A system and method for monitoring the dynamic environment and equipment status of a high-voltage distribution station

Abstract

The present invention discloses a dynamic environment and equipment status monitoring system and method for a high-voltage distribution station. The system includes an acquisition and integration module for multi-dimensional data perception, a heterogeneous data feature extraction and fusion transmission module, a pre-analysis module based on composite weight distribution, an association mapping module combined with historical data, a core diagnosis module based on an expert system rule base, and a multi-modal visualization output module. By collecting multi-source data such as temperature and humidity, gas concentration, vibration, voltage and current, after feature extraction and fusion, the potential relationship of the data is mined using composite weight distribution and historical data association, and then the equipment status is evaluated and faults are diagnosed based on the expert system rule base. Finally, the system is visualized and an early warning is output in a multi-modal form. The system and method can comprehensively monitor the distribution station environment and equipment status, accurately diagnose faults, and effectively improve the operational safety and operation and maintenance efficiency of high-voltage distribution stations.

Description

Technical Field

[0001] The present invention relates to the field of high-voltage distribution station monitoring, and in particular to a system and method for monitoring the dynamic environment and equipment status of a high-voltage distribution station. Background Art

[0002] As the scale of power systems continues to expand, high-voltage distribution substations, as key hubs for power transmission and distribution, are crucial for their operational safety and stability. High-voltage distribution substations house complex equipment and operate in a volatile environment. Equipment failures can not only cause localized power outages but can also trigger chain reactions, causing widespread blackouts and severely impacting social production and life. Furthermore, with the advancement of smart grid development, the need for monitoring high-voltage distribution substations is shifting from traditional single-parameter monitoring to comprehensive monitoring of the dynamic environment and equipment status, enabling early warning and accurate diagnosis of potential equipment failures.

[0003] However, existing high-voltage distribution substation monitoring technology has significant shortcomings. For one thing, data processing capabilities are limited. Traditional monitoring systems can often only collect and analyze a few key parameters of equipment, failing to fully integrate dynamic environmental parameters with equipment status parameters, and struggling to uncover potential correlations between different parameters. For example, when analyzing equipment failures, the impact of environmental factors such as temperature, humidity, and gas concentration on the equipment's operating status is often overlooked, resulting in an inability to promptly detect equipment hazards caused by environmental changes. Furthermore, fault diagnosis is inaccurate and time-sensitive. Most monitoring systems rely on fixed thresholds to determine whether equipment is faulty, lacking dynamic assessment of the equipment's operating status and in-depth analysis based on expert knowledge. When equipment exhibits early signs of failure or complex faults, it is difficult to quickly and accurately determine the type and cause of the fault, failing to meet the requirements for efficient operation and maintenance of high-voltage distribution substations. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the prior art, the present invention provides a system and method for monitoring the dynamic environment and equipment status of a high-voltage distribution station.

[0005] The technical solution adopted by the present invention is a high-voltage distribution station dynamic environment and equipment status monitoring system, including: a multi-dimensional data perception acquisition and integration module, which is connected to subsequent modules through a wired or wireless communication link and is used to collect temperature and humidity, gas concentration, vibration, voltage, current dynamic environment and equipment status parameters in the high-voltage distribution station; a heterogeneous data feature extraction and fusion transmission module, which receives data from the acquisition and integration module, extracts features from different types of data, and transmits the fused data through a preset data transmission protocol; a pre-analysis module based on composite weight distribution, which receives data from the fusion transmission module and performs preliminary analysis and processing on the equipment status parameters according to a preset weight distribution rule; a correlation mapping module combined with historical data, which correlates and maps the results of the pre-analysis module with historical data to explore potential relationships between data; a core diagnosis module based on an expert system rule base, which receives data from the correlation mapping module and uses the rule base in the expert system model to evaluate and diagnose the equipment status; a multimodal visualization output module, which receives the diagnosis results of the core module, visualizes the results in a multimodal form, and outputs early warning information.

[0006] Furthermore, in the pre-analysis module based on composite weight allocation, a composite weight allocation model is constructed using the equipment status assessment and fault diagnosis algorithm, and the formula is: ,in, is the final weight vector, For the The empirical weight coefficient of each device status parameter in the expert system model is in the range of , determined by experts based on historical failure data and equipment operating characteristics; For the The dynamic change weight coefficient of each device status parameter is calculated based on the change amplitude and frequency of the parameter within the set time window; For the The original data vector of the device state parameters, is the total number of device status parameters; in the association mapping module combining historical data, a dynamic environment and device status correlation model is constructed: ,in, is the correlation between dynamic environment parameters and equipment status parameters, For the Dynamic environment parameter sample values, is the sample mean of dynamic environmental parameters, For the Sample values ​​of device status parameters, is the sample mean of the equipment status parameter, is the sample size.

[0007] Furthermore, the pre-analysis module based on composite weight distribution also constructs a parameter fluctuation abnormality discrimination model: ,in, is the parameter fluctuation abnormality discrimination value, For the The device status parameters are The value of the moment, For the The device status parameters are The value of the moment, For the The historical mean of the device status parameters, For the The historical standard deviation of the device status parameters; the correlation mapping module combined with historical data further constructs a correlation trend prediction model: ,in, for The predicted value of the equipment status parameter at the moment, For the The prediction weight coefficient corresponding to the correlation degree is For the The correlation between dynamic environment parameters and equipment status parameters, for Dynamic environmental parameter values ​​at the moment, is the number of historical correlations considered, is the prediction error term.

[0008] Furthermore, the core diagnosis module based on the expert system rule base constructs an equipment failure probability assessment model: ,in, is the probability of equipment failure, The first The weight of the rule, The current device status parameter set Satisfy the The value is 1 when the rule is met, and the value is is the total number of rules in the expert system rule base; at the same time, a fault type discrimination model is constructed:

[0009]

[0010] in, is the fault type, is the set of all possible fault types, For the The influence coefficient of each device status parameter on the fault type judgment, The device status parameter For fault type If there is an impact, the value is 1; if there is no impact, the value is The number of device status parameters.

[0011] Furthermore, the multi-dimensional data perception acquisition integration module collects temperature and humidity parameters for analyzing the impact of the distribution station environment on equipment operation, and its acquisition frequency is dynamically adjusted according to the equipment operating status and environmental changes; the collected gas concentration parameters include oxygen, carbon monoxide, and sulfur hexafluoride gas concentrations, which are used to monitor whether there are potential equipment leakage faults; the collected vibration parameters obtain the vibration signal of the equipment during operation through a preset sensor to analyze the mechanical operation status of the equipment; the collected voltage and current parameters are used to evaluate the electrical operation status of the equipment; in the pre-analysis module based on composite weight distribution, a weight adjustment model based on the equipment operation stage is constructed: ,in, is the adjusted weight vector, For devices in The weight adjustment coefficient during the operation phase is set to different values ​​according to the different stages of equipment startup, normal operation, and shutdown.

[0012] Furthermore, in the core diagnosis module based on the expert system rule base, a comprehensive equipment health evaluation model is constructed: ,in, For device health, For the The correction coefficient of the impact of each fault type on the equipment health, For the The probability of occurrence of each fault type, is the total number of fault types; in the multimodal visualization output module, a warning level classification model is constructed: ,in, For the warning level, is the comprehensive evaluation index value of equipment failure, Thresholds for different warning levels.

[0013] Furthermore, the pre-analysis module based on composite weight distribution includes a data cleaning unit for removing noise data and abnormal data in the collected data; a weight calculation unit for calculating the weight of the equipment status parameters according to preset rules and the constructed model; a parameter sorting unit for sorting the equipment status parameters according to the weight size; and an analysis result temporary storage unit for temporarily storing the result data of the pre-analysis for calling by different modules.

[0014] Furthermore, the association mapping module combined with historical data includes a historical data storage unit for storing historical data of the dynamic environment and equipment status during the long-term operation of the high-voltage distribution station; a data matching unit for matching the currently collected data with the historical data; an association relationship mining unit for mining the association relationship between the dynamic environment and equipment status data using a preset algorithm; and a mapping result generation unit for generating an association mapping result between the dynamic environment and the equipment status based on the mined association relationship.

[0015] Furthermore, the core diagnosis module based on the expert system rule base includes a rule base management unit, which is used to add, delete, and modify the expert system rule base; a rule matching unit, which matches the processed data with the rules in the rule base; a fault diagnosis reasoning unit, which performs fault diagnosis reasoning based on the matching results; and a diagnosis result output unit, which outputs the final equipment status evaluation and fault diagnosis results.

[0016] A method for monitoring the dynamic environment and equipment status of a high-voltage distribution station comprises the following steps:

[0017] Step S1: Using a multi-dimensional data sensing acquisition integration module, different parameters of the dynamic environment and equipment status in the high-voltage distribution station are collected according to the set acquisition strategy and frequency;

[0018] Step S2: The collected parameters are transmitted to the heterogeneous data feature extraction and fusion transmission module to extract features from the data and perform fusion transmission;

[0019] Step S3: using a pre-analysis module based on composite weight allocation to perform preliminary analysis and processing on the device status parameters according to the preset weight allocation rules and the constructed model;

[0020] Step S4: transmitting the pre-analysis results to the association mapping module combined with historical data, performing association mapping with the historical data, and mining potential relationships between the data;

[0021] Step S5: Input the association mapping result into the core diagnosis module based on the expert system rule base, and use the rule base in the expert system model to evaluate the equipment status and diagnose the fault;

[0022] Step S6: The diagnosis result is transmitted to the multimodal visualization output module, which is visualized in a multimodal form and outputs warning information.

[0023] Beneficial effects: The present invention proposes a dynamic environment and equipment status monitoring system and method for high-voltage distribution station. In response to the problem of limited data processing capabilities of traditional monitoring systems, this system comprehensively collects dynamic environment and equipment status parameters such as temperature and humidity, gas concentration, vibration, voltage and current through an acquisition integration module based on multi-dimensional data perception, and uses a heterogeneous data feature extraction and fusion transmission module to achieve deep fusion and efficient transmission of multiple types of data. At the same time, the pre-analysis module based on composite weight distribution and the association mapping module combined with historical data work together, which can not only assign different weights according to parameter change characteristics and expert experience, but also deeply explore the potential connection between environment and equipment status parameters, changing the limitations of previous single parameter analysis and greatly improving the comprehensiveness and relevance of data processing. In terms of fault diagnosis, the problems of poor accuracy and timeliness of traditional technologies are solved. The core module of this system is based on the expert system rule base, combined with equipment status evaluation and fault diagnosis algorithms, and by constructing multiple evaluation and discrimination models, from fault probability assessment to fault type discrimination, and then to comprehensive equipment health assessment, a complete diagnostic system is formed. The system dynamically assesses equipment operating status based on historical data and expert experience, accurately identifying early signs of faults and the causes of complex failures. Furthermore, a multimodal visual output module presents diagnostic results in an intuitive, multimodal format and categorizes warning levels according to pre-set rules, ensuring that operations and maintenance personnel can quickly access critical information and take timely action. This significantly improves the accuracy and timeliness of fault diagnosis and ensures the safe and stable operation of high-voltage distribution stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a diagram of the system module composition of the present invention;

[0025] Figure 2 The flowchart of the method of the present invention is shown. DETAILED DESCRIPTION

[0026] It should be noted that, unless there is a conflict, the embodiments in this application and the features described in the embodiments can be combined with each other. The application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] like Figure 1 As shown, a high-voltage distribution substation dynamic environment and equipment status monitoring system includes: a multi-dimensional data perception acquisition integration module, which is connected to subsequent modules via a wired or wireless communication link and is used to collect temperature and humidity, gas concentration, vibration, voltage, current dynamic environment and equipment status parameters in the high-voltage distribution substation;

[0028] Specifically, the multi-dimensional data perception acquisition and integration module is the fundamental unit of data for the entire monitoring system. This module is designed to comprehensively collect all dynamic environmental and equipment status parameters related to equipment operation within the high-voltage distribution substation, including but not limited to temperature and humidity, gas concentrations (such as oxygen, carbon monoxide, and sulfur hexafluoride), equipment vibration parameters, and electrical parameters such as voltage and current. The collection of these parameters is crucial for subsequent analysis of equipment operating conditions, assessment of equipment health, and prediction of potential failures. Their integrity and accuracy directly impact the effectiveness of the entire monitoring system. This module establishes a comprehensive data acquisition network by rationally deploying various sensors at different locations within the distribution substation, targeting different monitoring targets.

[0029] In its implementation, the module utilizes a variety of sensors. For example, temperature and humidity sensors collect temperature and humidity data within the distribution substation, with accuracy requirements of ±0.5°C for temperature and ±3% RH for humidity. The acquisition frequency can be dynamically adjusted based on equipment operating status and environmental changes. Gas concentration sensors monitor the levels of various gases, with a detection limit of ppm for key gases. Vibration sensors, using accelerometers and other devices, acquire vibration signals during equipment operation, with a sampling frequency of at least 10kHz. Voltage and current sensors utilize high-precision transformers to ensure measurement errors within ±0.2%. Data collected by these sensors is transmitted to subsequent modules via wired or wireless communication links, following specific data protocols to ensure stable and real-time data transmission.

[0030] Heterogeneous data feature extraction and fusion transmission module receives data from the collection and integration module, extracts features from different types of data, and transmits the fused data through a preset data transmission protocol;

[0031] Specifically, the heterogeneous data feature extraction and fusion transmission module mainly solves the problem of processing and transmitting collected multi-source heterogeneous data. Since the data types acquired by the acquisition integration module are diverse, including analog signals, digital signals, etc., and the data formats, units and features are all different, this module uses specific algorithms and techniques to extract features from different types of data and converts them into data forms with unified feature representations to facilitate subsequent analysis and processing. At the same time, the module performs fusion processing on the data after feature extraction, integrates data information of different dimensions, and forms a more comprehensive and representative data set. It also transmits the fused data efficiently to the next module through a specific data transmission protocol, ensuring the accuracy and integrity of the data during transmission, and providing reliable data support for equipment status assessment and fault diagnosis.

[0032] During implementation, this module first applies corresponding feature extraction algorithms for different types of data. For continuous data such as temperature and humidity, sliding window statistics are used to extract features such as mean and variance. For gas concentration data, spectrum analysis is used to extract its changing trend characteristics. For vibration signals, methods such as Fourier transform are used to extract frequency domain features. For electrical parameters such as voltage and current, features such as effective value and peak value are extracted. During the data fusion stage, a fusion strategy based on weighted average is adopted, assigning weights based on the importance and reliability of different data types to generate fused data. Finally, the fused data is packaged and transmitted to a pre-analysis module based on composite weight allocation via mature data transmission protocols such as TCP / IP. Data verification mechanisms, such as CRC, are used during transmission to ensure the accuracy of data transmission.

[0033] The pre-analysis module based on composite weight distribution receives data from the fusion transmission module and performs preliminary analysis and processing on the device status parameters according to the preset weight distribution rules;

[0034] Specifically, the pre-analysis module, based on composite weight allocation, is a crucial step in the initial analysis and processing of data transmitted from the fusion transmission module. Its core function is to assign appropriate weights to each device status parameter based on pre-set weight allocation rules, taking into account the empirical importance of device status parameters in the expert system model and the dynamic characteristics of the parameters themselves. This helps highlight the role of key parameters in device status assessment. This module also performs preliminary anomaly detection and analysis on device status parameters, screening out potentially abnormal parameters and providing guidance for subsequent, more in-depth analysis. This reduces data processing blindness and improves the efficiency and targeted nature of the entire monitoring system.

[0035] During actual implementation, the module first determines the empirical weight coefficient for each device status parameter based on the historical fault data and equipment operating characteristics accumulated in the expert system model. This coefficient reflects the a priori importance of the parameter's impact on the equipment failure. Then, by calculating indicators such as the amplitude and frequency of parameter changes within a set time window, its dynamic change weight coefficient is determined to measure the impact of the parameter's real-time changes on the equipment status. The empirical weight coefficient and the dynamic change weight coefficient are combined to calculate the final weight of each parameter. These weights are then used to weight the device status parameters and compare them with historical data and preset thresholds to determine whether there are any abnormal fluctuations in the parameters. If an abnormal parameter is found, it is marked and transmitted, along with the weight information, to the association mapping module that combines historical data for further analysis.

[0036] Combined with the historical data association mapping module, the results of the pre-analysis module are associated and mapped with the historical data to explore the potential relationships between the data;

[0037] Specifically, the association mapping module, which integrates historical data, focuses on exploring potential relationships between device status parameters and dynamic environmental parameters, as well as between current data and historical data. By analyzing large amounts of historical data and establishing association models between environmental and device status parameters, this module can identify the impact of environmental factors on device operating status, as well as the trends and patterns of device status parameters over time. Uncovering these associations facilitates a more comprehensive understanding of device operating mechanisms, proactively identifying potential failure risks caused by environmental changes or gradual changes in the device itself, and providing richer and more in-depth information for device status assessment and fault diagnosis.

[0038] During specific implementation, the module first retrieves historical environmental and equipment status data related to the currently collected data from the historical data storage unit to form a data set. Then, data mining algorithms, such as correlation analysis and cluster analysis, are used to process the environmental parameters and equipment status parameters in the data set, calculate the correlation between different parameters, and find closely related parameter combinations. For these related parameters, a time series model is established to analyze their changing trends and mutual influence relationships at different time scales. By learning and training historical data, an association mapping model that can predict changes in equipment status parameters is constructed. After receiving abnormal parameter information from the pre-analysis module based on composite weight distribution, the model is used to further analyze the correlation between the abnormal parameters and other related parameters, determine the potential cause of the abnormality, and transmit the analysis results to the core diagnosis module based on the expert system rule base.

[0039] The core diagnosis module based on the expert system rule base receives data from the association mapping module and uses the rule base in the expert system model to evaluate the equipment status and diagnose faults;

[0040] Specifically, the core diagnostic module based on the expert system rule base is the key decision-making unit of the entire monitoring system. Based on the expert system rule base, this module combines equipment status assessment and fault diagnosis algorithms to conduct in-depth analysis of the data transmitted by the association mapping module to achieve accurate assessment of the equipment's operating status and fault diagnosis. The expert system rule base stores a large number of rules summarized based on expert experience and historical fault cases. These rules cover the normal state, abnormal state and fault type judgment basis of different types of equipment under various operating conditions. By matching and reasoning the processed data with the rules in the rule base, the module can determine whether the current operating status of the equipment is normal. If there is an abnormality, it can further determine the fault type, fault location and probability of the fault, providing a scientific basis for the operation and maintenance management of the equipment.

[0041] During implementation, the module first converts and standardizes the received data to ensure it meets the input requirements of the rules in the rule base. Then, using a rule-matching algorithm, it matches the processed data against each rule in the rule base to determine whether the data satisfies the rule conditions. When a rule that meets the conditions is found, it infers the probability of equipment failure based on the logic defined by the rule and determines the possible fault type based on the fault type defined in the rule. For complex fault scenarios, the module can also utilize an uncertain reasoning algorithm to comprehensively consider the influence of multiple relevant rules and parameters, thereby improving the accuracy of fault diagnosis. Finally, the equipment status assessment results and fault diagnosis conclusions are compiled and transmitted to the multimodal visualization output module.

[0042] The multimodal visualization output module receives the diagnostic results of the core module, visualizes the results in a multimodal form and outputs warning information.

[0043] Specifically, the multimodal visualization output module, as the final display and interaction unit of the entire monitoring system, undertakes the important task of presenting equipment status assessment and fault diagnosis results to operation and maintenance personnel in an intuitive and easy-to-understand manner. This module uses multimodal display technology to display diagnostic results in various forms such as graphics, charts, and text, allowing operation and maintenance personnel to quickly obtain key information on the equipment's operating status. At the same time, the module classifies the diagnostic results according to the severity of the equipment failure and the preset warning level classification rules. When an abnormality or failure is detected in the equipment, the module promptly issues a warning message of the corresponding level to remind the operation and maintenance personnel to take appropriate measures to ensure the safe and stable operation of the high-voltage distribution station.

[0044] During specific implementation, the module first receives the equipment status assessment and fault diagnosis result data from the core module based on the expert system rule base. Then, according to the data type and content, a suitable visualization method is selected for display. For example, for the trend data of the equipment operating status, a line chart or a curve chart is used for display; for the equipment fault type and location information, a schematic diagram or a topology diagram is used for display. In terms of early warning output, the threshold is divided according to the preset early warning level, and the comprehensive evaluation index value of the equipment fault is compared with the threshold to determine the early warning level. When an abnormality occurs in the equipment, early warning information is issued through various methods such as sound and light alarms, SMS notifications, and email reminders. The early warning information contains key content such as the equipment name, fault type, fault location, and early warning level to ensure that operation and maintenance personnel can understand the equipment status in a timely manner and take effective treatment measures.

[0045] Preferably, in the pre-analysis module based on composite weight allocation, a composite weight allocation model is constructed using an equipment status assessment and fault diagnosis algorithm, and the formula is: ,in, is the final weight vector, For the The empirical weight coefficient of each device status parameter in the expert system model is in the range of , determined by experts based on historical failure data and equipment operating characteristics; For the The dynamic change weight coefficient of each device status parameter is calculated based on the change amplitude and frequency of the parameter within the set time window; For the The original data vector of the device state parameters, is the total number of device status parameters; in the association mapping module combining historical data, a dynamic environment and device status correlation model is constructed: ,in, is the correlation between dynamic environment parameters and equipment status parameters, For the Dynamic environment parameter sample values, is the sample mean of dynamic environmental parameters, For the Sample values ​​of device status parameters, is the sample mean of the equipment status parameter, is the sample size.

[0046] Specifically, the pre-analysis module, based on composite weight allocation, constructs a composite weight allocation model to accurately calculate the weights of equipment status parameters. This model combines empirical weight coefficients determined by the expert system based on historical fault data and equipment operating characteristics with dynamic weight coefficients derived from the amplitude and frequency of parameter changes within a set time window. This model comprehensively considers the a priori importance of the parameters and their real-time fluctuations to assign appropriate weights to each equipment status parameter. During implementation, the expert team first sets empirical weight coefficients based on historical cases and equipment characteristics. Then, an algorithm calculates the dynamic weight coefficients in real time. The two coefficients are multiplied and combined with the original data vector to produce the final weight vector, highlighting the role of key parameters in equipment status assessment. The correlation model, constructed by the correlation mapping module based on historical data, aims to uncover potential connections between dynamic environmental parameters and equipment status parameters. The relationship between sample values ​​and the mean of these two parameter types is calculated to quantify the degree of correlation. During implementation, the system continuously collects samples of dynamic environmental and equipment status parameters, calculates their mean, and substitutes these mean values ​​into the model formula to derive the correlation value, providing data support for subsequent environmental-based prediction of equipment status changes and troubleshooting of potential faults.

[0047] Preferably, the pre-analysis module based on composite weight distribution further constructs a parameter fluctuation abnormality discrimination model: ,in, is the parameter fluctuation abnormality discrimination value, For the The device status parameters are The value of the moment, For the The device status parameters are The value of the moment, For the The historical mean of the device status parameters, For the The historical standard deviation of the device status parameters; the correlation mapping module combined with historical data further constructs a correlation trend prediction model: ,in, for The predicted value of the equipment status parameter at the moment, For the The prediction weight coefficient corresponding to the correlation degree is For the The correlation between dynamic environment parameters and equipment status parameters, for Dynamic environmental parameter values ​​at the moment, is the number of historical correlations considered, is the prediction error term.

[0048] Specifically, within the pre-analysis module based on composite weight allocation, the parameter fluctuation anomaly discrimination model compares the values ​​of the device status parameter at adjacent moments and combines the parameter's historical mean and standard deviation to determine whether the parameter fluctuation is abnormal. This model effectively identifies sudden changes or sustained deviations from the normal range, providing early warning signals for faults. During implementation, the system obtains the current and previous values ​​of the device status parameter in real time, retrieves the historical mean and standard deviation data, and substitutes these values ​​into the model formula to calculate the fluctuation anomaly discrimination value. If this value exceeds a preset threshold, the parameter is flagged as experiencing abnormal fluctuation. Combined with the historical data correlation mapping module's correlation trend prediction model, this model uses the mined correlations between the dynamic environment and the device status parameter, as well as the historical environmental parameter values, to predict the future values ​​of the device status parameter. The system first determines the number of historical correlations to consider, obtains the corresponding historical environmental parameter values ​​and correlations, and then, combining them with the prediction weight coefficients, calculates the predicted value of the device status parameter using the model formula. This helps operations and maintenance personnel proactively identify trends in device status changes and take timely maintenance measures.

[0049] Preferably, the core diagnosis module based on the expert system rule base constructs an equipment failure probability assessment model: ,in, is the probability of equipment failure, The first The weight of the rule, The current device status parameter set Satisfy the The value is 1 when the rule is met, and the value is is the total number of rules in the expert system rule base; at the same time, a fault type discrimination model is constructed:

[0050]

[0051] in, is the fault type, is the set of all possible fault types, For the The influence coefficient of each device status parameter on the fault type judgment, The device status parameter For fault type If there is an impact, the value is 1; if there is no impact, the value is The number of device status parameters.

[0052] Specifically, the equipment failure probability assessment model is based on the rules in the expert system's rule base. It calculates the number of instances in which different rules are satisfied and, combined with the weights of each rule, determines the probability of equipment failure. This model transforms expert experience into quantifiable evaluation metrics, comprehensively considering multiple criteria for fault diagnosis. During implementation, the system compares the processed set of equipment status parameters with each rule in the rule base to determine whether the parameters meet the rule conditions. If so, the corresponding rule value is 1; otherwise, it is 0. The model then calculates the equipment failure probability using a formula based on the weights of each rule. The fault type discrimination model, starting from the set of all possible fault types, determines the current fault type of the equipment based on the influence coefficients of the equipment status parameters on different fault types. The system iterates through all equipment status parameters, determines the influence of each parameter on different fault types, and, combining the influence coefficients, calculates a comprehensive score for each fault type using a formula. The fault type with the highest score is selected as the diagnosis result, achieving accurate fault type discrimination.

[0053] Preferably, the multi-dimensional data perception acquisition integration module collects temperature and humidity parameters for analyzing the impact of the distribution station room environment on equipment operation, and its acquisition frequency is dynamically adjusted according to the equipment operation status and environmental changes; the collected gas concentration parameters include oxygen, carbon monoxide, and sulfur hexafluoride gas concentrations, which are used to monitor whether there is a potential equipment leakage fault; the collected vibration parameters obtain the vibration signal of the equipment during operation through a preset sensor to analyze the mechanical operation status of the equipment; the collected voltage and current parameters are used to evaluate the electrical operation status of the equipment; in the pre-analysis module based on composite weight distribution, a weight adjustment model based on the equipment operation stage is constructed: ,in, is the adjusted weight vector, For devices in The weight adjustment coefficient during the operation phase is set to different values ​​according to the different stages of equipment startup, normal operation, and shutdown.

[0054] Specifically, the multi-dimensional data perception acquisition integration module clarifies the specific uses and acquisition strategies of various acquisition parameters. The frequency of temperature and humidity parameter acquisition will be dynamically adjusted according to the equipment's operating status and environmental changes to ensure that the impact of environmental changes on the equipment can be captured in a timely manner; gas concentration parameters monitor a variety of key gases and can effectively detect potential faults such as equipment leaks; vibration parameters are used to analyze the mechanical operating status of the equipment; and voltage and current parameters evaluate the electrical operating status of the equipment. The weight adjustment model based on the equipment operation stage is constructed based on the pre-analysis module based on the composite weight distribution. It takes into account the different importance of each parameter to the equipment status assessment during the different stages of equipment startup, normal operation, and shutdown. During implementation, the system first determines the operating stage of the equipment, retrieves the weight adjustment coefficient for the corresponding stage, combines it with the weight vector calculated by the original composite weight distribution model, and derives the adjusted weight vector through the new model formula, so that the weight distribution is more in line with the actual operation of the equipment and improves the accuracy of parameter analysis.

[0055] Preferably, in the core diagnosis module based on the expert system rule base, a comprehensive equipment health evaluation model is constructed: ,in, For device health, For the The correction coefficient of the impact of each fault type on the equipment health, For the The probability of occurrence of each fault type, is the total number of fault types; in the multimodal visualization output module, a warning level classification model is constructed: ,in, For the warning level, is the comprehensive evaluation index value of equipment failure, Thresholds for different warning levels.

[0056] Specifically, the comprehensive equipment health assessment model based on the core diagnostic module of the expert system rule base comprehensively considers the probability of occurrence of multiple fault types and the correction coefficient of the impact of each fault type on equipment health to derive the overall equipment health index. This model evaluates the operating status of the equipment from a macro perspective and provides an important reference for the formulation of equipment maintenance strategies. During implementation, the system first calculates the probability of occurrence of each fault type, combines the corresponding correction coefficient, substitutes it into the model formula, and obtains the equipment health value. The warning level classification model of the multimodal visualization output module divides warnings into different levels based on the comprehensive evaluation index value of the equipment fault. The system obtains the comprehensive evaluation index value of the equipment fault in real time, compares it with the preset threshold value of each level, and determines the corresponding warning level. When the equipment is abnormal, different levels of warning information are issued in various ways according to the warning level, so that operation and maintenance personnel can quickly understand the severity of the fault and take appropriate countermeasures.

[0057] Preferably, the pre-analysis module based on composite weight distribution includes a data cleaning unit for removing noise data and abnormal data in the collected data; a weight calculation unit for calculating the weights of the equipment status parameters according to preset rules and the constructed model; a parameter sorting unit for sorting the equipment status parameters according to the weight size; and an analysis result temporary storage unit for temporarily storing the result data of the pre-analysis for calling by different modules.

[0058] Specifically, the internal unit structure and function of the pre-analysis module based on composite weight distribution are as follows: the data cleaning unit uses specific algorithms and rules to process the collected data, identify and remove noise data and abnormal data, improve data quality, and provide a reliable data basis for subsequent analysis; the weight calculation unit calculates the weight of each device status parameter based on the preset weight distribution rules and the constructed related models, taking into account the empirical importance and dynamic change characteristics of the parameters; the parameter sorting unit sorts the device status parameters according to the calculated weight, highlights the key parameters, and facilitates subsequent targeted analysis; the analysis result temporary storage unit provides a temporary storage function for the pre-analysis results, ensuring that the data is not lost before being transmitted to the next module and can be quickly called, thereby ensuring the consistency and efficiency of data processing of the entire monitoring system.

[0059] Preferably, the association mapping module combined with historical data includes a historical data storage unit for storing historical data of the dynamic environment and equipment status during the long-term operation of the high-voltage distribution station; a data matching unit for matching the currently collected data with the historical data; an association relationship mining unit for mining the association relationship between the dynamic environment and the equipment status data using a preset algorithm; and a mapping result generation unit for generating an association mapping result between the dynamic environment and the equipment status based on the mined association relationship.

[0060] Specifically, the internal units of the association mapping module combined with historical data are explained. The historical data storage unit is responsible for long-term preservation of the dynamic environment and equipment status historical data during the operation of the high-voltage distribution station, forming a huge database, and providing rich data resources for data association analysis; the data matching unit accurately matches the currently collected real-time data with the historical data, and finds the corresponding historical data records through key information such as timestamps and equipment identification; the association mining unit uses data mining algorithms to conduct in-depth analysis of the matched data, and mines the hidden associations between the dynamic environment and equipment status data; the mapping result generation unit generates intuitive data association mapping results based on the mined associations, and presents them in the form of structured data or visual charts, etc., to provide strong data support for subsequent equipment status evaluation and fault diagnosis.

[0061] Preferably, the core diagnosis module based on the expert system rule base includes a rule base management unit, which is used to add, delete, and modify the expert system rule base; a rule matching unit, which matches the processed data with the rules in the rule base; a fault diagnosis reasoning unit, which performs fault diagnosis reasoning based on the matching results; and a diagnosis result output unit, which outputs the final equipment status evaluation and fault diagnosis results.

[0062] Specifically, based on the internal unit functions of the core diagnosis module of the expert system rule base, the rule base management unit supports management operations on the expert system rule base, including adding new fault judgment rules, deleting invalid rules, and modifying the content of existing rules to ensure the accuracy and timeliness of the rule base; the rule matching unit compares the processed equipment status data with the rules in the rule base one by one to determine whether the data meets the rule conditions, providing a basis for fault diagnosis reasoning; the fault diagnosis reasoning unit uses logical reasoning algorithms based on the rule matching results to analyze the equipment status and infer whether the equipment has a fault and the type of fault; the diagnosis result output unit organizes the final equipment status evaluation and fault diagnosis results, and outputs them to the multimodal visualization output module in accordance with the specified data format and communication protocol so that operation and maintenance personnel can obtain relevant information.

[0063] like Figure 2 As shown, a method for monitoring the dynamic environment and equipment status of a high-voltage distribution station includes the following steps:

[0064] Step S1: Using a multi-dimensional data sensing acquisition integration module, different parameters of the dynamic environment and equipment status in the high-voltage distribution station are collected according to the set acquisition strategy and frequency;

[0065] Step S2: The collected parameters are transmitted to the heterogeneous data feature extraction and fusion transmission module to extract features from the data and perform fusion transmission;

[0066] Step S3: using a pre-analysis module based on composite weight allocation to perform preliminary analysis and processing on the device status parameters according to the preset weight allocation rules and the constructed model;

[0067] Step S4: transmitting the pre-analysis results to the association mapping module combined with historical data, performing association mapping with the historical data, and mining potential relationships between the data;

[0068] Step S5: Input the association mapping result into the core diagnosis module based on the expert system rule base, and use the rule base in the expert system model to evaluate the equipment status and diagnose the fault;

[0069] Step S6: The diagnosis result is transmitted to the multimodal visualization output module, which is visualized in a multimodal form and outputs warning information.

[0070] A system and method for monitoring the dynamic environment and equipment status of high-voltage distribution substations. This system integrates comprehensive environmental and equipment operating parameters, such as temperature and humidity, gas concentration, vibration, voltage and current, into its acquisition scope through an integrated acquisition module based on multi-dimensional data perception. Furthermore, a heterogeneous data feature extraction and fusion transmission module enables deep fusion and efficient transmission of multi-source data. A pre-analysis module based on composite weight allocation assigns weights based on expert experience and parameter variation characteristics. Combined with a correlation mapping module for historical data, this module further explores potential connections between parameters, overcoming the limitations of previous single-parameter analysis and significantly improving the comprehensiveness and relevance of data processing.

[0071] At the fault diagnosis level, traditional technologies rely on fixed thresholds, making accuracy and timeliness difficult to guarantee for early-stage and complex faults. This system's core diagnostic module, based on an expert system rule base, builds a complete diagnostic system encompassing fault probability assessment, fault type identification, and comprehensive equipment health assessment. This module combines equipment status assessment with fault diagnosis algorithms, dynamically evaluating equipment operating status based on historical data and expert experience. It can accurately identify early signs of faults and quickly and accurately determine the causes of complex faults, significantly improving the accuracy and timeliness of fault diagnosis.

[0072] In addition, this monitoring system and method also excel in information presentation and operation and maintenance response. The multimodal visualization output module displays the diagnostic results in an intuitive multimodal format and scientifically divides the warning levels according to preset rules, so that operation and maintenance personnel can quickly obtain key information and take timely measures. Each module is subdivided into multiple functional units, such as the data cleaning and weight calculation unit of the pre-analysis module, the historical data storage and association relationship mining unit of the association mapping module, etc., each performing its duties and operating in coordination, from data collection and analysis to fault diagnosis and result output, forming a complete and efficient monitoring system, which comprehensively guarantees the safe and stable operation of high-voltage distribution stations and significantly improves the operation and maintenance management level of the power system.

[0073] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "disposed," "installed," "connected," "connected," and "fixed" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0074] While embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A high-voltage distribution station dynamic environment and equipment status monitoring system, characterized in that: include: A multi-dimensional data sensing acquisition integration module, which is connected to subsequent modules via wired or wireless communication links, is used to collect temperature and humidity, gas concentration, vibration, voltage, current dynamic environment and equipment status parameters in the high-voltage distribution station room; Heterogeneous data feature extraction and fusion transmission module receives data from the collection and integration module, extracts features from different types of data, and transmits the fused data through a preset data transmission protocol; The pre-analysis module based on composite weight distribution receives data from the fusion transmission module and performs preliminary analysis and processing on the device status parameters according to the preset weight distribution rules; Combined with the historical data association mapping module, the results of the pre-analysis module are associated and mapped with the historical data to explore the potential relationships between the data; The core diagnosis module based on the expert system rule base receives data from the association mapping module and uses the rule base in the expert system model to evaluate the equipment status and diagnose faults; Construct an equipment failure probability assessment model based on the core diagnosis module of the expert system rule base: in, is the probability of equipment failure, The first The weight of the rule, The current device status parameter set Satisfy the The value is 1 when the rule is met, and the value is is the total number of rules in the expert system rule base; At the same time, a fault type discrimination model is constructed: in, is the fault type, is the set of all possible fault types, For the The influence coefficient of each device status parameter on the fault type judgment, The device status parameter For fault type If there is an impact, the value is 1; if there is no impact, the value is is the number of device status parameters; The multimodal visualization output module receives the diagnostic results of the core module, visualizes the results in a multimodal form and outputs warning information.

2. A high-voltage distribution station dynamic environment and equipment status monitoring system according to claim 1, characterized in that: In the pre-analysis module based on composite weight allocation, a composite weight allocation model is constructed using the equipment status assessment and fault diagnosis algorithm, and the formula is: in, is the final weight vector, For the The empirical weight coefficient of each device status parameter in the expert system model is in the range of , determined by experts based on historical failure data and equipment operating characteristics; For the The dynamic change weight coefficient of each device status parameter is calculated based on the change amplitude and frequency of the parameter within the set time window; For the The original data vector of the device state parameters, is the total number of device status parameters; In the association mapping module combining historical data, a dynamic environment and device status association model is constructed: in, is the correlation between dynamic environment parameters and equipment status parameters, For the Dynamic environment parameter sample values, is the sample mean of dynamic environmental parameters, For the Sample values ​​of device status parameters, is the sample mean of the equipment status parameter, is the sample size.

3. A high-voltage distribution station dynamic environment and equipment status monitoring system according to claim 1, characterized in that: The pre-analysis module based on composite weight distribution also constructs a parameter fluctuation abnormality discrimination model: in, is the parameter fluctuation abnormality discrimination value, For the The device status parameters are The value of the moment, For the The device status parameters are The value of the moment, For the The historical mean of the device status parameters, For the The historical standard deviation of each device status parameter; The association mapping module combined with historical data also constructs an association trend prediction model: in, for The predicted value of the equipment status parameter at the moment, For the The prediction weight coefficient corresponding to the correlation degree is For the The correlation between dynamic environment parameters and equipment status parameters, for Dynamic environmental parameter values ​​at the moment, is the number of historical correlations considered, is the prediction error term.

4. A high-voltage distribution station dynamic environment and equipment status monitoring system according to claim 1, characterized in that: The multi-dimensional data perception acquisition integration module collects temperature and humidity parameters for analyzing the impact of the distribution station environment on equipment operation. The acquisition frequency is dynamically adjusted according to the equipment operating status and environmental changes. The collected gas concentration parameters include oxygen, carbon monoxide, and sulfur hexafluoride gas concentrations, which are used to monitor whether there are potential equipment leakage faults. The collected vibration parameters obtain the vibration signal of the equipment during operation through a preset sensor to analyze the mechanical operation status of the equipment. The collected voltage and current parameters are used to evaluate the electrical operation status of the equipment. In the pre-analysis module based on composite weight distribution, a weight adjustment model based on the equipment operation stage is constructed: in, is the adjusted weight vector, For devices in The weight adjustment coefficient during the operation phase is set to different values ​​according to the different stages of equipment startup, normal operation, and shutdown.

5. A high-voltage distribution station dynamic environment and equipment status monitoring system according to claim 1, characterized in that: In the core diagnosis module based on the expert system rule base, a comprehensive equipment health evaluation model is constructed: in, For device health, For the The correction coefficient of the impact of each fault type on the equipment health, For the The probability of occurrence of each fault type, is the total number of fault types; In the multimodal visualization output module, a warning level classification model is constructed: in, For the warning level, is the comprehensive evaluation index value of equipment failure, Thresholds for different warning levels.

6. A high-voltage distribution station dynamic environment and equipment status monitoring system according to claim 1, characterized in that: The pre-analysis module based on composite weight allocation includes a data cleaning unit for removing noise data and abnormal data from the collected data; a weight calculation unit for calculating the weight of the device status parameters according to preset rules and the constructed model; The parameter sorting unit sorts the device status parameters according to the weight; the analysis result temporary storage unit is used to temporarily store the pre-analysis result data for calling different modules.

7. A high-voltage distribution station dynamic environment and equipment status monitoring system according to claim 1, characterized in that: The association mapping module combined with historical data includes a historical data storage unit for storing the dynamic environment and equipment status historical data during the long-term operation of the high-voltage distribution station; a data matching unit for matching the current collected data with the historical data; and an association mining unit for mining the association between the dynamic environment and equipment status data using a preset algorithm. The mapping result generating unit generates the association mapping result between the dynamic environment and the device status according to the mined association relationship.

8. A high-voltage distribution station dynamic environment and equipment status monitoring system according to claim 1, characterized in that: The core diagnosis module based on the expert system rule base includes a rule base management unit for adding, deleting, and modifying the expert system rule base; a rule matching unit for matching the processed data with the rules in the rule base; and a fault diagnosis reasoning unit for performing fault diagnosis reasoning based on the matching results. The diagnostic result output unit outputs the final equipment status evaluation and fault diagnosis results.

9. A method for monitoring the dynamic environment and equipment status of a high-voltage distribution station applied by the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step S1: Using a multi-dimensional data sensing acquisition integration module, different parameters of the dynamic environment and equipment status in the high-voltage distribution station are collected according to the set acquisition strategy and frequency; Step S2: The collected parameters are transmitted to the heterogeneous data feature extraction and fusion transmission module to extract features from the data and perform fusion transmission; Step S3: using a pre-analysis module based on composite weight allocation to perform preliminary analysis and processing on the device status parameters according to the preset weight allocation rules and the constructed model; Step S4: transmitting the pre-analysis results to the association mapping module combined with historical data, performing association mapping with the historical data, and mining potential relationships between the data; Step S5: Input the association mapping result into the core diagnosis module based on the expert system rule base, and use the rule base in the expert system model to evaluate the equipment status and diagnose the fault; Step S6: The diagnosis result is transmitted to the multimodal visualization output module, which is visualized in a multimodal form and outputs warning information.

Images