In-site safety inspection system for thermal power plant

Abstract

The invention provides an in-field safety inspection system for a thermal power plant, which belongs to the technical field of industrial intelligence and comprises an object determination module for determining an object to be inspected and a corresponding inspection index based on the importance and operation state of power plant equipment; the object matching module is used for determining an inspection task and a task type based on an inspection object and a corresponding inspection index, and matching a corresponding inspection robot and an unmanned aerial vehicle based on the task type; the path determination module determines a first inspection path of the inspection robot and a second inspection path of the unmanned aerial vehicle based on a path planning algorithm and the inspection task, and performs monitoring based on the corresponding inspection paths; and the strategy generation module is used for analyzing the monitoring data and generating an inspection optimization strategy based on an analysis result. And intelligent optimization and management of equipment operation of the power plant can be realized.

Description

Technical Field

[0001] The present invention relates to the field of industrial intelligent technologies, and in particular to an on-site safety inspection system for thermal power plants. Background Art

[0002] As key energy facilities, thermal power plants need to conduct regular safety inspections to ensure stable operation of equipment, reduce the occurrence of failures and safety risks.

[0003] Currently, the inspections of thermal power plants usually consist of manual inspections, fixed-point detections, and regular maintenance. Mainly, data is collected using devices such as sensors and cameras, and then the equipment status and safety conditions are judged through manual analysis or simple algorithms, mainly relying on the limitations of manual judgment and having limited efficiency and accuracy.

[0004] Therefore, the present invention proposes an on-site safety inspection system for thermal power plants. Summary of the Invention

[0005] The present invention provides an on-site safety inspection system for thermal power plants, which automatically determines inspection objects, tasks, and paths through algorithms, reduces the demand for human resources, improves inspection efficiency, monitors different inspection objects and indicators, realizes multi-dimensional equipment status evaluation, uses devices such as inspection robots and drones for inspections, analyzes the monitored data, and generates inspection optimization strategies to help the power plant achieve intelligent optimization and management of equipment operation.

[0006] The present invention provides an on-site safety inspection system for thermal power plants, including:

[0007] Object determination module: Based on the importance and operating status of the power plant equipment, determine the objects to be inspected and the corresponding inspection indicators;

[0008] Object matching module: Based on the inspected objects and the corresponding inspection indicators, determine the inspection tasks and task types, and match the corresponding inspection robots and drones based on the task types;

[0009] Path determination module: Based on the path planning algorithm and the inspection tasks, determine the first inspection path of the inspection robot and the second inspection path of the drone, and perform monitoring based on the corresponding inspection paths respectively;

[0010] Strategy generation module: Analyze the monitored data and generate inspection optimization strategies based on the analysis results.

[0011] The present invention provides an on-site safety inspection system for thermal power plants, and the object determination module includes:

[0012] First processing unit: Collect historical operation data of power plant equipment, perform first processing on the historical operation data, and obtain the importance level of each device;

[0013] Second processing unit: Deploy sensors on each device to collect real-time operation data of the device, perform second processing on the real-time operation data, and obtain the real-time operation status of each device;

[0014] Device screening unit: Screen devices based on the importance level and real-time operation status of all devices to determine the first inspection objects;

[0015] First analysis unit: Perform first analysis on the historical operation data to obtain the first faults of each device;

[0016] Second analysis unit: Perform second analysis on the real-time operation status of each first inspection object to determine the probability of occurrence of the first faults of each first inspection object;

[0017] Fault determination unit: Determine the second faults of each first inspection object based on the probability of occurrence of the first faults of each first inspection object and a preset probability threshold;

[0018] Index determination unit: Determine corresponding inspection indexes according to the second faults of each first inspection object and a preset fault-index database.

[0019] The present invention provides an on-site safety inspection system for a thermal power plant. The first processing unit includes:

[0020] Index acquisition sub-unit: Collect historical operation data of power plant equipment, process the historical operation data to obtain several operation indexes of each device;

[0021] Index processing sub-unit: Obtain the importance level of each device based on several operation indexes of each device:

[0022]

[0023] where, E i is the importance level of the i-th device, w j (t) is the preset weight of the j-th operation index of the i-th device at time t, φ j (t, X i ) is the index processing function of the i-th device at time t regarding the preset first operation state of the device, X i is the preset operation state of the i-th device, n is the number of operation indexes of the i-th device, N is the number of time instants up to time t, w jk(t) is the preset weight of the k-th sub-index of the i-th device based on the j-th operation index at time t, θ jk (t, X i ) is the index processing function of the i-th device at time t regarding the preset second operation state of the device. p is the number of sub-indices of the i-th device under the j-th operation index, R ijk (t) is the membership degree of the k-th sub-index of the i-th device based on the j-th operation index at time t, S ijc (t) is the c-th additional variable of the i-th device based on the j-th operation index, σ jc (t) is the preset weight corresponding to the c-th additional variable of the i-th device based on the j-th operation index. q is the number of additional variables of the i-th device based on the j-th operation index, β jkl is the mutual influence coefficient between the k-th and l-th sub-indices of the i-th device based on the j-th operation index, R ijl (t) is the membership degree of the l-th sub-index of the i-th device based on the j-th operation index at time t, γ j (t) is the time decay coefficient of the j-th operation index at time t, E i0 is the preset initial importance level of the i-th device.

[0024] The present invention provides an on-site safety inspection system for a thermal power plant, and the operation indices include: equipment status index, equipment safety facility index, equipment electrical safety index, and equipment environment index.

[0025] The present invention provides an on-site safety inspection system for a thermal power plant, and the second processing unit includes:

[0026] Data acquisition sub-unit: Deploy sensors on each device to collect the operation data of the device in real time;

[0027] Data processing sub-unit: Perform second processing on the real-time operation data;

[0028] Status acquisition first sub-unit: Obtain the real-time operation status value of each device based on the second processing data of each device:

[0029] Among them, S i (t) is the operation status value of the i-th device at time t, r0 i (t) is the preset maximum value of the sensor data of the i-th device at time t, α h (t) is the preset weight of the corresponding second processing data collected by the h-th sensor at time t. m is the number of corresponding second processing data collected by the sensors on the i-th device, d ih(t) is the corresponding second processed data collected by the h-th sensor of the i-th device at time t, f h is the preset processing function for the corresponding second processed data collected by the h-th sensor, δ hg (t) is the preset mutual influence coefficient at time t for the corresponding second processed data collected by the h-th and g-th sensors, d ig (t) is the corresponding second processed data collected by the g-th sensor of the i-th device at time t, f g is the preset processing function for the corresponding second processed data collected by the g-th sensor;

[0030] State acquisition second sub-unit: Based on the preset state value - state data table and the real-time operating state value of each device, determine the real-time operating state of each device.

[0031] The present invention provides an on-site safety inspection system for a thermal power plant. The path determination module includes:

[0032] Map processing unit: Collect map data of the area to be inspected, and determine the positions of obstacles in the map based on image processing technology;

[0033] Area determination unit: Determine the inspection area according to the inspection task and map information;

[0034] Position determination unit: Based on the inspection area boundary, determine the starting positions of the robot and the drone, and based on the inspection task, determine the end positions of the robot and the drone;

[0035] Path planning unit: Based on the path planning algorithm, with the goal of minimizing the inspection time or the coverage area, respectively determine the first path corresponding to the robot from the starting point to the end point avoiding the obstacle positions and the second path corresponding to the drone;

[0036] Path optimization unit: Optimize and adjust the first path and the second path, and then determine the first inspection path of the inspection robot and the second inspection path of the drone.

[0037] The present invention provides an on-site safety inspection system for a thermal power plant. The path optimization unit includes:

[0038] Initialization sub-unit: Generate a number of initial paths based on the preset optimal path algorithm;

[0039] Fitness acquisition sub-unit: Obtain the fitness of each initial path based on the preset fitness function;

[0040] Path selection sub-unit: Determine a number of first initial paths based on the preset selection algorithm and the fitness of each initial path;

[0041] Path combination subunit: Combine several first initial paths pairwise to obtain several path combinations;

[0042] Path crossover subunit: Generate corresponding several second initial paths based on a preset algorithm and the first initial paths in each combination;

[0043] Path adjustment subunit: Adjust each second initial path based on a preset adjustment strategy to obtain a third initial path;

[0044] Fitness calculation subunit: Obtain the fitness of each third initial path based on a preset fitness function;

[0045] Dynamic iteration subunit: Obtain the fitness change rate based on the fitness data after each iteration, and set a dynamic iteration convergence control mechanism based on the fitness change rate;

[0046] Iterative calculation subunit: Obtain the maximum number of iterations based on the dynamic iteration convergence control mechanism, and repeat the above operations until the maximum number of iterations is reached to generate several optimal paths;

[0047] Path determination subunit: Perform the above path planning operations on the robot and the UAV respectively to obtain several optimal paths corresponding to the robot and the UAV, and then obtain the first path and the second path.

[0048] The present invention provides an in-plant safety inspection system for a thermal power plant, and the preset algorithms include: single-point crossover algorithm, multi-point crossover algorithm, uniform crossover algorithm, order crossover algorithm, partial mapping crossover algorithm, and cycle crossover algorithm, etc.

[0049] The in-plant safety inspection system for a thermal power plant provided by the present invention automatically determines the inspection objects, tasks, and paths through algorithms, reduces the demand for human resources, improves the inspection efficiency, monitors different inspection objects and indicators, realizes multi-dimensional equipment status evaluation, uses equipment such as inspection robots and UAVs for inspection, analyzes the monitoring data, and generates inspection optimization strategies to help the power plant achieve intelligent optimization and management of equipment operation. Brief Description of the Drawings

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1It is a schematic structural diagram of an on-site safety inspection system for a thermal power plant provided by an embodiment of the present invention. Detailed implementation manners

[0052] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment 1

[0054] The present invention provides an on-site safety inspection system for a thermal power plant, as Figure 1 shown, including:

[0055] Object determination module: Based on the importance and operating status of power plant equipment, determine the objects to be inspected and the corresponding inspection indicators;

[0056] Object matching module: Based on the objects to be inspected and the corresponding inspection indicators, determine the inspection tasks and task types, and match the corresponding inspection robots and unmanned aerial vehicles based on the task types;

[0057] Path determination module: Based on the path planning algorithm and the inspection tasks, determine the first inspection path of the inspection robot and the second inspection path of the unmanned aerial vehicle, and monitor based on the corresponding inspection paths respectively;

[0058] Strategy generation module: Analyze the monitoring data and generate an inspection optimization strategy based on the analysis results.

[0059] In this embodiment, the importance of power plant equipment refers to the evaluation of the importance of various equipment in a thermal power plant. For example, core equipment such as boilers, steam turbines, and generators is crucial for the operation of the power plant, while auxiliary equipment such as water pumps and valves, although not core equipment, also supports the operation of the entire system;

[0060] In this embodiment, the operating status refers to various states of the equipment during operation, including normal operation, abnormal operation, shutdown for maintenance, etc. By monitoring the operating status of the equipment, problems can be discovered in a timely manner and corresponding measures can be taken to ensure the safe operation of the equipment;

[0061] In this embodiment, the inspection indicators are specific parameters or indicators used to evaluate the operating status of the equipment. For example, parameters such as temperature, pressure, vibration, and smoke concentration are used to determine whether the equipment is operating normally.

[0062] In this embodiment, the patrol inspection task is to determine the necessary patrol inspection work based on the importance and operating status of the equipment. The task types include regular patrol inspections, inspections triggered by specific events, etc. For example, regularly performing vibration inspections on generators belongs to one type of task;

[0063] In this embodiment, matching the corresponding patrol inspection robots and drones based on the task type. According to different patrol inspection task types, the system will select suitable patrol inspection tools, which may be patrol inspection robots, drones, etc. For example, for the patrol inspection task of high-altitude equipment, a drone can be selected for inspection. If the task type is ground equipment inspection, a patrol inspection robot needs to be used instead of a drone.

[0064] In this embodiment, the path planning algorithm refers to an algorithm that determines the patrol inspection paths of patrol inspection robots and drones based on the location of the equipment, patrol inspection tasks, and other conditions. For example, in a thermal power plant, assuming that a patrol inspection task of high-altitude equipment needs to be carried out, this task type requires the use of a drone for inspection, including: identifying the task type as high-altitude equipment inspection, so a drone needs to be used for inspection, obtaining the location data of the high-altitude equipment, including its coordinate information. Based on the location data of the high-altitude equipment, the path planning algorithm will determine the optimal patrol inspection path so that the drone can effectively cover all equipment and minimize flight time and energy consumption. According to the optimal path generated by the path planning algorithm, the drone is guided to perform the patrol inspection task. The drone will fly along the predetermined path while using the sensors carried to monitor the equipment in real time and collect data;

[0065] In this embodiment, the first patrol inspection path and the second patrol inspection path respectively refer to the patrol inspection paths of the patrol inspection robot and the drone. In actual patrol inspections, two patrol inspection tools need to cooperate. The first patrol inspection path is usually executed by the robot, while the second patrol inspection path is executed by the drone;

[0066] In this embodiment, the patrol inspection optimization strategy is to generate targeted patrol inspection optimization strategies based on the analysis results of the monitoring data, such as adjusting the patrol inspection frequency, optimizing the patrol inspection path, etc., to improve the patrol inspection efficiency and accuracy, prioritize the patrol inspection tasks to ensure that key equipment is inspected first, and according to the real-time monitoring data, the system dynamically adjusts the patrol inspection tasks and patrol inspection paths to optimize the patrol inspection efficiency.

[0067] The beneficial effects of the above technical solutions are as follows: By automatically determining the patrol inspection objects, tasks, and paths through algorithms, the demand for human resources is reduced, and the patrol inspection efficiency is improved. Monitoring different patrol inspection objects and indicators realizes multi-dimensional equipment status assessment. Using equipment such as patrol inspection robots and drones for patrol inspection, analyzing the monitoring data, and generating patrol inspection optimization strategies to help the power plant achieve intelligent optimization and management of equipment operation.

[0068] Embodiment 2

[0069] The present invention provides an on-site safety inspection system for a thermal power plant. The object determination module includes:

[0070] The first processing unit: collects the historical operation data of the power plant equipment, performs a first processing on the historical operation data, and obtains the importance level of each equipment;

[0071] The second processing unit: deploys sensors on each equipment to collect the operation data of the equipment in real time, performs a second processing on the real-time operation data, and obtains the real-time operation status of each equipment;

[0072] The equipment screening unit: screens the equipment based on the importance level and real-time operation status of all equipment, and determines the first inspection object;

[0073] The first analysis unit: performs a first analysis on the historical operation data to obtain the first faults of each equipment;

[0074] The second analysis unit: performs a second analysis on the real-time operation status of each first inspection object to determine the probability of occurrence of the first faults of each first inspection object;

[0075] The fault determination unit: determines the second faults of each first inspection object based on the probability of occurrence of the first faults of each first inspection object and a preset probability threshold;

[0076] The index determination unit: determines the corresponding inspection indexes according to the second faults of each first inspection object and a preset fault-index database.

[0077] In this embodiment, the historical operation data refers to the operation data records of the power plant equipment in the past period of time, including various parameters such as temperature, pressure, vibration, fault data, operation critical data, maintenance cost data, etc. By analyzing the historical operation data, the operation condition and trend of the equipment can be understood, providing a reference for subsequent fault prediction and inspection;

[0078] In this embodiment, the importance level of each equipment processed for the first time is processed for the historical operation data of each equipment to determine its importance level in the entire power plant system. According to the historical operation data of the thermal power plant equipment, the importance of each equipment is evaluated and graded.

[0079] In this embodiment, the second processing is to process the operation data of the equipment collected in real time by the sensors to obtain the real-time operation status of the equipment, including current indexes such as temperature, pressure, vibration, etc., for judging whether the equipment is operating normally.

[0080] In this embodiment, the first inspection objects are screened according to the importance level and real-time operation status of the equipment to determine the objects that need to be inspected for the first time, including the equipment that shows abnormalities in historical data.

[0081] In this embodiment, the first analysis is to analyze the historical operation data to identify the first possible faults of each equipment, including common problems of the equipment or abnormal situations that repeatedly appear in the historical data. For example, equipment type: boiler, first fault: burner blockage.

[0082] In this embodiment, the second analysis is to analyze the real-time operation status of each first inspection object to determine the probability of occurrence of the first fault for each first inspection object, which is determined based on the comparison of real-time data and historical data.

[0083] In this embodiment, the preset probability threshold is determined based on empirical judgment and is used to judge whether the probability of occurrence of the first fault has reached the standard for triggering an inspection. The second fault is the possible next fault determined based on the preset probability threshold on the basis of the first inspection object having the first fault.

[0084] In this embodiment, the preset fault-index database is a database that contains various faults and their corresponding inspection indexes. After the second fault is determined, the corresponding inspection indexes are determined according to the preset fault-index database for subsequent inspection work. For example, if the second fault is lubricating oil leakage, the corresponding inspection indexes may include the lubricating oil leakage volume, the lubricating oil pressure of the equipment, etc.

[0085] The beneficial effects of the above technical solutions are as follows: By obtaining the importance level of each equipment, then combining the real-time operation status to screen the equipment, determining the first inspection objects, using historical data for fault analysis, and then performing probabilistic fault prediction through real-time data, further determining the potential faults of the inspection objects, and determining the corresponding inspection indexes according to the fault types, the inspection efficiency is improved. It realizes data-driven intelligent inspection and can prevent and discover potential equipment faults in a timely manner.

[0086] Embodiment 3

[0087] The present invention provides an on-site safety inspection system for a thermal power plant, and a first processing unit, including:

[0088] Index acquisition sub-unit: Collect the historical operation data of the power plant equipment, and process the historical operation data to obtain several operation indexes of each equipment;

[0089] Index processing sub-unit: Obtain the importance level of each equipment based on several operation indexes of each equipment:

[0090]

[0091] Among them, E i is the importance level of the i-th device, w j (t) is the preset weight of the j-th operating index of the i-th device at time t, φ j (t, X i ) is the index processing function of the i-th device at time t regarding the preset first operating state of the device, X i is the preset operating state of the i-th device, n is the number of operating indexes of the i-th device, N is the number of time instants up to time t, w jk (t) is the preset weight of the k-th sub-index of the i-th device based on the j-th operating index at time t, θ jk (t, X i ) is the index processing function of the i-th device at time t regarding the preset second operating state of the device, p is the number of sub-indexes of the j-th operating index of the i-th device, R ijk (t) is the membership degree of the k-th sub-index of the i-th device based on the j-th operating index at time t, S ijc (t) is the c-th additional variable of the i-th device based on the j-th operating index, σ jc (t) is the preset weight corresponding to the c-th additional variable of the i-th device based on the j-th operating index, q is the number of additional variables of the i-th device based on the j-th operating index, β jkl is the mutual influence coefficient between the k-th and l-th sub-indexes of the i-th device based on the j-th operating index, R ijl (t) is the membership degree of the l-th sub-index of the i-th device based on the j-th operating index at time t, γ j (t) is the time decay coefficient of the j-th operating index at time t, E i0 is the preset initial importance level of the i-th device.

[0092] In this embodiment, the preset operating state refers to the expected operating state or operating status preset for each device in advance, which varies according to the type and function of the device. For example, for a generator, the preset operating states are normal operation, shutdown, and maintenance;

[0093] In this embodiment, the index processing function is a function used to process various collected operating indexes to determine the importance level of the device, including steps such as data analysis, weighted calculation, and logical judgment, so as to evaluate its importance according to the operating state and indexes of the device. For example, the operating state and importance of the device are evaluated according to the weights and actual values of different indexes;

[0094] In this embodiment, membership degree refers to the degree of association between a certain value or state and a specific range or category, and is used to evaluate the state of device operation indicators. For example, the membership degree of a certain temperature indicator can represent the degree to which the temperature is within the normal range. For another example, the membership degree of the temperature indicator can represent the degree of association between the current temperature and the normal temperature range;

[0095] In this embodiment, an additional variable refers to other factors or parameters that may affect the operation state or performance of the device, including environmental conditions, load conditions, maintenance records, etc.

[0096] The beneficial effects of the above technical solutions are as follows: By collecting and processing the historical operation data of the device, multiple operation indicators of each device are determined, and the importance level of the device is evaluated accordingly. By presetting various parameters such as weights and index processing functions, comprehensively considering the operation state, operation conditions, and additional factors of the device, the inspection efficiency and accuracy are improved, ensuring the safe operation of the thermal power plant.

[0097] Embodiment 4

[0098] The present invention provides an on-site safety inspection system for a thermal power plant, and the operation indicators include: device status indicators, device safety facility indicators, device electrical safety indicators, and device environmental indicators.

[0099] In this embodiment, the device status indicator is used to evaluate the operation state of the device, such as temperature, pressure, rotation speed, etc. For example, for a boiler, the device status indicators may include boiler water level, burner status, flue gas emission, etc.;

[0100] In this embodiment, the device safety facility indicator is used to evaluate the state of the safety devices and protection facilities of the device to ensure that the device can take appropriate safety measures in a timely manner in case of abnormalities. For example, the opening state of the safety valve, the functionality of the emergency stop button, etc.;

[0101] In this embodiment, the device electrical safety indicator is used to evaluate the safety performance of the electrical system of the device, including voltage, current, insulation resistance, etc. For example, the output voltage of the generator, the temperature of the transformer, the leakage protection of the distribution box, etc.;

[0102] In this embodiment, the device environmental indicator is used to evaluate the conditions of the environment where the device is located, including temperature, humidity, air pressure, etc. For example, the temperature in the generator room, the air quality around the machinery and equipment, the smell in the chemical storage area, etc.

[0103] The beneficial effects of the above technical solution are as follows: By comprehensively monitoring the operating status and environmental conditions of thermal power generation equipment through equipment status indicators, equipment safety facility indicators, equipment electrical safety indicators, and equipment environmental indicators, the safety performance of the equipment is evaluated in all aspects and multi-dimensions, abnormal conditions are monitored in real time, and the safety management level of thermal power plants is improved.

[0104] Example 5

[0105] The present invention provides an on-site safety inspection system for a thermal power plant. The second processing unit includes:

[0106] Data acquisition sub-unit: Deploy sensors on each device to collect the operating data of the device in real time;

[0107] Data processing sub-unit: Perform second processing on the real-time operating data;

[0108] Status acquisition first sub-unit: Obtain the real-time operating status value of each device based on the second processed data of each device:

[0109] Wherein, S u (t) is the operating status value of the i-th device at time t, r0 u (t) is the maximum allowable value of the data collected by the sensor in the preset threshold of the system. For example, if the preset maximum value of the temperature sensor of a certain device in a thermal power plant is 100 degrees Celsius, then the system will issue an alarm or take measures once the temperature reaches or exceeds this value; h (t) is the preset weight of the corresponding second processed data collected by the h-th sensor at time t, m is the number of corresponding second processed data collected by the sensors on the i-th device, d ih (t) is the corresponding second processed data collected by the h-th sensor of the i-th device at time t, f h is the preset processing function of the corresponding second processed data collected by the h-th sensor, δ hg (t) is the preset mutual influence coefficient of the corresponding second processed data collected by the h-th and g-th sensors at time t, d ig (t) is the corresponding second processed data collected by the g-th sensor of the i-th device at time t, f g is the preset processing function of the corresponding second processed data collected by the g-th sensor;

[0110] Status acquisition second sub-unit: Determine the real-time operating status of each device based on the preset status value - status data table and the real-time operating status value of each device.

[0111] In this embodiment, the preset maximum sensor data refers to the maximum allowable value of the data collected by the sensor in the preset threshold of the system. For example, if the preset maximum value of the temperature sensor of a certain device in a thermal power plant is 100 degrees Celsius, then the system will issue an alarm or take measures once the temperature reaches or exceeds this value;

[0112] In this embodiment, the preset processing function refers to a method or algorithm preset in the system for processing the collected data, which performs processing such as smoothing, filtering, and anomaly detection on the collected data to improve the accuracy and usability of the data. For example, for the data collected by a temperature sensor, the preset processing function is a moving average algorithm for smoothing the temperature data;

[0113] In this embodiment, the preset interaction coefficient refers to a preset weight or coefficient for the interaction between the data collected by different sensors, which reflects the correlation or dependence between the data of different sensors. For example, if a certain device in a thermal power plant has both a temperature sensor and a pressure sensor, the preset interaction coefficient can be used to determine the relationship between temperature and pressure under different operating states, so as to more accurately evaluate the operating state of the device.

[0114] The beneficial effects of the above technical solutions are as follows: By collecting the device operation data in real time, performing the second processing, obtaining the real-time operation state value, and combining with the preset state value - state data table, the real-time operation state of each device is accurately determined. Multidimensional data is efficiently processed, and combined with the preset weights and processing functions, the device state is effectively identified. The system can quickly respond to device anomalies, improving the safety and production efficiency of the thermal power plant.

[0115] Embodiment 6

[0116] The present invention provides an on-site safety inspection system for a thermal power plant, and a path determination module, including:

[0117] Map processing unit: Collecting map data of the area to be inspected, and determining the positions of obstacles in the map based on image processing technology;

[0118] Area determination unit: Determining the inspection area according to the inspection task and map information;

[0119] Position determination unit: Determining the starting positions of the robot and the drone based on the boundary of the inspection area, and determining the end positions of the robot and the drone based on the inspection task;

[0120] Path planning unit: Based on the path planning algorithm, with the goal of minimizing the inspection time or the coverage area, respectively determining the first path corresponding to the robot from the starting point to the end point avoiding the obstacle positions and the second path corresponding to the drone;

[0121] Path optimization unit: Optimizing and adjusting the first path and the second path, and then determining the first inspection path of the inspection robot and the second inspection path of the drone.

[0122] In this embodiment, the map data refers to the data collected containing information related to the inspection area, usually including geographical features, obstacle positions, etc. For example, the map data can be ground images or point cloud data collected by lidar or cameras;

[0123] In this embodiment, the inspection area refers to the specific area range that needs to be inspected determined according to the map data and the inspection task. For example, the inspection area of a thermal power plant can be the entire plant area or the surrounding area of specific equipment;

[0124] In this embodiment, optimizing and adjusting the first path and the second path means further optimizing the inspection paths of the robot and the drone to improve efficiency or meet specific inspection requirements, including: Dynamic path planning: Dynamically adjust the paths of the robot and the drone according to real-time environmental changes and obstacle conditions. For example, when new obstacles appear or the area changes, the system can automatically re-plan the path to avoid obstacles or adapt to the new environment. Path smoothing: Smooth the paths of the robot and the drone to make the paths more continuous and stable. This can reduce path reversals and sharp turns, improve inspection efficiency and reduce energy consumption. For example, use spline curves or Bezier curves to smooth the paths. Considering resource limitations: Consider the actual capabilities and resource limitations of the robot and the drone, such as battery power, flight range, etc., and optimize the path. For example, by reasonably arranging charging stations or supply points, make the path more economical and reasonable to ensure that the robot and the drone can complete the entire inspection task. Multi-objective optimization: Consider two objectives of minimizing the inspection time and maximizing the coverage area at the same time, and adjust the path through a multi-objective optimization algorithm. For example, use a multi-objective genetic algorithm or a multi-objective particle swarm algorithm to optimize the path to balance the trade-off between time and coverage area. Suppose in the inspection task of a thermal power plant, the robot and the drone need to avoid some equipment or pipelines and need to cover the entire area as soon as possible. The path optimization unit can adjust the paths of the robot and the drone according to real-time obstacle information, making them avoid obstacles as much as possible and cover the entire area in an optimal way, thereby improving the inspection efficiency and coverage.

[0125] The beneficial effects of the above technical solutions are: By using image processing technology to identify obstacles, determine the inspection area and start and end positions, and plan the paths of the robot and the drone with the goal of minimizing time or coverage area, ensure that the inspection paths of the robot and the drone are optimal, improve the inspection efficiency and coverage, and thus enhance the safety management level of the thermal power plant.

[0126] Embodiment 7

[0127] The present invention provides an in-plant safety inspection system for a thermal power plant, and a path optimization unit, including:

[0128] Initialization subunit: Generate a number of initial paths based on a preset optimal path algorithm;

[0129] Fitness acquisition subunit: Obtain the fitness of each initial path based on a preset fitness function;

[0130] Path selection subunit: Determine a number of first initial paths based on a preset selection algorithm and the fitness of each initial path;

[0131] Path combination subunit: Combine the first initial paths pairwise to obtain a number of path combinations;

[0132] Path crossover subunit: Generate corresponding second initial paths based on a preset algorithm and the first initial paths in each combination;

[0133] Path adjustment subunit: Adjust each second initial path based on a preset adjustment strategy to obtain a third initial path;

[0134] Fitness calculation subunit: Obtain the fitness of each third initial path based on a preset fitness function;

[0135] Dynamic iteration subunit: Obtain the fitness change rate based on the fitness data after each iteration, and set a dynamic iteration convergence control mechanism based on the fitness change rate;

[0136] Iterative calculation subunit: Obtain the maximum number of iterations based on the dynamic iteration convergence control mechanism, and repeat the above operations until the maximum number of iterations is reached to generate a number of optimal paths;

[0137] Path determination subunit: Perform the above path planning operations on the robot and the drone respectively to obtain a number of optimal paths corresponding to the robot and the drone, and then obtain the first path and the second path.

[0138] In this embodiment, the preset optimal path algorithm refers to an algorithm preset in the system for generating initial paths, which is based on heuristic methods, empirical rules, or mathematical optimization models, etc. For example, genetic algorithms, simulated annealing algorithms, or ant colony algorithms can be used to generate initial paths;

[0139] In this embodiment, the preset fitness function is a function used to evaluate the quality of each initial path, which is designed according to the characteristics and objectives of the inspection task, including considering factors such as time, resource utilization rate, and safety. For example, the fitness function can evaluate the fitness of a path according to indicators such as the length of the path, the number of times of avoiding obstacles, and the area covered by the inspection;

[0140] In this embodiment, the fitness of the initial path refers to the fitness value obtained after each initial path is evaluated by the fitness function, which is used for subsequent path selection and optimization. The higher the fitness value, the better the path. For example, if the fitness value obtained by an initial path after being calculated by the fitness function is 0.85, it indicates that this path performs well under the objective function.

[0141] In this embodiment, the preset adjustment strategy refers to the strategy or method used in the path optimization process to adjust the second initial path to generate the third initial path. This can be some heuristic adjustment rules or methods based on mathematical optimization. For example, using a local search algorithm to fine-tune the second initial path, changing the positions of nodes in the path or adjusting path segments, as well as randomly inserting or deleting nodes in the path.

[0142] In this embodiment, the dynamic iterative convergence control mechanism refers to dynamically adjusting the number of iterations or other parameters during the iterative optimization process, dynamically adjusting according to the fitness change situation after each iteration, and monitoring the fitness change situation after each iteration. If the change rate of the fitness is small, it indicates that the algorithm is approaching convergence, and the number of iterations or the iteration step size can be appropriately reduced to save computing resources and accelerate the convergence speed. For example, if the change rate of the fitness is already very small, the iteration can be stopped in advance to save computing resources.

[0143] The beneficial effects of the above technical solutions are as follows: The optimal path is obtained through preset algorithms and functions for generating, selecting, combining, crossing, and adjusting the initial path, followed by dynamic iterative optimization. The technical features include flexible path generation and adjustment strategies, as well as a dynamic iterative convergence control mechanism, which effectively improves the efficiency and coverage of on-site safety inspections in thermal power plants.

[0144] Embodiment 8

[0145] The present invention provides an on-site safety inspection system for a thermal power plant, and preset algorithms, including: single-point crossover algorithm, multi-point crossover algorithm, uniform crossover algorithm, order crossover algorithm, partially mapped crossover algorithm, and cycle crossover algorithm, etc.

[0146] In this embodiment, the single-point crossover algorithm randomly selects a crossover point from two parent individuals, and swaps the genes of the two parent individuals after the crossover point to generate two offspring individuals. If parent individual 1 is ABCDEF and parent individual 2 is UVWXYZ, and the selected crossover point is the third position, then the offspring individuals after crossover are ABCUVW and XYZDEF. Suppose there are two parent individuals ABCDEF and UVWXYZ respectively, and the single-point crossover algorithm is selected with the crossover point at the fourth position, then the offspring individuals after crossover are ABCWXYZ and UVXDEF. For example, suppose the parent path 1 is A->B->C->D->E and the parent path 2 is W->X->Y->Z, and the selected crossover point is the second node, then the offspring paths after crossover are A->X->Y->Z and W->B->C->D->E;

[0147] In this embodiment, the multi-point crossover algorithm exchanges genes at multiple crossover points. If the selected crossover points are the second and fifth positions, then the offspring individuals after crossover are AUWXYZ and VBCDEF. For example, if the selected crossover points are the first node and the third node, then the offspring paths after crossover are A->X->C->D->E and W->B->Y->Z;

[0148] In this embodiment, the uniform crossover algorithm exchanges the genes of the two parent individuals according to a certain probability. For example, a probability of 0.5 is set, and a random probability sequence is generated as a mask, and whether to exchange genes is determined according to the mask;

[0149] In this embodiment, the order crossover algorithm retains the partial gene order of one parent individual and fills in the genes that do not appear in the other parent individual. If parent individual 1 is ABCDEF and parent individual 2 is UVWXYZ, and the order of the first three genes is selected for retention and then the remaining genes are filled in, then the offspring individual after crossover is ABCXYZ. For example, if the order of the first two nodes is selected for retention and then the remaining nodes are filled in, then the offspring paths after crossover are A->B->X->Y->Z and W->C->D->E;

[0150] In this embodiment, the partially mapped crossover algorithm selects two crossover points, swaps the genes between these two crossover points, and retains the gene positions in the corresponding interval of parent path 1. For example, if the selected crossover points are the second node and the fourth node, then the offspring paths after crossover are A->X->C->Y->E and W->B->D->Z;

[0151] In this embodiment, the cycle crossover algorithm exchanges genes by cyclically traversing the genes of two parent paths. For example, starting from the first node, the node corresponding to the position of this node is used as the exchange target, and then the node at the target position is continuously found until returning to the first node.

[0152] The beneficial effects of the above technical solution are as follows: path optimization is achieved through preset algorithms, including single-point, multi-point, uniform, sequential, partial mapping, and cycle crossover algorithms. These algorithms can effectively improve the efficiency and safety of the inspection path, optimize resource utilization, reduce inspection time, and lower the human error rate, thereby enhancing the safety management level and operation efficiency of thermal power plants.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An on-site safety inspection system for a thermal power plant, characterized in that, Including: Object determination module: Based on the importance and operating status of power plant equipment, determine the objects to be inspected and the corresponding inspection indicators; Object matching module: Based on the objects to be inspected and the corresponding inspection indicators, determine the inspection tasks and task types, and match the corresponding inspection robots and drones based on the task types; Path determination module: Based on the path planning algorithm and the inspection tasks, determine the first inspection path of the inspection robot and the second inspection path of the drone, and monitor based on the corresponding inspection paths respectively; Strategy generation module: Analyze the monitoring data and generate inspection optimization strategies based on the analysis results.

2. The on-site safety inspection system for a thermal power plant according to claim 1, wherein Object determination module, including: First processing unit: Collect the historical operating data of power plant equipment, perform first processing on the historical operating data, and obtain the importance level of each equipment; Second processing unit: Deploy sensors on each equipment to collect the operating data of the equipment in real time, and perform second processing on the real-time operating data to obtain the real-time operating status of each equipment; Equipment screening unit: Screen the equipment based on the importance level and real-time operating status of all equipment to determine the first inspection objects; First analysis unit: Perform first analysis on the historical operating data to obtain the first faults of each equipment; Second analysis unit: Perform second analysis on the real-time operating status of each first inspection object to determine the probability of occurrence of the first faults of each first inspection object; Fault determination unit: Based on the probability of occurrence of the first faults of each first inspection object and a preset probability threshold, determine the second faults of each first inspection object; Indicator determination unit: Determine the corresponding inspection indicators according to the second faults of each first inspection object and a preset fault-indicator database.

3. The on-site safety inspection system for a thermal power plant according to claim 2, wherein, First processing unit, including: Indicator acquisition sub-unit: Collect the historical operating data of power plant equipment, process the historical operating data to obtain several operating indicators of each equipment; Indicator processing sub-unit: Based on the several operating indicators of each equipment, obtain the importance level of each equipment: Among them, E i is the importance level of the i-th device, w j (t) is the preset weight of the j-th operating index of the i-th device at time t, φ j (t, X i ) is the index processing function of the i-th device at time t regarding the preset first operating state of the device, X i is the preset operating state of the i-th device, n is the number of operating indexes of the i-th device, N is the number of time instances up to time t, w jk (t) is the preset weight of the k-th sub-index of the i-th device based on the j-th operating index at time t, θ jk (t, X i ) is the index processing function of the i-th device at time t regarding the preset second operating state of the device, p is the number of sub-indexes of the i-th device under the j-th operating index, R ijk (t) is the membership degree of the k-th sub-index of the i-th device based on the j-th operating index at time t, S ijc (t) is the c-th additional variable of the i-th device based on the j-th operating index, σ jc (t) is the preset weight corresponding to the c-th additional variable of the i-th device based on the j-th operating index, q is the number of additional variables of the i-th device based on the j-th operating index, β jkl is the mutual influence coefficient between the k-th and l-th sub-indexes of the i-th device based on the j-th operating index, R ijl (t) is the membership degree of the l-th sub-index of the i-th device based on the j-th operating index at time t, γ j (t) is the time decay coefficient of the j-th operating index at time t, E i0 is the preset initial importance level of the i-th device.

4. The on-site safety inspection system for a thermal power plant according to claim 3, characterized in that, Operating indicators include: equipment status indicators, equipment safety facility indicators, equipment electrical safety indicators, and equipment environment indicators.

5. The on-site safety inspection system for a thermal power plant according to claim 2, wherein, Second processing unit, including: Data acquisition sub-unit: Deploy sensors on each equipment to collect the operating data of the equipment in real time; Data processing sub-unit: Perform second processing on the real-time operating data; Status acquisition first sub-unit: Based on the second processed data of each equipment, obtain the real-time operating status value of each equipment: Among them, S i (t) is the operating state value of the i-th device at time t, r0 i (t) is the preset maximum value of the sensor data of the i-th device at time t, α h (t) is the preset weight of the corresponding second processed data collected by the h-th sensor at time t, m is the number of the corresponding second processed data collected by the sensors on the i-th device, d ih (t) is the corresponding second processed data collected by the h-th sensor of the i-th device at time t, f h is the preset processing function of the corresponding second processed data collected by the h-th sensor, δ hg (t) is the preset mutual influence coefficient of the corresponding second processed data collected by the h-th and g-th sensors at time t, d ig (t) is the corresponding second processed data collected by the g-th sensor of the i-th device at time t, f g is the preset processing function of the corresponding second processed data collected by the g-th sensor; Status acquisition second sub-unit: Based on a preset status value-status data table and the real-time operating status value of each equipment, determine the real-time operating status of each equipment.

6. The on-site safety inspection system for a thermal power plant according to claim 2, wherein, Path determination module, including: Map processing unit: Collect the map data of the area to be inspected, and determine the positions of obstacles in the map based on image processing technology; Area determination unit: Determine the inspection area according to the inspection tasks and map information; Position determination unit: Based on the inspection area boundary, determine the starting positions of the robot and the drone, and based on the inspection tasks, determine the end positions of the robot and the drone; Path planning unit: Based on path planning algorithms, with the goal of minimizing inspection time or coverage area, respectively determine the first path corresponding to the robot from the starting point to the end point while avoiding obstacle positions and the second path corresponding to the drone. Path optimization unit: Optimize and adjust the first path and the second path, and then determine the first inspection path of the inspection robot and the second inspection path of the drone.

7. The on-site safety inspection system for a thermal power plant according to claim 6, wherein Path optimization unit, including: Initialization subunit: Generate several initial paths based on a preset optimal path algorithm. Fitness acquisition subunit: Obtain the fitness of each initial path based on a preset fitness function. Path selection subunit: Determine several first initial paths based on a preset selection algorithm and the fitness of each initial path. Path combination subunit: Pairwise combine several first initial paths to obtain several path combinations. Path crossover subunit: Generate corresponding several second initial paths based on a preset algorithm and the first initial paths in each combination. Path adjustment subunit: Adjust each second initial path based on a preset adjustment strategy to obtain a third initial path. Fitness calculation subunit: Obtain the fitness of each third initial path based on a preset fitness function. Dynamic iteration subunit: Obtain the fitness change rate based on the fitness data after each iteration, and set a dynamic iteration convergence control mechanism based on the fitness change rate. Iterative calculation subunit: Obtain the maximum number of iterations based on the dynamic iteration convergence control mechanism, and repeat the above operations until the maximum number of iterations is reached to generate several optimal paths. Path determination subunit: Perform the above path planning operations on the robot and the drone respectively to obtain several optimal paths corresponding to the robot and the drone, and then obtain the first path and the second path.

8. The in-plant safety inspection system for a thermal power plant according to claim 7, characterized in that Preset algorithms include: single-point crossover algorithm, multi-point crossover algorithm, uniform crossover algorithm, order crossover algorithm, partially mapped crossover algorithm, and cycle crossover algorithm, etc.

Images