A substation power fire fighting early warning method and system

Abstract

The present application relates to the technical field of power fire warning, and provides a substation power fire warning method and system, aiming at solving the problem of reduced warning accuracy caused by limited coverage of the large substation fire warning system. By acquiring substation operation data and position data, combining digital twin technology to construct a virtual image of the substation, and realizing global monitoring, the system dynamically plans the inspection path according to the information of the monitoring equipment and mobile inspection equipment, distinguishes between known and unknown monitoring ranges, generates a mobile inspection path for the unknown area, collects data and substitutes it into the warning model for analysis, and obtains the fire warning result. The present application improves the warning accuracy, effectively guarantees the safe and stable operation of the power system, and promotes the intelligent development of substation fire warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power fire warning, and in particular to a substation power fire warning method and system. BACKGROUND

[0002] Due to equipment aging, short circuit or improper operation, etc., power fire may occur in a substation. Substation fire fighting mainly relies on advanced fire alarm system and efficient fire extinguishing equipment, including gas fire extinguishing system and automatic water spray fire extinguishing system. When fire occurs, the fire source is extinguished by releasing fire extinguishing agent or spraying water, so as to prevent the spread of fire.

[0003] The existing substation power fire warning method mainly relies on a fire detection and alarm system, which includes a smoke detector, a temperature detector and a flame detector, which can monitor the smoke, temperature and flame in the substation in real time. When the detector detects an anomaly, the system will immediately trigger an alarm, and the fire information will be quickly conveyed to the on-duty personnel and the relevant emergency departments through sound and light signals, SMS notification or remote monitoring system.

[0004] In the actual application process of the substation power fire warning system, there are the following technical pain points: the coverage range of the large substation fire warning system is limited, the whole domain of the large substation cannot be monitored, the large substation fire warning system cannot obtain accurate data, and the accuracy of the large substation fire warning is reduced. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a substation power fire warning method and system, which solves the problem of the coverage range limitation of the large substation fire warning system, the whole domain of the large substation cannot be monitored, the large substation fire warning system cannot obtain accurate data, and the accuracy of the large substation fire warning is reduced.

[0006] To solve the above technical problems, the specific technical solutions of the present application are as follows:

[0007] In a first aspect, the present application provides a substation power fire warning method, comprising:

[0008] Step S101, obtaining substation operation data and substation position data, the substation operation data including monitoring standard data, equipment state data, video monitoring data and meteorological data, the monitoring standard data including monitoring equipment monitoring range information, smoke concentration standard, temperature standard and flame standard, the equipment state data including electrical equipment operation parameters, monitoring equipment information and fire extinguishing equipment state, and the meteorological data including weather data, temperature data, humidity data and wind speed data;

[0009] Step S102, substitute the substation location data into the preset digital twin model to obtain substation digital twin image data, perform feature recognition on the substation digital twin image data to obtain buildings and roads in the substation digital twin image data, call monitoring device information, the monitoring device information includes monitoring device model data and monitoring device location data, coordinate configure the monitoring device location data in the buildings and roads in the substation digital twin image data to obtain monitoring device coordinate information and power plant equipment coordinate information in the substation digital twin image data;

[0010] Step S103, collect monitoring device model information, match the monitoring device model information with the monitoring device coordinate information to obtain the corresponding monitoring device model on the monitoring device coordinate information, generate substation fire monitoring device monitoring range image data in the substation digital twin image based on the monitoring device model corresponding monitoring device monitoring range information and monitoring device coordinate information through the preset digital twin model, receive fire warning monitoring demand and mobile inspection device information, substitute the fire warning monitoring demand and mobile inspection device information into the preset inspection path generation model to obtain the mobile inspection device inspection path, match the mobile inspection device inspection path with the substation fire monitoring device monitoring range image data to obtain the monitoring device monitoring range coinciding with the mobile inspection device inspection path, take the monitoring device monitoring range coinciding with the mobile inspection device inspection path as the known monitoring range, and take the monitoring device monitoring range not coinciding with the mobile inspection device inspection path as the unknown monitoring range;

[0011] Step S104, obtain power plant equipment coordinate information in the unknown monitoring range in the substation digital twin image data, substitute the power plant equipment coordinate information in the unknown monitoring range into the preset inspection path generation model to generate a mobile inspection path of the unknown monitoring range, and send the mobile inspection path of the unknown monitoring range to the mobile inspection device;

[0012] Step S105, receive inspection data of the mobile inspection device in the unknown monitoring range, also receive monitoring data of the monitoring device in the known monitoring range, substitute the inspection data of the mobile inspection device in the unknown monitoring range and the monitoring data of the monitoring device in the known monitoring range into the preset substation power fire warning model in sequence to obtain a substation power fire warning result, and if the substation power fire warning result is a fire risk, generate fire warning information.

[0013] Further, the substation power fire warning method provided by the present application, the step S103, comprising:

[0014] The fire early warning monitoring requirement and the mobile inspection equipment information are taken as the model input information; the mobile inspection equipment information includes the mobile inspection equipment type, the mobile inspection equipment speed and the mobile inspection equipment endurance;

[0015] The optimal or suboptimal mobile inspection equipment inspection path is taken as the model output information;

[0016] The monitoring range of the monitoring equipment, the physical limitation of the mobile inspection equipment and the road traffic condition are taken as the constraint condition;

[0017] The objective function is received and set, and the objective function includes minimizing the total length of the inspection path and maximizing the monitoring coverage range;

[0018] The model input information and the constraint condition are input into the selected path planning algorithm, the ant colony algorithm is run, and the potential inspection path is generated;

[0019] The potential inspection path is evaluated according to the objective function, and the optimal path is obtained as the final inspection path.

[0020] Further, the substation power fire early warning method, the step S104, comprises:

[0021] The power plant equipment coordinate information in the unknown monitoring range is extracted from the substation digital twin image data, and the power plant equipment coordinate information includes the position of the power plant equipment and the type of the power plant equipment;

[0022] The power plant equipment coordinate information is taken as the input data and substituted into the inspection path generation model, and the inspection path generation model generates the optimal inspection path according to the input equipment coordinate, the inspection requirement of the equipment and the performance parameter of the mobile inspection equipment;

[0023] The optimal inspection path is sent to the mobile inspection equipment in the form of a GPS coordinate sequence and a map label.

[0024] Further, the substation power fire early warning method, the step S105, comprises:

[0025] The substation power fire early warning features are extracted from the substation operation data, and the substation power fire early warning features include the temperature feature, the smoke concentration feature, the flame detection feature and the equipment operation state feature;

[0026] The neural network model is trained using the substation power fire early warning features, and the substation power fire early warning model is obtained;

[0027] The data set of the substation operation data is divided into a training set, a verification set and a test set;

[0028] The substation power fire early warning model is trained using a training set, verified using a verification set, and compared with the verification set to obtain the verified substation power fire early warning model.

[0029] In a second aspect, the present application provides a substation power fire early warning system, which applies the substation power fire early warning method as described, comprising:

[0030] A data acquisition unit is configured to acquire substation operation data and substation location data, wherein the substation operation data comprises monitoring standard data, equipment state data, video monitoring data, and meteorological data, the monitoring standard data comprises monitoring equipment monitoring range information, smoke concentration standards, temperature standards, and flame standards, the equipment state data comprises electrical equipment operation parameters, monitoring equipment information, and fire extinguishing equipment state, and the meteorological data comprises weather data, temperature data, humidity data, and wind speed data.

[0031] A coordinate generation unit is configured to substitute the substation location data into a preset digital twin model to obtain substation digital twin image data, perform feature recognition on the substation digital twin image data to obtain buildings and roads in the substation digital twin image data, call monitoring equipment information comprising monitoring equipment model data and monitoring equipment location data, and perform coordinate configuration on the monitoring equipment location data and the buildings and roads in the substation digital twin image data to obtain monitoring equipment coordinate information and power plant equipment coordinate information in the substation digital twin image data.

[0032] A monitoring range determination unit is configured to acquire monitoring equipment model information, match the monitoring equipment model information with the monitoring equipment coordinate information to obtain corresponding monitoring equipment models on the monitoring equipment coordinate information, generate substation fire monitoring equipment monitoring range image data in the substation digital twin image based on the monitoring equipment model corresponding monitoring equipment monitoring range information and the monitoring equipment coordinate information through a preset digital twin model, receive fire early warning monitoring requirements and mobile inspection equipment information, substitute the fire early warning monitoring requirements and the mobile inspection equipment information into a preset inspection path generation model to obtain a mobile inspection equipment inspection path, match the mobile inspection equipment inspection path with the substation fire monitoring equipment monitoring range image data to obtain monitoring equipment monitoring ranges coinciding with the mobile inspection equipment inspection path, take the monitoring equipment monitoring ranges coinciding with the mobile inspection equipment inspection path as known monitoring ranges, and take the monitoring equipment monitoring ranges not coinciding with the mobile inspection equipment inspection path as unknown monitoring ranges.

[0033] a data analysis unit, configured to obtain power plant equipment coordinate information in an unknown monitoring range in substation digital twin image data, substitute the power plant equipment coordinate information in the unknown monitoring range into a preset inspection path generation model, generate a mobile inspection path in the unknown monitoring range, and send the mobile inspection path in the unknown monitoring range to a mobile inspection device;

[0034] a monitoring and early warning unit, configured to receive inspection data of the mobile inspection device in the unknown monitoring range, receive monitoring data of a monitoring device in a known monitoring range, substitute the inspection data of the mobile inspection device in the unknown monitoring range and the monitoring data of the monitoring device in the known monitoring range into a preset substation power fire early warning model in sequence, obtain a substation power fire early warning result, and generate fire early warning information if the substation power fire early warning result is a fire risk.

[0035] Further, the substation power fire early warning system, the monitoring range determination unit is further configured to:

[0036] the fire early warning monitoring requirement and the mobile inspection device information are used as model input information; the mobile inspection device information includes a mobile inspection device type, a mobile inspection device speed and a mobile inspection device endurance;

[0037] the optimal or suboptimal mobile inspection device inspection path is used as model output information;

[0038] the monitoring range of the monitoring device, the physical limit of the mobile inspection device and the road traffic condition are used as constraint conditions;

[0039] a target function is received and set, and the target function includes minimizing the total length of the inspection path and maximizing the monitoring coverage range;

[0040] the model input information and the constraint conditions are input into a selected path planning algorithm, an ant colony algorithm is run, and a potential inspection path is generated;

[0041] the potential inspection path is evaluated according to the target function, and the optimal path is obtained as the final inspection path.

[0042] Further, the substation power fire early warning system, the data analysis unit is further configured to:

[0043] the power plant equipment coordinate information in the unknown monitoring range is extracted from the substation digital twin image data, and the power plant equipment coordinate information includes a position of a power plant equipment and a type of the power plant equipment;

[0044] The power plant equipment coordinate information is taken as input data to substitute into the inspection path generation model, and the inspection path generation model generates the optimal inspection path according to the input equipment coordinates, the inspection requirements of the equipment, and the performance parameters of the mobile inspection equipment;

[0045] The optimal inspection path is sent to the mobile inspection equipment in the form of a GPS coordinate sequence and a map label.

[0046] Further, the substation power fire fire-fighting early warning system, the monitoring and early warning unit is further used for:

[0047] The substation power fire early warning features are extracted from the substation operation data, including temperature features, smoke concentration features, flame detection features and equipment operation state features;

[0048] The substation power fire early warning model is obtained by training the neural network model using the substation power fire early warning features;

[0049] The data set of the substation operation data is divided into a training set, a validation set and a test set;

[0050] The substation power fire early warning model is trained using the training set, and the substation power fire early warning model is verified using the validation set, the prediction result of the substation power fire early warning model is compared with the validation set, and the verified substation power fire early warning model is obtained.

[0051] The beneficial effects of the present application are:

[0052] The present application constructs a virtual image of the substation through digital twin technology, realizes accurate monitoring of the whole substation, solves the problem of limited coverage of traditional monitoring means, and improves the accuracy of fire warning. Comprehensive utilization of data of fixed monitoring equipment and mobile inspection equipment for all-round, multi-level monitoring of the substation reduces the monitoring blind area and further improves the reliability of the warning.

[0053] The present application can dynamically generate the optimal or suboptimal inspection path according to the fire warning monitoring requirements and the performance parameters of the mobile inspection equipment, improve the flexibility and adaptability of the system. Through intelligent path planning and dynamic adjustment mechanism, the system can respond to the changes of the substation environment and equipment state in real time, improve the timeliness and effectiveness of monitoring. Through intelligent path planning, the invalid inspection and repeated inspection of the mobile inspection equipment are reduced, and the inspection efficiency is improved. Comprehensive data analysis and application of the warning model enable the system to quickly identify fire risks and generate warning information, shorten the response time and improve the efficiency of emergency disposal.

[0054] The substation power fire fighting early warning method and system provided by the application can monitor the operation state of the substation in real time, discover and handle potential fire risks in time, effectively prevent the occurrence of fire accidents, and ensure the safe and stable operation of the power system. Through early warning and rapid response, the damage of the fire to the substation equipment and the influence on power supply are reduced, and the economic loss and social influence are reduced.

[0055] The implementation of the application promotes the intelligent development of the substation fire fighting early warning system, improves the automation level and intelligent degree of the system. Through the application of digital twin technology, neural network model and other advanced technologies, a new idea and direction are provided for the technical innovation and development of the substation fire fighting early warning field.

[0056] In summary, the application provides a new solution for substation power fire fighting early warning through improving the early warning accuracy, ensuring the safe and stable operation of the power system, and promoting the intelligent development. BRIEF DESCRIPTION OF DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the application, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor on the basis of the drawings.

[0058] Figure 1 The substation power fire fighting early warning method flowchart provided by the embodiment of the application. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical scheme and advantages of the application clearer, the technical scheme of the application will be described clearly and completely in combination with the specific embodiments of the application and corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application. The technical scheme provided by each embodiment of the application will be described in detail below in combination with the drawings.

[0060] In order to better understand the purpose of the application, the application will be further described in detail as follows.

[0061] In a first aspect, the application provides a substation power fire fighting early warning method, comprising:

[0062] Step S101, obtain the substation operation data and the substation position data, the substation operation data includes monitoring standard data, equipment state data, video monitoring data and meteorological data, the monitoring standard data includes monitoring equipment monitoring range information, smoke concentration standard, temperature standard and flame standard, the equipment state data includes electrical equipment operation parameter, monitoring equipment information and fire extinguishing equipment state, and the meteorological data includes weather data, temperature data, humidity data and wind speed data;

[0063] The substation operation data: the monitoring standard data includes the monitoring range information of the monitoring equipment, the smoke concentration standard, the temperature standard and the flame standard, defines the benchmark value or range for the system to judge the fire risk, and is an important basis for subsequent analysis and early warning.

[0064] The equipment state data covers the operation parameters of electrical equipment, monitoring equipment information (such as model, location, etc.) and fire extinguishing equipment state. These data reflect the real-time condition of the equipment in the substation, which is of great significance for timely discovery of abnormalities and potential risks.

[0065] Video monitoring data: data obtained through the video monitoring system, which can intuitively understand the situation in the substation and provide visual support for fire warning.

[0066] Meteorological data includes weather data, temperature data, humidity data and wind speed data, etc. Meteorological conditions have important influence on the occurrence and development of fire, so these data are also one of the factors that must be considered by the early warning system.

[0067] Substation position data: used to locate the position of the substation in the digital twin model, so as to construct the digital twin image of the substation. By combining the substation position data and operation data, the system can more accurately simulate and analyze the actual condition of the substation.

[0068] Step S101 is the data entry of the early warning system, which provides a solid data foundation for subsequent analysis, processing and early warning by comprehensively collecting substation operation and position data. By obtaining detailed monitoring standard data, equipment state data, video monitoring data and meteorological data, the system can more accurately judge the fire risk and improve the accuracy and reliability of early warning. Rich data resources provide more analysis dimensions and decision-making basis for the system, so that the system can make more intelligent and reasonable early warning decisions.

[0069] Step S102, substitute the substation location data into the preset digital twin model to obtain the substation digital twin image data, perform feature recognition on the substation digital twin image data, obtain the buildings and roads in the substation digital twin image data, call the monitoring equipment information, the monitoring equipment information includes monitoring equipment model data and monitoring equipment location data, coordinate the monitoring equipment location data with the buildings and roads in the substation digital twin image data to obtain the monitoring equipment coordinate information and the power plant equipment coordinate information in the substation digital twin image data;

[0070] Step S102 maps the actual physical environment of the substation into the digital world and processes and analyzes it through digital twin technology. Here is a detailed analysis of this step:

[0071] The system substitutes the substation location data (such as latitude and longitude coordinates, geographic location information, etc.) into the preset digital twin model. This step is the basis for building the substation digital twin image, making the digital twin image consistent with the actual geographic location of the substation. Through digital twin technology, the system generates digital twin image data of the substation according to the substation location data and the preset model. The image data is a virtual representation of the substation, containing information such as buildings, roads, and equipment layout within the substation.

[0072] In the generated digital twin image data, the system performs feature recognition to distinguish and extract the buildings and roads within the substation, which helps the system more accurately understand the layout and structure of the substation and provides a basis for subsequent device coordinate configuration.

[0073] The system calls relevant information of the monitoring equipment from the database, including the model data and location data of the monitoring equipment.

[0074] Coordinate the location data of the monitoring equipment with the buildings and roads in the substation digital twin image data. Specifically, the system will find the corresponding coordinate point in the digital twin image according to the actual location of the monitoring equipment, and associate the monitoring equipment with the coordinate point. At the same time, the system will also configure the coordinate information of the power plant equipment within the substation to facilitate subsequent analysis and early warning.

[0075] Through digital twin technology, the system can build a digital image covering the entire substation, solving the problem of limited coverage of traditional monitoring methods. Accurate device coordinate configuration enables the system to more accurately locate fire risk points, thereby improving the accuracy and reliability of early warning. The digital twin image provides the system with an intuitive and comprehensive view of the substation, which helps the system make more intelligent and reasonable early warning decisions.

[0076] Step S103, collect the model information of the monitoring device, match the model information of the monitoring device with the coordinate information of the monitoring device, obtain the corresponding monitoring device model on the monitoring device coordinate information, generate the substation fire monitoring device monitoring range image data in the substation digital twin image based on the monitoring device model corresponding monitoring device monitoring range information and monitoring device coordinate information through the preset digital twin model, receive the fire warning monitoring demand and mobile inspection device information, substitute the fire warning monitoring demand and mobile inspection device information into the preset inspection path generation model to obtain the mobile inspection device inspection path, match the mobile inspection device inspection path with the substation fire monitoring device monitoring range image data, obtain the monitoring range coinciding with the mobile inspection path, take the monitoring range coinciding with the mobile inspection path as the known monitoring range, and take the monitoring range not coinciding with the mobile inspection path as the unknown monitoring range.

[0077] First, collect the model information of all monitoring devices in the substation. The model information of the monitoring device is a key basis for determining the monitoring range of the monitoring device.

[0078] Match the collected monitoring device model information with the monitoring device coordinate information obtained in step S102. Through this step, the system can determine the specific position of each monitoring device in the substation digital twin image.

[0079] Based on the monitoring range information (such as monitoring radius, monitoring angle, etc.) corresponding to the monitoring device model and the coordinate information of the monitoring device, the system generates the monitoring range image data of the substation fire monitoring device in the substation digital twin image through the preset digital twin model. These data intuitively show the monitoring area that each monitoring device can cover.

[0080] The system receives the specific requirements of fire warning monitoring, including the priority of monitoring, the key areas of concern, etc. At the same time, the system also receives the relevant information of the mobile inspection device, such as the type of the device, the speed, the endurance, etc.

[0081] Substitute the fire warning monitoring demand and mobile inspection device information into the preset inspection path generation model. This model generates the optimal or suboptimal mobile inspection device inspection path according to the monitoring demand, device performance, and substation layout, etc. Match the generated mobile inspection device inspection path with the substation fire monitoring device monitoring range image data. Through this step, the system can determine which monitoring range of the monitoring device coincides with the inspection path (i.e. known monitoring range) and which monitoring range of the monitoring device does not coincide with the inspection path (i.e. unknown monitoring range).

[0082] By determining the known and unknown monitoring ranges, the system can more reasonably allocate monitoring resources, focusing on key areas for sufficient monitoring. The planned mobile inspection path takes into account monitoring needs and equipment performance, helping to improve the efficiency and accuracy of inspection. The identification and subsequent inspection of unknown monitoring ranges fill the monitoring blind area and enhance the overall early warning capability of the system.

[0083] In step S104, the coordinates of the power plant equipment in the unknown monitoring range are obtained from the digital twin image data of the substation, and the coordinates of the power plant equipment in the unknown monitoring range are substituted into the preset inspection path generation model to generate a mobile inspection path for the unknown monitoring range, and the mobile inspection path for the unknown monitoring range is sent to the mobile inspection device;

[0084] In the digital twin image data of the substation, the system first identifies the areas that are not covered by the known monitoring range (i.e. the areas covered by the fixed monitoring equipment and the previously planned mobile inspection path), i.e. the unknown monitoring range. The system extracts the coordinates of the power plant equipment from these unknown monitoring ranges. These coordinate information includes the specific location of the power plant equipment in the substation and the type of equipment and other information.

[0085] The extracted coordinates of the power plant equipment in the unknown monitoring range are used as input data and substituted into the preset inspection path generation model. The model takes into account the inspection needs of the equipment, the performance parameters of the mobile inspection device (such as speed, endurance, turning radius, etc.) and the road traffic conditions in the substation, etc.

[0086] The inspection path generation model calculates and generates the optimal or suboptimal mobile inspection path according to the input data using advanced algorithms such as ant colony algorithm, genetic algorithm, etc. These paths aim to enable the mobile inspection device to efficiently cover all power plant equipment in the unknown monitoring range while minimizing inspection time and energy consumption.

[0087] When the mobile inspection path for the unknown monitoring range is generated, the system sends the path information (usually in the form of GPS coordinate sequence and map annotation) to the corresponding mobile inspection device. After receiving the path information, the mobile inspection device can perform the inspection operation according to the planned path. Through the processing of the unknown monitoring range and the generation of the mobile inspection path, the system can effectively monitor all electrical equipment in the substation, thereby filling the monitoring blind area and improving the coverage of the early warning system.

[0088] The dynamically generated mobile inspection path takes into account a variety of factors, improving the efficiency and relevance of the inspection operation. This helps to reduce unnecessary inspection time and waste of resources. The implementation of step S104 enables the system to dynamically adjust the inspection plan according to the actual situation and changes in the substation, thereby enhancing the flexibility and adaptability of the system.

[0089] In summary, step S104 further improves the coverage of the early warning system through processing of unknown monitoring ranges and dynamic generation of mobile inspection paths.

[0090] In step S105, the inspection data of the mobile inspection device in the unknown monitoring range and the monitoring data of the monitoring device in the known monitoring range are received, and the inspection data of the mobile inspection device in the unknown monitoring range and the monitoring data of the monitoring device in the known monitoring range are sequentially substituted into the preset substation power fire early warning model to obtain a substation power fire early warning result. If the substation power fire early warning result is a fire risk, fire warning information is generated.

[0091] The system receives the inspection data collected by the mobile inspection device in the unknown monitoring range, including video, image, temperature, smoke concentration and other types of information, for evaluating the fire risk of the area. At the same time, the system also receives real-time monitoring data provided by the fixed monitoring device in the known monitoring range. These data help to understand the overall operation of the substation, especially the parameters related to fire.

[0092] The inspection data of the mobile inspection device in the unknown monitoring range and the monitoring data of the monitoring device in the known monitoring range are sequentially substituted into the preset substation power fire early warning model. The substation power fire early warning model is an intelligent system based on advanced algorithms such as machine learning or deep learning, which can automatically analyze data, identify fire characteristics and predict fire risks.

[0093] By running the early warning model, the system processes and analyzes the input data, and finally obtains the substation power fire early warning result, which is a probability value or risk level indicating the nature or severity of the fire in the current substation.

[0094] If the substation power fire early warning result indicates a fire risk (i.e. the probability value exceeds the preset threshold or the risk level reaches the alarm level), the system will automatically generate fire warning information. These information includes the specific location of the fire, the risk level, the recommended response measures and other contents.

[0095] The generated fire warning information will be timely delivered to the on-duty personnel, relevant emergency departments and affected areas. At the same time, the system will also trigger a series of automatic response measures, such as starting fire extinguishing equipment, cutting off power supply, etc., to minimize the loss and impact caused by the fire.

[0096] Through real-time processing and analysis of inspection and monitoring data, the system can accurately assess the fire risk of the substation, providing strong support for timely response measures. When detecting fire risk, the system can quickly generate warning information and start the response mechanism, effectively controlling the spread of fire and reducing losses. By implementing the substation power fire fighting and warning method provided by the present application, the safe operation level of the substation can be significantly improved, and the safety of personnel and equipment can be ensured.

[0097] The specific steps of the substation virtual image are mainly realized by digital twinning technology in the present application, and the specific steps are as follows:

[0098] Obtain the location data of the substation, including latitude and longitude, geographic location information, etc.

[0099] Obtain the operation data of the substation, including monitoring standard data, equipment state data, video monitoring data, and meteorological data, etc.

[0100] Substitute the location data of the substation into the preset digital twinning model, and the digital twinning model generates a preliminary virtual structure of the substation according to the location data. Feature recognition is performed on the preliminary generated virtual structure to distinguish key elements such as buildings and roads. According to the recognition result, the virtual image is further refined to make it closer to the layout and structure of the real substation.

[0101] Retrieve the information of the monitoring equipment, including the model and location data of the monitoring equipment. Coordinate the location data of the monitoring equipment with the buildings and roads in the virtual image, so that each monitoring equipment has an accurate position representation in the virtual image.

[0102] After the above steps, the substation digital twinning image data containing the entire substation, all key equipment and facilities are generated. The virtual image not only contains the physical layout of the substation, but also integrates the information of the monitoring equipment, providing an intuitive and comprehensive view for subsequent analysis and warning. Through these steps, the virtual image of the substation is successfully constructed, realizing accurate monitoring and visual display of the entire substation, and providing strong support for subsequent fire warning and emergency handling.

[0103] Specifically, the substation power fire fighting and warning method provided by the present application comprises the following steps:

[0104] The fire warning monitoring requirements and mobile inspection equipment information are used as model input information; the mobile inspection equipment information includes mobile inspection equipment type, mobile inspection equipment speed and mobile inspection equipment endurance;

[0105] The optimal or suboptimal mobile inspection equipment inspection path is used as the model output information;

[0106] The monitoring range of the monitoring device, the physical limitations of the mobile inspection device, and the road traffic conditions are taken as constraint conditions.

[0107] The objective function is received and set, which includes minimizing the total length of the inspection path and maximizing the monitoring coverage range.

[0108] The model input information and the constraint conditions are input into the selected path planning algorithm, the ant colony algorithm is run, and the potential inspection path is generated.

[0109] The potential inspection path is evaluated according to the objective function, and the optimal path is obtained as the final inspection path.

[0110] The input information includes: the fire warning monitoring requirements include the fire warning requirements for specific areas or devices in the substation;

[0111] The mobile inspection device information includes the type, speed, and endurance of the mobile inspection device. Different types of devices have different application ranges and performance characteristics, while speed and endurance directly affect the planning and execution efficiency of the inspection path.

[0112] The output information includes: the optimal or suboptimal mobile inspection device inspection path: this is the core output of step S103, which aims to provide one or more inspection paths for the mobile inspection device that can efficiently cover the monitoring range and meet the fire warning requirements;

[0113] The constraint conditions include:

[0114] The monitoring range of the monitoring device limits the key areas that the mobile inspection device needs to focus on, so that the inspection path can cover the area.

[0115] The physical limitations of the mobile inspection device: such as the turning radius, climbing ability, etc. of the device, which will affect the feasibility of the inspection path.

[0116] Road traffic conditions: factors such as road conditions and obstacles in the substation also need to be considered to ensure smooth execution of the inspection path.

[0117] The objective function includes:

[0118] Minimizing the total length of the inspection path: this helps to reduce the energy consumption and inspection time of the mobile inspection device, and improves the inspection efficiency.

[0119] Maximizing the monitoring coverage range: so that the inspection path can cover as many key areas and devices as possible, improving the accuracy and timeliness of fire warning.

[0120] The input information and constraint conditions are integrated, and the fire warning monitoring requirements, mobile inspection device information, and constraint conditions are integrated together as the input of the path planning algorithm.

[0121] Run the ant colony algorithm, select the ant colony algorithm as the path planning algorithm, the algorithm can gradually find the optimal or suboptimal path from the starting point to the ending point through simulating the pheromone accumulation and release mechanism in the foraging process of ants.

[0122] Generate potential inspection paths: the ant colony algorithm generates multiple potential inspection paths during the iteration process, and these paths are evaluated according to the objective function.

[0123] Path evaluation and selection: the potential inspection paths are evaluated according to the objective function (minimizing the total length of the inspection path, maximizing the monitoring coverage), and the optimal or suboptimal path is selected as the final inspection path.

[0124] Specifically, the substation power fire fighting and early warning method comprises the following steps:

[0125] Extract the power plant equipment coordinate information in the unknown monitoring range from the substation digital twin image data, which includes the location of the power plant equipment and the type of the power plant equipment;

[0126] Put the power plant equipment coordinate information as input data into the inspection path generation model, and the inspection path generation model generates the optimal inspection path according to the input equipment coordinates, equipment inspection requirements, and performance parameters of the mobile inspection equipment;

[0127] Send the optimal inspection path to the mobile inspection equipment in the form of GPS coordinate sequence and map annotation.

[0128] In the substation power fire fighting and early warning method, step S104 focuses on processing the unknown monitoring range and generating the corresponding inspection path for the mobile inspection equipment. The following is a detailed analysis of this step:

[0129] Extract information from the substation digital twin image data. The digital twin image data is a virtual copy of the physical environment of the substation, containing the precise location and type information of all key equipment and facilities in the substation. Extract the coordinate information of the power plant equipment in the unknown monitoring range. These coordinate information not only includes the location of the equipment (such as latitude and longitude or relative coordinates), but also includes the type of the equipment (such as transformers, switch devices, etc.), so that subsequent inspection strategies can be developed according to the type of the equipment.

[0130] The extracted power plant equipment coordinate information is taken as input data, while considering the inspection requirements of the equipment (such as inspection frequency, inspection focus, etc.) and the performance parameters of the mobile inspection equipment (such as speed, endurance, turning radius, etc.). The inspection path generation model automatically generates the optimal inspection path based on these input data, combined with the preset algorithm and logic. The path aims to enable the mobile inspection equipment to efficiently and comprehensively cover all key equipment within the unknown monitoring range.

[0131] The generated optimal inspection path is represented in the form of a GPS coordinate sequence and map annotations. The GPS coordinate sequence provides specific path points that the mobile inspection equipment needs to follow, while the map annotations help inspection personnel intuitively understand the path and target location. The optimal inspection path is sent to the mobile inspection equipment through wireless communication or other means. In this way, the mobile inspection equipment can perform automatic or manual assisted inspection work according to the received path.

[0132] Through step S104, the present application can automatically generate a reasonable inspection path for the unknown monitoring range in the substation, enabling the mobile inspection equipment to efficiently and accurately complete the inspection task, thereby timely discovering and handling potential fire risks.

[0133] Specifically, the substation power fire prevention and warning method of the present application, step S105, includes:

[0134] Extracting substation power fire warning features from substation operation data, including temperature features, smoke concentration features, flame detection features, and equipment operation state features;

[0135] Training the neural network model using the substation power fire warning features to obtain a substation power fire warning model;

[0136] Dividing the data set of substation operation data into a training set, a validation set, and a test set;

[0137] Training the substation power fire warning model using the training set, validating the substation power fire warning model using the validation set, comparing the prediction results of the substation power fire warning model with the validation set, and obtaining the validated substation power fire warning model.

[0138] Extract relevant information from substation operation data. These data include various parameters monitored by monitoring equipment in real time, such as temperature, smoke concentration, flame state, and electrical equipment operation state, etc. The power fire warning related features include temperature features (such as abnormal high temperature area), smoke concentration features (such as sudden increase in smoke concentration), flame detection features (such as appearance of flame), and equipment operation state features (such as abnormal operation parameters of equipment).

[0139] A neural network model is used as the early warning model. The neural network model performs well in complex system prediction and classification tasks due to its powerful non-linear mapping ability and self-learning ability. The extracted substation power fire early warning features are used as input to train the neural network model. During the training process, the model will continuously adjust its internal parameters to minimize the difference between the predicted results and the actual results.

[0140] The data set of substation operation data is divided into training set, validation set and test set. The training set is used to train the model, the validation set is used to verify the performance of the model during the training process, and the test set is used to finally evaluate the generalization ability of the model. The trained substation power fire early warning model is verified using the validation set. The predicted results of the model are compared with the actual results in the validation set to evaluate the accuracy, robustness and generalization ability of the model.

[0141] According to the verification results, the model is adjusted and optimized as necessary to improve its early warning performance, including adjusting the structure, parameters or training strategy of the model. After sufficient training and verification, the verified substation power fire early warning model is obtained. The model can accurately and reliably predict potential power fire risks in substations, providing strong support for fire prevention early warning.

[0142] Through step S105, the substation power fire early warning model based on neural network is successfully constructed, and its early warning performance is improved through strict verification process. The application of substation power fire early warning model will significantly improve the power fire early warning ability of substation, reduce the occurrence of fire accident, and ensure the safe and stable operation of power system.

[0143] In the second aspect, the present application provides a substation power fire early warning system, which applies the substation power fire early warning method as described, comprising:

[0144] A data acquisition unit is used to acquire substation operation data and substation location data. The substation operation data includes monitoring standard data, equipment state data, video monitoring data and meteorological data. The monitoring standard data includes monitoring equipment monitoring range information, smoke concentration standard, temperature standard and flame standard. The equipment state data includes electrical equipment operation parameters, monitoring equipment information and fire extinguishing equipment state. The meteorological data includes weather data, temperature data, humidity data and wind speed data.

[0145] The coordinate generation unit is used for substituting the substation position data into the preset digital twin model to obtain substation digital twin image data, performing feature recognition on the substation digital twin image data to obtain buildings and roads in the substation digital twin image data, calling monitoring device information, the monitoring device information including monitoring device model data and monitoring device position data, performing coordinate configuration on the monitoring device position data and the buildings and roads in the substation digital twin image data to obtain monitoring device coordinate information and power plant device coordinate information in the substation digital twin image data;

[0146] The monitoring range determination unit is used for collecting monitoring device model information, matching the monitoring device model information with the monitoring device coordinate information to obtain corresponding monitoring device models on the monitoring device coordinate information, generating substation fire monitoring device monitoring range image data in the substation digital twin image based on the monitoring device model corresponding monitoring device monitoring range information and the monitoring device coordinate information through the preset digital twin model, receiving fire warning monitoring requirements and mobile inspection device information, substituting the fire warning monitoring requirements and the mobile inspection device information into the preset inspection path generation model to obtain a mobile inspection device inspection path, matching the mobile inspection device inspection path with the substation fire monitoring device monitoring range image data to obtain monitoring device monitoring ranges coinciding with the mobile inspection device inspection path, taking the monitoring device monitoring ranges coinciding with the mobile inspection device inspection path as known monitoring ranges, and taking the monitoring device monitoring ranges not coinciding with the mobile inspection device inspection path as unknown monitoring ranges.

[0147] The data analysis unit is used for obtaining power plant device coordinate information in the unknown monitoring range in the substation digital twin image data, substituting the power plant device coordinate information in the unknown monitoring range into the preset inspection path generation model to generate a mobile inspection path of the unknown monitoring range, and sending the mobile inspection path of the unknown monitoring range to the mobile inspection device.

[0148] The monitoring and warning unit is used for receiving inspection data of the mobile inspection device in the unknown monitoring range and receiving monitoring data of the monitoring device in the known monitoring range, substituting the inspection data of the mobile inspection device in the unknown monitoring range and the monitoring data of the monitoring device in the known monitoring range into the preset substation power fire warning model in sequence to obtain a substation power fire warning result, and generating fire warning information if the substation power fire warning result is a fire risk.

[0149] Specifically, the monitoring range determination unit of the substation power fire fighting and warning system is further used for:

[0150] The fire early warning monitoring requirement and the mobile inspection equipment information are taken as model input information; the mobile inspection equipment information includes mobile inspection equipment type, mobile inspection equipment speed and mobile inspection equipment endurance;

[0151] The optimal or suboptimal mobile inspection equipment inspection path is taken as model output information;

[0152] The monitoring range of the monitoring equipment, the physical limitation of the mobile inspection equipment and the road traffic condition are taken as constraint conditions;

[0153] The objective function is received and set, and the objective function includes minimizing the total length of the inspection path and maximizing the monitoring coverage range;

[0154] The model input information and the constraint conditions are input into the selected path planning algorithm, the ant colony algorithm is run, and the potential inspection path is generated;

[0155] The potential inspection path is evaluated according to the objective function, and the optimal path is obtained as the final inspection path.

[0156] Specifically, the power substation fire early warning system provided by the application, the data analysis unit is also used for:

[0157] The power plant equipment coordinate information in the unknown monitoring range is extracted from the power substation digital twin image data, and the power plant equipment coordinate information includes the position of the power plant equipment and the type of the power plant equipment;

[0158] The power plant equipment coordinate information is taken as input data and substituted into the inspection path generation model, and the inspection path generation model generates the optimal inspection path according to the input equipment coordinate, the inspection requirement of the equipment and the performance parameter of the mobile inspection equipment;

[0159] The optimal inspection path is sent to the mobile inspection equipment in the form of a GPS coordinate sequence and a map label.

[0160] Specifically, the power substation fire early warning system provided by the application, the monitoring and early warning unit is also used for:

[0161] The power substation fire early warning features are extracted from the power substation operation data, and the power substation fire early warning features include temperature features, smoke concentration features, flame detection features and equipment operation state features;

[0162] The neural network model is trained using the power substation fire early warning features, and the power substation fire early warning model is obtained;

[0163] The data set of the power substation operation data is divided into a training set, a validation set and a test set;

[0164] The substation power fire early warning model is trained using the training set, and the substation power fire early warning model is verified using the verification set. The prediction result of the substation power fire early warning model is compared and verified with the verification set, and the verified substation power fire early warning model is obtained.

[0165] The technical scheme of the present application effectively solves the problems of coverage range limitation of the large substation fire warning system, inability to monitor the whole substation, and low warning accuracy through the following key steps and mechanisms:

[0166] The substation location data is substituted into the preset digital twin model to generate substation digital twin image data. This step constructs a virtual mirror of the substation through digital twin technology, so that the entire substation and its internal equipment, roads, buildings, etc. can be completely presented in a virtual environment. Digital twin technology enables the system to comprehensively cover the whole substation, without being limited by the layout and number of physical monitoring devices, thereby solving the problem of insufficient coverage.

[0167] By matching the monitoring device model information and coordinate information, substation fire monitoring device monitoring range image data is generated, and is matched with the mobile inspection device inspection path to distinguish known monitoring ranges and unknown monitoring ranges. For unknown monitoring ranges, a special mobile inspection path is generated and sent to the mobile inspection device, so that the area can also be effectively monitored. Through the organic combination of fixed monitoring devices and mobile inspection devices, the system can realize flexible monitoring of the whole substation, ensuring the comprehensiveness of monitoring and improving the efficiency and accuracy of monitoring.

[0168] In steps S103 and S104, the optimal or suboptimal mobile inspection path is generated by combining the fire warning monitoring demand, mobile inspection device information and monitoring device monitoring range through the preset inspection path generation model. With the change of the substation environment and device state, the system can dynamically adjust the inspection path to improve the timeliness and effectiveness of monitoring. Intelligent path planning improves the work efficiency of the mobile inspection device, and the dynamic adjustment mechanism enables the system to adapt to the complex and variable substation environment, further improving the accuracy of the warning.

[0169] The data of the mobile inspection device and the monitoring device is received, substituted into the preset substation power fire early warning model for analysis, and fire warning information is generated. By collecting substation operation data, meteorological data, etc., extracting fire warning features, training and optimizing the neural network model, a high-accuracy warning model is obtained. Comprehensive data analysis and advanced warning model improve the accuracy and reliability of fire warning, so that the system can timely and effectively discover potential fire risks.

[0170] To sum up, the technical scheme of the present application effectively solves the coverage range limitation problem of the fire warning system of large-scale substations by means of digital twinning technology, cooperation of monitoring equipment and mobile inspection equipment, intelligent path planning and dynamic adjustment, and comprehensive data analysis and early warning model, etc., and improves the accuracy and reliability of the early warning.

Claims

1. A substation electric power fire fighting pre-warning method, characterized in that, The method comprises the following steps: Step S101, obtaining substation operation data and substation position data, the substation operation data comprising monitoring standard data, equipment state data, video monitoring data and meteorological data, the monitoring standard data comprising monitoring equipment monitoring range information, smoke concentration standard, temperature standard and flame standard, the equipment state data comprising electrical equipment operation parameters, monitoring equipment information and fire extinguishing equipment state, and the meteorological data comprising weather data, temperature data, humidity data and wind speed data; Step S102, substituting the substation position data into a preset digital twin model to obtain substation digital twin image data, performing feature recognition on the substation digital twin image data to obtain buildings and roads in the substation digital twin image data, calling monitoring equipment information comprising monitoring equipment model data and monitoring equipment position data, and performing coordinate configuration on the monitoring equipment position data and the buildings and roads in the substation digital twin image data to obtain monitoring equipment coordinate information and power plant equipment coordinate information in the substation digital twin image data; Step S103, collecting monitoring equipment model information, matching the monitoring equipment model information with the monitoring equipment coordinate information to obtain corresponding monitoring equipment models on the monitoring equipment coordinate information, generating substation fire monitoring equipment monitoring range image data in the substation digital twin image based on the monitoring equipment model corresponding monitoring equipment monitoring range information and the monitoring equipment coordinate information through a preset digital twin model, receiving fire warning monitoring requirements and mobile inspection equipment information, substituting the fire warning monitoring requirements and the mobile inspection equipment information into a preset inspection path generation model to obtain a mobile inspection equipment inspection path, matching the mobile inspection equipment inspection path with the substation fire monitoring equipment monitoring range image data to obtain monitoring equipment monitoring ranges coinciding with the mobile inspection path, taking the monitoring equipment monitoring ranges coinciding with the mobile inspection path as known monitoring ranges, and taking the monitoring equipment monitoring ranges not coinciding with the mobile inspection path as unknown monitoring ranges; Step S104, obtaining power plant equipment coordinate information in the unknown monitoring ranges in the substation digital twin image data, substituting the power plant equipment coordinate information in the unknown monitoring ranges into the preset inspection path generation model to generate a mobile inspection path of the unknown monitoring ranges, and sending the mobile inspection path of the unknown monitoring ranges to the mobile inspection equipment; Step S105, receiving inspection data of the mobile inspection equipment in the unknown monitoring ranges, and also receiving monitoring data of the monitoring equipment in the known monitoring ranges, substituting the inspection data of the mobile inspection equipment in the unknown monitoring ranges and the monitoring data of the monitoring equipment in the known monitoring ranges into a preset substation power fire warning model in sequence to obtain a substation power fire warning result, and generating fire warning information if the substation power fire warning result indicates a fire risk.

2. The substation electric fire fighting pre-alarm method according to claim 1, characterized in that, The step S103 comprises: The fire warning monitoring demand and the mobile inspection equipment information are taken as the model input information; the mobile inspection equipment information includes the mobile inspection equipment type, the mobile inspection equipment speed and the mobile inspection equipment endurance; The optimal or suboptimal mobile inspection equipment inspection path is taken as the model output information; The monitoring range of the monitoring equipment, the physical limitation of the mobile inspection equipment and the road traffic condition are taken as the constraint conditions; The target function is received and set, and the target function includes minimizing the total length of the inspection path and maximizing the monitoring coverage range; The model input information and the constraint conditions are input into the selected path planning algorithm, the ant colony algorithm is run, and the potential inspection path is generated; The optimal path is obtained as the final inspection path according to the target function evaluation of the potential inspection path.

3. The substation electric fire fighting pre-alarm method according to claim 1, characterized in that, The step S104 comprises: The power plant equipment coordinate information in the unknown monitoring range is extracted from the digital twin image data of the transformer substation, and the power plant equipment coordinate information includes the position of the power plant equipment and the type of the power plant equipment; The power plant equipment coordinate information is taken as the input data and substituted into the inspection path generation model, and the inspection path generation model generates the optimal inspection path according to the input equipment coordinate, the inspection demand of the equipment and the performance parameter of the mobile inspection equipment; The optimal inspection path is sent to the mobile inspection equipment in the form of a GPS coordinate sequence and a map label.

4. The substation electric fire fighting pre-alarm method of claim 1, wherein, The step S105 comprises: The transformer substation power fire warning features are extracted from the transformer substation operation data, and the transformer substation power fire warning features include the temperature feature, the smoke concentration feature, the flame detection feature and the equipment operation state feature; The transformer substation power fire warning model is obtained by training the neural network model using the transformer substation power fire warning features; The data set of the transformer substation operation data is divided into a training set, a verification set and a test set; The transformer substation power fire warning model is trained using the training set, and the transformer substation power fire warning model is verified using the verification set; the prediction result of the transformer substation power fire warning model is compared with the verification set to obtain the verified transformer substation power fire warning model.

5. A substation electric power fire fighting early warning system applying the substation electric power fire fighting early warning method as claimed in any one of claims 1 to 4, characterized by, Comprise: The data acquisition unit is used for acquiring the transformer substation operation data and the transformer substation position data, the transformer substation operation data includes the monitoring standard data, the equipment state data, the video monitoring data and the meteorological data, the monitoring standard data includes the monitoring equipment monitoring range information, the smoke concentration standard, the temperature standard and the flame standard, the equipment state data includes the electrical equipment operation parameter, the monitoring equipment information and the fire extinguishing equipment state, and the meteorological data includes the weather data, the temperature data, the humidity data and the wind speed data; The coordinate generation unit is configured to substitute the substation position data into the preset digital twin model to obtain substation digital twin image data, perform feature recognition on the substation digital twin image data to obtain buildings and roads in the substation digital twin image data, call monitoring device information including monitoring device model data and monitoring device position data, and perform coordinate configuration on the monitoring device position data and the buildings and roads in the substation digital twin image data to obtain monitoring device coordinate information and power plant device coordinate information in the substation digital twin image data; The monitoring range determination unit is configured to collect monitoring device model information, match the monitoring device model information with the monitoring device coordinate information to obtain corresponding monitoring device models on the monitoring device coordinate information, generate substation fire monitoring device monitoring range image data in the substation digital twin image based on monitoring device monitoring range information corresponding to the monitoring device models and the monitoring device coordinate information through the preset digital twin model, receive fire warning monitoring requirements and mobile inspection device information, substitute the fire warning monitoring requirements and the mobile inspection device information into a preset inspection path generation model to obtain a mobile inspection device inspection path, match the mobile inspection device inspection path with the substation fire monitoring device monitoring range image data to obtain monitoring device monitoring ranges coinciding with the mobile inspection device inspection path, take the monitoring device monitoring ranges coinciding with the mobile inspection device inspection path as known monitoring ranges, and take monitoring device monitoring ranges not coinciding with the mobile inspection device inspection path as unknown monitoring ranges; The data analysis unit is configured to obtain power plant device coordinate information in the unknown monitoring ranges in the substation digital twin image data, substitute the power plant device coordinate information in the unknown monitoring ranges into the preset inspection path generation model to generate a mobile inspection path for the unknown monitoring ranges, and send the mobile inspection path for the unknown monitoring ranges to the mobile inspection device. The monitoring and warning unit is configured to receive inspection data of the mobile inspection device in the unknown monitoring ranges and monitoring data of the monitoring device in the known monitoring ranges, substitute the inspection data of the mobile inspection device in the unknown monitoring ranges and the monitoring data of the monitoring device in the known monitoring ranges into a preset substation power fire warning model in sequence to obtain a substation power fire warning result, and generate fire warning information if the substation power fire warning result indicates a fire risk.

6. The substation electrical fire-prevention warning system of claim 5, wherein, The monitoring range determination unit is further configured to: take the fire warning monitoring requirements and the mobile inspection device information as model input information, the mobile inspection device information including a mobile inspection device type, a mobile inspection device speed, and a mobile inspection device endurance; take an optimal or suboptimal mobile inspection device inspection path as model output information; take monitoring ranges of the monitoring devices, physical limitations of the mobile inspection devices, and road traffic conditions as constraint conditions; receive and set a target function including minimizing a total length of the inspection path and maximizing a monitoring coverage range; The model input information and constraint conditions are input into a selected path planning algorithm, the ant colony algorithm is run, and a potential inspection path is generated; The potential inspection path is evaluated according to a target function, and an optimal path is obtained as a final inspection path.

7. The substation electrical fire-prevention warning system of claim 5, wherein, The data analysis unit is further configured to: extract power plant equipment coordinate information in an unknown monitoring range from the digital twin image data of the substation, the power plant equipment coordinate information including positions of power plant equipment and types of the power plant equipment; input the power plant equipment coordinate information as input data into the inspection path generation model, and generate an optimal inspection path according to the input equipment coordinates, inspection requirements of the equipment, and performance parameters of the mobile inspection equipment; send the optimal inspection path to the mobile inspection equipment in the form of a GPS coordinate sequence and a map label.

8. The substation electrical fire-prevention warning system of claim 5, wherein, The monitoring and early warning unit is further configured to: extract substation power fire early warning features from the substation operation data, the substation power fire early warning features including temperature features, smoke concentration features, flame detection features, and equipment operation state features; train a neural network model using the substation power fire early warning features to obtain a substation power fire early warning model; divide a data set of the substation operation data into a training set, a validation set, and a test set; train the substation power fire early warning model using the training set, validate the substation power fire early warning model using the validation set, compare a prediction result of the substation power fire early warning model with the validation set, and obtain a validated substation power fire early warning model.

Images