Factory electrical equipment explosion-proof safety intelligent monitoring system and method based on Internet of Things

Abstract

The invention relates to the technical field of electrical equipment monitoring, and discloses a factory electrical equipment explosion-proof safety intelligent monitoring system and method based on the Internet of Things, and the system comprises a certificate intelligent management module, an environment perception and corrosion analysis module, an equipment loss Internet of Things monitoring module, an edge computing gateway, a cloud platform and an explosion-proof judgment module. Through the implementation of the method, the multi-source data collaborative analysis technology is utilized to integrate multi-level technical data such as equipment qualification, environmental corrosion, operation loss and digital twinborn deduction, data features are converted into a clear safety state result, and the generated evaluation result has a complete technical deduction link; the system provides an accurate technical decision basis for plant explosion-proof safety management, and solves the problems that a current plant electrical equipment explosion-proof safety intelligent monitoring system based on the Internet of Things is difficult to achieve full-process data linkage, dynamic accurate evaluation and multi-system collaborative management and control, and cannot fully meet the high-level safety management and control requirements of dangerous plants.

Description

Technical Field

[0001] This invention relates to the field of electrical equipment monitoring technology, specifically to an intelligent monitoring system and method for explosion-proof safety of electrical equipment in factory areas based on the Internet of Things. Background Technology

[0002] In flammable and explosive hazardous areas such as petrochemical, coal mining, and metallurgical smelting plants, electrical equipment is the core power support for production operations. Its explosion-proof safety status directly affects the plant's production safety, the safety of personnel and property, and the ecological environment. With the deepening of Industry 4.0 and increasingly stringent safety production standards, most large-scale plants have gradually phased out the traditional purely manual inspection mode and introduced technologies such as sensor monitoring and preliminary data networking. The level of explosion-proof monitoring of electrical equipment has been significantly improved. However, facing the trend of larger and more complex equipment and the demand for higher-level safety management, the existing monitoring system still has many shortcomings that need to be improved. The current mainstream explosion-proof monitoring mode for electrical equipment in plants mainly relies on fixed-point sensor networking plus periodic manual verification. Among them, sensor networking can realize the real-time acquisition of core operating parameters such as temperature, current, and insulation resistance, which to some extent makes up for the shortcomings of traditional manual inspections, such as long cycles and high risks in hazardous areas. Manual verification focuses on scenarios that are difficult for sensors to cover, such as the integrity of the equipment's appearance and the status of its explosion-proof sealing, forming a complementary relationship. However, this model still has inherent limitations: on the one hand, sensor monitoring focuses on local parameters of a single device, and data from different monitoring dimensions (such as environmental corrosion, equipment wear and tear, and qualification compliance) are independent of each other and do not form a linkage analysis; on the other hand, manual review is greatly affected by differences in experience, and has limited ability to identify hidden problems such as equipment aging and insulation deterioration, making it difficult to predict potential hazards in advance.

[0003] To further improve monitoring efficiency, many factories have introduced multi-sensor fusion and preliminary edge computing technologies, achieving centralized acquisition of multiple parameters and basic data preprocessing, significantly improving early warning response speed compared to earlier models. However, from the perspective of actual application effects, key technical bottlenecks still exist: First, the integration of multi-source data is insufficient. Existing systems can mostly achieve centralized data storage, but lack in-depth correlation analysis of equipment qualification data, environmental corrosion data, operating parameter data, and historical fault data. It is difficult to extract the core correlation characteristics of equipment explosion-proof safety from massive amounts of data, and the data value is not fully explored. Second, the dynamic adaptation capability is insufficient. Existing early warning thresholds are mostly fixed values ​​set based on equipment manuals. Although some systems can be slightly adjusted, they cannot adaptively adjust according to fluctuations in factory environmental parameters such as salt spray, humidity changes, equipment operating years, and historical fault patterns. This leads to insufficient early warning accuracy under complex operating conditions, making it difficult to accurately support operation and maintenance decisions. In recent years, technologies such as the Internet of Things, digital twins, and artificial intelligence have been gradually piloted in the field of explosion-proof monitoring. Some advanced factories have attempted to build digital twin models of equipment to achieve visualized display of equipment status. However, existing pilot programs often prioritize demonstration over application. For example, existing digital twin models focus on restoring the physical form of equipment. While they can correlate some real-time operational data, they do not deeply integrate key information such as equipment corrosion and aging data and historical failure cases, making it impossible to dynamically extrapolate and predict the explosion-proof safety status of equipment. In the safety assessment phase, although existing programs have attempted to introduce multi-dimensional indicators, the assessment weights are mostly fixed based on experience and are not dynamically optimized in conjunction with historical failure data. Furthermore, they do not fully correlate with fundamental indicators such as equipment explosion-proof qualification compliance, resulting in insufficient scientific rigor and relevance of the assessment results, making it difficult to meet the precise requirements of high-level safety management. Furthermore, existing monitoring systems suffer from common problems of insufficient collaboration: on the one hand, the collaboration between the edge and cloud is not deep enough. Most systems can only upload edge data to cloud storage, failing to fully leverage the real-time computing advantages of the edge, leading to data processing delays under complex operating conditions and making it difficult to meet the real-time requirements for early detection and handling of explosion-proof hazards. On the other hand, there is a disconnect between equipment explosion-proof qualification management and operational monitoring. Existing qualification management systems are mostly independent document management systems, which are not linked to and verified with real-time equipment operating status and corrosion loss data. This makes it difficult to dynamically control equipment explosion-proof compliance from the source, creating hidden dangers for safe operation in the plant area. In summary, although current explosion-proof monitoring technology for electrical equipment in plants has moved beyond the traditional outdated model and achieved basic intelligent upgrades, it still suffers from shortcomings such as insufficient data correlation, insufficient dynamic adaptability, inaccurate assessment systems, and poor collaboration among multiple systems. It is difficult to fully meet the needs of hazardous plants for full-process, high-precision, and forward-looking control of electrical equipment explosion-proof safety.

[0004] In summary, current IoT-based intelligent monitoring systems for explosion-proof safety of electrical equipment in factories have limitations in achieving full-process data linkage, dynamic and accurate assessment, and multi-system collaborative management, thus failing to fully meet the high-level safety management needs of hazardous factory areas. Summary of the Invention

[0005] In view of this, the present invention provides an intelligent monitoring system and method for explosion-proof safety of electrical equipment in factories based on the Internet of Things (IoT), in order to solve the problem that current IoT-based intelligent monitoring systems for explosion-proof safety of electrical equipment in factories are unable to achieve full-process data linkage, dynamic and accurate assessment and multi-system collaborative management, and thus cannot fully meet the high-level safety management requirements of hazardous factories.

[0006] In a first aspect, the present invention provides an intelligent monitoring system for explosion-proof safety of electrical equipment in a factory based on the Internet of Things (IoT). This system includes: a certificate intelligent management module, used to acquire explosion-proof qualification certificate data for electrical equipment, parse the explosion-proof qualification certificate data to obtain certificate information, perform dual-dimensional verification based on the certificate information to obtain equipment explosion-proof qualification status data, and output the certificate information and equipment explosion-proof qualification status data as first-layer monitoring data; and an environmental perception and corrosion analysis module, used to collect a set of environmental parameters using an intrinsically safe distributed sensor network, determine the corrosion degree data and remaining life data of the electrical equipment based on a preset coupled simulation model and the environmental parameter set, and output the corrosion degree data and remaining life data as second-layer monitoring data. Data; Equipment Loss IoT Monitoring Module, used to collect operating parameters of electrical equipment using monitoring terminals, determine the loss characteristics and fault location information of electrical equipment based on the operating parameters, and output the loss characteristics and fault location information as third-layer monitoring data; Edge Computing Gateway, used to determine target monitoring data and graded anomaly warning trigger information based on first-layer, second-layer, and third-layer monitoring data; Cloud Platform, used to build a digital twin model of equipment based on target monitoring data, graded anomaly warning trigger information, and historical monitoring data; Explosion-proof Judgment Module, used to conduct explosion-proof safety status assessment based on target monitoring data, graded anomaly warning trigger information, historical monitoring data, and digital twin model, and obtain assessment results.

[0007] The IoT-based intelligent monitoring system for explosion-proof safety of electrical equipment in factories provided in this embodiment firstly performs structured parsing of explosion-proof qualification certificate data through a certificate intelligent management module. This transforms unstructured certificate text and image data into standardized structured information. Combined with dual-dimensional verification technology, key certificate information is validated to ensure the authenticity and completeness of the qualification data. The generated equipment explosion-proof qualification status data can be directly used as the basic input parameters for subsequent safety assessments, achieving technical compatibility between qualification information and monitoring data and avoiding data integration obstacles caused by inconsistent qualification data formats. Secondly, the environmental perception and corrosion analysis module employs an intrinsically safe distributed sensor network. Distributed node deployment enables comprehensive data collection of the surrounding environment of the equipment. Coupled with a coupled simulation model, multi-dimensional environmental parameters are analyzed, overcoming the challenge that single environmental parameter monitoring cannot reflect complex corrosion mechanisms. This accurately quantifies the correlation between environmental factors and equipment corrosion. The generated corrosion degree data and remaining life data have clear technical traceability, providing precise technical support for equipment corrosion prevention and control. Simultaneously, the intrinsically safe circuit design meets explosion-proof requirements in hazardous environments, ensuring the technical safety of the data acquisition process. Subsequently, the equipment loss IoT monitoring module collects operating parameters using monitoring terminals. Combined with feature extraction algorithms, it extracts equipment loss characteristics from the raw operating data, achieving a quantitative representation of hidden equipment losses. Then, a fault location algorithm collaboratively analyzes electrical discharge parameters and loss characteristics to accurately pinpoint the spatial location of the fault. This overcomes the limitations of traditional operation monitoring, which can only detect anomalies but not locate the root cause. The generated loss characteristic data and fault location information possess high-dimensional technical correlation. Next, an edge computing gateway performs centralized preprocessing and feature filtering on the three layers of monitoring data. Edge-side real-time computing technology reduces the data transmission and processing pressure on the cloud, improving the real-time performance of data processing. Combined with a tiered early warning algorithm, the filtered target monitoring data is used to determine anomalies, achieving accurate differentiation between different levels of anomalies. This overcomes the technical shortcomings of traditional single-threshold early warning systems. The generated tiered anomaly early warning trigger information has clear technical judgment criteria, providing accurate anomaly data input for subsequent evaluation modules and ensuring the technical reliability of the entire system's early warning response. Furthermore, by integrating target monitoring data, hierarchical anomaly warning trigger information, and historical monitoring data through a cloud platform, a digital twin model of the equipment is constructed. Data fusion technology is used to achieve deep correlation between real-time data and historical data, restoring the full life cycle status of the equipment. Combined with digital mapping technology, the physical state of the equipment is transformed into a quantifiable and predictable digital model, breaking through the technical limitations of traditional monitoring that can only reflect the real-time status and cannot predict trends. This provides a technical carrier for the dynamic extrapolation and trend prediction of equipment status, and enhances the technical foresight of safety management.Finally, an explosion-proof safety status assessment is conducted by integrating multi-dimensional data through an explosion-proof judgment module. Multi-source data collaborative analysis technology integrates technical data from multiple levels, including equipment qualifications, environmental corrosion, operational losses, and digital twin simulations, improving the comprehensiveness and scientific rigor of the assessment results. Combined with assessment algorithms, data characteristics are transformed into clear safety status results, overcoming the limitations of traditional single-dimensional assessments. The generated assessment results possess a complete technical derivation chain, providing precise technical decision-making basis for explosion-proof safety management in the plant area, and improving the overall precision of explosion-proof safety control from a technical perspective. This addresses the problem that current IoT-based intelligent monitoring systems for explosion-proof safety of electrical equipment in plants struggle to achieve full-process data linkage, dynamic and accurate assessment, and multi-system collaborative control, thus failing to fully meet the high-level safety control requirements of hazardous plant areas.

[0008] In one optional implementation, the certificate intelligent management module, the environmental perception and corrosion analysis module, and the equipment loss IoT monitoring module are each communicatively connected to an edge computing gateway. The edge computing gateway is communicatively connected to a cloud platform, and the cloud platform is communicatively connected to an explosion-proof determination module. The certificate intelligent management module includes: a certificate input unit, used to receive paper scan data, electronic file data, and manually entered data of explosion-proof qualification certificates to obtain explosion-proof qualification certificate data; a parsing unit, used to perform structured parsing of the explosion-proof qualification certificate data and extract certificate information, including certificate number information, equipment model information, equipment 3D model information, issuing authority information, validity period information, explosion-proof level information, material characteristic information, equipment explosion-proof structure information, and applicable environmental parameter information; a first verification unit, used to generate an expiration warning signal based on system time and validity period information; a second verification unit, used to verify the certificate number information, equipment model information, and issuing authority information based on the official database and the issuing authority traceability platform to determine the qualification determination result; and a qualification determination unit, used to generate equipment explosion-proof qualification status data based on the expiration warning signal and the qualification determination result. The certificate association unit is used to establish a unique association between explosion-proof qualification certificates and electrical equipment, obtain association mapping data, and integrate the association mapping data into the certificate information; the first-level data output unit is used to integrate the certificate information and equipment explosion-proof qualification status data into first-level monitoring data and output it.

[0009] In one optional implementation, the environmental perception and corrosion analysis module includes: an intrinsically safe sensing unit, used to collect environmental parameters of the corresponding factory area of ​​the electrical equipment using an intrinsically safe distributed sensing network, to obtain an environmental parameter set, which includes salt spray parameters, humidity parameters, temperature parameters, harmful gas concentration parameters, dust concentration parameters, and ultraviolet intensity parameters; an environmental data preprocessing unit, used to perform wavelet transform denoising, standardization, and time series alignment processing on the environmental parameter set to obtain an environmental parameter time series dataset; a coupled simulation unit, used to calculate the corrosion degree data and remaining corrosion life data of various parts of the electrical equipment using a preset coupled simulation model, based on the environmental parameter time series dataset and material characteristic information, whereby the material characteristic information includes material thermal conductivity, electrochemical corrosion constant, and insulation medium parameters; an environmental risk level classification unit, used to determine the aging corrosion risk level of the equipment based on preset risk judgment rules, corrosion degree data, and remaining corrosion life data, generate risk level identification data, and integrate the risk level identification data and environmental parameter time series dataset into the corrosion degree data; and a second-layer data output unit, used to integrate the corrosion degree data and remaining life data into second-layer monitoring data and output it.

[0010] In one optional implementation, the coupled simulation unit includes: a parameter coordination subunit, used to decompose the environmental parameter time-series dataset into several time-series segments according to the time dimension, and perform parameter normalization and coordination mapping processing on each time-series segment based on the material thermal conductivity, electrochemical corrosion constant, and insulating medium parameters to generate a multi-dimensional time-series parameter set; and a coordination calculation subunit, used to construct multi-field coupled control equations based on the multi-dimensional time-series parameter set using a preset coupled simulation model, and calculate the dynamic corrosion rate, corrosion depth distribution data, and insulation performance degradation curve of each part of the electrical equipment according to the multi-field coupled control equations. The preset coupled simulation model is used to simulate the salt spray humidity and temperature coordinated corrosion field and the equipment operating electric field temperature. The system employs a field coupling mechanism; a dynamic prediction subunit for remaining life, which extracts dynamic corrosion rate, corrosion depth distribution data, and characteristic inflection points of insulation performance degradation curves. Based on these inflection points and equipment explosion-proof structure information, it constructs a corrosion life mapping model and obtains prediction results based on the corrosion life mapping model and environmental parameter time series datasets. A result correction subunit is used to correct the prediction results using preset environmental parameter fluctuation coefficients, obtaining remaining corrosion life data and life decay trend prediction curves for various parts of the electrical equipment. A data determination subunit is used to determine corrosion degree data based on corrosion depth distribution data, integrate the life decay trend prediction curves into the remaining corrosion life data, and output corrosion degree data and remaining corrosion life data.

[0011] In one optional implementation, the equipment loss IoT monitoring module includes: an intrinsically safe monitoring terminal unit, employing a three-level energy isolation intrinsically safe circuit architecture, which includes a power isolation subunit, a signal isolation subunit, and an energy storage isolation subunit; an operating parameter acquisition unit, used to acquire operating parameters of the electrical equipment, including mechanical vibration parameters, electrical discharge parameters, temperature parameters, current parameters, insulation resistance parameters, and rotational speed parameters; an operating parameter preprocessing unit, used to perform noise reduction, format standardization, and outlier screening on the operating parameters, outputting a standardized operating parameter dataset; a loss feature extraction unit, used to extract equipment loss features based on the standardized operating parameter dataset; a fault location unit, used to determine the three-dimensional location information of the fault point and generate fault location information based on electrical discharge parameters and electrical loss features using a partial discharge acoustic signature localization algorithm and an ultrasonic collaborative localization algorithm; a location transmission unit, used to transmit the equipment loss features and fault location information to an edge computing gateway, the location transmission unit being configured with the national cryptographic SM4 encryption protocol; and a third-layer data output unit, used to integrate the equipment loss features and fault location information into third-layer monitoring data and output it.

[0012] In one optional implementation, the standardized operating parameter dataset includes standardized mechanical vibration parameters, standardized electrical discharge parameters, standardized temperature parameters, standardized current parameters, standardized insulation resistance parameters, and standardized rotational speed parameters. The loss feature extraction unit includes: a mechanical loss feature extraction subunit, used to process the standardized mechanical vibration parameters using a ratio analysis algorithm to obtain mechanical vibration features, and to synchronously calibrate the mechanical vibration features based on the standardized rotational speed parameters to obtain mechanical loss features; an electrical loss feature extraction subunit, used to process the standardized electrical discharge parameters using a spectrum analysis algorithm to obtain discharge parameter features, and to verify the insulation degradation degree of the discharge parameter features based on the standardized insulation resistance parameters to obtain electrical loss features; and an operating loss feature extraction subunit, used to calculate the temperature-current correlation coefficient based on the standardized temperature and standardized current parameters, determine the load heating matching relationship data based on the temperature-current correlation coefficient, and obtain the operating loss features based on the load heating matching relationship data.

[0013] In one optional implementation, the edge computing gateway includes: a hierarchical preprocessing unit, used to extract certificate-type text data, environmental time-series data, and device operating parameter data from the first-layer monitoring data, the second-layer monitoring data, and the third-layer monitoring data; perform semantic alignment and format standardization processing on the certificate-type text data; perform wavelet threshold denoising and timestamp calibration processing on the environmental time-series data; perform Kalman filtering denoising and outlier screening processing on the device operating parameters; and output a standardized multi-source dataset; an edge-side real-time computing unit, used to perform cross-dimensional data association matching on the standardized multi-source dataset based on a preset feature weight matrix and weighted Euclidean distance; filter to obtain target monitoring data; and use a sliding window mechanism to perform trend fitting on the target monitoring data to generate data change slope and fluctuation coefficient; and an early warning triggering unit, used to generate hierarchical anomaly early warning triggering information based on a preset dynamic threshold model, target monitoring data, data change slope, and fluctuation coefficient using a dual judgment mechanism of absolute threshold and trend threshold.

[0014] In one optional implementation, the historical monitoring data includes historical first-layer monitoring data, historical second-layer monitoring data, historical third-layer monitoring data, historical target monitoring data, and historical graded anomaly warning trigger information. The cloud platform includes: a multi-source data storage unit, used to classify and store target monitoring data, graded anomaly warning trigger information, and historical monitoring data using a distributed time-series database, and constructing a data index based on the device's unique identifier; a digital twin model construction unit, used to construct a digital twin model based on the device's 3D model information, target monitoring data, historical monitoring data, corrosion degree data, remaining lifespan data, and device wear characteristics, and output digital twin mapping data, which includes the device's full state correlation and device state inference results; and a dynamic threshold training unit, used to train and update a preset dynamic threshold model based on historical monitoring data, and distribute the updated preset dynamic threshold model to the edge computing gateway.

[0015] In one optional implementation, the explosion-proof determination module includes: an indicator system matching unit, used to match the corresponding explosion-proof safety assessment indicator system based on equipment model information. The assessment indicator system is constructed based on the analytic hierarchy process (AHP) and includes certificate compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators. These indicators are associated with corresponding data items in the target monitoring data. A dynamic weight configuration unit, used to adjust the weight coefficients of the certificate compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators based on fault case data in historical monitoring data, to obtain updated weight coefficients. An evidence fusion unit, used to fuse the target monitoring data and graded abnormality early warning trigger information using evidence theory, based on the updated weight coefficients and the full-state correlation of the equipment, to generate a credibility distribution result. A safety level determination unit, used to map the credibility distribution result to the corresponding explosion-proof safety level based on preset level determination rules and state deduction results, and output the safety level determination result.

[0016] Secondly, this invention provides an intelligent monitoring method for explosion-proof safety of electrical equipment in a factory based on the Internet of Things (IoT). The method includes: acquiring explosion-proof qualification certificate data for electrical equipment; parsing the explosion-proof qualification certificate data to obtain certificate information; performing dual-dimensional verification based on the certificate information to obtain equipment explosion-proof qualification status data; and outputting the certificate information and equipment explosion-proof qualification status data as first-layer monitoring data; collecting a set of environmental parameters using an intrinsically safe distributed sensor network; determining the corrosion degree data and remaining lifespan data of the electrical equipment based on a preset coupled simulation model and the set of environmental parameters; and outputting the corrosion degree data and remaining lifespan data as second-layer monitoring data; collecting operating parameters of the electrical equipment using a monitoring terminal; determining the wear characteristics and fault location information of the electrical equipment based on the operating parameters; and outputting the wear characteristics and fault location information as third-layer monitoring data; determining target monitoring data and graded anomaly warning trigger information based on the first-layer, second-layer, and third-layer monitoring data; constructing a digital twin model of the equipment based on the target monitoring data, graded anomaly warning trigger information, and historical monitoring data; and conducting an explosion-proof safety status assessment based on the target monitoring data, graded anomaly warning trigger information, historical monitoring data, and the digital twin model to obtain the assessment result. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a specific example of an IoT-based intelligent monitoring system for explosion-proof safety of electrical equipment in a factory, according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating a specific example of an IoT-based intelligent monitoring method for explosion-proof safety of electrical equipment in a factory, according to an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] In flammable and explosive hazardous areas such as petrochemical, coal mining, and metallurgical smelting plants, electrical equipment is the core power support for production operations. Its explosion-proof safety status directly affects the plant's production safety, the safety of personnel and property, and the ecological environment. With the deepening of Industry 4.0 and increasingly stringent safety production standards, most large-scale plants have gradually phased out the traditional purely manual inspection mode and introduced technologies such as sensor monitoring and preliminary data networking. The level of explosion-proof monitoring of electrical equipment has been significantly improved. However, facing the trend of larger and more complex equipment and the demand for higher-level safety management, the existing monitoring system still has many shortcomings that need to be improved. The current mainstream explosion-proof monitoring mode for electrical equipment in plants mainly relies on fixed-point sensor networking plus periodic manual verification. Among them, sensor networking can realize the real-time acquisition of core operating parameters such as temperature, current, and insulation resistance, which to some extent makes up for the shortcomings of traditional manual inspections, such as long cycles and high risks in hazardous areas. Manual verification focuses on scenarios that are difficult for sensors to cover, such as the integrity of the equipment's appearance and the status of its explosion-proof sealing, forming a complementary relationship. However, this model still has inherent limitations: on the one hand, sensor monitoring focuses on local parameters of a single device, and data from different monitoring dimensions (such as environmental corrosion, equipment wear and tear, and qualification compliance) are independent of each other and do not form a linkage analysis; on the other hand, manual review is greatly affected by differences in experience, and has limited ability to identify hidden problems such as equipment aging and insulation deterioration, making it difficult to predict potential hazards in advance.

[0021] To further improve monitoring efficiency, many factories have introduced multi-sensor fusion and preliminary edge computing technologies, achieving centralized acquisition of multiple parameters and basic data preprocessing, significantly improving early warning response speed compared to earlier models. However, from the perspective of actual application effects, key technical bottlenecks still exist: First, the integration of multi-source data is insufficient. Existing systems can mostly achieve centralized data storage, but lack in-depth correlation analysis of equipment qualification data, environmental corrosion data, operating parameter data, and historical fault data. It is difficult to extract the core correlation characteristics of equipment explosion-proof safety from massive amounts of data, and the data value is not fully explored. Second, the dynamic adaptation capability is insufficient. Existing early warning thresholds are mostly fixed values ​​set based on equipment manuals. Although some systems can be slightly adjusted, they cannot adaptively adjust according to fluctuations in factory environmental parameters such as salt spray, humidity changes, equipment operating years, and historical fault patterns. This leads to insufficient early warning accuracy under complex operating conditions, making it difficult to accurately support operation and maintenance decisions. In recent years, technologies such as the Internet of Things, digital twins, and artificial intelligence have been gradually piloted in the field of explosion-proof monitoring. Some advanced factories have attempted to build digital twin models of equipment to achieve visualized display of equipment status. However, existing pilot programs often prioritize demonstration over application. For example, existing digital twin models focus on restoring the physical form of equipment. While they can correlate some real-time operational data, they do not deeply integrate key information such as equipment corrosion and aging data and historical failure cases, making it impossible to dynamically extrapolate and predict the explosion-proof safety status of equipment. In the safety assessment phase, although existing programs have attempted to introduce multi-dimensional indicators, the assessment weights are mostly fixed based on experience and are not dynamically optimized in conjunction with historical failure data. Furthermore, they do not fully correlate with fundamental indicators such as equipment explosion-proof qualification compliance, resulting in insufficient scientific rigor and relevance of the assessment results, making it difficult to meet the precise requirements of high-level safety management. Furthermore, existing monitoring systems suffer from common problems of insufficient collaboration: on the one hand, the collaboration between the edge and cloud is not deep enough. Most systems can only upload edge data to cloud storage, failing to fully leverage the real-time computing advantages of the edge, leading to data processing delays under complex operating conditions and making it difficult to meet the real-time requirements for early detection and handling of explosion-proof hazards. On the other hand, there is a disconnect between equipment explosion-proof qualification management and operational monitoring. Existing qualification management systems are mostly independent document management systems, which are not linked to and verified with real-time equipment operating status and corrosion loss data. This makes it difficult to dynamically control equipment explosion-proof compliance from the source, creating hidden dangers for safe operation in the plant area. In summary, although current explosion-proof monitoring technology for electrical equipment in plants has moved beyond the traditional outdated model and achieved basic intelligent upgrades, it still suffers from shortcomings such as insufficient data correlation, insufficient dynamic adaptability, inaccurate assessment systems, and poor collaboration among multiple systems. It is difficult to fully meet the needs of hazardous plants for full-process, high-precision, and forward-looking control of electrical equipment explosion-proof safety.

[0022] In summary, current IoT-based intelligent monitoring systems for explosion-proof safety of electrical equipment in factories have limitations in achieving full-process data linkage, dynamic and accurate assessment, and multi-system collaborative management, thus failing to fully meet the high-level safety management needs of hazardous factory areas.

[0023] To address the technical problems mentioned in the background section, this application provides an IoT-based intelligent monitoring system for explosion-proof safety of electrical equipment in factories. See details below. Figure 1 As shown, the system includes: a certificate intelligent management module, an environmental perception and corrosion analysis module, an equipment loss IoT monitoring module, an edge computing gateway, a cloud platform, and an explosion-proof determination module. The functions of each of the above components are described below.

[0024] The intelligent certificate management module is used to acquire explosion-proof qualification certificate data of electrical equipment, parse the explosion-proof qualification certificate data to obtain certificate information, perform dual-dimensional verification based on the certificate information to obtain equipment explosion-proof qualification status data, and output the certificate information and equipment explosion-proof qualification status data as the first-level monitoring data.

[0025] Specifically, the certificate intelligent management module, environmental perception and corrosion analysis module, and equipment loss IoT monitoring module are each connected to the edge computing gateway. The edge computing gateway is connected to the cloud platform, and the cloud platform is connected to the explosion-proof determination module. The certificate intelligent management module includes: The certificate entry unit is used to receive paper scan data, electronic file data, and manually entered data of explosion-proof qualification certificates to obtain explosion-proof qualification certificate data.

[0026] Furthermore, paper scan data is obtained by scanning paper-based explosion-proof certificates using a scanner; electronic document data is obtained by receiving electronic certificate files in formats such as PDF and images; and manually entered data is obtained by manually inputting certificate-related information through an interactive interface. The certificate entry unit recognizes the format of the different types of data received, converts paper scan data and electronic document data into editable text data, and then performs preliminary integration of various data types, removing duplicate or invalid information, ultimately forming standardized explosion-proof certificate data.

[0027] The parsing unit is used to perform structured parsing of explosion-proof certificate data and extract certificate information, including certificate number information, equipment model information, equipment 3D model information, issuing authority information, validity period information, explosion-proof level information, material characteristic information, equipment explosion-proof structure information, and applicable environmental parameter information.

[0028] Furthermore, the parsing unit invokes a pre-defined parsing rule base, which contains feature identifiers and extraction logic for common fields in explosion-proof qualification certificates. By matching rules, key fields in the certificate data are located, including certificate number information, equipment model information, equipment 3D model information, issuing authority information, validity period information, explosion-proof rating information, material property information, equipment explosion-proof structure information, and applicable environmental parameter information. The content corresponding to each field is extracted and verified to ensure the completeness of the extracted information, ultimately forming structured certificate information.

[0029] The first verification unit is used to generate an expiration warning signal based on system time and validity period information.

[0030] Furthermore, the system time is the standard time synchronized in real time by the monitoring system. The first verification unit periodically obtains the system time and compares it with the validity period information in the certificate information. A preset warning threshold is set. When the system time is less than the expiration date, a level one expiration warning signal is generated; when the system time exceeds the expiration date, a level two expiration warning signal is generated.

[0031] The second verification unit is used to verify the certificate number information, equipment model information, and issuing agency information based on the official database and the issuing agency traceability platform to determine the qualification assessment result.

[0032] Furthermore, the second verification unit is used to verify the certificate number, equipment model, and issuing authority information based on the official database and the issuing authority traceability platform to determine the qualification assessment result. By accessing the official database and the issuing authority traceability platform via a network interface, the information to be verified is uploaded to the platform for comparison. If the uploaded information completely matches the platform's registration information, the qualification assessment result is compliant; if there is an information mismatch or the registration information cannot be found, the qualification assessment result is non-compliant.

[0033] The qualification assessment unit is used to generate equipment explosion-proof qualification status data based on the expiration warning signal and the qualification assessment result.

[0034] Furthermore, the expiration warning signal generated by the first verification unit and the qualification judgment result determined by the second verification unit are comprehensively encoded. Different combinations of warning signals and qualification judgment results correspond to different qualification status codes. For example, compliance with no expiration warning corresponds to code 01, compliance with a level 1 expiration warning corresponds to code 02, and non-compliance corresponds to code 03. The encoding results are integrated with the corresponding status descriptions to form equipment explosion-proof qualification status data.

[0035] The certificate association unit is used to establish a unique association between explosion-proof qualification certificates and electrical equipment, obtain association mapping data, and integrate the association mapping data into the certificate information.

[0036] Furthermore, the certificate association unit is used to establish a unique association between explosion-proof certificates and electrical equipment, obtaining association mapping data. By extracting the unique identification information of the electrical equipment and the certificate number information of the explosion-proof certificate, a one-to-one correspondence between the two is established, generating association mapping data containing the equipment's unique identifier and the certificate number. This association mapping data is then added to the extended fields of the certificate information, completing the integration of certificate information.

[0037] The first-level data output unit is used to integrate certificate information and equipment explosion-proof qualification status data into first-level monitoring data and output them.

[0038] Furthermore, following the preset data format specifications, the integrated certificate information and equipment explosion-proof qualification status data are packaged together, and data collection timestamps and unique equipment identifiers are added to form standardized first-layer monitoring data. This first-layer monitoring data is then transmitted to the edge computing gateway via a preset communication interface, ensuring the integrity and timeliness of data transmission.

[0039] The environmental perception and corrosion analysis module is used to collect a set of environmental parameters using an intrinsically safe distributed sensor network. Based on a preset coupled simulation model and the set of environmental parameters, it determines the corrosion degree data and remaining life data of electrical equipment, and outputs the corrosion degree data and remaining life data as the second-level monitoring data.

[0040] Specifically, the environmental sensing and corrosion analysis module includes: The intrinsically safe sensing unit is used to collect environmental parameters of the corresponding factory area of ​​the electrical equipment using an intrinsically safe distributed sensing network, and obtain a set of environmental parameters, including salt spray parameters, humidity parameters, temperature parameters, harmful gas concentration parameters, dust concentration parameters, and ultraviolet intensity parameters.

[0041] Furthermore, an intrinsically safe distributed sensor network refers to a distributed data acquisition network composed of multiple sensor nodes with intrinsically safe performance. Intrinsic safety performance refers to the ability of the equipment, under both normal and fault conditions, to prevent the electrical sparks and thermal effects generated from igniting explosive mixtures. Sensor nodes are distributed in the plant area corresponding to the electrical equipment according to a pre-defined deployment strategy, achieving comprehensive coverage of the surrounding environment. The environmental parameter set refers to a collection of various parameters that reflect the corrosive effects of the plant environment on electrical equipment, specifically including salt spray parameters, humidity parameters, temperature parameters, harmful gas concentration parameters, dust concentration parameters, and ultraviolet radiation intensity parameters. Each sensor node collects one or more environmental parameters. During the collection process, the physical signals are converted into electrical signals by the sensor's built-in signal conversion module. The signals collected by each node are then transmitted to the gateway node through the internal communication link of the sensor network. The gateway node aggregates and integrates all signals to ultimately form the environmental parameter set.

[0042] The environmental data preprocessing unit is used to perform wavelet transform denoising, standardization, and time series alignment on the environmental parameter set to obtain a time series dataset of environmental parameters.

[0043] Furthermore, wavelet transform denoising refers to using wavelet analysis to filter noise signals in environmental parameters. By selecting appropriate wavelet basis functions, the environmental parameter signals are decomposed at multiple scales to separate the effective and noise components. The effective components are then reconstructed to achieve denoising. Standardization refers to transforming environmental parameters of different dimensions and orders of magnitude into data under a unified standard, eliminating the impact of dimensional differences on subsequent analysis. Time series alignment refers to calibrating the acquisition time of each environmental parameter using a unified time reference to ensure that all environmental parameters remain synchronized in the time dimension. The environmental parameter time series dataset refers to the set of environmental parameters arranged in chronological order after the above processing. The specific processing flow is as follows: first, wavelet transform denoising is performed on each parameter in the environmental parameter set to remove noise interference; then, the denoised parameters are standardized; finally, the standardized parameters are aligned in time series using system time as a unified time reference to ultimately form the environmental parameter time series dataset.

[0044] The coupled simulation unit is used to calculate the corrosion degree data and remaining corrosion life data of various parts of electrical equipment based on the environmental parameter time series dataset and material property information using a preset coupled simulation model. The material property information includes the material thermal conductivity, electrochemical corrosion constant and insulation medium parameters.

[0045] Furthermore, the coupled simulation unit includes: The parameter coordination subunit is used to decompose the environmental parameter time series dataset into several time series segments according to the time dimension. Based on the material thermal conductivity, electrochemical corrosion constant and insulating medium parameters, parameter normalization and coordination mapping are performed on each time series segment to generate a multi-dimensional time series parameter set.

[0046] Furthermore, the environmental parameter time-series dataset refers to a preprocessed set of environmental parameters arranged in chronological order, while a time-series segment refers to segmented data formed by splitting the continuous environmental parameter time-series dataset into time intervals. The thermal conductivity of the material characterizes the material's ability to transfer heat, the electrochemical corrosion constant characterizes the material's characteristic parameters during the electrochemical corrosion process, and the insulating medium parameters characterize the relevant parameters of the material's insulation performance. The environmental parameter time-series dataset is decomposed into several groups of continuous time-series segments at preset time intervals, ensuring that each time-series segment reflects the environmental parameter variation characteristics within a certain time period. A parameter co-mapping matrix is ​​then constructed based on the material thermal conductivity, electrochemical corrosion constant, and insulating medium parameters. This matrix is ​​used to establish the correlation between environmental parameters and material characteristic parameters. The environmental parameters in each time-series segment are normalized using this matrix, transforming environmental parameters with different dimensions into dimensionless data under a unified standard, eliminating the influence of dimensional differences on subsequent calculations. Subsequently, co-mapping processing is performed to correlate and fuse the normalized environmental parameters with the corresponding material characteristic parameters, ultimately generating a multi-dimensional time-series parameter set containing the correlation information between environmental parameters and material characteristic parameters.

[0047] The collaborative computing subunit is used to construct multi-field coupled control equations based on a multi-dimensional time-series parameter set using a preset coupled simulation model. It calculates the dynamic corrosion rate, corrosion depth distribution data and insulation performance degradation curves of various parts of the electrical equipment by calculating the multi-field coupled control equations. The preset coupled simulation model is used to simulate the coupling effect of the salt spray humidity and temperature synergistic corrosion field and the equipment operation electric field and temperature field.

[0048] Furthermore, the pre-built coupled simulation model refers to a mathematical model that is pre-constructed and stored to simulate the corrosion process of equipment under multi-field coupling. The multi-field coupling includes the interaction between the salt spray, humidity, and temperature synergistic corrosion field and the equipment operating electric and temperature fields. This model pre-stores the corresponding field coupling calculation logic. The multi-dimensional time-series parameter set refers to a segmented time-series data set containing information on the correlation between environmental parameters and material property parameters. The multi-field coupling control equation refers to the differential equation used to describe the equipment corrosion process under multi-field coupling. During implementation, the multi-dimensional time-series parameter set is input into the pre-built coupled simulation model. Based on the input parameters, the model automatically identifies the type and range of each field parameter and constructs the multi-field coupling control equation according to the pre-built field coupling calculation logic. The equation includes key variables such as salt spray, humidity, and temperature synergistic corrosion field parameters, equipment operating electric and temperature field parameters, and material property parameters. Numerical solution methods are used to solve the constructed multi-field coupling control equation. During the solution process, the three-dimensional structural information of the electrical equipment is combined to divide the equipment into several calculation units, and the corrosion-related data corresponding to each unit in different time segments are calculated separately. Ultimately, we obtain the dynamic corrosion rate of various parts of the electrical equipment over time, the distribution of corrosion depth in different parts (i.e., corrosion depth distribution data), and the curve of the equipment's insulation performance over time (i.e., insulation performance degradation curve).

[0049] The remaining life dynamic prediction subunit is used to extract the dynamic corrosion rate, corrosion depth distribution data and characteristic inflection points of the insulation performance degradation curve. Based on the characteristic inflection points and the equipment explosion-proof structure information, a corrosion life mapping model is constructed, and the prediction results are obtained based on the corrosion life mapping model and the time series dataset of environmental parameters.

[0050] Furthermore, dynamic corrosion rate refers to the change in equipment corrosion rate over time; corrosion depth distribution data refers to the distribution of corrosion depth in different parts of the equipment; insulation performance degradation curve refers to the curve showing the change in equipment insulation performance over time; characteristic inflection points refer to key points that reflect significant changes in the data trend; and equipment explosion-proof structure information refers to relevant information describing the explosion-proof design and construction of the equipment, including explosion-proof sealing structure and shell strength structure. A data inflection point identification algorithm is used to analyze the dynamic corrosion rate curve, corrosion depth distribution data curve, and insulation performance degradation curve, extracting characteristic inflection points in each curve where corrosion rate, corrosion depth, and insulation performance change significantly, and recording the time information and parameter values ​​corresponding to each characteristic inflection point. Combined with the equipment explosion-proof structure information, the critical corrosion depth and critical insulation performance threshold for equipment explosion-proof performance failure are determined. The critical corrosion depth refers to the critical value at which the equipment's corrosion depth reaches the point where explosion-proof performance fails, and the critical insulation performance threshold refers to the critical value at which the equipment's insulation performance decreases to the point where explosion-proof performance fails. A corrosion life mapping model is constructed based on extracted feature inflection points, critical corrosion depth, critical insulation performance threshold, and equipment explosion-proof structure information. This model is a mathematical model used to correlate changes in environmental parameters, corrosion state, and the remaining lifespan of the equipment. By inputting a time-series dataset of environmental parameters into this model, the model predicts the time required for the equipment to reach the explosion-proof performance failure threshold from its current state, based on historical trends in environmental parameter changes and corrosion state variations, thus obtaining preliminary prediction results.

[0051] The result correction subunit is used to correct the prediction results using a preset environmental parameter fluctuation coefficient, so as to obtain the remaining corrosion life data and life decay trend prediction curves of various parts of the electrical equipment.

[0052] Furthermore, the prediction result refers to the preliminary remaining lifespan data of the equipment output by the corrosion life mapping model. The environmental parameter fluctuation coefficient refers to the coefficient used to characterize the degree of influence of environmental parameter fluctuations on corrosion lifespan. This coefficient is pre-calibrated based on historical environmental parameter fluctuation data and corresponding lifespan impact data. Different types of environmental parameters correspond to different fluctuation coefficients. Pre-set environmental parameter fluctuation coefficients are obtained, including fluctuation coefficients corresponding to various environmental parameters such as salt spray parameter fluctuation coefficient, humidity parameter fluctuation coefficient, and temperature parameter fluctuation coefficient. The fluctuation of each environmental parameter in the current environmental parameter time series dataset is analyzed to determine the combination of fluctuation coefficients corresponding to the current environment. The preliminary prediction result is weighted and calculated with the corresponding fluctuation coefficient combination. During the weighting calculation, the coefficient proportion is allocated according to the weight of the influence of different environmental parameters on corrosion lifespan, correcting the preliminary prediction result and eliminating the influence of environmental parameter fluctuations on the lifespan prediction result. After correction, accurate remaining corrosion lifespan data for each part of the electrical equipment is obtained. The remaining corrosion lifespan data refers to the remaining time that the equipment can maintain normal explosion-proof performance under the current environmental conditions. Meanwhile, based on the corrected remaining corrosion life data and the changing trends of historical corrosion data, a curve of the equipment's remaining life decaying over time is generated, namely the life decay trend prediction curve, which intuitively reflects the changing pattern of the equipment's remaining life.

[0053] The data determination subunit is used to determine the corrosion degree data based on the corrosion depth distribution data, integrate the lifetime decay trend prediction curve into the remaining corrosion lifetime data, and output the corrosion degree data and the remaining corrosion lifetime data.

[0054] Furthermore, corrosion depth distribution data refers to the distribution of corrosion depth in different parts of the equipment; corrosion severity data refers to data characterizing the current severity of corrosion on the equipment; the lifespan decay trend prediction curve refers to a curve reflecting the decay law of the equipment's remaining lifespan over time; and remaining corrosion lifespan data refers to the remaining time the equipment can maintain normal explosion-proof performance under current environmental conditions. A preset corrosion severity grading standard is established, based on the correlation between corrosion depth and the equipment's explosion-proof performance, including corrosion severity levels corresponding to different corrosion depth ranges. The corrosion depth of each part in the corrosion depth distribution data is compared with the preset corrosion severity grading standard. Based on the comparison results, the corrosion severity level corresponding to each part is determined. Then, the corrosion severity levels of each part are summarized and integrated to generate corrosion severity data that includes the overall equipment and the corrosion severity of each part. Subsequently, the lifespan decay trend prediction curve is added as extended data to the remaining corrosion lifespan data, establishing a correlation between the remaining corrosion lifespan data and the lifespan decay trend prediction curve, so that the remaining corrosion lifespan data not only includes specific lifespan time but also reflects the lifespan change trend. Finally, the integrated corrosion degree data and remaining corrosion life data are standardized to ensure that the data format meets the requirements of subsequent data processing steps. After processing, the corrosion degree data and remaining corrosion life data are output.

[0055] The environmental risk level classification unit is used to determine the aging corrosion risk level of equipment based on preset risk judgment rules, corrosion degree data and remaining corrosion life data, generate risk level identification data, and integrate the risk level identification data and environmental parameter time series dataset into the corrosion degree data.

[0056] Furthermore, the preset risk assessment rule refers to the pre-defined criteria used to classify the risk levels of equipment aging and corrosion. This rule is based on the correlation between the degree of corrosion and the remaining corrosion life. The equipment aging and corrosion risk level refers to the level of risk of equipment aging and damage due to environmental corrosion. Risk level identification data refers to the coded or symbolic data used to identify different risk levels. In specific implementation, the calculated corrosion degree data and remaining corrosion life data are first compared with the thresholds in the preset risk assessment rule. Based on the comparison results, the corresponding aging and corrosion risk level of the equipment is determined. Then, corresponding identification data is assigned to different risk levels to generate risk level identification data. Subsequently, the risk level identification data and the time-series dataset of environmental parameters are added to the extended fields of the corrosion degree data to achieve data integration.

[0057] The second-layer data output unit is used to integrate corrosion level data and remaining life data into second-layer monitoring data and output them.

[0058] Furthermore, the second-layer monitoring data refers to a standardized set of monitoring data used to reflect the impact of environmental corrosion on electrical equipment. During the integration process, the integrated corrosion level data and remaining lifespan data are structured and packaged according to preset data format specifications, and data acquisition timestamps and unique equipment identifiers are added. The unique equipment identifier is a unique identifier used to distinguish different electrical devices. After packaging, the second-layer monitoring data is transmitted to the edge computing gateway through a preset communication interface to ensure the integrity and timeliness of data transmission.

[0059] The IoT monitoring module for equipment loss is used to collect operating parameters of electrical equipment using monitoring terminals, determine the loss characteristics and fault location information of electrical equipment based on the operating parameters, and output the loss characteristics and fault location information as third-layer monitoring data.

[0060] Specifically, the equipment loss IoT monitoring module includes: The intrinsically safe monitoring terminal unit adopts a three-level energy isolation intrinsically safe circuit architecture, which includes a power isolation subunit, a signal isolation subunit, and an energy storage isolation subunit.

[0061] Furthermore, the three-level energy isolation intrinsically safe circuit architecture refers to a circuit structure that achieves intrinsic safety performance through a three-level isolation design. Intrinsic safety performance refers to the ability of the equipment, under normal operating and fault conditions, to prevent the electrical sparks and thermal effects generated from igniting explosive mixtures. This architecture includes a power isolation subunit, a signal isolation subunit, and an energy storage isolation subunit. The power isolation subunit uses an isolation transformer to achieve electrical isolation between the input power supply and the internal circuitry of the terminal. An isolation transformer is a transformer capable of achieving electrical isolation between the primary and secondary circuits. It transfers energy through electromagnetic induction, blocking the transmission of interference signals from the power supply side to the internal circuitry of the terminal, preventing power supply side interference from affecting the normal operation of the terminal. The signal isolation subunit uses an optocoupler to achieve isolated transmission of the acquired signal. An optocoupler is a device that transmits electrical signals using light as a medium. It converts the input electrical signal into an optical signal, and then converts the optical signal back into an electrical signal for output, achieving electrical isolation between the input and output circuits during signal transmission, ensuring the stability and safety of signal transmission. The energy storage isolation subunit adopts a combination structure of energy storage capacitor and current limiting resistor. The energy storage capacitor is used to store a small amount of working electrical energy, and the current limiting resistor is used to limit the current in the circuit. By limiting the total energy stored in the circuit through this combination structure, the risk of the circuit generating excessive energy and igniting an explosive mixture under fault conditions is avoided, thus ensuring the intrinsic safety performance of the terminal in hazardous factory environments.

[0062] The operating parameter acquisition unit is used to collect the operating parameters of electrical equipment, including mechanical vibration parameters, electrical discharge parameters, temperature parameters, current parameters, insulation resistance parameters, and speed parameters.

[0063] Furthermore, operating parameters refer to various parameters that reflect the operating status and wear of electrical equipment, including mechanical vibration parameters, electrical discharge parameters, temperature parameters, current parameters, insulation resistance parameters, and rotational speed parameters. Mechanical vibration parameters refer to vibration-related data generated during the operation of electrical equipment; electrical discharge parameters refer to parameters related to discharge phenomena occurring inside or on the surface of the equipment; insulation resistance parameters refer to resistance data characterizing the insulation performance of the equipment; and rotational speed parameters refer to the rotational speed data of rotating parts of the equipment. During data acquisition, various specialized sensors equipped in the intrinsically safe monitoring terminal unit are used to collect parameters. Mechanical vibration parameters are collected using vibration sensors, electrical discharge parameters using partial discharge sensors, temperature parameters using temperature sensors, current parameters using current sensors, insulation resistance parameters using insulation resistance testers, and rotational speed parameters using rotational speed sensors. Each sensor is precisely connected to its corresponding monitoring part of the electrical equipment to ensure that the sensors can accurately capture the equipment's operating status signals. Each sensor collects its corresponding parameters in real time and transmits the collected signals to the operating parameter preprocessing unit.

[0064] The runtime parameter preprocessing unit is used to perform noise reduction, format standardization, and outlier screening on the runtime parameters, and output a standardized runtime parameter dataset.

[0065] Furthermore, denoising refers to the process of eliminating irrelevant interference signals contained in the operating parameters. This is achieved using the Kalman filter algorithm, which is an algorithm that uses the state equation of a linear system to make an optimal estimate of the system state through system input and output observation data. This algorithm filters out random noise in each operating parameter, retaining the valid signal. Format standardization refers to converting parameters of different formats collected by different sensors into a preset digital format, eliminating data integration obstacles caused by differences in the output formats of different sensors. Outlier screening refers to removing extreme data in the operating parameters that do not conform to normal operating conditions, using the 3σ criterion to retain valid data. The specific processing flow is as follows: first, Kalman filtering is performed on each of the collected operating parameters to remove noise interference; then, format standardization is performed on each denoised parameter to unify the data format; finally, outlier screening is performed to remove invalid extreme data. After the above processing, a standardized operating parameter dataset is formed, containing standardized mechanical vibration parameters, standardized electrical discharge parameters, standardized temperature parameters, standardized current parameters, standardized insulation resistance parameters, and standardized speed parameters. The standardized operating parameter dataset refers to the set of operating parameters with unified format and valid data after preprocessing.

[0066] The loss feature extraction unit is used to extract equipment loss features based on a standardized operating parameter dataset.

[0067] Furthermore, the standardized operating parameter dataset includes standardized mechanical vibration parameters, standardized electrical discharge parameters, standardized temperature parameters, standardized current parameters, standardized insulation resistance parameters, and standardized rotational speed parameters. The loss feature extraction unit includes: The mechanical loss feature extraction subunit is used to process the standardized mechanical vibration parameters using the order ratio analysis algorithm to obtain mechanical vibration features. Based on the standardized rotational speed parameters, the mechanical vibration features are synchronously calibrated to obtain mechanical loss features.

[0068] Furthermore, standardized mechanical vibration parameters refer to parameters that, after preprocessing such as denoising and format standardization, can reflect the mechanical vibration state of electrical equipment; standardized rotational speed parameters refer to parameters that, after preprocessing, characterize the rotational speed of rotating parts of the equipment; and the order ratio analysis algorithm is a signal analysis algorithm that can eliminate the influence of rotational speed fluctuations on vibration signals. Its core is to convert non-stationary time-domain vibration signals into stationary order ratio domain signals through resampling, thereby accurately extracting vibration features. In specific implementation, the standardized mechanical vibration parameters are first input into the order ratio analysis algorithm. The algorithm determines the correlation between the signal sampling period and rotational speed changes, performs adaptive resampling processing on the vibration signal to eliminate signal frequency drift caused by rotational speed fluctuations, and then performs spectral analysis on the resampled signal to extract characteristic parameters such as peak amplitude, frequency distribution, and harmonic components, forming mechanical vibration characteristics. Subsequently, time-series variation data of standardized rotational speed parameters are obtained. Using the variation curve of rotational speed parameters as a benchmark, the extracted mechanical vibration characteristics are calibrated on a time scale. The time nodes corresponding to each vibration characteristic parameter are adjusted to ensure that the vibration characteristics accurately match the actual operating state of the equipment under different rotational speed conditions. After calibration, mechanical loss characteristics that can comprehensively characterize mechanical losses such as bearing wear, gear meshing loss, and component loosening are obtained.

[0069] The electrical loss feature extraction subunit is used to process standardized electrical discharge parameters using a spectrum analysis algorithm to obtain discharge parameter features. Based on standardized insulation resistance parameters, the insulation degradation degree of the discharge parameter features is verified to obtain electrical loss features.

[0070] Furthermore, standardized electrical discharge parameters refer to electrical signal parameters that, after preprocessing, reflect partial discharge phenomena inside or on the surface of equipment; standardized insulation resistance parameters refer to parameters that, after preprocessing, characterize the ability of the equipment's insulation material to prevent current flow; and spectrum analysis algorithms refer to algorithms that convert time-domain electrical signals into frequency-domain signals and extract effective information by analyzing frequency distribution characteristics. In implementation, the standardized electrical discharge parameters are first input into the spectrum analysis algorithm. Fourier transform is used to convert the time-domain discharge signal into a frequency-domain signal, extracting parameters such as peak frequency, frequency bandwidth, amplitude spectral density, and phase spectral characteristics to form discharge parameter features. Then, time-series data of standardized insulation resistance parameters are acquired and correlated with the discharge parameter features. During verification, the trend of change in discharge parameter features is judged to be consistent with the trend of change in insulation resistance. Interference features unrelated to insulation degradation are eliminated, while effective features reflecting electrical loss states such as insulation aging, winding partial discharge, and terminal leakage are retained. Finally, these are integrated to form electrical loss features.

[0071] The operating loss feature extraction subunit is used to calculate the temperature-current correlation coefficient based on standardized temperature and current parameters, determine the load heating matching relationship data based on the temperature-current correlation coefficient, and obtain the operating loss features based on the load heating matching relationship data.

[0072] Furthermore, standardized temperature parameters refer to parameters that, after preprocessing, characterize the temperature changes of various parts during the operation of electrical equipment; standardized current parameters refer to parameters that, after preprocessing, reflect the magnitude and changes of current during equipment operation; the temperature-current correlation coefficient refers to an index that quantifies the correlation between temperature and current changes, used to characterize the degree of influence of load current changes on equipment heating; and load heating matching relationship data refers to correlation data reflecting whether the actual load current of the equipment and the corresponding heating temperature conform to normal operating rules. In practice, the Pearson correlation analysis method is first used to calculate the temperature-current correlation coefficient based on the time-series data of standardized temperature and standardized current parameters, clarifying the sensitivity of temperature to current changes. Then, combined with the equipment's rated current, rated operating temperature, and other technical parameters, a load heating matching model is constructed. The actually collected temperature-current correlation coefficient is input into the model to analyze the reasonable fluctuation range of equipment temperature under different current loads, determining the load heating matching relationship data of temperature and current in actual operation, including parameters such as matching degree level, abnormal heating threshold, and temperature-current lag time. Finally, characteristic parameters such as abnormal heating frequency, temperature-current mismatch amplitude, and overheating duration are extracted from the load heating matching relationship data to form operating loss characteristics that can characterize the operating losses of equipment caused by overload operation, excessive contact resistance, poor heat dissipation, etc.

[0073] The fault location unit is used to determine the three-dimensional location information of the fault point and generate fault location information based on electrical discharge parameters and electrical loss characteristics by employing a partial discharge acoustic fingerprint location algorithm and an ultrasonic collaborative location algorithm.

[0074] Furthermore, the partial discharge acoustic signature localization algorithm refers to an algorithm that determines the discharge location by analyzing the acoustic signature signals generated during the electrical discharge process. The acoustic signature signal refers to the sound wave signal generated during the discharge process, and the acoustic signature signals generated by discharges at different locations have different characteristics. The ultrasonic collaborative localization algorithm refers to an algorithm that uses ultrasonic sensors to collect ultrasonic signals generated by the discharge and calculates the signal propagation time difference to assist in locating the fault point. In implementation, the partial discharge acoustic signature localization algorithm first performs feature analysis on the acoustic signature signals corresponding to the electrical discharge parameters, extracting information such as the propagation direction and frequency characteristics of the acoustic signature signals to initially determine the source direction of the discharge signal. Simultaneously, an ultrasonic sensor array collects the ultrasonic signals generated by the discharge, and the ultrasonic collaborative localization algorithm calculates the time difference of the signals received by different sensors. Combined with the deployment location information of the sensor array, the fault point range is further narrowed down. Finally, electrical loss characteristics are combined to determine the possible location of the fault. These electrical loss characteristics include the loss status of various parts of the equipment, and locations with abnormal losses are likely to be the location of the fault. Based on the above analysis results, the three-dimensional coordinates of the fault point are calculated using the spatial positioning model, i.e., the three-dimensional location information of the fault point. The three-dimensional coordinates are then associated with the three-dimensional model information of the equipment to clarify the specific part of the equipment corresponding to the fault point, and fault location information containing the three-dimensional coordinates of the fault point, the corresponding equipment part, and the associated information of the fault type is generated.

[0075] The positioning transmission unit is used to transmit equipment wear characteristics and fault location information to the edge computing gateway. The positioning transmission unit is configured with the national cryptographic SM4 encryption protocol.

[0076] Furthermore, the SM4 encryption protocol, a block cipher algorithm independently developed in my country, is used to encrypt and protect transmitted data, ensuring security and confidentiality during data transmission. In specific implementation, the equipment wear characteristics and fault location information are first integrated into a data format, and then packaged into a transmission data packet according to a preset format. The transmission data packet is encrypted using the SM4 encryption module built into the positioning transmission unit. During encryption, a preset key is used to perform block encryption operations on the data packet, generating encrypted ciphertext data. After encryption, the ciphertext data is transmitted to the edge computing gateway via a wireless communication module. The wireless communication module uses a communication method that complies with the communication standards for hazardous plant areas to ensure the stability and reliability of data transmission. After receiving the ciphertext data, the edge computing gateway uses the corresponding decryption module with the same key to decrypt it, restoring the equipment wear characteristics and fault location information, ensuring secure and controllable data transmission throughout the entire process.

[0077] The third-layer data output unit is used to integrate equipment loss characteristics and fault location information into third-layer monitoring data and output it.

[0078] Furthermore, the third-layer monitoring data refers to a standardized set of monitoring data used to reflect the operating loss status and fault conditions of electrical equipment. During the integration process, the equipment loss characteristics and fault location information are first structured, and the two types of data are classified and organized according to preset data field specifications, clarifying the attributes and meanings of each data item. The structured equipment loss characteristics and fault location information are then linked and integrated to establish a correspondence between loss characteristics and fault location information, enabling the data to corroborate each other. Data acquisition timestamps and unique equipment identifiers are added. The data acquisition timestamp records the time information of data acquisition, and the unique equipment identifier is a unique identifier used to distinguish different electrical equipment, ensuring data traceability. The integrated data is packaged according to a preset transmission format to form standardized third-layer monitoring data. The third-layer monitoring data is transmitted to the edge computing gateway through a preset communication interface, ensuring the integrity, timeliness, and standardization of data transmission, providing high-quality operating loss and fault dimension data support for subsequent multi-source data integration and analysis.

[0079] Edge computing gateways are used to determine target monitoring data and hierarchical anomaly warning trigger information based on first-layer monitoring data, second-layer monitoring data, and third-layer monitoring data.

[0080] Specifically, the edge computing gateway includes: The hierarchical preprocessing unit is used to extract certificate-type text data, environmental time-series data, and equipment operation parameter data from the first-level monitoring data, the second-level monitoring data, and the third-level monitoring data. It performs semantic alignment and format standardization on the certificate-type text data, wavelet threshold denoising and timestamp calibration on the environmental time-series data, and Kalman filtering denoising and outlier screening on the equipment operation parameters, outputting a standardized multi-source dataset.

[0081] Furthermore, certificate-type text data refers to structured text information related to explosion-proof qualification certificates extracted from the first-level monitoring data; environmental time-series data refers to environmental corrosion-related data arranged in chronological order extracted from the second-level monitoring data; and equipment operating parameter data refers to various parameter data reflecting the operating status of electrical equipment extracted from the third-level monitoring data. Semantic alignment processing unifies certificate-type text information with different expressions but the same meaning into a preset standard semantic expression, eliminating semantic ambiguity; format standardization processing converts the semantically aligned certificate-type text data into a unified structured format, ensuring data fields are standardized and structurally consistent. Wavelet threshold denoising processing decomposes environmental time-series data using wavelet transform, sets thresholds to filter out noise components, and retains effective signals; timestamp calibration processing adjusts the collection timestamps of environmental time-series data based on the unified system time of the edge computing gateway, ensuring time synchronization of environmental data from different sources. Kalman filter denoising processing uses the Kalman filter algorithm to construct system state equations and observation equations, filters random interference signals in equipment operating parameters, and obtains optimal estimates; outlier screening identifies and removes extreme data in equipment operating parameters that deviate from normal operating patterns. In practice, the corresponding data types are first extracted from the three layers of monitoring data, and then the above-mentioned targeted preprocessing operations are performed on each type of data. Finally, the processed certificate text data, environmental time series data and equipment operation parameter data are integrated to form a standardized multi-source dataset containing multi-dimensional, high-quality standardized data.

[0082] The edge-side real-time computing unit is used to perform cross-dimensional data association matching on standardized multi-source datasets based on preset feature weight matrices and weighted Euclidean distance, filter out target monitoring data, and use a sliding window mechanism to perform trend fitting on the target monitoring data to generate data change slope and fluctuation coefficient.

[0083] Furthermore, the preset feature weight matrix refers to a pre-constructed matrix used to characterize the importance of different dimensions of data on the explosion-proof safety of equipment. The matrix contains weight values ​​corresponding to each data feature, determined based on historical safety data and expert experience. Weighted Euclidean distance refers to a calculation method that incorporates feature weights when calculating the distance between data from different dimensions, used to accurately measure the similarity of cross-dimensional data. Cross-dimensional data association matching refers to the process of selecting data combinations with high similarity and inherent correlation by calculating the weighted Euclidean distance between data from different dimensions. Target monitoring data refers to data selected from standardized multi-source datasets that plays a crucial role in characterizing the explosion-proof safety status of equipment. The sliding window mechanism refers to a mechanism that sets a fixed-length time window and processes target monitoring data segment by segment in chronological order; trend fitting refers to fitting the target monitoring data within the window using a mathematical model to reconstruct the data change patterns. Data change slope refers to an indicator characterizing the speed of change of target monitoring data within the window, and fluctuation coefficient refers to an indicator characterizing the dispersion of target monitoring data within the window. In practice, a pre-set feature weight matrix is ​​first loaded. Then, the data from each dimension of the standardized multi-source dataset are substituted into the weighted Euclidean distance formula for calculation to obtain the similarity values ​​between different dimensions of data. Based on the similarity values, data combinations with high correlation are selected to form target monitoring data that can accurately reflect the explosion-proof safety status of equipment. A sliding window mechanism is used to segment the target monitoring data. A linear fitting method is used to fit the trend of the data in each window. The corresponding data change slope is calculated through the linear equation obtained by fitting. The fluctuation coefficient is calculated based on the mean and dispersion of the data in the window. Finally, the target monitoring data, the data change slope and fluctuation coefficient corresponding to each window are output.

[0084] The early warning triggering unit is used to generate graded anomaly early warning triggering information based on a preset dynamic threshold model, target monitoring data, data change slope and fluctuation coefficient, using a dual judgment mechanism of absolute threshold and trend threshold.

[0085] Furthermore, the absolute threshold refers to the pre-defined normal range boundary of various target monitoring data, used to determine whether the data exceeds the normal range. The trend threshold refers to the pre-defined normal range boundary of the data change slope and fluctuation coefficient, used to determine whether the data change trend is abnormal. The dual judgment mechanism refers to simultaneously judging anomalies based on both absolute and trend thresholds, ensuring the comprehensiveness and accuracy of anomaly identification. The preset dynamic threshold model refers to a pre-built model that stores absolute and trend thresholds for different equipment types and operating conditions. The model can dynamically adjust the thresholds according to the real-time operating status of the equipment and historical data to adapt to different scenario requirements. The graded anomaly warning trigger information refers to a standardized warning signal that includes information such as warning level, abnormal data item, anomaly occurrence time, and anomaly type. The warning level is divided according to the severity of the anomaly. In practice, the system first retrieves the absolute and trend thresholds from the preset dynamic threshold model that match the current equipment type and operating conditions. It then compares the target monitoring data with each absolute threshold to determine if any data item exceeds the normal range. Next, it compares the data change slope and fluctuation coefficient with the corresponding trend thresholds to determine if the data change trend is abnormal. Finally, it performs a dual judgment based on the two comparison results. If the target monitoring data exceeds the absolute threshold, or the data change slope or fluctuation coefficient exceeds the trend threshold, it is judged as abnormal. Based on the degree of deviation from the threshold, it classifies the abnormality into three warning levels: minor, moderate, and severe, corresponding to low-level, medium-level, and high-level warnings, respectively. Finally, it integrates information such as the warning level, abnormal data item, abnormal occurrence time, and abnormal judgment criteria to generate standardized hierarchical abnormality warning trigger information.

[0086] The cloud platform is used to build digital twin models of equipment based on target monitoring data, hierarchical anomaly warning trigger information, and historical monitoring data.

[0087] Specifically, historical monitoring data includes historical first-level monitoring data, historical second-level monitoring data, historical third-level monitoring data, historical target monitoring data, and historical graded anomaly warning trigger information. The cloud platform includes: The multi-source data storage unit is used to classify and store target monitoring data, hierarchical anomaly warning trigger information and historical monitoring data using a distributed time-series database, and to build a data index based on the device's unique identifier.

[0088] Furthermore, a distributed time-series database refers to a database system that employs a distributed architecture and is specifically designed for storing and managing time-series data. Time-series data refers to continuous data generated in chronological order. This database possesses characteristics such as high throughput, high reliability, and fast time-series data retrieval. Target monitoring data refers to data selected from standardized multi-source datasets that plays a crucial role in characterizing the explosion-proof safety status of equipment. Graded anomaly warning trigger information refers to standardized warning signals containing information such as warning levels and anomaly data items. Historical monitoring data refers to a stored collection of past monitoring data, including historical first-level monitoring data, historical second-level monitoring data, historical third-level monitoring data, historical target monitoring data, and historical graded anomaly warning trigger information. Unique equipment identifier refers to a unique identifier used to distinguish different electrical devices. Data index refers to an index structure built to improve data query efficiency, through which various types of monitoring data for corresponding equipment can be quickly located. The system categorizes and labels target monitoring data, hierarchical anomaly warning trigger information, and historical monitoring data, clarifying the ownership type and associated device information of each data type. The categorized data is then written into a distributed time-series database in chronological order, with a timestamp and a unique device identifier tag added to each data type. A multi-level data index is built based on the unique device identifier, with index levels including device identifier, data type, and time range. The constructed index is optimized using an index optimization algorithm to ensure that subsequent retrieval of various monitoring data for the corresponding device within any time range can be achieved quickly using the unique device identifier, realizing the orderly storage and efficient access of multi-source monitoring data.

[0089] The digital twin model building unit is used to construct a digital twin model based on the equipment's 3D model information, target monitoring data, historical monitoring data, corrosion degree data, remaining life data, and equipment wear characteristics. It outputs digital twin mapping data, which includes the equipment's full state correlation and equipment state projection results.

[0090] Furthermore, the equipment 3D model information refers to the digital model data extracted from the explosion-proof certificate to characterize the equipment's 3D structure; corrosion degree data refers to data characterizing the current severity of corrosion of the equipment; remaining life data refers to the remaining time the equipment can maintain normal explosion-proof performance under current environmental conditions; equipment loss characteristics refer to comprehensive loss data including mechanical loss characteristics, electrical loss characteristics, and operational loss characteristics. A digital twin model refers to a virtual model constructed digitally that corresponds one-to-one with the physical equipment, capable of mapping the physical equipment's operating state in real time and realizing state prediction; digital twin mapping data refers to the relevant data on the state mapping between the digital twin model and the physical equipment; equipment full-state correlation refers to the inherent correlation data between the states, operating parameters, and environmental parameters of various parts of the equipment; and equipment state prediction results refer to the predictive data on future state changes of the equipment based on the model. The process involves loading the equipment's 3D model information as the foundational framework for the digital twin model; extracting key data items from target monitoring data, historical monitoring data, corrosion level data, remaining lifespan data, and equipment wear characteristics to establish mapping relationships between each data item and different parts of the equipment's 3D model; fusing real-time target monitoring data with historical monitoring data using a time-series data fusion algorithm to generate a complete and real-time equipment status dataset; inputting the fused equipment status dataset into the foundational framework and combining it with a physics engine to construct dynamic mapping rules for equipment operating status, enabling real-time mapping of the digital twin model to the physical equipment status; training a status inference algorithm based on historical monitoring data and integrating it into the digital twin model to infer the equipment's status changes over a future period; and finally outputting digital twin mapping data containing the full state correlation of the equipment and the equipment status inference results.

[0091] The dynamic threshold training unit is used to train and update the preset dynamic threshold model based on historical monitoring data, and then distribute the updated preset dynamic threshold model to the edge computing gateway.

[0092] Furthermore, the preset dynamic threshold model refers to a model that stores absolute and trend thresholds for different device types and operating conditions, used for anomaly detection in edge computing gateways; the absolute threshold refers to the normal range boundary of various target monitoring data, and the trend threshold refers to the normal range boundary of data change slope and fluctuation coefficient. Historical monitoring data is extracted from a distributed time-series database. This data is then categorized and filtered according to equipment type and operating conditions, removing invalid and redundant data to create training datasets for different equipment operating conditions. Labels are added to these training datasets, including information such as whether the data is abnormal and the level of abnormality. A machine learning algorithm is selected as the training algorithm, and the labeled training dataset is input into a preset dynamic threshold model for training. The absolute and trend threshold parameters in the model are iteratively optimized to ensure the model accurately adapts to the anomaly detection requirements of different equipment types and operating conditions. After training, the model is validated using a validation dataset to evaluate metrics such as accuracy and false alarm rate. If the metrics do not meet the preset requirements, the algorithm parameters are adjusted and retrained until the metrics meet the standards. The updated preset dynamic threshold model is then converted to a format suitable for the edge computing gateway's operating environment. Finally, the updated preset dynamic threshold model is distributed to the edge computing gateway via an encrypted communication link, simultaneously triggering the edge computing gateway's model update mechanism to ensure the new model takes effect promptly.

[0093] The explosion-proof assessment module is used to evaluate the explosion-proof safety status based on target monitoring data, graded abnormality early warning trigger information, historical monitoring data, and digital twin model, and obtain the assessment results.

[0094] Specifically, the explosion-proof determination module includes: The indicator system matching unit is used to match the corresponding explosion-proof safety assessment indicator system based on the equipment model information. The assessment indicator system is constructed based on the analytic hierarchy process and includes certificate and qualification compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators. The certificate and qualification compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators are respectively associated with the corresponding data items in the target monitoring data.

[0095] Furthermore, equipment model information refers to the identification information extracted from explosion-proof qualification certificates used to distinguish equipment specifications and types; the explosion-proof safety assessment index system refers to the set of indicators used to comprehensively assess the explosion-proof safety status of equipment; the analytic hierarchy process (AHP) refers to a systematic analysis method that decomposes complex decision-making problems into multiple levels and factors, and determines the weights of factors through pairwise comparisons; certificate qualification compliance indicators refer to indicators used to assess whether the explosion-proof qualification of equipment meets the standards; environmental corrosion risk indicators refer to indicators used to assess the degree of impact of environmental corrosion on the explosion-proof performance of equipment; equipment wear and deterioration indicators refer to indicators used to assess the degree of equipment operating wear and performance deterioration; and early warning trigger level indicators refer to indicators used to assess the severity of abnormal early warnings. In practice, the equipment model information of the current equipment is first extracted. Then, the explosion-proof safety assessment indicator system corresponding to the equipment model information is retrieved from the preset indicator system database. The indicator system database stores the mapping relationship between different equipment models and corresponding assessment indicator systems in advance. After confirming that the retrieved assessment indicator system includes certificate qualification compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators, the association relationship between each indicator and the corresponding data item in the target monitoring data is established, and the data source field corresponding to each indicator is clarified to ensure that the data required for each indicator can be accurately extracted in the subsequent assessment process.

[0096] The dynamic weight configuration unit is used to adjust the weight coefficients of certificate compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators based on fault case data in historical monitoring data, and obtain updated weight coefficients.

[0097] Furthermore, historical monitoring data refers to the stored collection of various past monitoring data; fault case data refers to the collection of cases extracted from historical monitoring data, containing information such as equipment fault occurrence time, fault type, fault cause, and corresponding indicator data; weight coefficients refer to values ​​used to characterize the importance of each evaluation indicator in explosion-proof safety assessment; updated weight coefficients refer to the adjusted weight values ​​of each indicator adapted to the current equipment operating scenario. Fault case data is extracted from historical monitoring data in a distributed time-series database. This data is cleaned and organized, removing incomplete cases and retaining valid cases containing complete indicator data and fault details. The correlation between each evaluation indicator data and the occurrence of the fault in valid fault cases is statistically analyzed to determine the contribution of different indicators to the occurrence of the fault. Based on the correlation analysis results and combined with the weight calculation logic of the analytic hierarchy process (AHP), the initial weight coefficients are adjusted. Indicators with a high correlation to the occurrence of the fault have their weight coefficients increased, while those with a low correlation have their weight coefficients appropriately decreased. After adjustment, the weight coefficients are normalized to ensure that the sum of the weight coefficients of each indicator is one, ultimately yielding the updated weight coefficients.

[0098] The evidence fusion unit is used to fuse target monitoring data and hierarchical anomaly warning trigger information by employing evidence theory, based on updated weight coefficients and the correlation of the full state of the equipment, to generate a credibility distribution result.

[0099] Furthermore, evidence theory refers to a fusion reasoning method for handling uncertain information, which achieves the fusion of multi-source information by defining a basic probability allocation function, a trust function, and a likelihood function; updated weight coefficients refer to the adjusted values ​​that characterize the importance of each evaluation indicator; equipment full-state correlation refers to the intrinsic correlation data between the states, operating parameters, environmental parameters, etc. of various parts of the equipment; target monitoring data refers to the data reflecting the current explosion-proof safety critical state of the equipment; graded anomaly early warning trigger information refers to standardized early warning signals containing information such as early warning level and abnormal data items; and credibility distribution results refer to the credibility value distribution corresponding to each explosion-proof safety state obtained through evidence fusion. In practice, the target monitoring data and graded anomaly warning trigger information are first classified and decomposed according to the evaluation index system to obtain the original evidence data corresponding to each evaluation index. Based on the updated weight coefficient, weights are assigned to the original evidence data corresponding to each index, and the weight assignment results are consistent with the updated weight coefficient. Combining the correlation of the equipment's full state, contradictory evidence data is eliminated, and consistent valid evidence is retained. The valid evidence is substituted into the fusion formula of the evidence theory, and the support degree of each piece of evidence for different explosion-proof safety states is calculated through the basic probability allocation function. Then, the support degree of multi-source evidence is synthesized through the trust function and the likelihood function, and finally, the credibility distribution results corresponding to each explosion-proof safety state are generated.

[0100] The safety level determination unit is used to map the confidence distribution results to the corresponding explosion-proof safety level based on the preset level determination rules and state inference results, and output the safety level determination result.

[0101] Furthermore, the preset level determination rules refer to the pre-defined criteria for converting the credibility distribution results into specific safety levels, including the mapping relationship between different credibility ranges and corresponding safety levels; the state inference results refer to the prediction data of the digital twin model for future changes in the state of the equipment; the credibility distribution results refer to the credibility value distribution corresponding to each explosion-proof safety state; the explosion-proof safety level refers to the level classification used to characterize the current and future explosion-proof safety level of the equipment; and the safety level determination results refer to standardized result data containing information such as specific safety levels, determination criteria, and risk warnings. The system acquires the preset level determination rules and the state simulation results output by the digital twin model; compares the maximum confidence value in the confidence distribution results with the confidence range in the level determination rules to initially determine the corresponding explosion-proof safety level; verifies the initially determined safety level in conjunction with the state simulation results. If the state simulation results show that the equipment's safety status is likely to deteriorate in the future, the safety level is appropriately lowered; if the simulation results show that the safety status is stable, the initial determination level is maintained; after verification, the system integrates the confidence distribution results, state simulation results, and level determination criteria to generate a safety level determination result that includes the specific safety level, a description of the determination process, and risk warning information, and outputs it to the relevant monitoring terminal through a preset interface.

[0102] This embodiment provides an IoT-based intelligent monitoring method for explosion-proof safety of electrical equipment in factory areas. First, the intelligent certificate management module performs structured parsing of explosion-proof qualification certificate data, transforming unstructured certificate text and image data into standardized structured information. Combined with dual-dimensional verification technology, key certificate information is validated to ensure the authenticity and completeness of the qualification data. The generated equipment explosion-proof qualification status data can be directly used as the basic input parameters for subsequent safety assessments, achieving technical compatibility between qualification information and monitoring data and avoiding data integration obstacles caused by inconsistent qualification data formats. Second, the environmental perception and corrosion analysis module employs an intrinsically safe distributed sensor network. Distributed node deployment enables comprehensive data collection of the equipment's surrounding environment. Coupled with a coupled simulation model, multi-dimensional environmental parameters are analyzed, overcoming the challenge that single environmental parameter monitoring cannot reflect complex corrosion mechanisms. The method accurately quantifies the correlation between environmental factors and equipment corrosion. The generated corrosion degree data and remaining life data have clear technical traceability, providing precise technical support for equipment corrosion prevention and control. Simultaneously, the intrinsically safe circuit design meets explosion-proof requirements in hazardous environments, ensuring the technical safety of the data acquisition process. Subsequently, the equipment loss IoT monitoring module collects operating parameters using monitoring terminals. Combined with feature extraction algorithms, it extracts equipment loss characteristics from the raw operating data, achieving a quantitative representation of hidden equipment losses. Then, a fault location algorithm collaboratively analyzes electrical discharge parameters and loss characteristics to accurately pinpoint the spatial location of the fault. This overcomes the limitations of traditional operation monitoring, which can only detect anomalies but not locate the root cause. The generated loss characteristic data and fault location information possess high-dimensional technical correlation. Next, an edge computing gateway performs centralized preprocessing and feature filtering on the three layers of monitoring data. Edge-side real-time computing technology reduces the data transmission and processing pressure on the cloud, improving the real-time performance of data processing. Combined with a tiered early warning algorithm, the filtered target monitoring data is used to determine anomalies, achieving accurate differentiation between different levels of anomalies. This overcomes the technical shortcomings of traditional single-threshold early warning systems. The generated tiered anomaly early warning trigger information has clear technical judgment criteria, providing accurate anomaly data input for subsequent evaluation modules and ensuring the technical reliability of the entire system's early warning response. Furthermore, by integrating target monitoring data, hierarchical anomaly warning trigger information, and historical monitoring data through a cloud platform, a digital twin model of the equipment is constructed. Data fusion technology is used to achieve deep correlation between real-time data and historical data, restoring the full life cycle status of the equipment. Combined with digital mapping technology, the physical state of the equipment is transformed into a quantifiable and predictable digital model, breaking through the technical limitations of traditional monitoring that can only reflect the real-time status and cannot predict trends. This provides a technical carrier for the dynamic extrapolation and trend prediction of equipment status, and enhances the technical foresight of safety management.Finally, an explosion-proof safety status assessment is conducted by integrating multi-dimensional data through an explosion-proof judgment module. Multi-source data collaborative analysis technology integrates technical data from multiple levels, including equipment qualifications, environmental corrosion, operational losses, and digital twin simulations, improving the comprehensiveness and scientific rigor of the assessment results. Combined with assessment algorithms, data characteristics are transformed into clear safety status results, overcoming the limitations of traditional single-dimensional assessments. The generated assessment results possess a complete technical derivation chain, providing precise technical decision-making basis for explosion-proof safety management in the plant area, and improving the overall precision of explosion-proof safety control from a technical perspective. This addresses the problem that current IoT-based intelligent monitoring systems for explosion-proof safety of electrical equipment in plants struggle to achieve full-process data linkage, dynamic and accurate assessment, and multi-system collaborative control, thus failing to fully meet the high-level safety control requirements of hazardous plant areas.

[0103] The above are embodiments of the IoT-based intelligent monitoring method for explosion-proof safety of electrical equipment in factories provided in this application. Other embodiments of the IoT-based intelligent monitoring method for explosion-proof safety of electrical equipment in factories provided in this application are described below.

[0104] This invention also discloses an intelligent monitoring method for explosion-proof safety of electrical equipment in factories based on the Internet of Things, such as... Figure 2 As shown, the method includes: acquiring explosion-proof qualification certificate data of electrical equipment; parsing the explosion-proof qualification certificate data to obtain certificate information; performing dual-dimensional verification based on the certificate information to obtain equipment explosion-proof qualification status data; and outputting the certificate information and equipment explosion-proof qualification status data as first-layer monitoring data; collecting environmental parameter sets using an intrinsically safe distributed sensor network; determining the corrosion degree data and remaining life data of the electrical equipment based on a preset coupled simulation model and environmental parameter set; and outputting the corrosion degree data and remaining life data as second-layer monitoring data; collecting operating parameters of the electrical equipment using a monitoring terminal; determining the electrical equipment wear characteristics and fault location information based on the operating parameters; and outputting the wear characteristics and fault location information as third-layer monitoring data; determining target monitoring data and graded anomaly warning trigger information based on the first-layer monitoring data, second-layer monitoring data, and third-layer monitoring data; constructing a digital twin model of the equipment based on the target monitoring data, graded anomaly warning trigger information, and historical monitoring data; and conducting an explosion-proof safety status assessment based on the target monitoring data, graded anomaly warning trigger information, historical monitoring data, and digital twin model to obtain the assessment result.

[0105] The technical solution of this invention has the following advantages: This embodiment provides an IoT-based intelligent monitoring method for explosion-proof safety of electrical equipment in factory areas. First, the intelligent certificate management module performs structured parsing of explosion-proof qualification certificate data, transforming unstructured certificate text and image data into standardized structured information. Combined with dual-dimensional verification technology, key certificate information is validated to ensure the authenticity and completeness of the qualification data. The generated equipment explosion-proof qualification status data can be directly used as the basic input parameters for subsequent safety assessments, achieving technical compatibility between qualification information and monitoring data and avoiding data integration obstacles caused by inconsistent qualification data formats. Second, the environmental perception and corrosion analysis module employs an intrinsically safe distributed sensor network. Distributed node deployment enables comprehensive data collection of the equipment's surrounding environment. Coupled with a coupled simulation model, multi-dimensional environmental parameters are analyzed, overcoming the challenge that single environmental parameter monitoring cannot reflect complex corrosion mechanisms. The method accurately quantifies the correlation between environmental factors and equipment corrosion. The generated corrosion degree data and remaining life data have clear technical traceability, providing precise technical support for equipment corrosion prevention and control. Simultaneously, the intrinsically safe circuit design meets explosion-proof requirements in hazardous environments, ensuring the technical safety of the data acquisition process. Subsequently, the equipment loss IoT monitoring module collects operating parameters using monitoring terminals. Combined with feature extraction algorithms, it extracts equipment loss characteristics from the raw operating data, achieving a quantitative representation of hidden equipment losses. Then, a fault location algorithm collaboratively analyzes electrical discharge parameters and loss characteristics to accurately pinpoint the spatial location of the fault. This overcomes the limitations of traditional operation monitoring, which can only detect anomalies but not locate the root cause. The generated loss characteristic data and fault location information possess high-dimensional technical correlation. Next, an edge computing gateway performs centralized preprocessing and feature filtering on the three layers of monitoring data. Edge-side real-time computing technology reduces the data transmission and processing pressure on the cloud, improving the real-time performance of data processing. Combined with a tiered early warning algorithm, the filtered target monitoring data is used to determine anomalies, achieving accurate differentiation between different levels of anomalies. This overcomes the technical shortcomings of traditional single-threshold early warning systems. The generated tiered anomaly early warning trigger information has clear technical judgment criteria, providing accurate anomaly data input for subsequent evaluation modules and ensuring the technical reliability of the entire system's early warning response. Furthermore, by integrating target monitoring data, hierarchical anomaly warning trigger information, and historical monitoring data through a cloud platform, a digital twin model of the equipment is constructed. Data fusion technology is used to achieve deep correlation between real-time data and historical data, restoring the full life cycle status of the equipment. Combined with digital mapping technology, the physical state of the equipment is transformed into a quantifiable and predictable digital model, breaking through the technical limitations of traditional monitoring that can only reflect the real-time status and cannot predict trends. This provides a technical carrier for the dynamic extrapolation and trend prediction of equipment status, and enhances the technical foresight of safety management.Finally, an explosion-proof safety status assessment is conducted by integrating multi-dimensional data through an explosion-proof judgment module. Multi-source data collaborative analysis technology integrates technical data from multiple levels, including equipment qualifications, environmental corrosion, operational losses, and digital twin simulations, improving the comprehensiveness and scientific rigor of the assessment results. Combined with assessment algorithms, data characteristics are transformed into clear safety status results, overcoming the limitations of traditional single-dimensional assessments. The generated assessment results possess a complete technical derivation chain, providing precise technical decision-making basis for explosion-proof safety management in the plant area, and improving the overall precision of explosion-proof safety control from a technical perspective. This addresses the problem that current IoT-based intelligent monitoring systems for explosion-proof safety of electrical equipment in plants struggle to achieve full-process data linkage, dynamic and accurate assessment, and multi-system collaborative control, thus failing to fully meet the high-level safety control requirements of hazardous plant areas.

[0106] The implementation methods of each step in the IoT-based intelligent monitoring method for explosion-proof safety of electrical equipment in factories provided in this embodiment of the invention have been described in detail in any of the above system embodiments, and therefore will not be repeated here.

[0107] In this embodiment, the IoT-based intelligent monitoring system for explosion-proof safety of electrical equipment in the factory is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

Claims

1. An intelligent monitoring system for explosion-proof safety of electrical equipment in a factory based on the Internet of Things, characterized in that, The system includes: The certificate intelligent management module is used to acquire explosion-proof qualification certificate data of electrical equipment, parse the explosion-proof qualification certificate data to obtain certificate information, perform dual-dimensional verification based on the certificate information to obtain equipment explosion-proof qualification status data, and output the certificate information and equipment explosion-proof qualification status data as first-level monitoring data. The environmental perception and corrosion analysis module is used to collect a set of environmental parameters using an intrinsically safe distributed sensor network, determine the corrosion degree data and remaining life data of the electrical equipment based on a preset coupled simulation model and the set of environmental parameters, and output the corrosion degree data and remaining life data as second-layer monitoring data. The IoT monitoring module for equipment loss is used to collect the operating parameters of the electrical equipment using a monitoring terminal, determine the loss characteristics and fault location information of the electrical equipment based on the operating parameters, and output the loss characteristics and fault location information as third-layer monitoring data. An edge computing gateway is used to determine target monitoring data and hierarchical anomaly warning trigger information based on the first layer monitoring data, the second layer monitoring data and the third layer monitoring data; A cloud platform is used to construct a digital twin model of the equipment based on the target monitoring data, the hierarchical anomaly warning trigger information, and historical monitoring data; The explosion-proof determination module is used to assess the explosion-proof safety status based on the target monitoring data, the graded abnormality early warning trigger information, the historical monitoring data, and the digital twin model, and obtain the assessment result.

2. The system according to claim 1, characterized in that, The certificate intelligent management module, environmental perception and corrosion analysis module, and equipment loss IoT monitoring module are each communicatively connected to the edge computing gateway. The edge computing gateway is communicatively connected to the cloud platform. The cloud platform is communicatively connected to the explosion-proof determination module. The certificate intelligent management module includes: The certificate entry unit is used to receive paper scan data, electronic file data and manually entered data of explosion-proof qualification certificates to obtain explosion-proof qualification certificate data; The parsing unit is used to perform structured parsing of the explosion-proof certificate data and extract certificate information, which includes certificate number information, equipment model information, equipment three-dimensional model information, issuing authority information, validity period information, explosion-proof level information, material characteristic information, equipment explosion-proof structure information, and applicable environmental parameter information. The first verification unit is used to generate an expiration warning signal based on the system time and the validity period information; The second verification unit is used to verify the certificate number information, the equipment model information, and the issuing authority information based on the official database and the issuing authority traceability platform to determine the qualification assessment result; The qualification assessment unit is used to generate equipment explosion-proof qualification status data based on the expiration warning signal and the qualification assessment result; The certificate association unit is used to establish a unique association between the explosion-proof certificate and the electrical equipment, obtain the association mapping data, and integrate the association mapping data into the certificate information; The first-level data output unit is used to integrate the certificate information and the equipment explosion-proof qualification status data into first-level monitoring data and output them.

3. The system according to claim 2, characterized in that, The environmental sensing and corrosion analysis module includes: The intrinsically safe sensing unit is used to collect environmental parameters of the corresponding factory area of ​​the electrical equipment using an intrinsically safe distributed sensing network, and obtain a set of environmental parameters, which includes salt spray parameters, humidity parameters, temperature parameters, harmful gas concentration parameters, dust concentration parameters and ultraviolet intensity parameters. An environmental data preprocessing unit is used to perform wavelet transform denoising, standardization and time series alignment on the environmental parameter set to obtain a time series dataset of environmental parameters. The coupled simulation unit is used to calculate the corrosion degree data and remaining corrosion life data of various parts of the electrical equipment based on the environmental parameter time series dataset and the material characteristic information using a preset coupled simulation model. The material characteristic information includes the material thermal conductivity, electrochemical corrosion constant and insulating medium parameters. An environmental risk level classification unit is used to determine the aging corrosion risk level of equipment based on preset risk judgment rules, the corrosion degree data, and the remaining corrosion life data, generate risk level identification data, and integrate the risk level identification data and the environmental parameter time series dataset into the corrosion degree data. The second-layer data output unit is used to integrate the corrosion degree data and the remaining lifespan data into second-layer monitoring data and output them.

4. The system according to claim 3, characterized in that, The coupled simulation unit includes: The parameter coordination subunit is used to decompose the environmental parameter time series dataset into several time series segments according to the time dimension, and perform parameter normalization and coordination mapping on each time series segment based on the material thermal conductivity, the electrochemical corrosion constant and the insulating medium parameter to generate a multi-dimensional time series parameter set. The collaborative computing subunit is used to construct multi-field coupled control equations based on the multi-dimensional time series parameter set using a preset coupled simulation model, and calculate the dynamic corrosion rate, corrosion depth distribution data and insulation performance degradation curve of each part of the electrical equipment by calculating the multi-field coupled control equations. The preset coupled simulation model is used to simulate the coupling effect of the salt spray humidity and temperature synergistic corrosion field and the equipment operation electric field and temperature field. The remaining lifetime dynamic prediction subunit is used to extract the feature inflection points of the dynamic corrosion rate, corrosion depth distribution data and insulation performance degradation curve, construct a corrosion lifetime mapping model based on the feature inflection points and the equipment explosion-proof structure information, and obtain the prediction result based on the corrosion lifetime mapping model and the environmental parameter time series dataset. The result correction subunit is used to correct the prediction result using a preset environmental parameter fluctuation coefficient, so as to obtain the remaining corrosion life data and life decay trend prediction curve of each part of the electrical equipment. The data determination subunit is used to determine the corrosion degree data based on the corrosion depth distribution data, integrate the lifetime decay trend prediction curve into the remaining corrosion lifetime data, and output the corrosion degree data and the remaining corrosion lifetime data.

5. The system according to claim 4, characterized in that, The device loss IoT monitoring module includes: The intrinsically safe monitoring terminal unit adopts a three-level energy isolation intrinsically safe circuit architecture, which includes a power isolation subunit, a signal isolation subunit, and an energy storage isolation subunit. An operating parameter acquisition unit is used to acquire the operating parameters of the electrical equipment, including mechanical vibration parameters, electrical discharge parameters, temperature parameters, current parameters, insulation resistance parameters, and rotational speed parameters. The running parameter preprocessing unit is used to perform noise reduction, format standardization, and outlier screening on the running parameters, and output a standardized running parameter dataset. The loss feature extraction unit is used to extract equipment loss features based on the standardized operating parameter dataset. The fault location unit is used to determine the three-dimensional location information of the fault point and generate fault location information based on the electrical discharge parameters and the electrical loss characteristics using a partial discharge acoustic signature location algorithm and an ultrasonic collaborative location algorithm. A positioning transmission unit is used to transmit the device wear characteristics and the fault location information to the edge computing gateway. The positioning transmission unit is configured with the national cryptographic SM4 encryption protocol. The third-layer data output unit is used to integrate the equipment loss characteristics and the fault location information into third-layer monitoring data and output it.

6. The system according to claim 5, characterized in that, The standardized operating parameter dataset includes standardized mechanical vibration parameters, standardized electrical discharge parameters, standardized temperature parameters, standardized current parameters, standardized insulation resistance parameters, and standardized rotational speed parameters. The loss feature extraction unit includes: The mechanical loss feature extraction subunit is used to process the standardized mechanical vibration parameters using the order ratio analysis algorithm to obtain mechanical vibration features, and to synchronously calibrate the mechanical vibration features based on the standardized rotational speed parameters to obtain mechanical loss features. The electrical loss feature extraction subunit is used to process the standardized electrical discharge parameters using a spectrum analysis algorithm to obtain discharge parameter features, and to verify the degree of insulation degradation of the discharge parameter features based on the standardized insulation resistance parameters to obtain electrical loss features. The operating loss feature extraction subunit is used to calculate the temperature-current correlation coefficient based on the standardized temperature parameters and the standardized current parameters, determine the load heating matching relationship data based on the temperature-current correlation coefficient, and obtain the operating loss features based on the load heating matching relationship data.

7. The system according to claim 6, characterized in that, The edge computing gateway includes: The hierarchical preprocessing unit is used to extract certificate-type text data, environmental time-series data, and equipment operating parameter data from the first-layer monitoring data, the second-layer monitoring data, and the third-layer monitoring data. It performs semantic alignment and format standardization processing on the certificate-type text data, wavelet threshold denoising processing and timestamp calibration processing on the environmental time-series data, and Kalman filtering denoising processing and outlier screening processing on the equipment operating parameters, and outputs a standardized multi-source dataset. The edge-side real-time computing unit is used to perform cross-dimensional data association matching on the standardized multi-source dataset based on a preset feature weight matrix and weighted Euclidean distance, filter out target monitoring data, and use a sliding window mechanism to perform trend fitting on the target monitoring data to generate data change slope and fluctuation coefficient. The early warning triggering unit is used to generate graded anomaly early warning triggering information based on a preset dynamic threshold model, the target monitoring data, the data change slope, and the fluctuation coefficient, using a dual judgment mechanism of absolute threshold and trend threshold.

8. The system according to claim 7, characterized in that, The historical monitoring data includes historical first-level monitoring data, historical second-level monitoring data, historical third-level monitoring data, historical target monitoring data, and historical graded anomaly warning trigger information. The cloud platform includes: A multi-source data storage unit is used to classify and store the target monitoring data, the hierarchical anomaly warning trigger information and historical monitoring data using a distributed time-series database, and to build a data index based on the unique device identifier; The digital twin model construction unit is used to construct a digital twin model based on the equipment's three-dimensional model information, the target monitoring data, the historical monitoring data, the corrosion degree data, the remaining lifespan data, and the equipment wear characteristics, and output digital twin mapping data, which includes the equipment's full state correlation and equipment state inference results. The dynamic threshold training unit is used to train and update the preset dynamic threshold model based on the historical monitoring data, and to send the updated preset dynamic threshold model to the edge computing gateway.

9. The system according to claim 8, characterized in that, The explosion-proof determination module includes: The indicator system matching unit is used to match the corresponding explosion-proof safety assessment indicator system based on the equipment model information. The assessment indicator system is constructed based on the analytic hierarchy process and includes certificate and qualification compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators. The certificate and qualification compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators are respectively associated with the corresponding data items in the target monitoring data. The dynamic weight configuration unit is used to adjust the weight coefficients of certificate compliance indicators, environmental corrosion risk indicators, equipment wear and deterioration indicators, and early warning trigger level indicators based on fault case data in historical monitoring data, and obtain updated weight coefficients. The evidence fusion unit is used to apply evidence theory, based on the updated weight coefficients and the full-state correlation of the device, to fuse the target monitoring data and the hierarchical anomaly warning trigger information to generate a credibility distribution result. The safety level determination unit is used to map the confidence distribution result to the corresponding explosion-proof safety level based on the preset level determination rules and the state inference result, and output the safety level determination result.

10. A method for intelligent monitoring of explosion-proof safety of electrical equipment in a factory based on the Internet of Things, characterized in that, The method includes: Obtain explosion-proof qualification certificate data of electrical equipment, parse the explosion-proof qualification certificate data to obtain certificate information, perform dual-dimensional verification based on the certificate information to obtain equipment explosion-proof qualification status data, and output the certificate information and equipment explosion-proof qualification status data as first-level monitoring data; An intrinsically safe distributed sensor network is used to collect a set of environmental parameters. Based on a preset coupled simulation model and the set of environmental parameters, the corrosion degree data and remaining life data of the electrical equipment are determined, and the corrosion degree data and remaining life data are output as the second layer of monitoring data. The operating parameters of the electrical equipment are collected using a monitoring terminal. Based on the operating parameters, the loss characteristics and fault location information of the electrical equipment are determined, and the loss characteristics and fault location information are output as third-layer monitoring data. Based on the first-layer monitoring data, the second-layer monitoring data, and the third-layer monitoring data, target monitoring data and graded anomaly warning trigger information are determined. A digital twin model of the equipment is constructed based on the target monitoring data, the hierarchical anomaly early warning triggering information, and historical monitoring data; An explosion-proof safety status assessment is conducted based on the target monitoring data, the graded anomaly warning triggering information, the historical monitoring data, and the digital twin model to obtain the assessment results.

Images