculation. If not, return to step S23 for iteration.

[0009] Furthermore, the steps to eliminate singularity interference in multi-source monitoring data include: S221: Monitoring data within a certain time period Normalization is performed: (1-1) Where X is the normalized value. ; and These represent the maximum and minimum values ​​of the monitored data, respectively. S222: Calculate the beta distribution of the monitoring data: (1-2) in, The midpoint value within the distribution interval of each sample; The shape parameters of the data are solved using the least squares method; The number of sample data points falling within interval j; The width of each sample interval j; n is the number of sample data contained in each sample interval j; N is the total number of sample intervals. S223: Calculate the two-sided quantiles of the sample interval: (1-3) (1-4) in, , They represent the lower side respectively. and the upper side quantiles; For sample data Data set after normalization; Pick Given a value of 0.05, find the normal data range: (1-5) S224: Determine whether the sampled value at time t is an interference value: (1-6) in, The sampled value predicted at time t. , These are the sampled values ​​at times t-1 and t-2, respectively. (1-7) (1-8) (1-9) in, Let be the sampled value monitored at time t. , These are the upper and lower limits of the error value, respectively. These are characteristic parameters, and their values ​​can be selected according to the characteristics of the data model; If the data value used at time t satisfies Formula 1-7, then the data is marked; if the data value used at time t satisfies Formula 1-8, then it is judged as heavily interfered and needs to be removed, and the data is used again. replace The search continues, and once a point that satisfies Formulas 1-9 is found, the starting point is selected again to continue the search.

[0010] Furthermore, the steps for building a multi-source data fusion fault diagnosis model include: S31: Set the population size of the artificial bee colony (the number of hired bees and observation bees are equal), the maximum number of iterations, and the maximum number of times the food source can be used; S32: Use the temperature data of each compartment after partitioning in step S2 as the initial population, and each combination of temperature data corresponds to a food source; S33: Search for new food sources near existing food sources, calculate the fitness value of the new food source, and if the new fitness value is greater than the fitness value of the original food source, replace the original food source and set the loop count to 0; otherwise, keep the original food source and increment the loop count by 1. S34: The hired bee transmits food source information to the observer bee. The observer bee selects a food source according to the probability P_i=fit_i / Σ (i=1 to SN)fit_i, where SN is the total number of food sources and fit_i is the fitness value of the i-th food source. The observer bee performs a neighborhood search around the selected food source and updates the food source according to the greedy principle. S35: If the number of times a food source can be used reaches the limit, the corresponding observation bee becomes a scout bee, randomly searches for and generates a new food source, and sets the initial number of uses to 0; S36: Repeat steps S33 to S35 until the maximum number of iterations is reached, and output the predicted temperature curves for each compartment.

[0011] Systems suitable for fault monitoring and early warning in 10kV switchgear include: The data acquisition module is used to collect real-time operating status data of multiple compartments of the switchgear, wherein the multiple compartments include at least a busbar compartment, a cable compartment, a functional unit compartment, and a relay instrument compartment; A data processing module, connected to the data acquisition module, is used to preprocess the acquired data. The preprocessing includes: partitioning the temperature monitoring data of the multiple compartments using a spectral clustering algorithm, and eliminating outliers in each partition using a beta distribution dynamic threshold algorithm. The fault diagnosis module, connected to the data processing module, is used to perform fault diagnosis and prediction based on preprocessed data using a multi-source data fusion model optimized by the artificial bee colony algorithm, and to generate fault early warning information; and, The early warning output module is connected to the fault diagnosis module and is used to receive the fault early warning information and provide a visual early warning interface.

[0012] Furthermore, it also includes a data transmission module, which transmits the collected data to the server via a field receiver through a 433MHz radio frequency. The server then analyzes the operating status of the switchgear using a multi-data fusion algorithm.

[0013] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the program, implements the above-described method for fault monitoring and early warning of 10KV switchgear.

[0014] A computer-readable storage medium having a computer program stored thereon, characterized in that, when executed by a processor, the program implements the above-described method for fault monitoring and early warning of 10KV switchgear. The beneficial effects of this invention are reflected in: This invention analyzes the structural characteristics and fault mechanisms of switchgear and constructs a four-dimensional monitoring system comprising a busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment. It uses wireless temperature measurement technology to collect temperature data of each compartment in real time, providing a solution for temperature monitoring of switchgear.

[0015] This invention employs a spectral clustering algorithm to perform preliminary classification of the collected temperature data. Based on this, a beta distribution dynamic threshold algorithm is used to process the unbalanced dataset, which solves the problem of singularity interference in the data, enhances the robustness of the algorithm to unbalanced data and noise interference, effectively improves the accuracy of switchgear fault diagnosis, and provides effective data support for fault early warning.

[0016] This invention utilizes a multi-data fusion algorithm based on artificial bee colonies to fully leverage the multi-dimensional information of switchgear, predict future data using historical operating data, and achieves early warning of faults. The algorithm was simulated and verified using the MATLAB platform.

[0017] This invention, based on a multi-data fusion artificial bee colony algorithm, establishes a visualized switchgear fault early warning interface. It collects switchgear status and environmental parameters via a built-in temperature sensor and remotely transmits them to a server. Big data analytics algorithms are then used to intelligently assess the switchgear's status. By combining historical and real-time data, the operating status of the switchgear is calculated and analyzed, meeting the requirements for "unattended operation" and providing effective technical support for the safe operation of the switchgear. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the fault monitoring and early warning method for 10KV switchgear according to the present invention; Figure 2 This is a schematic diagram illustrating the steps involved in building a multi-source data fusion fault diagnosis model according to the present invention. Figure 3 This is a schematic diagram of the 10KV switchgear monitoring system of the present invention; Figure 4 This is a schematic diagram of the fault monitoring and early warning system for 10KV switchgear according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1-3 This invention provides a method for fault monitoring and early warning of 10KV switchgear, including the following steps; S1: Temperature acquisition devices are installed in the busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment of the 10KV switchgear to collect temperature data of each compartment in real time. S2: Perform partitioning processing on the collected temperature data and eliminate singular interference in multi-source monitoring data; S3: Using preprocessed multi-source monitoring data as input, construct a multi-source input feature vector with temperature as the main component and electrical quantity as the auxiliary component, build a multi-source data fusion fault diagnosis model, generate predicted temperature curves for each compartment, perform fault diagnosis and temperature prediction, and generate and output fault early warning information.

[0021] This embodiment achieves end-to-end monitoring through a four-step closed loop: data acquisition, preprocessing, model building, and early warning judgment. First, a four-dimensional monitoring system covering key compartments of the switchgear is constructed, simultaneously collecting multi-source data on temperature and electrical quantities to ensure comprehensive monitoring dimensions. Then, spectral clustering and beta distribution dynamic threshold algorithms are used to partition and denoise the raw data, resolving data imbalance and singularity interference issues. Next, based on the preprocessed data, a feature vector is constructed with temperature as the primary factor and electrical quantities as secondary factors. The global optimization capability of the artificial bee colony algorithm is used to build a fusion diagnostic model to predict temperature curves. Finally, a comprehensive analysis combining historical and real-time data is performed to accurately identify fault risks and output early warnings.

[0022] This embodiment breaks through the limitations of traditional single-parameter monitoring and manual inspection, achieving deep fusion and intelligent analysis of multi-source data. The four-step process is logically rigorous, forming a complete link from data acquisition to early warning output. It effectively solves the problems of low monitoring efficiency, delayed early warning, and insufficient diagnostic accuracy in existing technologies, enabling early detection of potential faults, significantly improving the safety and stability of switchgear operation, reducing maintenance costs, and balancing reliability and economy.

[0023] In one embodiment, the temperature acquisition method for the busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment is as follows: Wireless temperature sensors are installed at the busbar crimp points in the busbar compartment to measure the temperature; The cable compartment uses a current transformer to measure the outgoing line current, and together with the bus voltage, obtains the active power, reactive power, and power factor. A wireless temperature sensor is installed at the cable joint to measure the temperature. The functional unit room uses current transformers to measure the opening and closing coil current of the bus circuit breaker and the breaking current of the circuit breaker, and installs wireless temperature sensors at the moving contact of the circuit breaker to measure the temperature. The relay instrument room uses non-contact infrared sensing temperature measurement.

[0024] This embodiment tailors a differentiated temperature acquisition scheme based on the structural characteristics and temperature measurement requirements of each compartment in the four-dimensional monitoring system. In the busbar compartment, the busbar contact points are critical current-carrying areas; wireless temperature sensors are directly installed here to achieve accurate temperature measurement of key heat-generating points. In the functional unit compartment, the circuit breaker moving contacts operate frequently and are prone to overheating; simultaneously, current parameters need to be monitored. Therefore, wireless temperature sensors and current transformers are combined to simultaneously collect temperature and key current data. In the cable compartment, cable joints are weak points in the connection; temperature is monitored using wireless temperature sensors, and power parameters are calculated using current transformers and busbar voltage to comprehensively reflect the operating status. In the relay instrument compartment, the space is enclosed and contact measurement is not advisable; therefore, non-contact infrared induction temperature measurement is used to avoid interfering with equipment operation.

[0025] This embodiment achieves a precise match between the temperature measurement methods of each compartment and its structural characteristics and operational requirements, ensuring both the accuracy and real-time nature of temperature data acquisition while also considering the safety and compatibility of equipment operation. The collaborative work of multiple types of acquisition devices provides high-quality foundational data for subsequent data fusion and analysis, effectively overcoming the shortcomings of traditional temperature measurement technologies, such as insufficient targeting and low data reliability.

[0026] In one embodiment, the step of partitioning the collected temperature data includes: S211: The temperature network of the switchgear is abstracted into a weighted undirected connected graph. A Laplace matrix is ​​constructed, and the second and third smallest eigenvalues ​​of the matrix are used to construct a two-dimensional space. Specifically, the temperature network of the switchgear is abstracted into a weighted undirected connected graph, where each node corresponds to a temperature acquisition point. The weight between nodes is the reciprocal of the temperature difference between the two points (the smaller the temperature difference, the larger the weight, representing a stronger correlation between the two points). Based on this graph, a Laplace matrix L=DW is constructed (where D is the degree matrix and W is the weight matrix). The second and third smallest eigenvalues ​​of the matrix are extracted through eigenvalue decomposition to construct a two-dimensional feature space, thus completing the dimensionality reduction of the high-dimensional temperature data.

[0027] S212: Randomly select k nodes from n nodes as initial cluster centers, and the value of k is consistent with the number of switch cabinet compartments; specifically, set the number of clusters k=4 (consistent with the number of four compartments), and randomly select 4 nodes from n=24 temperature nodes (6 collection points in each compartment) as initial cluster centers, denoted as C1, C2, C3, and C4, which correspond to the initial center nodes of the busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment, respectively.

[0028] S213: Calculate the distance from each node to each cluster center, and divide all nodes according to the minimum distance. Specifically, use Euclidean distance to calculate the distance from each temperature node to the four cluster centers. The distance formula is d(x,y)=√[(x1-y1)²+(x2-y2)²] (x and y are the node coordinates in the two-dimensional feature space). Divide each node into the corresponding cluster according to the minimum distance principle. Recalculate the center coordinates of each cluster. The formula is Cj=(1 / m)Σ(i=1 to m)xi (where m is the number of nodes in the j-th cluster, and xi is the coordinates of each node in the cluster).

[0029] S214: Recalculate the cluster centers; S215: Calculate the standard measure function. If the preset conditions are met, stop the calculation; otherwise, return to step S23 for iteration. Specifically, the standard measure function is calculated as the sum of squared errors within the cluster, SSE = Σ(j=1 to 4)Σ(xi∈Cj)||xi-Cj||². The preset conditions are that the change in SSE is less than 0.01 or the number of iterations reaches 50. If these conditions are met, stop the iteration and complete the temperature data partitioning.

[0030] In one embodiment, the step of eliminating singularity interference in multi-source monitoring data includes: S221: Monitoring data within a certain time period Normalization is performed: (1-1) Where X is the normalized value. ; and These represent the maximum and minimum values ​​of the monitored data, respectively. S222: Calculate the beta distribution of the monitoring data: (1-2) in, The midpoint value within the distribution interval of each sample; The shape parameters of the data are solved using the least squares method; The number of sample data points falling within interval j; The width of each sample interval j; n is the number of sample data contained in each sample interval j; N is the total number of sample intervals. S223: Calculate the two-sided quantiles of the sample interval: (1-3) (1-4) in, , They represent the lower side respectively. and the upper side quantiles; For sample data Data set after normalization; Pick Given a value of 0.05, find the normal data range: (1-5) S224: Determine whether the sampled value at time t is an interference value: (1-6) in, The sampled value predicted at time t. , These are the sampled values ​​at times t-1 and t-2, respectively. (1-7) (1-8) (1-9) in, Let be the sampled value monitored at time t. , These are the upper and lower limits of the error value, respectively. These are characteristic parameters, and their values ​​can be selected according to the characteristics of the data model; If the data value used at time t satisfies Formula 1-7, then the data is marked; if the data value used at time t satisfies Formula 1-8, then it is judged as heavily interfered and needs to be removed, and the data is used again. replace The search continues, and once a point that satisfies Formulas 1-9 is found, the starting point is selected again to continue the search.

[0031] Specifically, monitoring data (a total of 1440 sample values) within a certain 24-hour period were selected as the processing object and normalized, where Xi is the original monitoring data (temperature or electrical quantity). and These are the maximum and minimum values ​​of the data within the specified time period. After normalization, the data X is mapped to the interval [0,1] to eliminate dimensional differences.

[0032] Calculate the beta distribution of the monitoring data: Divide the normalized dataset into N=20 equal-width sample intervals, each with a width lj=0.05. Count the number of sample data in each interval kj and determine the midpoint value Xj of each interval. Solve for the shape parameters α and β of the beta distribution using the least squares method to maximize the fit between the distribution model and the actual data distribution. The final beta distribution formula is: .

[0033] With σ=0.05, the two-sided quantiles of the sample interval are calculated by integration: λ1=∫(0 to λ1)f(X;α,β)dX=0.05, λ2=∫(0 to λ2)f(X;α,β)dX=0.95, and the normal data interval is determined to be [λ1,λ2].

[0034] Determine whether the sampled value at time t is an interference value: (1) Predict the sampled value at time t according to the formula; (2) Set the upper limit of error W1=0.05 and the lower limit of error W2=0.1 (determined according to the characteristics of the 10KV switchgear operation data). If | - |> If | - |> If the data is heavily disturbed, it is judged as heavily disturbed data, and the predicted value is used. Replace actual sampled values Continue to judge the subsequent data; (3) Set the preset convergence condition as 5 consecutive sample values ​​all satisfy | - |≤ If a point that meets the condition is found, then that point is selected as the starting point again, and the above interference value judgment process is repeated to ensure the continuity and reliability of the data sequence.

[0035] In one embodiment, the steps of building a multi-source data fusion fault diagnosis model include: S31: Set the population size of the artificial bee colony (the number of hired bees and observation bees are equal), the maximum number of iterations, and the maximum number of times the food source can be used; S32: Use the temperature data of each compartment after partitioning in step S2 as the initial population, and each combination of temperature data corresponds to a food source; S33: Search for new food sources near existing food sources, calculate the fitness value of the new food source, and if the new fitness value is greater than the fitness value of the original food source, replace the original food source and set the loop count to 0; otherwise, keep the original food source and increment the loop count by 1. S34: The hired bee transmits food source information to the observer bee. The observer bee selects a food source according to the probability P_i=fit_i / Σ (i=1 to SN)fit_i, where SN is the total number of food sources and fit_i is the fitness value of the i-th food source. The observer bee performs a neighborhood search around the selected food source and updates the food source according to the greedy principle. S35: If the number of times a food source can be used reaches the limit, the corresponding observation bee becomes a scout bee, randomly searches for and generates a new food source, and sets the initial number of uses to 0; S36: Repeat steps S33 to S35 until the maximum number of iterations is reached, and output the predicted temperature curves for each compartment.

[0036] Please see Figure 4 The present invention also provides a system for fault monitoring and early warning of 10KV switchgear, comprising: The data acquisition module is used to collect real-time operating status data of multiple compartments of the switchgear. These multiple compartments include at least a busbar compartment, a cable compartment, a functional unit compartment, and a relay instrument compartment. At conductor connections (such as crimp points and contacts) in busbar compartments, cable compartments, and functional unit compartments, thermistors built into wireless temperature sensors (such as platinum resistance thermometers Pt100 or digital temperature sensors DS18B20) directly sense the contact temperature and convert it into resistance changes or digital signals. In relay instrument compartments, non-contact infrared sensors detect the intensity of infrared radiation emitted by object surfaces and, according to the Stefan-Boltzmann law, convert the radiant energy into a corresponding temperature reading, achieving non-invasive temperature measurement. Electrical quantity sensing: Current transformers (CTs) are based on the principle of electromagnetic induction, proportionally converting a large primary current into a small secondary current signal. Voltage transformers (PTs) are also based on electromagnetic induction, converting a high primary voltage into a low primary voltage signal. These analog signals are then converted into digital quantities by analog-to-digital converters (ADCs) after passing through signal conditioning circuits (such as filtering and amplification).

[0037] Each sensor node not only collects raw data but also possesses preliminary timestamping and local caching capabilities. In a self-organizing wireless sensor network, nodes transmit temperature data packets with location identifiers (e.g., "Cabinet 1 - Busbar Room - Phase A Connector") and timestamps to the field data concentrator according to a predetermined routing protocol (e.g., based on cluster heads or mesh networks). The concentrator simultaneously receives electrical quantity digital information from the CT / PT integrated measurement unit or protection and control device via a digital input / communication interface (e.g., RS-485 Modbus). The concentrator operates by performing time synchronization alignment and data packet encapsulation on these multi-source heterogeneous data from different physical layers and protocols, forming a unified data frame with a unified time reference and standard format (e.g., JSON or a specific binary format).

[0038] The data processing module, connected to the data acquisition module, is used to preprocess the acquired data. The preprocessing includes: partitioning the temperature monitoring data of multiple compartments using a spectral clustering algorithm, and eliminating outliers in each partition using a beta distribution dynamic threshold algorithm. Spectral clustering partitioning maps temperature monitoring points in the physical space to a "feature space" for clustering. First, the monitoring network is constructed as a graph, where vertices represent temperature measurement points, and edge weights represent the similarity of temperature changes between pairs of points (calculated using a Gaussian kernel function). The Laplacian matrix is ​​the mathematical representation of this graph, reflecting its connectivity structure. Eigenvalue decomposition is performed on this matrix; the eigenvectors corresponding to the smallest non-zero eigenvalues ​​(called spectra) constitute the data embedding in the low-dimensional space. These spectral vectors effectively preserve the global connectivity between data points, making data points that might be nonlinearly separable in the original temperature space more easily linearly partitioned in the spectral space (low-dimensional feature space). Finally, K-means clustering is performed in the low-dimensional spectral space. The principle is iterative optimization, minimizing the sum of squared Euclidean distances within the same cluster and maximizing the distances between different clusters, thereby automatically grouping closely related temperature measurement points (usually belonging to the same physical compartment or having similar thermal behavior) into one class.

[0039] Beta distribution dynamic threshold scrambling: The beta distribution is a continuous probability distribution defined in the interval [0,1]. Its shape is determined by two parameters, α and β, making it very flexible and capable of fitting various forms of unimodal data (such as skewed and symmetrical). Fitting normalized historical temperature data to a beta distribution essentially establishes a probabilistic model for the fluctuation range of normal data.

[0040] The boundaries of normal values ​​are determined by calculating the quantiles of this distribution (such as 5% and 95%). Compared to a fixed threshold, this boundary dynamically adjusts as the input data distribution (determined by the α and β parameters) changes, thereby adapting to normal temperature drift caused by variations in equipment load, ambient temperature, etc.

[0041] This method utilizes the autocorrelation of time series (such as AR models) to predict the current value from past values. The current measured value is compared with the predicted value, while simultaneously checking if the measured value falls within a dynamic threshold range. Only when the measured value significantly deviates from the predicted value and exceeds the dynamic probability boundary is it identified as a true singularity (such as a sudden disturbance), rather than a normal rapid fluctuation. Its working principle balances temporal continuity and statistical regularity, improving the robustness of the judgment.

[0042] The fault diagnosis module is connected to the data processing module. It is used to perform fault diagnosis and prediction based on the preprocessed data and a multi-source data fusion model optimized by the artificial bee colony algorithm, and to generate fault warning information. The early warning output module is connected to the fault diagnosis module to receive fault early warning information and provide a visual early warning interface.

[0043] In one embodiment, a data transmission module is also included, which transmits the collected data to the server via a field receiver through a 433MHz radio frequency. The server analyzes the operating status of the switchgear through a multi-data fusion algorithm.

[0044] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0045] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0046] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the dual-serial-bus-based switch matrix control systems described above.

[0047] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0048] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0049] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0050] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0051] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for fault monitoring and early warning of 10KV switchgear, characterized in that: Includes the following steps; S1: Temperature acquisition devices are installed in the busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment of the 10KV switchgear to collect temperature data of each compartment in real time. S2: Perform partitioning processing on the collected temperature data and eliminate singular interference in multi-source monitoring data; S3: Using preprocessed multi-source monitoring data as input, construct a multi-source input feature vector with temperature as the main component and electrical quantity as the auxiliary component, build a multi-source data fusion fault diagnosis model, generate predicted temperature curves for each compartment, perform fault diagnosis and temperature prediction, and generate and output fault early warning information.

2. The method for fault monitoring and early warning of 10KV switchgear as described in claim 1, characterized in that: The specific temperature acquisition methods for the busbar compartment, cable compartment, functional unit compartment, and relay instrument compartment are as follows: The busbar compartment will have wireless temperature sensors installed at the busbar crimp points to measure the temperature. The cable compartment uses a current transformer to measure the outgoing current, and together with the bus voltage, obtains the active power, reactive power, and power factor. A wireless temperature sensor is installed at the cable joint to measure the temperature. The functional unit room uses a current transformer to measure the opening and closing coil current of the bus circuit breaker and the breaking current of the circuit breaker, and installs a wireless temperature sensor at the moving contact of the circuit breaker to measure the temperature. The relay instrument compartment uses non-contact infrared sensing temperature measurement.

3. The method for fault monitoring and early warning of 10KV switchgear as described in claim 1, characterized in that: The steps for partitioning the collected temperature data include: S211: Abstract the temperature network of the switchgear into a weighted undirected connected graph, construct a Laplace matrix, and take the second and third smallest eigenvalues ​​of the matrix to construct a two-dimensional space; S212: Randomly select k nodes from n nodes as initial cluster centers, where the value of k is the same as the number of switch cabinet compartments; S213: Calculate the distance from each node to each cluster center, and divide all nodes according to the minimum distance; S214: Recalculate the cluster centers; S215: Calculate the standard measure function. If the preset conditions are met, stop the calculation. If not, return to step S23 for iteration.

4. The method for fault monitoring and early warning of 10KV switchgear as described in claim 1, characterized in that: The steps to eliminate singularity interference in multi-source monitoring data include: S221: Monitoring data within a certain time period Normalization is performed: (1-1) Where X is the normalized value. ; and These represent the maximum and minimum values ​​of the monitored data, respectively. S222: Calculate the beta distribution of the monitoring data: ；（1-2） in, The midpoint value within the distribution interval of each sample; The shape parameters of the data are solved using the least squares method; The number of sample data points falling within interval j; The width of each sample interval j; n is the number of sample data contained in each sample interval j; N is the total number of sample intervals. S223: Calculate the two-sided quantiles of the sample interval: ；（1-3） ；（1-4） in, , They represent the lower side respectively. and the upper side quantiles; For sample data Data set after normalization; Pick Given a value of 0.05, find the normal data range: ；（1-5） S224: Determine whether the sampled value at time t is an interference value: (1-6) in, The sampled value predicted at time t. , These are the sampled values ​​at times t-1 and t-2, respectively. ；（1-7） ；（1-8） ；（1-9） in, Let be the sampled value monitored at time t. , These are the upper and lower limits of the error value, respectively; These are characteristic parameters, and their values ​​can be selected according to the characteristics of the data model; If the data value used at time t satisfies Formula 1-7, then the data is marked; if the data value used at time t satisfies Formula 1-8, then it is judged as heavily interfered and needs to be removed, and the data is used again. replace The search continues, and once a point that satisfies Formulas 1-9 is found, the starting point is selected again to continue the search.

5. The method for fault monitoring and early warning of 10KV switchgear as described in claim 1, characterized in that: The steps for building a multi-source data fusion fault diagnosis model include: S31: Set the population size of the artificial bee colony (the number of hired bees and observation bees are equal), the maximum number of iterations, and the maximum number of times the food source can be used; S32: Use the temperature data of each compartment after partitioning in step S2 as the initial population, and each combination of temperature data corresponds to a food source; S33: Search for new food sources near existing food sources, calculate the fitness value of the new food sources, and replace the original food source if the new fitness value is greater than the original food source's fitness value. Set the loop count to 0. Otherwise, retain the original food source and increment the loop count by 1; S34: The hired bee transmits food source information to the observer bee. The observer bee selects a food source according to the probability P_i=fit_i / Σ (i=1 to SN)fit_i, where SN is the total number of food sources and fit_i is the fitness value of the i-th food source. The observer bee performs a neighborhood search around the selected food source and updates the food source according to the greedy principle. S35: If the number of times a food source can be used reaches the limit, the corresponding observation bee becomes a scout bee, randomly searches for and generates a new food source, and sets the initial number of uses to 0; S36: Repeat steps S33 to S35 until the maximum number of iterations is reached, and output the predicted temperature curves for each compartment.

6. A system for fault monitoring and early warning of 10KV switchgear, characterized in that, include: The data acquisition module is used to collect real-time operating status data of multiple compartments of the switchgear, wherein the multiple compartments include at least a busbar compartment, a cable compartment, a functional unit compartment, and a relay instrument compartment; A data processing module, connected to the data acquisition module, is used to preprocess the acquired data. The preprocessing includes: partitioning the temperature monitoring data of the multiple compartments using a spectral clustering algorithm, and eliminating outliers in each partition using a beta distribution dynamic threshold algorithm. The fault diagnosis module, connected to the data processing module, is used to perform fault diagnosis and prediction based on preprocessed data using a multi-source data fusion model optimized by the artificial bee colony algorithm, and to generate fault early warning information; and, The early warning output module is connected to the fault diagnosis module and is used to receive the fault early warning information and provide a visual early warning interface.

7. The system for fault monitoring and early warning of 10KV switchgear as described in claim 6, characterized in that: It also includes a data transmission module, which transmits the collected data to the server via a field receiver through a 433MHz radio frequency. The server then analyzes the operating status of the switchgear using a multi-data fusion algorithm.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the fault monitoring and early warning method for 10KV switchgear as described in any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the fault monitoring and early warning method for 10KV switchgear as described in any one of claims 1 to 5.

Images