Power distribution facility condition assessment system and method

Abstract

The application provides a power distribution facility state evaluation system and method, which comprises the following steps: collecting multi-dimensional operation data sets (including electrical operation parameters, equipment environment parameters and equipment mechanical state parameters) by using a data acquisition module; performing multi-modal fusion on the multi-dimensional operation data sets by using a multi-source feature fusion module to accurately extract a key feature set of the fused power distribution facility; and finally, a state evaluation module adopts a multi-modal feature extraction / fusion model to evaluate the key feature set of the power distribution facility to obtain an accurate comprehensive state index of the power distribution facility. The application not only effectively improves the accuracy and reliability of the evaluation results and future change trends, but also has strong generalization ability of the multi-modal model and can adapt to different operation environments and different types of power distribution facilities.

Description

Technical Field

[0001] This invention relates to the technical field of power system automation and intelligent operation and maintenance, and in particular to a power distribution facility condition assessment system and method. Background Technology

[0002] The power distribution facility condition assessment system is a key intelligent operation and maintenance tool that has emerged to address the core challenges of modern power distribution networks being "invisible, unmanageable, and inaccurate to repair," driven by the convergence of four forces: safety requirements, technological advancements, policy guidance, and economic incentives.

[0003] However, existing power distribution facility condition assessment systems typically rely on data from multiple single sensors (such as infrared temperature sensors) or use simple threshold alarms, making it difficult to comprehensively reflect the true health status of equipment. For example, they may not consider the impact of the environment, or the models may have weak generalization capabilities and be unable to account for risk factors. Furthermore, existing facility condition assessment systems lack an effective mechanism for fusing multi-source heterogeneous data, leading to problems such as delayed assessment results and insufficient accuracy. Therefore, an intelligent assessment system that integrates real-time operational data, environmental parameters, and historical operation and maintenance information is needed to achieve accurate quantification of the condition of power distribution facilities and their future trends. Summary of the Invention

[0004] This paper discloses a system and method for assessing the condition of power distribution facilities. It employs multimodal fusion of collected multi-dimensional operational datasets (including electrical operating parameters, equipment environmental parameters, and equipment mechanical condition parameters) to accurately extract the key feature set of the fused power distribution facilities. Then, a multimodal feature extraction / fusion model is used to evaluate the power distribution facilities based on this key feature set, resulting in a precise comprehensive condition index. This approach not only effectively improves the accuracy and reliability of the assessment results and future trends but also demonstrates strong generalization ability of the multimodal model, making it adaptable to different operating environments and different types of power distribution facilities.

[0005] In a first aspect, this disclosure provides a power distribution facility condition assessment system, including: The data acquisition module is used to collect electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters of power distribution facilities in real time through sensors, forming a multi-dimensional operating dataset; The multi-source feature fusion module is used to extract local feature sets and time-series feature sets of data in each dimension of the running dataset, and fuse the local feature sets and time-series feature sets of data in each dimension to obtain the key feature set of the power distribution facility. The key feature set of the power distribution facility includes the key feature set of the fused electrical operating state, the key feature set of the fused equipment environmental state, and the key feature set of the fused equipment mechanical state. The status assessment module is used to input the key feature set into a preset time series prediction model, and calculate the comprehensive status index of the power distribution facility through the preset time series prediction model. The preset time series prediction model is a model formed by fusing a long short-term memory network model and a fault key feature model.

[0006] In some embodiments, the multi-source feature fusion module includes an electrical operating state feature fusion unit, an equipment environmental state feature fusion unit, and an equipment mechanical state feature fusion unit: The electrical operating state feature fusion unit is used to perform Hadamard product operation on the local features of electrical operating parameters to obtain the key feature set of the fused electrical operating state. The equipment environment state feature fusion unit performs Hadamard product operation on the local features of the equipment environment parameters to obtain the key feature set of the fused equipment environment state. The equipment mechanical state feature fusion unit is used to perform Hadamard product operation on the local features of the equipment mechanical state parameters to obtain the key feature set of the fused equipment mechanical state.

[0007] In some embodiments, the preset time series prediction model is: in, It is a comprehensive state index; This refers to the electrical operating status index; This is the weight matrix for the electrical operating state of the fully connected layer; This is the key feature set of the electrical operating status after integration; This is the bias value for the electrical operating state; This refers to the equipment's environmental condition index. This is a weight matrix representing the environmental state of the fully connected layer devices. This is a set of key features representing the environmental state of the merged equipment. This is the bias value for the device's environmental conditions; This refers to the mechanical condition index of the equipment. This is the weight matrix for the mechanical state of the fully connected layer device; This is the key feature set of the mechanical state of the fused equipment; This is the bias value for the mechanical state of the equipment.

[0008] In some embodiments, the power distribution facility condition assessment system further includes an interception module: The interception module is used to control the comprehensive state index within a preset index range; wherein the preset index range is... .

[0009] In some embodiments, the power distribution facility condition assessment system further includes a data preprocessing module: The data preprocessing module is used to perform time-series alignment and normalization on the multi-dimensional running dataset to obtain a normalized multi-dimensional running dataset. The multi-source feature fusion module is also used to extract the local feature sets and time-series feature sets of each dimension of the normalized multi-dimensional running dataset, and to fuse the local feature sets and time-series feature sets of each dimension to obtain the key feature set of the power distribution facility.

[0010] In some embodiments, the power distribution facility status assessment system further includes a visualization and early warning module: The visualization and early warning module is used to determine the danger warning value corresponding to the comprehensive status index, and to provide early warning and visualization based on the danger warning value.

[0011] In some embodiments, the power distribution facility condition assessment system may further include a data storage module: The data storage module is used to store the comprehensive state index and the corresponding operational datasets for each dimension of the comprehensive state index.

[0012] Secondly, this disclosure provides a method for assessing the condition of power distribution facilities, including: By collecting electrical operating parameters, equipment environmental parameters, and equipment mechanical condition parameters of power distribution facilities in real time through sensors, a multi-dimensional operating dataset is formed. Local feature sets and time-series feature sets of each dimension of the running dataset are extracted, and the local feature sets and time-series feature sets of each dimension of the data are fused to obtain the key feature set of the power distribution facility. The key feature set of the power distribution facility includes the key feature set of the fused electrical operating state, the key feature set of the fused equipment environmental state, and the key feature set of the fused equipment mechanical state. The key feature set is input into a preset time series prediction model, and the comprehensive state index of the power distribution facility is calculated through the preset time series prediction model. The preset time series prediction model is a model formed by fusing a long short-term memory network model and a fault key feature model.

[0013] Thirdly, this disclosure provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is used to implement the power distribution facility status assessment system described in any of the above embodiments when running the computer program.

[0014] Fourthly, this disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the power distribution facility status assessment system described in any of the above embodiments.

[0015] This disclosure provides a power distribution facility condition assessment system and method. By performing multimodal fusion on a collected multi-dimensional operational dataset (including electrical operating parameters, equipment environmental parameters, and equipment mechanical condition parameters), the system accurately extracts the key feature set of the fused power distribution facility. Then, a multimodal feature extraction / fusion model is used to assess the power distribution facility based on this key feature set, resulting in an accurate comprehensive condition index. This not only effectively improves the accuracy and reliability of the assessment results and future trends, but also demonstrates strong generalization ability of the multimodal model, making it adaptable to different operating environments and different types of power distribution facilities. Attached Figure Description

[0016] The present disclosure will be described in more detail below with reference to embodiments and the accompanying drawings; Figure 1 This is a flowchart illustrating a power distribution facility status assessment method in one embodiment; Figure 2 This is a schematic diagram of the structure of a power distribution facility condition assessment system in one embodiment; Figure 3 This is a schematic diagram of the internal structure of a computer device in one embodiment.

[0017] In the accompanying drawings, the same parts are referred to by the same reference numerals, and the drawings are not drawn to scale. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions of this disclosure, and to fully understand and implement the process of how this disclosure applies technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, not all embodiments. The embodiments of this disclosure and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort should fall within the protection scope of this disclosure.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0021] Example 1 In this embodiment, as Figure 1 As shown, a power distribution facility condition assessment system is provided, comprising: a data acquisition module 111, a multi-source feature fusion module 112, and a condition assessment module 113. The data acquisition module 111 and the multi-source feature fusion module 112 are communicatively connected; the multi-source feature fusion module 112 and the condition assessment module 113 are communicatively connected; and the data acquisition module 111 and the condition assessment module 113 are communicatively connected.

[0022] In this embodiment, the data acquisition module 111 is used to collect electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters of the power distribution facility in real time through sensors, forming a multi-dimensional operating dataset.

[0023] The multi-source feature fusion module 112 is used to extract local feature sets and time-series feature sets of data from each dimension in the running dataset, and fuse the local feature sets and time-series feature sets of data from each dimension to obtain a key feature set of the power distribution facility. The key feature set of the power distribution facility includes a key feature set of the fused electrical operating state, a key feature set of the fused equipment environmental state, and a key feature set of the fused equipment mechanical state. The state assessment module 113 is used to input the key feature set into a preset time series prediction model, and calculate the comprehensive state index of the power distribution facility through the preset time series prediction model. The preset time series prediction model is a model formed by fusing a long short-term memory network model and a fault key feature model.

[0024] Specifically, the data acquisition module 111 is communicatively connected to each monitoring point of the power distribution facility. Various sensors are installed at each monitoring point to collect real-time electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters of the power distribution facility. These sensors transmit the collected electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters to the data acquisition module 111. The data acquisition module 111 receives the data information transmitted by each sensor, including electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters. Finally, the electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters are used as a multi-dimensional operational dataset. In this embodiment, electrical operating parameters may include voltage, current harmonics, power factor, leakage current, transformers, switchgear, or circuit breakers; equipment environmental parameters may include oil temperature, winding temperature, or ambient temperature and humidity; and equipment mechanical status parameters may include dissolved gas in oil (DGA) analysis data, infrared thermal imaging spectra, partial discharge (PD) location maps, or vibration acoustic signals.

[0025] After the data acquisition module 111 acquires the multi-dimensional operational dataset, it sends the dataset to the multi-source feature fusion module 112. The multi-source feature fusion module 112 receives the multi-dimensional operational dataset and then uses signal processing algorithms to extract local and temporal features from each dimension of the dataset, forming local and temporal feature sets. These sets are then further fused to obtain the key feature set of the power distribution facility. Specifically, the signal processing algorithm is a multi-modal fusion algorithm using a convolutional network model and a bidirectional LSTM (Long Short-Term Memory) model. The multi-source feature fusion module 112 performs temporal processing on the multi-dimensional operational dataset to obtain a multi-dimensional temporal dataset. It then uses a convolutional network model to perform convolution and pooling on the multi-dimensional temporal dataset to extract local feature sets. The multi-source feature fusion module 112 then inputs the local feature sets into the bidirectional LSTM model, using the model to fit the long-term temporal dependencies of the multi-source features of the power distribution facility corresponding to the local feature sets, outputting a temporal hidden state sequence H. ;in . These are local characteristics of the state of power distribution facilities; The convolution kernel weights correspond to the state of the positive power distribution facility; This refers to the local characteristics of the positive power distribution facility's status; The convolution kernel weights correspond to the reverse power distribution facility status; This refers to the local characteristics of the reverse power distribution facility's condition; This is the bias value for the state characteristics of the power distribution facility.

[0026] In this embodiment, , These are local characteristics of the electrical operating state; These are local characteristics of the equipment's environmental state. These are local characteristics of the equipment's mechanical state, specifically: in, These are local characteristics of the electrical operating state; These are the convolution kernel weights corresponding to the positive electrical operating state; These are local characteristics of a positive electrical operating state; The convolution kernel weights are the values ​​corresponding to the reverse electrical operating state. These are local characteristics of the reverse electrical operating state; This is the bias value for the electrical operating state characteristics.

[0027] in, These are local characteristics of the equipment's environmental state. These are the convolutional kernel weights corresponding to the positive device environment state; Local features representing the positive device environment state; The convolutional kernel weights are the values ​​corresponding to the reverse device environment state. These are local features of the reverse device's environmental state; This is the bias value for the environmental condition characteristics of the equipment.

[0028] in, These are local features of the equipment's mechanical state; These are the convolution kernel weights corresponding to the forward mechanical state of the device; These are local features of the mechanical state of the forward-moving equipment; These are the convolution kernel weights corresponding to the mechanical state of the reverse device; These are local features of the mechanical state of the reversed device; This is the bias value for the mechanical state characteristics of the equipment.

[0029] After the multi-source feature fusion module 112 outputs the temporal hidden state sequence H, a weight allocation calculation is further performed on the temporal hidden state sequence H (i.e., assigning greater weights to fault-sensitive key temporal features to weaken noise features) to obtain the global fused features. .in . For the key feature set of the nth type of power distribution facility, For the local characteristics of the state of the t-th type of power distribution facility, This is the Hadamard product operator; The weighted fusion feature parameters are the local features corresponding to the state of the t-th type of power distribution facility.

[0030] In this embodiment, , This is the key feature set of the electrical operating status after integration; This is a set of key features representing the environmental state of the merged equipment. This is the key feature set of the mechanical state of the equipment after fusion.

[0031] In this embodiment, both the multi-dimensional time-series dataset and the local feature set are represented using matrices.

[0032] In this embodiment, the key feature set of the power distribution facility may include the key feature set of the electrical operation after integration, the key feature set of the environmental state of the equipment after integration, and the key feature set of the mechanical state of the equipment after integration, etc., without specific limitations here.

[0033] After obtaining the key feature set of the power distribution facility, the multi-source feature fusion module 112 transmits the key feature set to the state assessment module 113. The state assessment module 113 inputs the key feature set into a preset time series prediction model and calculates the comprehensive state index of the power distribution facility using the preset time series prediction model. The specific steps for calculating the comprehensive state index of the power distribution facility are as follows: determine the number of types of key feature sets for the power distribution facility, substitute the key feature sets of different types of power distribution facilities into the preset time series prediction model, perform data calculation and processing analysis using the preset time series prediction model to obtain the comprehensive state index, and then output the comprehensive state index. In this embodiment, the preset time series prediction model is a model formed by fusing a long short-term memory network model and a fault key feature model, specifically: in, It is a comprehensive state index; This is the index for the first type of power distribution facilities; This is the index for the second type of power distribution facilities; This is the index for the nth type of power distribution facility.

[0034] in, This is the weight matrix for the state of the nth type of power distribution facility in the fully connected layer; This represents the key feature set for the nth type of power distribution facility. This is the bias value for the nth type of power distribution facility.

[0035] In practical applications, the preset time series forecasting model is as follows: in, It is a comprehensive state index; This refers to the electrical operating status index; This is the weight matrix for the electrical operating state of the fully connected layer; This is the key feature set of the electrical operating status after integration; This is the bias value for the electrical operating state; This refers to the equipment's environmental condition index. This is a weight matrix representing the environmental state of the fully connected layer devices. This is a set of key features representing the environmental state of the merged equipment. This is the bias value for the device's environmental conditions; This refers to the mechanical condition index of the equipment. This is the weight matrix for the mechanical state of the fully connected layer device; This is the key feature set of the mechanical state of the fused equipment; This is the bias value for the mechanical state of the equipment.

[0036] For example, in this embodiment, a key feature set of the fused electrical operating state is calculated using a preset time series prediction model. Corresponding electrical operating status index Then, the key feature set of the fused device environment status is calculated using a preset time series prediction model. Corresponding equipment environmental status index Furthermore, the key feature set of the fused equipment mechanical state is calculated using a preset time series prediction model. Corresponding equipment mechanical condition index Finally, based on the electrical operating status index... Equipment environmental status index and equipment mechanical condition index Calculate the comprehensive state index .

[0037] Specific limitations regarding the power distribution facility condition assessment system can be found in the limitations of the power distribution facility condition assessment method below, and will not be repeated here. Each unit in the aforementioned power distribution facility condition assessment system can be implemented entirely or partially through software, hardware, or a combination thereof. These units can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each unit.

[0038] In one embodiment, the multi-source feature fusion module includes an electrical operating state feature fusion unit, an equipment environmental state feature fusion unit, and an equipment mechanical state feature fusion unit.

[0039] The electrical operating state feature fusion unit performs Hadamard product operations on the local features of electrical operating parameters to obtain a key feature set of the fused electrical operating state. The equipment environmental state feature fusion unit performs Hadamard product operations on the local features of equipment environmental parameters to obtain a key feature set of the fused equipment environmental state. The equipment mechanical state feature fusion unit performs Hadamard product operations on the local features of equipment mechanical state parameters to obtain a key feature set of the fused equipment mechanical state.

[0040] In this embodiment, the key feature set of the fused electrical operating state, the key feature set of the fused equipment environmental state, and the key feature set of the fused equipment mechanical state are calculated using the Hadamard product operation method. This can enhance the common fault characteristics of various parameters, suppress independent noise, and retain the physical meaning of the features, thereby improving the purity of features and fault sensitivity, reducing the computational complexity of subsequent models, and improving the accuracy and real-time performance of state assessment.

[0041] In one embodiment, the power distribution facility condition assessment system further includes an interception module.

[0042] The interception module is used to control the comprehensive state index within a preset index range; wherein the preset index range is... .

[0043] In this embodiment, the interception module uses an interception function. The overall state index is controlled to remain within a preset range, and the preset range is [value missing]. Specifically: in, This is the extracted comprehensive state index; This is the initial comprehensive state index.

[0044] In this embodiment, a truncation function is used. Control the comprehensive state index at Within this scope, the evaluation results are standardized and made more intuitive; this facilitates state classification, equipment comparison, and system integration, thereby improving the stability of model training and the practicality of operation and maintenance.

[0045] In one embodiment, the power distribution facility condition assessment system further includes a data preprocessing module.

[0046] The data preprocessing module is used to perform time-series alignment and normalization on the multi-dimensional operational dataset to obtain a normalized multi-dimensional operational dataset. The multi-source feature fusion module 112 is also used to extract local feature sets and time-series feature sets of each dimension of the normalized multi-dimensional operational dataset, and fuse the local feature sets and time-series feature sets of each dimension to obtain the key feature set of the power distribution facility.

[0047] In this embodiment, multi-source data with non-equidistant, asynchronous, and different frequencies are unified to a timeline with the same time step and equal intervals. This provides standard data for subsequent signal processing, feature extraction, Hadamard product fusion, and model input. The multi-dimensional running dataset undergoes time-series alignment and normalization to obtain a normalized multi-dimensional running dataset. Specifically, after receiving the multi-dimensional data transmitted by the data acquisition module 111, the data preprocessing module performs resampling and interpolation for each category and dimension of data to achieve strict alignment with the target timeline. Finally, the Min-Max normalization method is used to normalize the aligned multi-dimensional running dataset.

[0048] In practical applications, after the data preprocessing module completes the time-series alignment and normalization of the multi-dimensional running dataset, a normalized multi-dimensional running dataset is obtained. This normalized multi-dimensional running dataset is then transmitted to the multi-source feature fusion module 112. The multi-source feature fusion module 112 extracts the local and temporal features of each dimension in the normalized multi-dimensional running dataset and fuses the local and temporal feature sets to obtain the key feature set of the power distribution facility. Specifically, the steps of the multi-source feature fusion module 112 in obtaining the key feature set of the power distribution facility based on the normalized multi-dimensional running dataset are the same as those in the multi-source feature fusion module 112 in obtaining the key feature set of the power distribution facility based on the unnormalized multi-dimensional running dataset, and will not be elaborated here.

[0049] In one embodiment, the power distribution facility status assessment system further includes a visualization and early warning module.

[0050] The visualization and early warning module is used to determine the danger warning value corresponding to the comprehensive status index, and to issue warnings and visualize the warnings based on the danger warning value.

[0051] In this embodiment, the power distribution facility status assessment system stores pre-defined danger warning values, such as 80-100 for normal conditions; 60-79 for minor anomalies; 30-59 for moderate anomalies; and 0-29 for severe anomalies. After determining the danger warning values, the visualization warning module judges the range of danger warning values ​​corresponding to the comprehensive status index to determine the danger warning value corresponding to the comprehensive status index. When the visualization warning module determines that the comprehensive status index is between 80-100, it visualizes the information indicating that the power distribution facility status assessment is normal; when it determines that the comprehensive status index is between 60-79, it visualizes the information indicating that the power distribution facility status assessment is minor anomalies; when it determines that the comprehensive status index is between 30-59, it visualizes the information indicating that the power distribution facility status assessment is moderate anomalies; and when it determines that the comprehensive status index is between 0-29, it visualizes the information indicating that the power distribution facility status assessment is severe anomalies.

[0052] In this embodiment, by determining the multi-level hazard warning values ​​corresponding to the comprehensive status index and performing graded warnings and visualization, it is possible to achieve quantitative grading, accurate warnings and intuitive presentation of equipment risks, reduce the efficiency of missed and false alarms, improve the efficiency of operation and maintenance response and handling, and ensure the safe operation of equipment.

[0053] In one embodiment, the power distribution facility condition assessment system may further include a data storage module.

[0054] The data storage module is used to store the comprehensive state index and the running datasets of each dimension corresponding to the comprehensive state index.

[0055] In this embodiment, storing the comprehensive status index and the corresponding operational datasets for each dimension is to enable traceability of equipment status, retrospective analysis of faults, iterative modeling, and predictable trends; providing complete data support for operation and maintenance decisions, fault analysis, and system optimization, thereby improving system reliability and intelligence.

[0056] Example 2 In this embodiment, as Figure 2 As shown, a method for assessing the condition of power distribution facilities is provided, including: Step S211: Real-time collection of electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters of power distribution facilities by sensors to form a multi-dimensional operating dataset.

[0057] In this embodiment, various sensors are installed at each monitoring point of the power distribution facility. The electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters of the power distribution facility are collected in real time through these sensors. Finally, the electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters are used as a multi-dimensional operating dataset.

[0058] Step S212: Extract the local feature set and time series feature set of each dimension of the running dataset, and fuse the local feature set and time series feature set of each dimension of the data to obtain the key feature set of the power distribution facility. The key feature set of the power distribution facility includes the key feature set of the fused electrical operating status, the key feature set of the fused equipment environmental status, and the key feature set of the fused equipment mechanical status.

[0059] After collecting multi-dimensional operational datasets, signal processing algorithms are used to extract local and temporal features of each dimension of the data, forming local and temporal feature sets for each dimension. These sets are then further fused to obtain the key feature set of the power distribution facility. Specifically, the signal processing algorithm is a multi-modal fusion algorithm using a convolutional network model and a bidirectional LSTM (Long Short-Term Memory) model. First, the multi-dimensional operational dataset undergoes temporal processing to obtain a multi-dimensional temporal dataset. The convolutional network model is then used to perform convolution and pooling on the multi-dimensional temporal dataset to extract local feature sets. These local feature sets are then input into the bidirectional LSTM model, which fits the long-term temporal dependencies of the multi-source features of the power distribution facility corresponding to the local feature sets, outputting a temporal hidden state sequence H. ;in . These are local characteristics of the state of power distribution facilities; The convolution kernel weights correspond to the state of the positive power distribution facility; This refers to the local characteristics of the positive power distribution facility's status; The convolution kernel weights correspond to the reverse power distribution facility status; This refers to the local characteristics of the reverse power distribution facility's condition; This is the bias value for the state characteristics of the power distribution facility.

[0060] In this embodiment, , These are local characteristics of the electrical operating state; These are local characteristics of the equipment's environmental state. These are local characteristics of the equipment's mechanical state, specifically: in, These are local characteristics of the electrical operating state; These are the convolution kernel weights corresponding to the positive electrical operating state; These are local characteristics of a positive electrical operating state; The convolution kernel weights are the values ​​corresponding to the reverse electrical operating state. These are local characteristics of the reverse electrical operating state; This is the bias value for the electrical operating state characteristics.

[0061] in, These are local characteristics of the equipment's environmental state. These are the convolutional kernel weights corresponding to the positive device environment state; Local features representing the positive device environment state; The convolutional kernel weights are the values ​​corresponding to the reverse device environment state. These are local features of the reverse device's environmental state; This is the bias value for the environmental condition characteristics of the equipment.

[0062] in, These are local features of the equipment's mechanical state; These are the convolution kernel weights corresponding to the forward mechanical state of the device; These are local features of the mechanical state of the forward-moving equipment; These are the convolution kernel weights corresponding to the mechanical state of the reverse device; These are local features of the mechanical state of the reversed device; This is the bias value for the mechanical state characteristics of the equipment.

[0063] After the multi-source feature fusion module 112 outputs the temporal hidden state sequence H, the temporal hidden state sequence H is further weighted (i.e., given greater weight according to the fault-sensitive key temporal features to weaken noise features) to obtain the global fused features. .in For the key feature set of the nth type of power distribution facility, For the local characteristics of the state of the t-th type of power distribution facility, This is the Hadamard product operator; The weighted fusion feature parameters are the local features corresponding to the state of the t-th type of power distribution facility.

[0064] In this embodiment, , This is the key feature set of the electrical operating status after integration; This is a set of key features representing the environmental state of the merged equipment. This is the key feature set of the mechanical state of the equipment after fusion.

[0065] Step S213 inputs the key feature set into the preset time series prediction model, and calculates the comprehensive state index of the power distribution facility through the preset time series prediction model, wherein the preset time series prediction model is a model formed by fusing the long short-term memory network model and the fault key feature model.

[0066] In this embodiment, key feature sets are input into a preset time series prediction model, and the comprehensive state index of the power distribution facility is calculated through the preset time series prediction model. The specific steps for calculating the comprehensive state index of the power distribution facility are as follows: determine the number of types of key feature sets of the power distribution facility, substitute the key feature sets of different types of power distribution facilities into the preset time series prediction model, perform data calculation and processing analysis through the preset time series prediction model to obtain the comprehensive state index, and then output the comprehensive state index. In this embodiment, the preset time series prediction model is a model formed by fusing a long short-term memory network model and a fault key feature model, specifically: in, It is a comprehensive state index; This is the index for the first type of power distribution facilities; This is the index for the second type of power distribution facilities; This is the index for the nth type of power distribution facility.

[0067] in, This is the weight matrix for the state of the nth type of power distribution facility in the fully connected layer; This represents the key feature set for the nth type of power distribution facility. This is the bias value for the nth type of power distribution facility.

[0068] In practical applications, the preset time series forecasting model is as follows: in, It is a comprehensive state index; This refers to the electrical operating status index; This is the weight matrix for the electrical operating state of the fully connected layer; This is the key feature set of the electrical operating status after integration; This is the bias value for the electrical operating state; This refers to the equipment's environmental condition index. This is a weight matrix representing the environmental state of the fully connected layer devices. This is a set of key features representing the environmental state of the merged equipment. This is the bias value for the device's environmental conditions; This refers to the mechanical condition index of the equipment. This is the weight matrix for the mechanical state of the fully connected layer device; This is the key feature set of the mechanical state of the fused equipment; This is the bias value for the mechanical state of the equipment.

[0069] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0070] In one embodiment, the fusion of the local feature sets and time-series feature sets of the data from each dimension to obtain a key feature set of the power distribution facility includes a fused key feature set of electrical operating status, a fused key feature set of equipment environmental status, and a fused key feature set of equipment mechanical status, including: 1-1) Perform Hadamard product operation on the local features of the electrical operating parameters to obtain the key feature set of the fused electrical operating state.

[0071] 1-2) Perform Hadamard product operation on the local features of the equipment environment parameters to obtain the key feature set of the fused equipment environment state.

[0072] 1-3) Perform Hadamard product operation on the local features of the equipment's mechanical state parameters to obtain the key feature set of the fused equipment mechanical state.

[0073] In one embodiment, the preset time series prediction model is: in, It is a comprehensive state index; This refers to the electrical operating status index; This is the weight matrix for the electrical operating state of the fully connected layer; This is the key feature set of the electrical operating status after integration; This is the bias value for the electrical operating state; This refers to the equipment's environmental condition index. This is a weight matrix representing the environmental state of the fully connected layer devices. This is a set of key features representing the environmental state of the merged equipment. This is the bias value for the device's environmental conditions; This refers to the mechanical condition index of the equipment. This is the weight matrix for the mechanical state of the fully connected layer device; This is the key feature set of the mechanical state of the fused equipment; This is the bias value for the mechanical state of the equipment.

[0074] In one embodiment, the power distribution facility status assessment method further includes: controlling the comprehensive status index within a preset index range; wherein the preset index range is... .

[0075] In one embodiment, the power distribution facility status assessment method further includes: performing time-series alignment and normalization on the multi-dimensional operational dataset to obtain a normalized multi-dimensional operational dataset.

[0076] In one embodiment, the power distribution facility status assessment method further includes: determining the danger warning value corresponding to the comprehensive status index, and issuing warnings and visualizations based on the danger warning value.

[0077] In one embodiment, the power distribution facility status assessment method further includes: storing the comprehensive status index and the operational datasets corresponding to each dimension of the comprehensive status index.

[0078] Example 3 In this embodiment, a computer device is provided. Its internal structure diagram can be shown as follows: Figure 3As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs, and it also contains a database for storing all data related to the power distribution facility condition assessment system and method. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used to communicate with other computer devices that have deployed application software. When the computer program is executed by the processor, it implements a power distribution facility condition assessment system and method. The display screen of the computer device can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0079] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0080] In one embodiment, an electronic device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the power distribution facility condition assessment system and method described in any of the above embodiments.

[0081] Example 4 In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the power distribution facility status assessment system and method described in any of the above embodiments.

[0082] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, all of which fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A power distribution facility condition assessment system, characterized in that, include: The data acquisition module is used to collect electrical operating parameters, equipment environmental parameters, and equipment mechanical status parameters of power distribution facilities in real time through sensors, forming a multi-dimensional operating dataset; The multi-source feature fusion module is used to extract local feature sets and time-series feature sets of data in each dimension of the running dataset, and fuse the local feature sets and time-series feature sets of data in each dimension to obtain the key feature set of the power distribution facility. The key feature set of the power distribution facility includes the key feature set of the fused electrical operating state, the key feature set of the fused equipment environmental state, and the key feature set of the fused equipment mechanical state. The status assessment module is used to input the key feature set into a preset time series prediction model, and calculate the comprehensive status index of the power distribution facility through the preset time series prediction model. The preset time series prediction model is a model formed by fusing a long short-term memory network model and a fault key feature model.

2. The power distribution facility condition assessment system according to claim 1, characterized in that, The multi-source feature fusion module includes an electrical operating status feature fusion unit, an equipment environmental status feature fusion unit, and an equipment mechanical status feature fusion unit. The electrical operating state feature fusion unit is used to perform Hadamard product operation on the local features of electrical operating parameters to obtain the key feature set of the fused electrical operating state. The equipment environment state feature fusion unit performs Hadamard product operation on the local features of the equipment environment parameters to obtain the key feature set of the fused equipment environment state. The equipment mechanical state feature fusion unit is used to perform Hadamard product operation on the local features of the equipment mechanical state parameters to obtain the key feature set of the fused equipment mechanical state.

3. The power distribution facility condition assessment system according to claim 1, characterized in that, The preset time series prediction model is: in, It is a comprehensive state index; This refers to the electrical operating status index; This is the weight matrix for the electrical operating state of the fully connected layer; This is the key feature set of the electrical operating status after integration; This is the bias value for the electrical operating state; This refers to the equipment's environmental condition index. This is a weight matrix representing the environmental state of the fully connected layer devices. This is a set of key features representing the environmental state of the merged equipment. This is the bias value for the device's environmental conditions; This refers to the mechanical condition index of the equipment. This is the weight matrix for the mechanical state of the fully connected layer device; This is the key feature set of the mechanical state of the fused equipment; This is the bias value for the mechanical state of the equipment.

4. The power distribution facility condition assessment system according to claim 1, characterized in that, The power distribution facility condition assessment system also includes an interception module: The interception module is used to control the comprehensive state index within a preset index range; wherein the preset index range is... .

5. The power distribution facility condition assessment system according to claim 1, characterized in that, The power distribution facility condition assessment system also includes a data preprocessing module: The data preprocessing module is used to perform time-series alignment and normalization on the multi-dimensional running dataset to obtain a normalized multi-dimensional running dataset. The multi-source feature fusion module is also used to extract the local feature sets and time-series feature sets of each dimension of the normalized multi-dimensional running dataset, and to fuse the local feature sets and time-series feature sets of each dimension to obtain the key feature set of the power distribution facility.

6. The power distribution facility condition assessment system according to any one of claims 1-5, characterized in that, The power distribution facility status assessment system also includes a visualization and early warning module: The visualization and early warning module is used to determine the danger warning value corresponding to the comprehensive status index, and to provide early warning and visualization based on the danger warning value.

7. The power distribution facility condition assessment system according to any one of claims 1-5, characterized in that, The power distribution facility condition assessment system may also include a data storage module: The data storage module is used to store the comprehensive state index and the corresponding operational datasets for each dimension of the comprehensive state index.

8. A method for assessing the condition of power distribution facilities, characterized in that, include: By collecting electrical operating parameters, equipment environmental parameters, and equipment mechanical condition parameters of power distribution facilities in real time through sensors, a multi-dimensional operating dataset is formed. Local feature sets and time-series feature sets of each dimension of the running dataset are extracted, and the local feature sets and time-series feature sets of each dimension of the data are fused to obtain the key feature set of the power distribution facility. The key feature set of the power distribution facility includes the key feature set of the fused electrical operating state, the key feature set of the fused equipment environmental state, and the key feature set of the fused equipment mechanical state. The key feature set is input into a preset time series prediction model, and the comprehensive state index of the power distribution facility is calculated through the preset time series prediction model. The preset time series prediction model is a model formed by fusing a long short-term memory network model and a fault key feature model.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, The processor is used to implement the power distribution facility condition assessment system according to any one of claims 1 to 7 when running the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the power distribution facility condition assessment system as described in any one of claims 1 to 7.

Images