Unmanned inspection task planning method and system for thermal power plant
By calculating the task weight and status data of the inspection vehicle, optimizing the connection location and dispatching mode of the nanny vehicle, the problem of unreasonable resource allocation in the unmanned inspection system of the thermal power plant is solved, the efficiency of inspection tasks and system stability are improved, and the efficiency of inspection tasks and system stability are ensured. Equipment safety and efficient monitoring.
Patent Information
- Application Number
- CN202411024709.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-12-03
AI Technical Summary
In the existing unmanned patrol system of thermal power plants, the dispatching method between the nanny car and the patrol car lacks dynamic response capabilities, resulting in unreasonable resource allocation and low patrol task efficiency, affecting system stability and equipment safety.
By recording and analyzing the task and status data of the inspection vehicle, calculating the task weight, determining the weighted center point position, and adjusting the connection position of the nanny vehicle through iterative calculations, optimizing the resource allocation and scheduling mode, and flexibly responding to task changes and emergencies through flexibly .
It improves the execution efficiency of inspection tasks and system operation efficiency, reduces the movement and waiting time of the nanny car, ensures the safety and continuous monitoring of the equipment, and improves the efficient and stable operation of the unmanned inspection system.
Smart Images

Figure CN119067286A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of unmanned inspection of thermal power plants, and more specifically, to a method and system for planning unmanned inspection tasks of thermal power plants. Background Art
[0002] The unmanned inspection system of a thermal power plant is an unmanned intelligent monitoring system that deploys inspection vehicles to conduct automatic inspections within the plant area, thereby achieving real-time monitoring and data collection of equipment and environment. The nanny car, as a support platform, is mainly responsible for providing power supply, equipment maintenance, troubleshooting and transfer services for one or more inspection vehicles to ensure that the inspection vehicles can continuously and efficiently perform inspection tasks. The relationship between the nanny car and the inspection vehicle is close. The nanny car makes reasonable parking and dispatching arrangements based on the status and task requirements of the inspection vehicle, thereby ensuring the smooth progress of the inspection task.
[0003] The existing coordination between the nanny car and the inspection car has certain shortcomings. Traditional dispatching methods often lack dynamic response capabilities and are difficult to adapt to changes in inspection tasks and emergencies in a timely manner, resulting in unreasonable resource allocation, frequent movement of the nanny car and long waiting time, affecting the overall inspection efficiency and stability of the system operation. At the same time, the lack of scientific decision-making basis makes it difficult to achieve the optimal selection of the nanny car parking mode, and it is impossible to give full play to the advantages of the unmanned inspection system to ensure the safety and continuous monitoring of thermal power plant equipment.
[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for planning unmanned inspection tasks in a thermal power plant. By recording and analyzing the tasks and status data of the inspection vehicle, calculating the task weight and weighting the task position, it is ensured that the importance and urgency of the task are accurately reflected in space. By initializing the task data and iteratively calculating, the adjusted weighted center point position tends to be stable, and the docking position of the nanny car is finally determined, thereby improving the service efficiency of the nanny car to the inspection vehicle, optimizing resource allocation, reducing movement and waiting time, and improving the overall execution efficiency of the inspection task and the system operation efficiency. The stability of the adjusted weighted center point position and the change in task urgency are analyzed, and the dynamic scheduling demand index is calculated to determine whether it is necessary to switch the nanny car parking mode. After obtaining the mode switching signal, the task distribution information and response information are extracted, and the parking mode decision coefficient is calculated to optimize resource allocation and task response efficiency. Combined with the spatiotemporal variation index of inspection tasks and the task emergency response coefficient, it is ensured that the nanny car can flexibly respond to task changes and emergencies according to the real-time dynamic scheduling mode or the predictive scheduling mode, improve the response speed and system adaptability, ensure the efficient and stable operation of the unmanned inspection system, and ensure the safety and continuous monitoring of thermal power plant equipment, so as to solve the problems raised in the above background technology.
[0006] To achieve the above object, the present invention provides the following technical solution: a method for planning unmanned inspection tasks in a thermal power plant, comprising the steps of:
[0007] S1, by collecting the task and status data of the inspection vehicle, assigning task weights, determining the weighted center point position, and adjusting the weighted center point position according to the status data, and determining the docking position of the nanny car through iterative calculation;
[0008] S2, analyzing whether it is necessary to switch the parking mode of the nanny car according to the change of the adjusted weighted center point position within the unit time and the change of the urgency of the task within the corresponding time;
[0009] S3, after determining that the parking mode of the nanny car needs to be switched, extract the task distribution information and task response information during the inspection car's task execution process to decide the parking mode of the nanny car.
[0010] In a preferred embodiment, step S1 includes the following contents:
[0011] a1, record the task and status data of each inspection vehicle, including the starting and ending locations of the task, the priority and urgency of the task, the power status and endurance;
[0012] a2. Consider the priority and urgency of the task and calculate the weight of each task: W i =PL i ×UL i;
[0013] Among them, W i is the task weight, PL i For task priority, UL i is the urgency of the task, i is the task number;
[0014] a3, weighted processing of task positions, calculation formula: (x i ,y i )W=(x i ×W i ,y i ×W i );
[0015] Among them, (x i ,y i ) is the latitude and longitude of the mission location, (x i ,y i )W is the weighted center point position;
[0016] a4. Initialize the weights, positions, and endurance data of all tasks, and calculate the new weighted center positions of all tasks. The calculation formulas for the longitude and latitude of the new weighted center points are as follows:
[0017] Among them, WC x and WC y are the longitude and latitude of the weighted center point, i.e., the new weighted center point position, n is the number of tasks, is the weighted sum of the longitudes of all mission locations, is the weighted sum of the latitudes of all task locations;
[0018] a5. Adjust the new weighted center point according to the endurance of the inspection vehicle: WC adj,x =WC x ±Δ x , W.C. adj,y =WC y ±Δ y ;
[0019] Among them, WC adj,x and WC adj,y is the adjusted latitude and longitude of the weighted center point, that is, the adjusted position of the weighted center point, Δ x and Δ y This is an adjustment based on battery life and power status.
[0020] In a preferred embodiment, a6, the docking position of the nanny car is obtained through multiple iterative calculations, and the specific steps of the iteration are as follows:
[0021] a6-1, set the initial weighted center point position and iterative convergence threshold, and set the maximum number of iterations;
[0022] a6-2, in each iteration, based on the current weighted center point, the positions and weights of all tasks relative to the current center point are recalculated through the iterative formula to obtain a new weighted center point position;
[0023] a6-3, based on the endurance of the inspection vehicle, fine-tune the new weighted center point position to obtain the adjusted weighted center point position;
[0024] a6-4, compare the difference between the adjusted weighted center point position and the adjusted weighted center point position calculated last time, if the difference is less than the convergence threshold, the iteration converges and the iteration stops; otherwise, the iteration continues until the maximum number of iterations is reached;
[0025] a6-5, when the convergence condition is met and the maximum number of iterations is reached, the final adjusted weighted center point position is the docking position of the nanny car;
[0026] Among them, the iteration formula is: |WC adj,k+1 -WC adj,k |<ε;
[0027] WC adj,k is the adjusted weighted center point position, WC adj,k+1 is the adjusted weighted center point position calculated last time, ε is the iteration convergence threshold, and k is the current iteration number;
[0028] In a preferred embodiment, step S2 includes the following contents:
[0029] S2-1, in each iteration, record the current adjusted weighted center point position, calculate the position change amplitude by using the Euclidean distance between the adjusted weighted center point position calculated in each iteration and the adjusted weighted center point position in the previous iteration, and then average all position change amplitudes to obtain the stability index;
[0030] S2-2, record the task urgency of each task at multiple time points, compare the task urgency of the same task between two consecutive time points, calculate its change range, collect all the task urgency change range data, and average them to obtain the task urgency index;
[0031] The stability index and the task urgency index are calculated comprehensively to obtain the dynamic scheduling demand index;
[0032] The dynamic scheduling demand index is compared with the mode switching trigger threshold to determine whether the nanny car parking mode needs to be switched; if the dynamic scheduling demand index is greater than or equal to the mode switching trigger threshold, it means that the nanny car parking mode needs to be switched, and a parking mode switching signal is generated; if the dynamic scheduling demand index is less than the mode switching trigger threshold, there is no need to switch the nanny car parking mode.
[0033] In a preferred embodiment, step S3 includes the following contents:
[0034] If a parking mode switching signal is obtained, the task distribution information and task response information are extracted based on the data during the inspection vehicle's task execution, wherein the task distribution information includes the inspection task spatiotemporal variation index, and the task response information includes the task emergency response coefficient. The inspection task spatiotemporal variation index and the task emergency response coefficient are comprehensively processed to obtain the parking mode decision coefficient; the parking mode decision coefficient is compared with the decision threshold, if the parking mode decision coefficient is greater than or equal to the decision threshold, the real-time dynamic scheduling mode needs to be adopted; otherwise, if the parking mode decision coefficient is less than the decision threshold, the predictive scheduling mode is adopted.
[0035] In a preferred embodiment, the logic for obtaining the spatiotemporal variation index of the inspection task is as follows:
[0036] Record the longitude, latitude, altitude, mission start time and end time of the inspection vehicle;
[0037] Divide the inspection route into fixed time periods and space segments;
[0038] In each time period, the distance between all position points and the average position point is calculated to obtain the spatial discreteness of the inspection vehicle position;
[0039] In each spatial segment, the time difference between all time points and the average time point is analyzed to obtain the time discreteness of the inspection task;
[0040] The square sum of all spatial discreteness in each time period is averaged and then the square root is taken to obtain the spatial variation index of the corresponding time period;
[0041] The sum of squares of all time discreteness in each spatial segment is averaged and then the square root is taken to obtain the time variation index of the corresponding spatial segment;
[0042] The spatiotemporal variation index of the inspection task is obtained by combining the square sum of the spatial variation index and the temporal variation index of all time periods and spatial segments and then taking the square root.
[0043] In a preferred embodiment, the logic for obtaining the mission emergency response coefficient is as follows:
[0044] Collect the priority and response time data of all tasks and construct an evaluation matrix. Let the evaluation matrix be A = [a 11 ,a 12 ,…,a 21 ,…,a ov …], where each element a ov represents the vth evaluation index value of the oth task, i.e., priority or response time;
[0045] The evaluation matrix A is standardized to eliminate the dimensional influence of different evaluation indicators, and the element r of the standardized evaluation matrix R is ov Calculated by the following formula: Where n is the number of tasks;
[0046] The hierarchical analysis method is used to assign weights to each evaluation index to form a weight vector U. The element u of the weight vector v satisfy Where m is the number of evaluation indicators;
[0047] Multiply each element in the standardized evaluation matrix R by the corresponding weight to obtain the weighted standardized evaluation matrix Z, whose element z ov The calculation is as follows: ov =u v × ov ;
[0048] Ideal solution A * and negative ideal solution A - Represent the best and worst mission emergency response situations respectively. The elements of the ideal solution and the negative ideal solution are: A * =(max o z ov |v∈J1,min o z ov |v∈J2),A - =(min o z ov |v∈J1,max o z ov |v∈J2);
[0049] Among them, J1 is a set of benefit indicators, and J2 is a set of cost indicators;
[0050] Calculate the difference between each task and the ideal solution A * and negative ideal solution A - The Euclidean distance between the oth task and the ideal solution The distance to the negative ideal solution The calculations are as follows: Where m represents the number of evaluation indicators;
[0051] Calculate the relative proximity C of each task o :
[0052] Count the relative proximity C of all tasks o Take the average to obtain the mission emergency response coefficient.
[0053] A thermal power plant unmanned inspection task planning system, comprising: a data acquisition module, a weighted center point calculation module, a mode switching analysis module and a parking mode decision module;
[0054] The data collection module collects the task data and status data of the inspection vehicle, and transmits the collected task data and status data to the weighted center point calculation module;
[0055] The weighted center point calculation module assigns task weights, determines the weighted center point position, and adjusts the weighted center point position according to the status data. It also determines the docking position of the nanny car through iterative calculation, and passes the determined docking position to the mode switching analysis module;
[0056] The mode switching analysis module analyzes whether it is necessary to switch the parking mode of the nanny car according to the change of the adjusted weighted center point position within the unit time and the change of the urgency of the task within the corresponding time. After determining that the mode switching is necessary, the switching signal and related data are transmitted to the parking mode decision module;
[0057] The parking mode decision module extracts the task distribution information and task response information during the inspection vehicle's task execution, performs comprehensive calculations, and decides the parking mode of the nanny car.
[0058] The technical effects and advantages of the method and system for planning unmanned inspection tasks in a thermal power plant of the present invention are as follows:
[0059] 1. The present invention systematically records and comprehensively analyzes the tasks and status data of the inspection vehicle, calculates the weight of each task, and performs weighted processing on the task position, ensuring that the importance and urgency of the task are accurately reflected in space. After initializing all task data, the initial weighted center point position of the task is calculated, and adjusted according to the endurance of the inspection vehicle. The weighted center point position is stabilized through iterative calculation to ensure that after the convergence conditions are met or the maximum number of iterations is reached, the final weighted center point position is the docking position of the nanny car. The service efficiency of the nanny car for the inspection vehicle is effectively improved, resource allocation is optimized, the movement and waiting time of the nanny car is reduced, the overall execution efficiency of the inspection task is improved, and it is ensured that the inspection vehicle receives timely support and maintenance within the endurance range, thereby improving the overall operation efficiency and safety of the unmanned inspection system of the thermal power plant.
[0060] 2. The present invention calculates the dynamic scheduling demand index by analyzing the stability of the weighted center point position and the change in task urgency to determine whether the nanny car parking mode needs to be switched. The stability of the weighted center point position is obtained by recording and analyzing the position change of the weighted center point in multiple iterative steps, calculating the position change amplitude through Euclidean distance; monitoring the urgency of each task, and recording its changes at different time points, calculating the average value of the task urgency change; comprehensively calculating these two indicators to obtain the dynamic scheduling demand index; comparing the dynamic scheduling demand index with the preset mode switching trigger threshold to determine whether the nanny car parking mode needs to be switched, thereby ensuring the efficient and stable operation of the inspection system.
[0061] 3. After obtaining the parking mode switching signal, the present invention analyzes the data during the inspection vehicle's execution of the task, extracts the task distribution information and task response information, comprehensively calculates the parking mode decision coefficient, and decides the parking mode of the nanny car, effectively optimizing resource allocation and task response efficiency. Combined with the inspection task spatiotemporal variation index and the task emergency response coefficient, it ensures that the nanny car can flexibly respond to task changes and emergencies according to the real-time dynamic scheduling mode or the predictive scheduling mode, improves the system's response speed and adaptability, and reduces unnecessary movement and waiting time, improves resource utilization efficiency and system operation stability, ensures the efficient and stable operation of the unmanned inspection system, and ensures the safety and continuous monitoring of thermal power plant equipment. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] Figure 1 A schematic diagram of a process flow of a method for planning unmanned inspection tasks in a thermal power plant according to the present invention;
[0063] Figure 2 The present invention is a schematic structural diagram of a thermal power plant unmanned inspection task planning system. DETAILED DESCRIPTION
[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0065] like Figure 1 As shown, a method for planning unmanned inspection tasks in a thermal power plant comprises the following steps:
[0066] S1, by collecting the task and status data of the inspection vehicle, assigning task weights, determining the weighted center point position, and adjusting the weighted center point position according to the status data, and determining the docking position of the nanny car through iterative calculation;
[0067] S2, analyzing whether it is necessary to switch the parking mode of the nanny car according to the change of the adjusted weighted center point position within the unit time and the change of the urgency of the task within the corresponding time;
[0068] S3, after determining that the parking mode of the nanny car needs to be switched, extract the task distribution information and task response information during the inspection car's task execution process to decide the parking mode of the nanny car.
[0069] In the unmanned inspection system of thermal power plants, inspection vehicles are widely used to detect and maintain various equipment in the plant to ensure the normal operation and safety of the equipment. Due to the vast area of thermal power plants and the dense and complex distribution of equipment, the temporal and spatial distribution of inspection tasks has great variability and complexity. When the inspection vehicle is performing tasks, its location and task nature directly affect the service efficiency and response speed of the nanny car. Therefore, it is necessary to determine the docking position of the nanny car according to the task planning of the inspection vehicle. A reasonable docking position can significantly improve the scheduling efficiency of the nanny car and the completion quality of the inspection task.
[0070] Different inspection tasks have different priorities and urgency, which leads to significant differences in the inspection vehicle's demand for nanny cars. By accurately determining the docking location of the nanny car, the waiting time and ineffective movement of the inspection vehicle during task execution can be minimized, ensuring that the nanny car can efficiently and quickly provide necessary support and services, improve the operating efficiency and reliability of the entire inspection system, and ultimately achieve the optimal allocation and management of resources. Such decisions help reduce the system's operating costs, reduce the risk of equipment failure, improve the safety and operating efficiency of thermal power plants, and ensure the accuracy and timeliness of inspection tasks.
[0071] The weighted center point location optimization algorithm aims to determine the optimal position of the nanny car when assigning tasks and execution paths to multiple inspection vehicles at the same time. By comprehensively considering the task location, task priority, task urgency and endurance of the inspection vehicle, the position of the nanny car is optimized to improve the efficiency of releasing and storing the inspection vehicle.
[0072] Step S1 includes the following contents:
[0073] a1, record the task and status data of each inspection vehicle, including the starting and ending locations (longitude and latitude) of the task, the priority and urgency of the task, the power status and endurance;
[0074] a2, comprehensively consider the priority and urgency of the task, calculate the weight of each task, which reflects the importance and urgency of the task. The calculation formula is: W i =PL i ×UL i ;
[0075] Among them, W i is the task weight, PL i For task priority, UL i is the urgency of the task, i is the task number;
[0076] Task priority (Priority Level, PL) is an indicator to measure the relative importance of each inspection task in the overall task system. In the unmanned inspection system of a thermal power plant, different equipment and areas have different importance. For example, the normal operation of key equipment such as generators, transformers and boilers is crucial to the safety and efficiency of the power plant. Therefore, the inspection tasks of these key equipment usually have a higher priority. In addition, some inspection tasks may involve the investigation of safety hazards, and the priority of such tasks will also be relatively high. The task priority is comprehensively evaluated and set by the system based on factors such as the importance of the equipment, the type of task and historical data, and is used to guide the scheduling and resource allocation of inspection tasks to ensure that important tasks can be given priority and responded to.
[0077] The urgency level (UL) of a task is an indicator of how quickly each inspection task needs to be responded to and completed. It reflects the time sensitivity and urgency of the task. In the unmanned inspection system of a thermal power plant, tasks with high urgency usually require immediate processing due to equipment abnormalities, fault warnings, or other emergencies. For example, if a device sensor detects an abnormally high temperature, it may indicate that the equipment is about to fail. At this time, the urgency of the inspection task will be very high, and the inspection vehicle will need to respond quickly and check the equipment status. The urgency of the task is dynamically evaluated and set by the system based on factors such as real-time monitoring data, fault warning information, and the nature of the task, and is used to optimize the priority sorting and scheduling strategies of the inspection tasks to ensure that emergency tasks can be responded to quickly and resolved in a timely manner.
[0078] Task priority and task urgency are key scheduling parameters. Task priority determines which tasks should be executed first when resources are limited, while task urgency determines the timeliness of task response. In actual operation, high-priority and high-urgency tasks, such as fault warnings of key equipment, will be given priority by the system to be executed by inspection vehicles to ensure the safe and stable operation of thermal power plants. Ordinary routine inspection tasks are appropriately arranged according to their importance and task urgency, so as to reasonably allocate inspection tasks and routes of inspection vehicles and optimize overall inspection efficiency. By reasonably evaluating and setting task priorities and urgency, the efficient operation of the unmanned inspection system of thermal power plants can be ensured, potential problems can be discovered and handled in a timely manner, and the normal operation of equipment and safe production of power plants can be guaranteed.
[0079] a3, weighting the task location to reflect the spatial impact of the importance and urgency of different tasks. The calculation formula is: (x i ,y i ) W =(x i ×W i ,y i ×W i );
[0080] Among them, (x i ,y i ) is the latitude and longitude of the mission location, (x i ,y i ) W is the weighted center point position.
[0081] a4. Initialize the weights, positions, and endurance data of all tasks, and calculate the new weighted center positions of all tasks. The calculation formulas for the longitude and latitude of the new weighted center points are as follows:
[0082] Among them, WC x and WC y are the longitude and latitude of the new weighted center point, i.e., the new weighted center point position, n is the number of tasks, is the weighted sum of the longitudes of all mission locations, is the weighted sum of the latitudes of all task locations.
[0083] a5. According to the endurance of the patrol vehicle, adjust the new weighted center point position so that the nanny car can serve all patrol vehicles within a reasonable range. By fine-tuning the center point position, it can better meet the actual operation needs. Adjustment formula: WC adj,x =WC x ±Δ x , W.C. adj,y =WC y ±Δ y ;
[0084] Among them, WC adj,x and WC adj,y are the adjusted longitude and latitude of the weighted center point, that is, the adjusted weighted center point position, Δ x and Δ y This is an adjustment based on battery life and power status.
[0085] a6, through multiple iterative calculations, the adjusted weighted center point position tends to be stable, and the docking position of the nanny car is obtained. The iterative process considers the effect of each adjustment and ensures that the final position is optimal. The specific steps of the iteration are as follows:
[0086] a6-1, set the initial weighted center point position and iterative convergence threshold, and set the maximum number of iterations to avoid infinite iterations;
[0087] a6-2, as shown in step a4, in each iteration, based on the current weighted center point position, the positions and weights of all tasks relative to the current center point are recalculated to obtain a new weighted center point position;
[0088] a6-3, as shown in step a5, the new weighted center point position is fine-tuned in combination with the endurance of the inspection vehicle to obtain the adjusted weighted center point position;
[0089] a6-4, compare the difference between the adjusted weighted center point position and the adjusted weighted center point position calculated last time, if the difference is less than the convergence threshold, the iteration converges and the iteration stops; otherwise, the iteration continues until the maximum number of iterations is reached;
[0090] a6-5, when the convergence condition is met and the maximum number of iterations is reached, the final adjusted weighted center point position is the docking position of the nanny car.
[0091] It should be noted that the formula for iterative calculation is: |WC adj,k+1 -WC adj,k |<ε;
[0092] Among them, WC adj,k is the adjusted weighted center point position, WC adj,k+1 is the adjusted weighted center point position calculated last time, ε is the iteration convergence threshold, and k is the current iteration number.
[0093] A nanny car is a special vehicle that supports and maintains unmanned inspection vehicles. It is mainly responsible for providing power supply, equipment maintenance, troubleshooting and transfer services for the inspection vehicles to ensure that the inspection vehicles can perform inspection tasks continuously and efficiently. In the unmanned inspection system of a thermal power plant, the inspection vehicle is responsible for conducting inspections along the predetermined routes within the plant area, monitoring and recording the equipment status in real time, and reporting abnormal situations. Since the battery life of the inspection vehicle is limited and it is prone to failure or requires regular maintenance in a complex inspection environment, the nanny car, as a support platform, can provide the necessary services to the inspection vehicle in a timely manner through reasonable scheduling and positioning, thereby ensuring the normal operation of the inspection vehicle and improving the efficiency and reliability of the entire inspection system. The collaborative working relationship between the nanny car and the inspection vehicle ensures that the unmanned inspection system can operate continuously, stably and efficiently in the thermal power plant.
[0094] The present invention systematically records and comprehensively analyzes the tasks and status data of the inspection vehicle, calculates the weight of each task, and performs weighted processing on the task position, ensuring that the importance and urgency of the task are accurately reflected in space. After initializing all task data, the initial weighted center point position of the task is calculated, and adjusted according to the endurance of the inspection vehicle. The adjusted weighted center point position is stabilized through iterative calculation, ensuring that after the convergence condition is met or the maximum number of iterations is reached, the final adjusted weighted center point position is the docking position of the nanny car. The service efficiency of the nanny car for the inspection vehicle is effectively improved, resource allocation is optimized, the movement and waiting time of the nanny car is reduced, the overall execution efficiency of the inspection task is improved, and it is ensured that the inspection vehicle is supported and maintained in a timely manner within the endurance range, thereby improving the overall operation efficiency and safety of the unmanned inspection system of the thermal power plant.
[0095] Step S2 includes the following contents:
[0096] In the unmanned inspection system of thermal power plants, it is crucial to ensure that the nanny car can efficiently support the operation of the inspection car. To this end, it is necessary to determine the docking position of the nanny car in order to quickly respond to and process the task requirements of the inspection car. However, it is not enough to simply determine the initial position. The inspection tasks and equipment status are dynamically changing, and the position of the nanny car also needs to be adjusted and optimized according to these changes. To this end, it is necessary to analyze the stability of the adjusted weighted center point position, and evaluate the system's operating stability and changes in task requirements by calculating the position change amplitude of the adjusted weighted center point position in multiple iterations. If the adjusted weighted center point position changes greatly, it means that the inspection task position and weight change frequently, and the system operation is unstable. At this time, it may be necessary to switch the parking mode of the nanny car and adopt the real-time dynamic scheduling mode to cope with the frequently changing needs; if the adjusted weighted center point position changes less, it means that the system runs relatively smoothly and the inspection task requirements are stable, and the predictive scheduling mode can continue to be used. Analyzing whether the parking mode selection is triggered can ensure that the scheduling method of the nanny car adapts to actual needs and improve the overall efficiency and reliability of the inspection system. Therefore, calculating and analyzing the stability of the adjusted weighted center point position becomes a key step in determining whether parking mode switching is necessary. Through this process, the most suitable nanny car parking mode can be scientifically selected to ensure the efficient operation of the unmanned inspection system.
[0097] S2-1, the stability calculation process of the adjusted weighted center point position involves recording and analyzing the changes in the adjusted weighted center point position in multiple iteration steps;
[0098] In each iteration, the current adjusted weighted center point position, including longitude and latitude, is recorded. Based on the inspection vehicle's task weight, task location, and endurance, these data are adjusted to calculate the new adjusted weighted center point position. The adjusted weighted center point position calculated in each iteration is compared with the adjusted weighted center point position of the previous iteration through Euclidean distance calculation to obtain the position change amplitude, and then all position change amplitudes are averaged to obtain the stability index;
[0099] It should be noted that the amplitude of change reflects the stability of the adjusted weighted center point position in multiple iterations. Through multiple consecutive iterations, observe the trend of the adjusted weighted center point position change. If the adjusted weighted center point position changes tend to be stable after multiple iterations, it means that the task position and weight change little, and the system runs stably; if the adjusted weighted center point position changes greatly, it means that the task position and weight change frequently, and the system runs unstable. This calculation process helps to determine whether it is necessary to switch the parking mode of the nanny car to ensure the efficiency and stability of the system operation;
[0100] In the unmanned inspection system of thermal power plants, the task urgency index is used to evaluate the urgency of each inspection task to ensure that the task can be responded to and processed in a timely manner. As the operating status of the equipment and environmental conditions continue to change, the urgency of different tasks is also changing dynamically. The purpose of evaluating the task urgency index is to identify urgent tasks that need to be handled first, thereby optimizing the allocation and scheduling of inspection resources and improving the overall response efficiency and reliability of the system. The task urgency index provides a scientific basis for judging the fluctuations in the emergency demand of the system operation by monitoring and analyzing the changes in the urgency of each task at different time points, and provides support for the efficient operation of the unmanned inspection system.
[0101] S2-2, record the task urgency of each task at multiple time points, compare the task urgency of the same task between two consecutive time points, calculate its variation, collect all task urgency variation data, and average them to get the task urgency index. This process can reveal the fluctuation of task urgency requirements. If the average variation is large, it means that the system needs to respond to emergency tasks frequently. Otherwise, it means that the system task requirements are relatively stable. Analyzing the changes in task urgency helps determine whether the parking mode of the nanny car needs to be adjusted to ensure timely response and processing of emergency tasks;
[0102] The stability index and the task urgency index are calculated comprehensively to obtain the dynamic scheduling demand index to analyze whether the nanny car parking mode needs to be switched, for example, through the following calculation formula: DDNI = α × ΔWC avg +β×ΔUL avg ;
[0103] Among them, DDNI represents the dynamic dispatch demand index. When the dynamic dispatch demand index value is high, it indicates that the nanny car docking position changes greatly and the task urgency fluctuates greatly, indicating that the system operation is unstable and it may be necessary to consider switching the nanny car parking mode to cope with the frequently changing demand; when the dynamic dispatch demand index value is low, it indicates that the nanny car docking position changes little and the task urgency fluctuates little, the task demand is relatively stable, and there is no need to switch the nanny car parking mode;
[0104] ΔWC avg Indicates the stability index, indicating the change range of the nanny car docking location; ΔUL avg represents the task urgency index, which indicates the fluctuation of the urgency of task execution; α and β are the weight coefficients of the stability index and the task urgency index respectively.
[0105] Compare the dynamic scheduling demand index with the mode switching trigger threshold to determine whether the nanny car parking mode needs to be switched. If the dynamic scheduling demand index is greater than or equal to the mode switching trigger threshold, it means that the system operation is unstable, the nanny car docking position changes greatly, and the task urgency fluctuates greatly, prompting the system to switch the nanny car parking mode and generate a parking mode switching signal to cope with frequently changing needs and ensure efficient response and processing of inspection tasks. Conversely, if the dynamic scheduling demand index is less than the mode switching trigger threshold, it means that the system operation is stable, the nanny car docking position changes little, and the task urgency fluctuates little, the task demand is relatively stable, and there is no need to switch the nanny car parking mode. The existing scheduling mode can effectively meet the needs of the inspection system and ensure the efficient and stable operation of the system.
[0106] Among them, the mode switching threshold is a preset value used to determine whether it is necessary to switch the nanny car parking mode to cope with changes in system operation and fluctuations in task requirements. The determination of the mode switching threshold is based on historical data analysis. Through long-term monitoring and statistics of system operation data, a critical point that can reflect changes in system stability and task urgency is determined. The specific calculation process includes collecting and analyzing the adjusted weighted center point position change and task urgency change data in multiple historical cycles, calculating the average and standard deviation of the dynamic scheduling demand index, and setting the threshold based on this to ensure that when the dynamic scheduling demand index exceeds this threshold, it indicates that the system operation has entered an unstable state and needs to switch to the nanny car parking mode; otherwise, the existing mode is maintained to ensure efficient and stable operation of the system.
[0107] Among them, the weight coefficient is a parameter used to assign the relative importance of different factors in the comprehensive calculation, reflecting the degree of influence of each factor in the final result. The calculation of the weight coefficient is based on historical data analysis and expert experience. By performing regression analysis and sensitivity analysis on the system operation data, the relative contribution of each factor to the system stability and task urgency changes is determined. The specific calculation process includes collecting a large amount of historical data, using a multivariate regression model to analyze the impact of the adjusted weighted center point position change and task urgency change on the system operation, calculating the regression coefficient of each factor, and determining the value of the weight coefficient based on these coefficients. Through this method, the weight coefficient of each factor can be scientifically determined, so that the comprehensive calculation results can more accurately reflect the system operation status and provide a reliable basis for the decision-making of the nanny car parking mode.
[0108] The present invention calculates the dynamic scheduling demand index by analyzing the stability of the adjusted weighted center point position and the change in task urgency to determine whether the nanny car parking mode needs to be switched. The stability of the adjusted weighted center point position is obtained by recording and analyzing the position change of the weighted center point in multiple iteration steps, calculating the position change amplitude by Euclidean distance; monitoring the urgency of each task, and recording its changes at different time points, calculating the average value of the task urgency change; comprehensively calculating these two indicators to obtain the dynamic scheduling demand index; comparing the dynamic scheduling demand index with the preset mode switching trigger threshold to determine whether the nanny car parking mode needs to be switched, thereby ensuring the efficient and stable operation of the inspection system.
[0109] Step S3 includes the following contents:
[0110] If a parking mode switching signal is obtained, the task distribution information and task response information are extracted based on the data during the inspection vehicle's task execution. The task distribution information includes the inspection task spatiotemporal variation index, and the task response information includes the task emergency response coefficient. The inspection task spatiotemporal variation index and the task emergency response coefficient are comprehensively processed to obtain the parking mode decision coefficient.
[0111] The decision on which mode to adopt for the nanny car is aimed at optimizing resource allocation and task response efficiency, which is of great significance for the unmanned inspection task planning of thermal power plants. By selecting the appropriate scheduling mode, it is ensured that the inspection car can get timely support and maintenance. In the real-time dynamic scheduling mode, it can flexibly respond to task changes and emergencies, and improve the response speed and adaptability of inspection tasks. In the predictive scheduling mode, by planning the location of the nanny car in advance, it can improve resource utilization efficiency and system stability, and reduce unnecessary movement and waiting time. The decision of the two modes is made according to the complexity of the system operating environment and the stability of the task requirements, so as to ensure the efficient and stable operation of the unmanned inspection system and ensure the safety and continuous monitoring of thermal power plant equipment.
[0112] The real-time dynamic scheduling mode dynamically adjusts the location and task allocation of the nanny car based on the real-time data of the inspection task and the current system status to ensure that the needs of the inspection car can be responded to quickly. In this mode, the coordination between the nanny car and the inspection car is highly flexible, which can quickly respond to emergencies and task changes, improve the response speed and system adaptability. However, this mode may cause the nanny car to move frequently, increase energy consumption and scheduling complexity.
[0113] The predictive scheduling mode plans the location and task allocation of the nanny car in advance based on historical data and preset task plans, ensuring that the nanny car is in the best position before the task begins. In this mode, the coordination between the nanny car and the inspection car has high planning and resource utilization efficiency, which can reduce unnecessary movement and waiting time and improve the overall stability and operation efficiency of the system. However, this mode has weak response capabilities to emergencies and is difficult to quickly adjust to meet the task requirements of immediate changes.
[0114] The logic for obtaining the spatiotemporal variation index of inspection tasks is as follows:
[0115] The longitude, latitude, altitude, task start and end time of the inspection vehicle are recorded, and the inspection route is segmented according to fixed time periods and spatial segments.
[0116] In each time period, the distance between all position points and the average position point is calculated to obtain the spatial discreteness of the inspection vehicle position.
[0117] In each spatial segment, the time difference between all time points and the average time point is analyzed to obtain the time discreteness of the inspection task.
[0118] The square sum of all spatial dispersions in each time period is averaged and the square root is taken to obtain the spatial variation index (SVI) of the corresponding time period.
[0119] The sum of the squares of all time discreteness in each spatial segment is averaged and then the square root is taken to obtain the time variation index TVI of the corresponding spatial segment.
[0120] The inspection task spatiotemporal variation index (STVI) is obtained by combining the square sum of the spatial variation index (SVI) and the temporal variation index (TVI) of all time periods and spatial segments and then taking the square root to quantify the overall variation degree of the inspection task in both time and space dimensions.
[0121] Specifically, the spatiotemporal variation index of inspection tasks is a key indicator for quantifying the overall variation of inspection tasks in both time and space. By recording the spatial position and task time data of the inspection vehicle and analyzing and processing these data in detail, the spatiotemporal variation index of inspection tasks can reflect the distribution law of inspection tasks in time and space. A high spatiotemporal variation index of inspection tasks indicates that the distribution of inspection tasks in time and space is relatively discrete and unstable, while a low spatiotemporal variation index of inspection tasks indicates that the distribution of inspection tasks in time and space is relatively concentrated and stable. Combined with the specific application scenarios of thermal power plants, the spatiotemporal variation index of inspection tasks provides a scientific basis for the selection of parking modes for nanny cars. If the spatiotemporal variation index of inspection tasks is high, it is recommended to adopt a real-time dynamic scheduling mode so that the nanny car can flexibly respond to frequently changing inspection needs and ensure rapid response and efficient service; if the spatiotemporal variation index of inspection tasks is low, it is recommended to adopt a predictive scheduling mode to improve resource utilization efficiency and system operation stability by planning and optimizing the location and tasks of the nanny car in advance. The application of the spatiotemporal variation index of inspection tasks helps to optimize the overall efficiency and reliability of the unmanned inspection system and ensure the safe and efficient operation of thermal power plants.
[0122] The logic for obtaining the mission emergency response coefficient is as follows:
[0123] Collect the priority and response time data of all tasks and construct an evaluation matrix. Let the evaluation matrix be A = [a 11 ,a 12 ,…,a 21 ,…,a ov …], where each element a ov represents the vth evaluation index value (i.e., priority or response time) of the oth task;
[0124] The evaluation matrix A is standardized to eliminate the dimensional effects of different evaluation indicators. The element r of the standardized evaluation matrix R ov Calculated by the following formula: Where n is the number of tasks.
[0125] The analytic hierarchy process is used to assign weights to each evaluation index to form a weight vector U. The element u of the weight vector v satisfy Where m is the number of evaluation indicators (here 2, i.e., priority and response time);
[0126] The analytic hierarchy process is a decision analysis method used to determine the weights of multiple indicators. The weights of each evaluation indicator are obtained by constructing a judgment matrix and performing consistency tests.
[0127] Multiply each element in the standardized evaluation matrix R by the corresponding weight to obtain the weighted standardized evaluation matrix Z, whose element zov The calculation is as follows: ov =u v × ov ;
[0128] Ideal solution A * and negative ideal solution A - Represent the best and worst mission emergency response situations respectively. The elements of the ideal solution and negative ideal solution are: A * =(max o z ov |v∈J1,min o z ov |v∈J2),A - =(min o z ov |v∈J1,max o z ov |v∈J2);
[0129] Among them, J1 is a set of benefit indicators (such as priority), and J2 is a set of cost indicators (such as response time).
[0130] Calculate the difference between each task and the ideal solution A * and negative ideal solution A - The Euclidean distance between the oth task and the ideal solution The distance to the negative ideal solution The calculations are as follows: Where m represents the number of evaluation indicators;
[0131] Calculate the relative proximity C of each task o :
[0132] Count the emergency response coefficient C of all tasks o Take the average to obtain the mission emergency response coefficient.
[0133] The task emergency response coefficient is used to represent and reflect the task in the system and its demand for rapid response. The larger the coefficient, the more rapid the response and processing. The smaller the coefficient, the less urgent the task and the more relaxed the response demand. The task emergency response coefficient provides important implications for the parking mode of the nanny car. A higher emergency response coefficient indicates that there are a large number of emergency tasks in the system. It is recommended to adopt a real-time dynamic scheduling mode to ensure that the nanny car can flexibly respond to emergency tasks and improve the response speed. A lower emergency response coefficient indicates that the system is running more stably. It is recommended to adopt a predictive scheduling mode to improve resource utilization efficiency and the overall stability of the system by planning and optimizing the location and tasks of the nanny car in advance.
[0134] The parking mode decision coefficient is calculated by comprehensively processing the inspection task spatiotemporal variation index and the task emergency response coefficient to help decide the parking mode of the nanny car. For example, it is calculated by the following calculation formula:
[0135] Among them, JC represents the parking mode decision coefficient, STVI and TRR are the inspection task spatiotemporal variation index and task emergency response coefficient, respectively, and f1 and f2 are the preset proportional coefficients of the inspection task spatiotemporal variation index and task emergency response coefficient, respectively.
[0136] The parking mode decision coefficient is used to represent and reflect the adaptability and optimization degree of the nanny car parking mode. The larger the coefficient, the higher the spatiotemporal variability and emergency response requirements of the inspection task, and the more complex and changeable the operating environment. It is necessary to adopt a real-time dynamic scheduling mode to flexibly respond to frequently changing task requirements and ensure timely response and processing of inspection tasks; the smaller the coefficient, the lower the spatiotemporal variability and emergency response requirements of the inspection task, and the more stable the operating environment. The predictive scheduling mode can be adopted to improve resource utilization efficiency and system operation stability by planning and optimizing the location and tasks of the nanny car in advance. The size of the parking mode decision coefficient provides a scientific basis for selecting the most suitable parking mode for the nanny car, ensuring that the unmanned inspection system can operate efficiently and stably under various operating environments.
[0137] The parking mode decision coefficient is compared with the decision threshold. If the parking mode decision coefficient is greater than or equal to the decision threshold, it means that the system operating environment is complex and changeable, and the spatiotemporal variability and emergency response requirements of the inspection tasks are high. It is necessary to adopt a real-time dynamic scheduling mode to flexibly respond to frequently changing task requirements and ensure timely response and processing of inspection tasks. On the contrary, if the parking mode decision coefficient is less than the decision threshold, it means that the system operating environment is relatively stable, and the spatiotemporal variability and emergency response requirements of the inspection tasks are low. The predictive scheduling mode can be adopted to improve resource utilization efficiency and the overall stability of the system by planning and optimizing the location and tasks of the nanny car in advance.
[0138] After obtaining the parking mode switching signal, the present invention analyzes the data during the inspection vehicle's execution of the task, extracts the task distribution information and task response information, comprehensively calculates the parking mode decision coefficient, and decides the parking mode of the nanny car, effectively optimizing resource allocation and task response efficiency. Combined with the inspection task spatiotemporal variation index and the task emergency response coefficient, it ensures that the nanny car can flexibly respond to task changes and emergencies according to the real-time dynamic scheduling mode or the predictive scheduling mode, improves the system's response speed and adaptability, and reduces unnecessary movement and waiting time, improves resource utilization efficiency and system operation stability, ensures the efficient and stable operation of the unmanned inspection system, and ensures the safety and continuous monitoring of thermal power plant equipment.
[0139] like Figure 2 The invention provides a thermal power plant unmanned inspection task planning system, comprising: a data acquisition module, a weighted center point calculation module, a mode switching analysis module and a parking mode decision module;
[0140] The data collection module collects the task data and status data of the inspection vehicle, and transmits the collected task data and status data to the weighted center point calculation module;
[0141] The weighted center point calculation module assigns task weights, determines the weighted center point position, and adjusts the weighted center point position according to the status data. It also determines the docking position of the nanny car through iterative calculation, and passes the determined docking position to the mode switching analysis module;
[0142] The mode switching analysis module analyzes whether it is necessary to switch the parking mode of the nanny car according to the change of the adjusted weighted center point position within the unit time and the change of the urgency of the task within the corresponding time. After determining that the mode switching is necessary, the switching signal and related data are transmitted to the parking mode decision module;
[0143] The parking mode decision module extracts the task distribution information and task response information during the inspection vehicle's task execution, performs comprehensive calculations, and decides the parking mode of the nanny car.
[0144] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0145] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.
[0146] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0147] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for planning unmanned inspection tasks in a thermal power plant, characterized in that: Includes steps: S1, by collecting the task and status data of the inspection vehicle, assigning task weights, determining the weighted center point position, and adjusting the weighted center point position according to the status data, and determining the docking position of the nanny car through iterative calculation; S2, analyzing whether it is necessary to switch the parking mode of the nanny car according to the change of the adjusted weighted center point position within the unit time and the change of the urgency of the task within the corresponding time; S3, after determining that the parking mode of the nanny car needs to be switched, extract the task distribution information and task response information during the inspection car's task execution process to decide the parking mode of the nanny car.
2. A method for planning unmanned inspection tasks in a thermal power plant according to claim 1, characterized in that: Step S1 includes the following contents: a1, record the task and status data of each inspection vehicle, including the starting and ending locations of the task, the priority and urgency of the task, the power status and endurance; a2. Consider the priority and urgency of the task and calculate the weight of each task: W i =PL i ×UL i ; Among them, W i is the task weight, PL i For task priority, UL i is the urgency of the task, i is the task number; a3, weighted processing of task positions, calculation formula: (x i ,y i ) W =(x i ×W i ,y i ×W i ); Among them, (x i ,y i ) is the latitude and longitude of the mission location, (x i ,y i ) W is the weighted center point position; a4. Initialize the weights, positions, and endurance data of all tasks, and calculate the new weighted center positions of all tasks. The calculation formulas for the longitude and latitude of the new weighted center points are as follows: Among them, WC x and WC y are the longitude and latitude of the new weighted center point, i.e., the new weighted center point position, n is the number of tasks, is the weighted sum of the longitudes of all mission locations, is the weighted sum of the latitudes of all task locations; a5. Adjust the new weighted center point according to the endurance of the inspection vehicle: WC adj,x =WC x ±Δ x , W.C. adj,y =WC y ±Δ y ; Among them, WC adj,x and WC adj,y is the adjusted latitude and longitude of the weighted center point, that is, the adjusted weighted center point position, Δ x and Δ y The adjustment amount is based on battery life and power status.
3. A method for planning unmanned inspection tasks in a thermal power plant according to claim 2, characterized in that: a6, through multiple iterative calculations, the docking position of the nanny car is obtained. The specific steps of the iteration are as follows: a6-1, set the initial weighted center point position and iterative convergence threshold, and set the maximum number of iterations; a6-2, in each iteration, based on the current weighted center point, the positions and weights of all tasks relative to the current center point are recalculated through the iterative formula to obtain a new weighted center point position; a6-3, based on the endurance of the inspection vehicle, fine-tune the new weighted center point position to obtain the adjusted weighted center point position; a6-4, compare the difference between the adjusted weighted center point position and the adjusted weighted center point position calculated last time, if the difference is less than the convergence threshold, the iteration converges and the iteration stops; otherwise, the iteration continues until the maximum number of iterations is reached; a6-5, when the convergence condition is met and the maximum number of iterations is reached, the final adjusted weighted center point position is the docking position of the nanny car; The iteration formula is: |WC adj,k+1 -WC adj,k |<ε; Among them, WC adj,k is the adjusted weighted center point position, WC adj,k+1 is the adjusted weighted center point position calculated last time, ε is the iteration convergence threshold, and k is the current iteration number.
4. A method for planning unmanned inspection tasks in a thermal power plant according to claim 3, characterized in that: Step S2 includes the following contents: S2-1, in each iteration, record the current adjusted weighted center point position, calculate the position change amplitude by using the Euclidean distance between the adjusted weighted center point position calculated in each iteration and the adjusted weighted center point position in the previous iteration, and then average all position change amplitudes to obtain the stability index; S2-2, record the task urgency of each task at multiple time points, compare the task urgency of the same task between two consecutive time points, calculate its change range, collect all the task urgency change range data, and average them to obtain the task urgency index; The stability index and the task urgency index are calculated comprehensively to obtain the dynamic scheduling demand index; Compare the dynamic dispatch demand index with the mode switching trigger threshold to determine whether the nanny car parking mode needs to be switched; If the dynamic dispatch demand index is greater than or equal to the mode switching trigger threshold, it means that the nanny car parking mode needs to be switched, and a parking mode switching signal is generated; If the dynamic scheduling demand index is less than the mode switching trigger threshold, there is no need to switch the nanny car parking mode.
5. A method for planning unmanned inspection tasks in a thermal power plant according to claim 4, characterized in that: Step S3 includes the following contents: If a parking mode switching signal is obtained, based on the data of the inspection vehicle during the task execution, the task distribution information and the task response information are extracted, wherein the task distribution information includes the inspection task spatiotemporal variation index, and the task response information includes the task emergency response coefficient. The inspection task spatiotemporal variation index and the task emergency response coefficient are comprehensively processed to obtain the parking mode decision coefficient; The parking mode decision coefficient is compared with the decision threshold. If the parking mode decision coefficient is greater than or equal to the decision threshold, the real-time dynamic scheduling mode needs to be adopted; otherwise, if the parking mode decision coefficient is less than the decision threshold, the predictive scheduling mode is adopted.
6. A method for planning unmanned inspection tasks in a thermal power plant according to claim 5, characterized in that: The logic for obtaining the spatiotemporal variation index of inspection tasks is as follows: Record the longitude, latitude, altitude, task start time and end time of the inspection vehicle; at the same time, divide the inspection route into fixed time periods and space segments; In each time period, the distance between all position points and the average position point is calculated to obtain the spatial discreteness of the inspection vehicle position; In each spatial segment, the time difference between all time points and the average time point is analyzed to obtain the time discreteness of the inspection task; The square sum of all spatial discreteness in each time period is averaged and then the square root is taken to obtain the spatial variation index of the corresponding time period; The sum of squares of all time discreteness in each spatial segment is averaged and then the square root is taken to obtain the time variation index of the corresponding spatial segment; The spatiotemporal variation index of the inspection task is obtained by combining the square sum of the spatial variation index and the temporal variation index of all time periods and spatial segments and then taking the square root.
7. A method for planning unmanned inspection tasks in a thermal power plant according to claim 5, characterized in that: The logic for obtaining the mission emergency response coefficient is as follows: Collect the priority and response time data of all tasks and construct the evaluation matrix A = [a 11 ,a 12 ,…,a 21 ,…,a ov …], where each element a ov represents the vth evaluation index value of the oth task, i.e., priority or response time; The evaluation matrix A is standardized to eliminate the dimensional influence of different evaluation indicators, and the element r of the standardized evaluation matrix R is ov Calculated by the following formula: Where n is the number of tasks; The hierarchical analysis method is used to assign weights to each evaluation index to form a weight vector U. The element u of the weight vector v satisfy Where m is the number of evaluation indicators; Multiply each element in the standardized evaluation matrix R by the corresponding weight to obtain the weighted standardized evaluation matrix Z, whose element z ov The calculation is as follows: ov =u v × ov ; Ideal solution A * and negative ideal solution A - Represent the best and worst mission emergency response situations respectively. The elements of the ideal solution and the negative ideal solution are: A * =(max o z ov |v∈J1,min o z ov |v∈J2),A - =(min o z ov |v∈J1,max o z ov |v∈J2); Among them, J1 is a set of benefit indicators, and J2 is a set of cost indicators; Calculate the difference between each task and the ideal solution A * and negative ideal solution A - The Euclidean distance between the oth task and the ideal solution The distance to the negative ideal solution The calculations are as follows: Where m represents the number of evaluation indicators; Calculate the relative proximity C of each task o : Count the relative proximity C of all tasks o Take the average to obtain the mission emergency response coefficient.
8. A system for planning unmanned inspection tasks in a thermal power plant, used to implement a method for planning unmanned inspection tasks in a thermal power plant as claimed in any one of claims 1 to 7, characterized in that: include: Data acquisition module, weighted center point calculation module, mode switching analysis module and parking mode decision module; The data collection module collects the task data and status data of the inspection vehicle, and transmits the collected task data and status data to the weighted center point calculation module; The weighted center point calculation module assigns task weights, determines the weighted center point position, and adjusts the weighted center point position according to the status data. It also determines the docking position of the nanny car through iterative calculation, and passes the determined docking position to the mode switching analysis module; The mode switching analysis module analyzes whether it is necessary to switch the parking mode of the nanny car according to the change of the adjusted weighted center point position within the unit time and the change of the urgency of the task within the corresponding time. After determining that the mode switching is necessary, the switching signal and related data are transmitted to the parking mode decision module; The parking mode decision module extracts the task distribution information and task response information during the inspection vehicle's task execution, performs comprehensive calculations, and decides the parking mode of the nanny car.
