New energy station inspection control method and system based on data mining

Abstract

The invention discloses a new energy station inspection control method and system based on data mining, and relates to the field of data mining, and the method comprises the steps: obtaining the environment information and operation parameter information of a target station; on the basis of the environment information and the operation parameter information, inspection task information of each to-be-inspected sub-region is determined; performing task allocation based on the position information of each to-be-inspected sub-region and the inspection task information of each inspection sub-region to determine inspection robot inspection sub-task information, inspection unmanned aerial vehicle sub-task information and inspection personnel sub-task information; the inspection robot inspection subtask information is sent to the inspection robot, the inspection unmanned aerial vehicle subtask information is sent to the inspection unmanned aerial vehicle, and the inspection personnel subtask information is sent to a mobile terminal corresponding to the inspection personnel; the inspection robot, the inspection unmanned aerial vehicle and the inspection personnel execute the corresponding inspection subtasks.

Description

Technical Field

[0001] This specification relates to the field of patrol control. More specifically, the present application relates to a patrol control method and system for new energy power stations based on data mining. Background Art

[0002] In new energy power stations, the patrol work is crucial for the normal operation of equipment and fault prevention. Traditional patrol methods mainly rely on manual patrols. Patrol personnel regularly go to the equipment area for inspections according to daily experience and the management requirements of the power station. However, with the expansion of the power station scale and the increase in equipment complexity, there are many problems with manual patrols:

[0003] 1. Low efficiency: Manual patrols take a long time and it is difficult to complete the patrol work of a large area or complex equipment in a short time. Especially in bad weather or emergencies, the patrol difficulty further increases.

[0004] 2. Lack of real-time performance: Due to the limitations of patrol frequency and personnel scheduling, it is difficult for traditional patrols to detect sudden equipment failures or drastic environmental changes in a timely manner.

[0005] 3. Low patrol accuracy: Manual patrols rely on the experience level of patrol personnel and are easily affected by subjective factors, which may lead to the omission of minor problems and affect the operation safety of equipment.

[0006] With the rapid development of intelligent patrol technology, automated patrol equipment (such as patrol robots and drones) has gradually been applied to power station patrols. These devices can collect environmental information and operating parameter information in real time through sensors. However, how to achieve reasonable task allocation and priority management in the case of multi-device patrols is still an urgent problem to be solved. Therefore, an intelligent-based patrol task allocation method to rationally and efficiently utilize various patrol devices to meet the patrol requirements of different regions has become the development direction of current technology. Summary of the Invention

[0007] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further elaborated in the Detailed Description section. The Summary of the Invention section of the present application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the protection scope of the claimed technical solution.

[0008] In a first aspect, the present application proposes a patrol control method for new energy power stations based on data mining, including:

[0009] Obtain the environmental information and operating parameter information of the target power station, where the environmental information includes rain and snow information, environmental temperature information, and humidity information, and the above-mentioned operating parameter information includes equipment temperature information, current information, and voltage information;

[0010] Based on the above environmental information and the above operating parameter information, determine the inspection task information for each sub-area to be inspected, where the above inspection task information includes inspection type information and inspection priority information;

[0011] Based on the location information of each of the above sub-areas to be inspected and the above inspection task information of each of the above inspection sub-areas, perform task allocation to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection UAV, and the inspection sub-task information of the inspection personnel;

[0012] Send the above inspection sub-task information of the inspection robot to the inspection robot, send the above inspection sub-task information of the inspection UAV to the inspection UAV, and send the above inspection sub-task information of the inspection personnel to the mobile terminal corresponding to the inspection personnel, so that the above inspection robot, the above inspection UAV, and the above inspection personnel execute the corresponding inspection sub-tasks.

[0013] In a feasible implementation manner, the above determining the inspection task information for each sub-area to be inspected based on the above environmental information and the above operating parameter information includes:

[0014] Divide the above target station into multiple sub-areas to be inspected based on a preset area;

[0015] Based on the above environmental information and the above operating parameter information of each of the above sub-areas to be inspected, calculate the inspection type score;

[0016] Based on the above inspection type score and the inspection type score threshold, determine the above inspection type information corresponding to the sub-area to be inspected, where the above inspection type information includes equipment failure inspection, environmental inspection, and regular inspection;

[0017] Based on the above environmental information and the above operating parameter information of the sub-area to be inspected, determine the above inspection priority information based on the fuzzy logic method and the particle swarm optimization method.

[0018] In a feasible implementation manner, determine the above inspection type score based on the following formula:

[0019]

[0020] In the formula, S type is the above inspection type score, W temp is the weight coefficient corresponding to the temperature factor, A is the attenuation factor of the temperature influence, T temp is the current environmental temperature information, T max is the maximum temperature reference information, W humidity is the weight information corresponding to the humidity factor, H humidity is the current humidity information, H maxis the maximum humidity reference information, W rain is the weight coefficient corresponding to the rain and snow factor, R rain is the current rain and snow information value, W devicetemp is the weight coefficient corresponding to the equipment factor, T device is the current temperature information of the equipment, T devicemax is the maximum reference temperature information of the equipment, W current is the weight information corresponding to the current factor, I current is the current current information, I norm is the normal current reference information, I max is the maximum current reference information, W voltage is the weight information corresponding to the voltage factor, V voltage is the current voltage information, V norm is the normal voltage reference information, V max is the maximum voltage reference information.

[0021] In a feasible implementation manner, the above-mentioned inspection priority information is determined based on the above-mentioned environmental information and the above-mentioned operating parameter information of the to-be-inspected sub-region by using the fuzzy logic method and the particle swarm optimization method, including:

[0022] Perform a preprocessing operation on the above-mentioned environmental information and the above-mentioned operating parameter information of each to-be-inspected sub-region to obtain the processed environmental information and the processed operating parameter information corresponding to each to-be-inspected sub-region;

[0023] Perform a fuzzification process on the processed environmental information and the processed operating parameter information corresponding to each to-be-inspected sub-region to obtain the fuzzy set corresponding to each to-be-inspected sub-region;

[0024] Perform a fuzzy mapping operation based on the fuzzy set and the fuzzy rule base corresponding to each to-be-inspected sub-region to obtain the fuzzy level of the inspection priority corresponding to each to-be-inspected sub-region;

[0025] Perform a defuzzification process on the fuzzy level of the inspection priority corresponding to each to-be-inspected sub-region to obtain the initial inspection priority score of each to-be-inspected sub-region;

[0026] Perform an initial particle swarm operation based on the initial inspection priority score of each to-be-inspected sub-region to obtain the initial position information of the particle swarm and the initial velocity information of the particle swarm;

[0027] Perform a fitness calculation on the initial position information of the particle swarm based on the target fitness function to obtain the fitness value information of each particle;

[0028] Based on the above particle swarm initial position information, the above particle swarm initial velocity information, and the fitness information of the above particles, perform particle swarm iterative update operations to obtain the inspection priority information for each of the above sub-regions to be inspected.

[0029] In a feasible implementation, the above objective fitness function Fitness(x) is:

[0030] Fitness(x) = w1·E(x) + w2·S(x) - w3·C(x)

[0031]

[0032] C(x) = B·Inspection distance + C·Inspection time

[0033] In the formula, E(x) is the environmental matching degree, S(x) is the importance of equipment status, C(x) is the inspection cost, w1, w2, and w3 are all weight coefficients corresponding to each factor, T(x) is the temperature information of the current sub-region, H(x) is the humidity information of the current sub-region, T 目标 is the desired temperature information, H 目标 is the desired humidity information, B is the weight coefficient corresponding to the inspection distance, and C is the weight coefficient corresponding to the inspection time.

[0034] In a feasible implementation, perform task allocation based on the position information of each of the above sub-regions to be inspected and the above inspection task information of each of the above inspection sub-regions to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection drone, and the inspection sub-task information of the inspection personnel, including:

[0035] Perform task allocation based on the position information of each of the above sub-regions to be inspected and the above inspection task information of each of the above inspection sub-regions based on the multi-objective particle swarm optimization algorithm to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection drone, and the inspection sub-task information of the inspection personnel, where the objective functions corresponding to the above multi-objective particle swarm optimization algorithm include a path length objective function, a task priority objective function, and an inspection characteristic matching objective function.

[0036] In a feasible implementation, the above path length objective function f1 is determined based on the following formula:

[0037]

[0038] where, distance(p i , p i+1 ) represents the Euclidean distance from point p i to point p i+1 , and d(p i ) is at p iThe residence time at the position, α is the decay factor for adjusting the path weight, θ(p i ,p i+1 ) is the turning angle between two positions, β is the turning penalty coefficient, p i and p i+1 are two adjacent inspection points, and n is the total number of inspection points.

[0039] In a feasible implementation manner, the above task priority objective function f2 is determined based on the following formula:

[0040]

[0041] Among them, priority(T j ) represents the priority of the task, time(T j ) is the expected completion time of the task, actual_time(T j ) is the actual completion time of the task, γ is the delay penalty coefficient, and m represents the total number of tasks.

[0042] In a feasible implementation manner, the above inspection feature matching objective function f3 is determined based on the following formula:

[0043]

[0044] Among them, D k represents the inspection execution body k. The above inspection execution body includes inspection robots, inspection drones, and inspection personnel. T j represents the task j. Compatibility(D k ,T j ) is the fitness of D k for T j , compatibility(D k ,T j ) ∈ [0,1], C energy (D k ) represents the unit energy consumption of D k . The above inspection drones and the above inspection robots have specific energy consumption values. The energy consumption of the above inspection personnel is calculated according to labor cost and time cost. C reliability (D k ,T j ) represents the reliability penalty coefficient when D k executes T j , and C cormplexity (D k ,T j ) represents the energy consumption when D k executes T jRegarding the task complexity matching, δ is the weight coefficient corresponding to the above reliability penalty coefficient, and η is the weight coefficient corresponding to the above complexity penalty coefficient.

[0045] In a second aspect, a new energy power station inspection and control system based on data mining according to the present application includes:

[0046] An acquisition unit configured to acquire the environmental information and operation parameter information of a target power station. The environmental information includes rain and snow information, environmental temperature information, and humidity information, and the above operation parameter information includes equipment temperature information, current information, and voltage information;

[0047] A first determination unit configured to determine inspection task information based on the above environmental information and the above operation parameter information. The inspection task information includes inspection type information and inspection priority information;

[0048] A second determination unit configured to perform task allocation based on the above inspection task information to determine inspection robot sub-task information, inspection drone sub-task information, and inspection personnel sub-task information;

[0049] A control unit configured to send the above inspection robot sub-task information to the inspection robot, send the above inspection drone sub-task information to the inspection drone, and send the above inspection personnel sub-task information to the mobile terminal corresponding to the inspection personnel, so that the above inspection robot, the above inspection drone, and the above inspection personnel execute the corresponding inspection sub-tasks.

[0050] In summary, compared with the traditional manual inspection method, this application realizes the automatic allocation of inspection tasks based on data mining through an intelligent method, and integrates the advantages of various inspection equipment (such as inspection robots, inspection drones and manual inspection personnel), realizing a more efficient and accurate inspection process for new energy stations. By dividing the station into multiple sub-areas to be inspected, and dynamically calculating the inspection type score and inspection priority information based on the environmental information and operating parameter information of each sub-area, the accurate allocation of tasks can be achieved. Compared with traditional manual inspections, it can quickly identify the inspection needs of different areas, automatically allocate suitable inspection equipment, and improve the inspection efficiency. Based on the environmental information and operating parameter information of each sub-area, this application uses fuzzy logic and particle swarm optimization to analyze the inspection tasks, intelligently judge the needs of equipment failure, environmental inspection problems or daily regular inspections, and ensure that high-risk or sudden failure areas are inspected first through inspection type scoring and priority judgment. This application can automatically allocate inspection subtasks of inspection robots, inspection drones and manual inspection personnel based on the inspection task type and sub-area location, and realize the collaborative work of multiple devices. Inspection robots are suitable for ground tasks, drones are suitable for high-altitude and large-scale inspection tasks, and manual inspection personnel are suitable for complex tasks or tasks with high safety requirements. In this way, various equipment can be used efficiently in the tasks they are good at, reducing inspection time and labor costs. The inspection subtask information is automatically sent to the corresponding inspection equipment. After receiving the task, the inspection robot, inspection drone and inspection personnel can immediately start the inspection task to achieve real-time response. Compared with traditional methods, the present application can quickly start the inspection equipment after the task is generated, reducing the response time of the inspection, and is especially suitable for scenarios that require emergency response. In summary, the present application realizes efficient, real-time and multi-device collaborative inspection of new energy stations through data-driven intelligent inspection task allocation, greatly improves the inspection quality and efficiency, optimizes the shortcomings of traditional inspection methods, and contributes to the safe and stable operation of station equipment.

[0051] The new energy station inspection and control method based on data mining proposed in this application, and other advantages, objectives and features of this application will be reflected in part through the following description, and in part will also be understood by technical personnel in this field through research and practice of this application. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0053] Figure 1A flowchart of a new energy power station inspection control method based on data mining provided by an embodiment of the present application;

[0054] Figure 2 A new energy power station inspection system based on data mining provided by an embodiment of the present application. Detailed implementation manners

[0055] The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order different from that shown or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily limit to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0056] Please refer to Figure 1 , a flowchart of a new energy power station inspection control method based on data mining provided by an embodiment of the present application, including:

[0057] S110. Obtain the environmental information and operation parameter information of the target power station. Among them, the environmental information includes rain and snow information, environmental temperature information and humidity information, and the above operation parameter information includes equipment temperature information, current information and voltage information.

[0058] Exemplarily, obtain the environmental information and operation parameter information of the target power station. The environmental information includes rain and snow information, environmental temperature information and humidity information of the power station, which is used to evaluate the impact of the current weather and environmental conditions on the inspection task. The operation parameter information includes equipment temperature information, current information and voltage information, which helps to judge the operation status of the equipment (such as blades, gearboxes, generators, busbar boxes and photovoltaic modules, etc.) and whether there are abnormal conditions, and will be an important basis for judging the inspection requirements.

[0059] Collect the environmental information and operation parameter information through the sensor devices deployed at the power station. For example, temperature sensors record the current environmental temperature information, humidity sensors monitor the air humidity, and current and voltage monitoring devices are used to obtain the current information and voltage information of the power station equipment. The equipment temperature information can be measured by thermocouples, thermistors and infrared temperature sensors set on the equipment.

[0060] S120. Based on the above environmental information and the above operating parameter information, determine the inspection task information for each sub-region to be inspected, where the above inspection task information includes inspection type information and inspection priority information.

[0061] Exemplarily, based on the obtained environmental information and operating parameter information, determine the inspection task information for each sub-region to be inspected. Among them, the inspection task information includes inspection type information and inspection priority information. The inspection type information indicates the specific inspection type required for this sub-region, such as equipment failure inspection, environmental inspection, or regular inspection. The inspection priority information is prioritized according to the importance or urgency of the task to ensure that high-priority inspection tasks are executed first.

[0062] If the equipment temperature information of a certain sub-region abnormally increases, the inspection task information of this region will be set as equipment failure inspection, and the priority will also be increased to promptly investigate potential equipment failure hazards; if the environmental information shows that the humidity in this region increases, the task type will be set as environmental inspection.

[0063] S130. Based on the location information of each of the above sub-regions to be inspected and the above inspection task information of each of the above inspection sub-regions, perform task allocation to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection drone, and the inspection sub-task information of the inspection personnel.

[0064] Exemplarily, through analyzing the location information of each sub-region to be inspected and the corresponding inspection task information, perform intelligent task allocation for the inspection tasks. The allocation results will generate the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection drone, and the inspection sub-task information of the inspection personnel. The allocation rules are based on the task type, equipment characteristics, and inspection priority. For example, for ground tasks or high-priority inspection tasks with fixed paths, they are preferentially allocated to the inspection robot; tasks suitable for high altitudes or large areas are preferentially allocated to the inspection drone; complex or high-risk tasks are allocated to the inspection personnel.

[0065] If the task of a certain sub-region is determined to be equipment failure inspection, and this region is on the ground and can be reached by the inspection robot, then this task is allocated to the inspection robot; for tasks with higher priority and larger coverage, the inspection drone will be preferentially selected to execute the task; when the task complexity is high or the equipment is not suitable, it is allocated to the manual inspection personnel.

[0066] S140. Send the above inspection sub-task information of the inspection robot to the inspection robot, send the above inspection sub-task information of the inspection drone to the inspection drone, and send the above inspection sub-task information of the inspection personnel to the mobile terminal corresponding to the inspection personnel, so that the above inspection robot, the above inspection drone, and the above inspection personnel executor perform the corresponding inspection sub-tasks.

[0067] Exemplarily, the generated inspection robot subtask information is sent to the inspection robot, the inspection drone subtask information is sent to the inspection drone, and the inspection personnel subtask information is sent to the mobile terminal corresponding to the inspection personnel, so that the executors corresponding to the inspection robot, inspection drone and inspection personnel respectively receive their respective inspection subtasks and immediately perform the inspection work according to the task information.

[0068] Send inspection subtask information, such as inspection area and task type, to the control module of the inspection robot, so that it can inspect the specified equipment along the specified path. After receiving the inspection task information, the drone automatically flies to the specified area and starts high-altitude inspection. The inspection personnel receive the inspection subtask information through the mobile terminal and go to the specified area to manually complete the inspection task.

[0069] In summary, compared with the traditional manual inspection method, this application realizes the automatic allocation of inspection tasks based on data mining through an intelligent method, and integrates the advantages of various inspection equipment (such as inspection robots, inspection drones and manual inspection personnel) to achieve a more efficient and accurate inspection process for new energy stations. By dividing the station into multiple sub-areas to be inspected, and dynamically calculating the inspection type score and inspection priority information based on the environmental information and operating parameter information of each sub-area, the accurate allocation of tasks can be achieved. Compared with traditional manual inspections, it can quickly identify the inspection needs of different areas, automatically allocate suitable inspection equipment, and improve the inspection efficiency. Based on the environmental information and operating parameter information of each sub-area, this application uses fuzzy logic and particle swarm optimization to analyze the inspection tasks, intelligently judge the needs of equipment failure, environmental problems or daily inspections, and ensure that high-risk or sudden failure areas are given priority inspections through inspection type scoring and priority judgment. This application can automatically allocate inspection subtasks of inspection robots, inspection drones and manual inspection personnel based on the inspection task type and sub-area location, and realize the collaborative work of multiple devices. Inspection robots are suitable for ground tasks, drones are suitable for high-altitude and large-scale inspection tasks, and manual inspection personnel are suitable for complex tasks or tasks with high safety requirements. In this way, various equipment can be used efficiently in the tasks they are good at, reducing inspection time and labor costs. The inspection subtask information is automatically sent to the corresponding inspection equipment. After receiving the task, the inspection robot, inspection drone and inspection personnel can immediately start the inspection task to achieve real-time response. Compared with traditional methods, the present application can quickly start the inspection equipment after the task is generated, reducing the response time of the inspection, and is especially suitable for scenarios that require emergency response. In summary, the present application realizes efficient, real-time and multi-device collaborative inspection of new energy stations through data-driven intelligent inspection task allocation, greatly improves the inspection quality and efficiency, optimizes the shortcomings of traditional inspection methods, and contributes to the safe and stable operation of station equipment.

[0070] In a feasible implementation manner, determining the inspection task information of each sub-area to be inspected based on the above environmental information and the above operation parameter information includes:

[0071] Dividing the above target station into multiple sub-areas to be inspected based on a preset area.

[0072] Calculating an inspection type score based on the above environmental information and the above operation parameter information of each of the above sub-areas to be inspected.

[0073] Determining the above inspection type information corresponding to the sub-area to be inspected based on the above inspection type score and an inspection type score threshold, where the above inspection type information includes equipment failure inspection, environmental inspection, and regular inspection.

[0074] Determining the above inspection priority information based on the above environmental information and the above operation parameter information of the sub-area to be inspected by using the fuzzy logic method and the particle swarm optimization method.

[0075] Exemplarily, according to the total area of the target station, the target station is divided into several sub-areas to be inspected. The division of each sub-area is based on a preset area to ensure that the distribution of inspection tasks is uniform and covers all areas of the station. This division enables the system to process each sub-area to be inspected one by one in subsequent inspection task allocation, facilitating the customization of inspection tasks according to the characteristics of each area. For example, the target station can be divided into 5 sub-areas to be inspected, and the area of each sub-area is the same, enabling inspection robots, inspection drones, and inspection personnel to perform inspection tasks in each area separately or jointly, thus avoiding waste of inspection resources and omission of inspections.

[0076] After the division is completed, for each sub-area to be inspected, the system calculates an inspection type score based on the environmental information and operation parameter information of that area. The environmental information includes rain and snow information, environmental temperature information, and humidity information, while the operation parameter information includes equipment temperature information, current information, and voltage information. The system inputs this information into a specific score calculation model and calculates an inspection type score according to the weights and influence degrees of different information to judge the type of inspection requirements for that area. For example: In a certain sub-area to be inspected, if there are significant changes in the environmental temperature information and humidity information of that area, the system may assign a higher inspection type score to indicate that the environment of that area may have problems and environmental inspection is required; relatively, if the current information or voltage information is abnormal, the inspection type score will also increase, tending to the need for equipment failure inspection.

[0077] Compare the inspection type score of each sub - area to be inspected with the preset inspection type score threshold to determine the inspection type information of the sub - area. If the inspection type score is higher than a certain inspection type score threshold, the inspection type information of the sub - area will be defined as the corresponding inspection type, which may include equipment failure inspection, environmental inspection, and regular inspection. For example, if the inspection type score of a sub - area to be inspected is higher than the score threshold of equipment failure inspection, the system will define the inspection type information of this area as equipment failure inspection. If the inspection type score is within the score threshold range of environmental inspection, the inspection type information is defined as environmental inspection. For areas that do not exceed any score threshold, the inspection type information is defined as regular inspection to ensure that all sub - areas have reasonable inspection frequencies.

[0078] After determining the inspection type information of each sub - area to be inspected, further determine the inspection priority information of each area according to the environmental information and operation parameter information of the area. In a certain sub - area to be inspected, the environmental information shows a sharp change in temperature, and the operation parameter information shows abnormal voltage fluctuations. These values can be processed into fuzzy sets through the fuzzy logic method to obtain preliminary priority information. Then, it is further optimized through the particle swarm optimization method to ensure that the inspection priority information of this sub - area is relatively high and the inspection resources are preferentially allocated.

[0079] In summary, through the division of the target station, score calculation, score threshold determination, application of the fuzzy logic method and the particle swarm optimization method in the embodiments of the present disclosure, the inspection type information and inspection priority information are effectively determined, providing strong support for the efficient allocation and execution of inspection tasks.

[0080] In a feasible implementation manner, the above - mentioned inspection type score is determined based on the following formula:

[0081]

[0082] In the formula, S type is the above - mentioned inspection type score, W temp is the weight coefficient corresponding to the temperature factor, A is the attenuation factor of the temperature influence, T temp is the current environmental temperature information, T max is the maximum temperature reference information, W humidity is the weight information corresponding to the humidity factor, H humidity is the current humidity information, H max is the maximum humidity reference information, W rain is the weight coefficient corresponding to the rain - snow factor, R rain is the current rain - snow information value, W devicetemp is the weight coefficient corresponding to the equipment factor, T device is the current temperature information of the equipment, T devicemaxis the maximum reference temperature information of the device, W current is the weight information corresponding to the current factor, I current is the current current information, I norm is the normal current reference information, I max is the maximum current reference information, W voltage is the weight information corresponding to the voltage factor, V voltage is the current voltage information, V norm is the normal voltage reference information, V max is the maximum voltage reference information.

[0083] Exemplarily, the above formula incorporates multiple environmental and device operation parameters such as temperature, humidity, rain, snow, current, and voltage into the calculation, comprehensively considering the key factors affecting the inspection requirements. Each factor has its unique manifestation form on the device operation and environmental changes. For example, the temperature uses exponential decay, the humidity uses square, and the device temperature uses logarithm, etc. By carefully modeling different parameters, the inspection requirements of each area can be accurately reflected, ensuring that the judgment of the inspection task type is more scientific and accurate.

[0084] Through accurate calculation of the inspection type score, potential device failures or environmental change risks can be warned in advance, and maintenance personnel or equipment can be arranged in advance for inspection. This not only improves the efficiency of inspection but also reduces the risk of downtime caused by sudden device failures. For new energy power stations with a large number of devices and scattered sites, quickly and accurately judging the inspection requirements can greatly reduce labor and time costs and improve the overall operation safety and maintenance efficiency.

[0085] In summary, the formula design has significant beneficial effects in terms of the accuracy, adaptability, real-time performance, and efficiency of the inspection task, providing a more intelligent and efficient inspection task allocation application for new energy power stations.

[0086] In a feasible implementation manner, determining the above inspection priority information based on the above environmental information and the above operation parameter information of the to-be-inspected sub-region by means of the fuzzy logic method and the particle swarm optimization method includes:

[0087] Perform preprocessing operations on the above environmental information and the above operation parameter information of each to-be-inspected sub-region to obtain the processed environmental information and processed operation parameter information corresponding to each to-be-inspected sub-region.

[0088] Perform fuzzy processing on the above processed environmental information and the above processed operation parameter information corresponding to each to-be-inspected sub-region to obtain the fuzzy set corresponding to each to-be-inspected sub-region.

[0089] Perform a fuzzy mapping operation based on the fuzzy sets and fuzzy rule bases corresponding to each of the above-mentioned sub-areas to be inspected, so as to obtain the fuzzy levels of the inspection priorities corresponding to each of the above-mentioned sub-areas to be inspected.

[0090] Defuzzify the fuzzy levels of the inspection priorities corresponding to each of the above-mentioned sub-areas to be inspected, so as to obtain the initial inspection priority scores for each of the above-mentioned sub-areas to be inspected.

[0091] Perform an initial particle swarm operation based on the initial inspection priority scores of each of the above-mentioned sub-areas to be inspected, so as to obtain the initial position information of the particle swarm and the initial velocity information of the particle swarm.

[0092] Calculate the fitness of the above-mentioned initial position information of the particle swarm based on the target fitness function, so as to obtain the fitness value information of each particle.

[0093] Perform a particle swarm iterative update operation based on the above-mentioned initial position information of the particle swarm, the above-mentioned initial velocity information of the particle swarm, and the fitness information of the above-mentioned particles, so as to obtain the inspection priority information of each of the above-mentioned sub-areas to be inspected.

[0094] Exemplarily, preprocess the environmental information and operating parameter information of each sub-area to be inspected to eliminate abnormal data and noise, thereby improving the accuracy and consistency of the data. The environmental information includes rain and snow information, environmental temperature information, and humidity information, while the operating parameter information includes equipment temperature information, current information, and voltage information. The result of the preprocessing operation is the processed environmental information and processed operating parameter information of each sub-area to be inspected, providing high-quality input data for subsequent fuzzy processing. For example, if there are large fluctuations in the equipment temperature information data of a certain sub-area to be inspected, the preprocessing operation can smooth the data through filtering or denoising methods, making the equipment temperature information more representative; abnormal values in the voltage information can be processed through median replacement or interpolation methods.

[0095] After the preprocessing is completed, perform a fuzzy processing on the processed environmental information and processed operating parameter information of each sub-area to be inspected. The fuzzy processing converts continuous numerical information into fuzzy sets to express uncertainty and ambiguity. The fuzzy processing result corresponding to each sub-area to be inspected is a fuzzy set, including the fuzzy levels (such as "high", "medium", "low") of indicators such as temperature, humidity, current, and voltage. For example, the processed equipment temperature information can be divided into three fuzzy levels: "high temperature", "moderate", and "low temperature"; the humidity information can be divided into three fuzzy levels: "humid", "normal", and "dry". This can represent the states of various parameters more flexibly.

[0096] After obtaining the fuzzy sets of each sub-region to be inspected, perform fuzzy mapping operations based on the fuzzy sets and the fuzzy rule base. The fuzzy rule base contains the relationship rules between environmental information, operating parameter information, and inspection priorities. Through the matching operation of the rule base, the fuzzy sets of each sub-region to be inspected are mapped to the fuzzy levels of the corresponding inspection priorities. The fuzzy levels can be divided into "high priority", "medium priority", "low priority", etc., to indicate the urgency of inspection in this area. For example, if the temperature of a sub-region to be inspected is "high", the humidity is "humid", the current is "slightly high", and the voltage is "normal", then according to the fuzzy rule base, this region can be mapped to "high priority", indicating that there is a certain inspection requirement in this area.

[0097] After completing the fuzzy mapping, defuzzify the fuzzy levels of the inspection priorities of each sub-region to be inspected, and convert the fuzzy levels into an initial inspection priority score in a numerical form. The defuzzification operation makes the inspection priority become a specific numerical score, which is convenient for subsequent particle swarm optimization processing. If the fuzzy level of the inspection priority of a sub-region to be inspected is "high priority", then after defuzzification, the initial inspection priority score of this region is set to a relatively high value (such as 0.8); while "low priority" may be defuzzified to a lower score value (such as 0.3).

[0098] Use the initial inspection priority scores of each sub-region to be inspected to perform the initialization operation of the particle swarm, and assign an initial position and an initial velocity to each particle. The initial position and velocity reflect the initial state of the inspection priority score in the particle swarm, providing a basis for subsequent iterative optimization. The initial inspection priority score of each sub-region can be used as the initial position of the particle, and the environmental complexity and importance of this region can affect the initial velocity of the particle.

[0099] Define an objective fitness function to evaluate the rationality of the particle's priority. Based on this objective fitness function, calculate the fitness of the initial position information of the particle swarm to obtain the fitness value information of each particle. The higher the fitness value, the more the position of the particle meets the requirements of the inspection priority. The fitness function can be designed according to factors such as the importance of inspection requirements and resource availability. For example, the fitness of the particle corresponding to the area with strong inspection requirements should be higher.

[0100] Using the initial positions, initial velocities, and fitness information of the particle swarm, iterative update operations are performed on the particle swarm. In each iteration, the particles adjust their positions and velocities according to their individual best positions and the global best position until convergence to the optimal position. The final iteration result is the inspection priority information for each sub-region to be inspected, determining the inspection priorities of each region. In each iteration, if a region has a higher priority and is far from the best position, the particle will accelerate towards the best position to ensure that the priority requirements are met.

[0101] In the embodiment of the present application, fuzzy logic method is used to process uncertain information, and combined with particle swarm optimization method to achieve accurate calculation of inspection priority information, ensuring reasonable priority allocation of inspection tasks and improving the response efficiency of the system.

[0102] In a feasible embodiment, the above-mentioned objective fitness function Fitness(x) is:

[0103] Fitness(x) = w1·E(x) + w2·S(x) - w3·C(x)

[0104]

[0105]

[0106] C(x) = B·inspection distance + C·inspection time

[0107] In the formula, E(x) is the environmental matching degree, S(x) is the importance of equipment status, C(x) is the inspection cost, w1, w2, and w3 are all weight coefficients corresponding to each factor, T(x) is the temperature information of the current sub-region, H(x) is the humidity information of the current sub-region, T 目标 is the expected temperature information, H 目标 is the expected humidity information, B is the weight coefficient corresponding to the inspection distance, and C is the weight coefficient corresponding to the inspection time.

[0108] Exemplarily, by combining the environmental matching degree, the importance of equipment status, and the inspection cost, the system of the present application can comprehensively consider various factors and perform optimal allocation of inspection tasks. The environmental matching degree ensures that the inspection work is carried out under the most suitable conditions, the importance of equipment status gives priority to ensuring the inspection of key equipment, and the inspection cost controls the reasonable utilization of inspection resources, improving the inspection efficiency.

[0109] The weight coefficients w1, w2, and w3 in the fitness function can be adjusted according to the actual situation to meet different inspection requirements. For example, under extreme weather conditions, the weight w1 of the environmental matching degree can be increased to preferentially select areas with suitable environmental conditions; in a station where the equipment is severely aged, the weight w2 of the importance of the equipment status can be increased to ensure that key equipment is maintained in a timely manner.

[0110] The importance S(x) of the equipment status can accurately reflect the urgency of the equipment through indicators such as the running duration, the fault interval duration, and the usage frequency, enabling high-priority inspection tasks to be preferentially allocated resources, thereby reducing the risk of equipment failure and improving the reliability of the equipment.

[0111] The inspection cost C(x) enables the system to preferentially select inspection tasks with short distances and less time, reducing the resource consumption and time cost of inspections and improving the utilization rate of inspection resources. Especially for large-scale new energy stations, this cost control helps to achieve efficient inspection management.

[0112] The fitness function structure proposed in this application is flexible and applicable to various inspection scenarios. The environmental matching degree is applicable to adjustments under different weather conditions, the importance of the equipment status is applicable to the health status of different equipment, and the inspection cost is applicable to inspection tasks of different station scales. Through the design of the above fitness function, this application realizes the intelligent allocation of inspection tasks in new energy stations, improves the scientificity, real-time performance, and resource utilization rate of inspections, and effectively enhances the operation reliability and maintenance efficiency of station equipment.

[0113] In a feasible implementation manner, the above task allocation is performed based on the location information of each of the above sub-areas to be inspected and the above inspection task information of each of the above inspection sub-areas to determine the inspection robot sub-task information, the inspection drone sub-task information, and the inspection personnel sub-task information, including:

[0114] Based on the location information of each of the above sub-areas to be inspected and the above inspection task information of each of the above inspection sub-areas, task allocation is performed based on the multi-objective particle swarm optimization algorithm to determine the inspection robot sub-task information, the inspection drone sub-task information, and the inspection personnel sub-task information, where the objective function corresponding to the above multi-objective particle swarm optimization algorithm includes a path length objective function, a task priority objective function, and an inspection characteristic matching objective function.

[0115] Exemplarily, the multi-objective particle swarm optimization algorithm realizes the optimal solution of task allocation by simulating the movement of particles in the search space. Each particle represents an inspection task allocation application, and its position and speed are continuously updated until the optimal task allocation result is found. The objective functions include the path length objective function, the task priority objective function, and the inspection feature matching objective function, ensuring that the task allocation result can be optimized in multiple aspects.

[0116] In this multi-objective particle swarm optimization algorithm, each particle represents an inspection task allocation application, and its position is comprehensively determined by the path length, task priority, and inspection feature matching. The specific optimization steps are as follows:

[0117] 1. Particle initialization: According to the position information and inspection task information of each sub-area to be inspected, randomly generate the position and speed of the particle. Each particle represents an inspection equipment allocation application.

[0118] 2. Calculate fitness: Use the path length objective function, the task priority objective function, and the inspection feature matching objective function to calculate the fitness value of each particle. The higher the fitness, the better the current particle's task allocation application.

[0119] It should be noted that the fitness value is the core index for evaluating the quality of the particle task allocation scheme. In the multi-objective particle swarm optimization algorithm, it is usually necessary to comprehensively process the results of multiple objective functions to generate a single fitness value. The specific steps include:

[0120] 2.1 Definition of objective functions:

[0121] The path length objective function (f path ) is used to measure the total length of the inspection path of the inspection equipment, and the shorter the better. The task priority objective function (f priority ) is used to measure whether the priority of the inspection task is satisfied, and the higher the priority, the better the task allocation. The inspection feature matching objective function (f match ): It is used to evaluate the matching degree between the inspection equipment (robot, drone, personnel) and the task requirements, and the higher the matching degree, the better.

[0122] 2.2 Normalization processing: The values of each objective function may have different dimensions or value ranges, so it is necessary to perform normalization processing on each objective function. Assuming that the normalized values are respectively and

[0123] The normalization formula is as follows:

[0124]

[0125] where and is the minimum and maximum values of the objective function f in the current iteration i of the objective function

[0126] 2.3 Weighted summation: The normalized objective function is weighted and summed according to the weight coefficients

[0127] ω path , ω priority , ω match to obtain a single fitness value:

[0128]

[0129] The value of the weight coefficient ω can be set according to the importance of the task. For example:

[0130] If path length is prioritized, ω path > ω priority , ω match

[0131] If matching degree is important, ω match can be set higher.

[0132] 3. Individual and global optimal update: In each iteration, each particle updates its position and velocity based on its historical optimal position (individual best) and the global optimal position (global best) among all particles to gradually approach the optimal solution.

[0133] 4. Iterative update: The particles continuously move in the search space and update their positions and velocities, iterating until convergence or reaching the set maximum number of iterations. Finally, the position of the optimal particle in the particle swarm is the optimal inspection task allocation application.

[0134] It should be noted that in each iteration, all particles calculate the fitness value based on the value of the objective function. The higher the fitness value, the closer the task allocation scheme of the particle is to the optimal solution. Each particle records its best position in history, that is, the position corresponding to the highest fitness value obtained by the particle in all previous iterations. Among all particles, the position of the particle with the highest fitness value is found as the current global optimal position. Through iterative update, the fitness value of each particle will continuously approach the optimal. Finally, the position of the particle with the highest fitness value is the optimal task allocation scheme.

[0135] After the optimization is completed, the system assigns the inspection tasks to different inspection devices according to the position of the optimal particle, thereby determining the inspection sub-task information of inspection robots, the inspection sub-task information of inspection drones, and the inspection sub-task information of inspection personnel. This can ensure the reasonable allocation of inspection tasks, make full use of the advantages of various devices, and ensure the efficient completion of inspection tasks.

[0136] It should be noted that the position of the optimal particle represents the optimal solution of the inspection task allocation scheme, which is obtained by weighing the path length, task priority, and feature matching. Specifically: the "position" of each particle corresponds to an inspection task allocation scheme, for example, allocating the tasks of a certain sub-region to robots, drones, or personnel. In the optimal particle, the path length is the shortest, the task priority matching degree is the highest, and the device feature matching is the best. The system reads the position data of the optimal particle and parses the specific task allocation information, such as:

[0137] The inspection robot is responsible for the tasks in certain specific sub-regions.

[0138] The inspection drone is responsible for the inspection of certain sub-regions.

[0139] The inspector is responsible for special tasks that require manual intervention.

[0140] In this way, the task allocation takes into account both the path and task priority, and also makes full use of the characteristics of the equipment, achieving the optimal allocation of resources and efficient execution.

[0141] In the embodiments of the present disclosure, the intelligent allocation of inspection tasks is realized through the multi-objective particle swarm optimization algorithm. The use of the path length objective function makes the movement path of the inspection equipment between different sub-regions the shortest, improving the inspection efficiency and reducing the time and energy consumption. The task priority objective function ensures that high-priority tasks are preferentially allocated, enhancing the urgency of the inspection response and reducing the potential risks in key areas. The inspection feature matching objective function ensures that tasks are assigned to the most suitable equipment, reducing the energy consumption and unnecessary failures during the inspection process, and improving the safety and effectiveness of the inspection tasks. Through the intelligent allocation of the multi-objective particle swarm optimization algorithm, multi-device collaborative work is achieved, improving the overall efficiency and effect of the inspection of new energy power stations.

[0142] In a feasible implementation manner, the above path length objective function f1 is determined based on the following formula:

[0143]

[0144] where, distance(p i ,p i+1 ) represents the Euclidean distance from point p i to point p i+1 , d(p i ) is the residence time at position p i , α is the decay factor for adjusting the path weight, θ(p i ,p i+1 ) is the turning angle between two positions, β is the turning penalty coefficient, p i and p i+1are two adjacent inspection points, and n is the total number of inspection points.

[0145] In a feasible implementation, the above task priority objective function f2 is determined based on the following formula:

[0146]

[0147] where priority(T j ) represents the priority of the task, time(T j ) is the expected completion time of the task, actual_time(T j ) is the actual completion time of the task, γ is the delay penalty coefficient, and m represents the total number of tasks.

[0148] In a feasible implementation, the above inspection feature matching objective function f3 is determined based on the following formula:

[0149]

[0150] where D k represents the inspection execution body k, and the above inspection execution body includes inspection robots, inspection drones, and inspection personnel. T j represents the task j. Compatibility(D k ,T j ) is the adaptability of D k to T j , and compatibility(D k ,T j ) ∈ [0, 1]. C energy (D k ) represents the unit energy consumption of D k . The above inspection drones and the above inspection robots have specific energy consumption values, and the energy consumption of the above inspection personnel is calculated according to labor costs and time costs. C reliability (D k ,T j ) represents the reliability penalty coefficient when D k executes T j . C cormplexity (D k ,T j ) represents the task complexity matching situation when D k executes T j . δ is the weight coefficient corresponding to the above reliability penalty coefficient, and η is the weight coefficient corresponding to the above complexity penalty coefficient.

[0151] Exemplarily, the inspection task allocation based on the multi-objective particle swarm optimization algorithm includes three main objective functions: the path length objective function, the task priority objective function, and the feature matching degree objective function. Each objective function optimizes the inspection task from a different perspective to ensure the reasonable allocation of inspection tasks and the efficient utilization of resources.

[0152] The path length objective function minimizes the path length of the inspection equipment among different regions, reduces the inspection time and energy consumption, and improves the overall inspection efficiency. The task priority objective function ensures that high-priority tasks are executed first, reduces the risk of inspection delay for critical equipment and important regions, and guarantees the safety of the new energy power station. The feature matching degree objective function, by comprehensively considering energy consumption, reliability, and task complexity, ensures that the inspection tasks are assigned to the most suitable equipment, thereby optimizing the inspection effect and reducing unnecessary energy consumption and risks. Through the allocation method of the multi-objective particle swarm optimization algorithm, this application realizes the collaborative inspection of multiple devices, significantly improving the inspection efficiency and task completion quality of the new energy power station.

[0153] In a second aspect, a new energy power station inspection control system based on data mining according to this application includes:

[0154] An acquisition unit 21, configured to acquire the environmental information and operation parameter information of the target power station, where the environmental information includes rain and snow information, environmental temperature information, and humidity information, and the above operation parameter information includes equipment temperature information, current information, and voltage information.

[0155] A first determination unit 22, configured to determine inspection task information based on the above environmental information and the above operation parameter information, where the above inspection task information includes inspection type information and inspection priority information

[0156] A second determination unit 23, configured to perform task allocation based on the above inspection task information to determine inspection robot sub-task information, inspection drone sub-task information, and inspection personnel sub-task information.

[0157] A control unit 24, configured to send the above inspection robot sub-task information to the inspection robot, send the above inspection drone sub-task information to the inspection drone, and send the above inspection personnel sub-task information to the mobile terminal corresponding to the inspection personnel, so that the above inspection robot, the above inspection drone, and the above inspection personnel execute the corresponding inspection sub-tasks.

[0158] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application 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 on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A new energy power station inspection and control method based on data mining, characterized in that, Including: Obtain the environmental information and operating parameter information of the target station yard. The environmental information includes rain and snow information, environmental temperature information, and humidity information, and the operating parameter information includes equipment temperature information, current information, and voltage information; Based on the environmental information and the operating parameter information, determine the inspection task information of each sub-area to be inspected. The inspection task information includes inspection type information and inspection priority information; Based on the location information of each sub-area to be inspected and the inspection task information of each inspection sub-area, perform task allocation to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection UAV, and the inspection sub-task information of the inspection personnel; Send the inspection sub-task information of the inspection robot to the inspection robot, send the inspection sub-task information of the inspection UAV to the inspection UAV, and send the inspection sub-task information of the inspection personnel to the mobile terminal corresponding to the inspection personnel, so that the inspection robot, the inspection UAV, and the inspection personnel execute the corresponding inspection sub-tasks.

2. The new energy power station inspection control method based on data mining according to claim 1, characterized in that The determining the inspection task information of each sub-area to be inspected based on the environmental information and the operating parameter information includes: Divide the target station yard into multiple sub-areas to be inspected based on a preset area; Based on the environmental information and the operating parameter information of each sub-area to be inspected, calculate the inspection type score; Based on the inspection type score and the inspection type score threshold, determine the inspection type information corresponding to the sub-area to be inspected. The inspection type information includes equipment failure inspection, environmental inspection, and regular inspection; Determine the inspection priority information according to the environmental information and the operating parameter information of the sub-area to be inspected based on the fuzzy logic method and the particle swarm optimization method.

3. The inspection control method for new energy power stations based on data mining according to claim 2, wherein, Determine the inspection type score based on the following formula: Wherein, S type is the inspection type score, W temp is the weight coefficient corresponding to the temperature factor, A is the attenuation factor of the temperature influence, T temp is the current ambient temperature information, T max is the maximum temperature reference information, W humidity is the weight information corresponding to the humidity factor, H humidity is the current humidity information, H max is the maximum humidity reference information, W rain is the weight coefficient corresponding to the rain and snow factor, R rain is the current rain and snow information value, W devicetemp is the weight coefficient corresponding to the equipment factor, T device is the current equipment temperature information, T devicemax is the maximum reference temperature information of the equipment, W current is the weight information corresponding to the current factor, I current is the current current information, I norm is the normal current reference information, I max is the maximum current reference information, W voltage is the weight information corresponding to the voltage factor, V voltage is the current voltage information, V norm is the normal voltage reference information, V max is the maximum voltage reference information.

4. The inspection control method for new energy power stations based on data mining according to claim 2, wherein, The determining the inspection priority information according to the environmental information and the operating parameter information of the sub-area to be inspected based on the fuzzy logic method and the particle swarm optimization method includes: Perform a preprocessing operation on the environmental information and the operating parameter information of each sub-area to be inspected to obtain the processed environmental information and the processed operating parameter information corresponding to each sub-area to be inspected; Perform a fuzzification process on the processed environmental information and the processed operating parameter information corresponding to each sub-area to be inspected to obtain the fuzzy set corresponding to each sub-area to be inspected; Perform a fuzzy mapping operation based on the fuzzy set and the fuzzy rule base corresponding to each sub-area to be inspected to obtain the fuzzy level of the inspection priority corresponding to each sub-area to be inspected; Perform a defuzzification process on the fuzzy level of the inspection priority corresponding to each sub-area to be inspected to obtain the initial inspection priority score of each sub-area to be inspected; Perform an initial particle swarm operation based on the initial inspection priority score of each sub-area to be inspected to obtain the initial position information of the particle swarm and the initial velocity information of the particle swarm; Perform a fitness calculation on the initial position information of the particle swarm based on the target fitness function to obtain the fitness value information of each particle; Perform particle swarm iterative update operations based on the initial position information of the particle swarm, the initial velocity information of the particle swarm, and the fitness information of the particle to obtain the inspection priority information for each sub-region to be inspected.

5. The inspection control method for new energy power stations based on data mining according to claim 4, wherein, The target fitness function Fitness(x) is: Fitness(x) = w1·E(x) + w2·S(x) - w3·C(x) C(x) = B·Inspection distance + C·Inspection time Where, E(x) is the environmental matching degree, S(x) is the importance of equipment status, C(x) is the inspection cost, w1, w2, and w3 are all weight coefficients corresponding to each factor, T(x) is the temperature information of the current sub-region, H(x) is the humidity information of the current sub-region, T 目标 is the expected temperature information, H 目标 is the expected humidity information, B is the weight coefficient corresponding to the inspection distance, and C is the weight coefficient corresponding to the inspection time.

6. The inspection control method for new energy power stations based on data mining according to claim 1, wherein Perform task allocation based on the position information of each sub-region to be inspected and the inspection task information of each inspection sub-region to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection drone, and the inspection sub-task information of the inspection personnel, including: Perform task allocation based on the position information of each sub-region to be inspected and the inspection task information of each inspection sub-region using the multi-objective particle swarm optimization algorithm to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection drone, and the inspection sub-task information of the inspection personnel. Among them, the objective functions corresponding to the multi-objective particle swarm optimization algorithm include a path length objective function, a task priority objective function, and an inspection characteristic matching objective function.

7. The inspection control method for new energy power stations based on data mining according to claim 6, characterized in that, The path length objective function f1 is determined based on the following formula: Among them, distance(p i , p i+1 ) represents the Euclidean distance from point p i to point p i+1 . d(p i ) is the residence time at the position of p i . α is the decay factor for adjusting the path weight. θ(p i , p i+1 ) is the turning angle between two positions. β is the turning penalty coefficient. p i and p i+1 are two adjacent inspection points, and n is the total number of inspection points.

8. The inspection and control method for new energy power stations based on data mining according to claim 6, characterized in that The task priority objective function f2 is determined based on the following formula: Among them, priority(T j ) represents the priority of the task, time(T j ) is the expected completion time of the task, actual_time(T j ) is the actual completion time of the task, γ is the delay penalty coefficient, and m represents the total number of tasks.

9. The inspection control method for new energy power stations based on data mining according to claim 6, wherein The inspection characteristic matching objective function f3 is determined based on the following formula: Among them, D k represents the inspection execution body k, and the inspection execution body includes an inspection robot, an inspection UAV, and an inspection personnel. T j represents the task j, and Compatibility(D k , T j ) is the adaptability of D k to T j , and compatibility(D k , T j ) ∈ [0, 1]. C energy (D k ) represents the unit energy consumption of D k . The inspection UAV and the inspection robot have specific energy consumption values, and the energy consumption of the inspection personnel is calculated according to the labor cost and the time cost. C reliability (D k , T j ) represents the reliability penalty coefficient of D k when executing T j . C cormplexity (D k , T j ) represents the task complexity matching situation of D k executing T j . δ is the weight coefficient corresponding to the reliability penalty coefficient, and η is the weight coefficient corresponding to the complexity penalty coefficient.

10. A new energy power station inspection and control system based on data mining, characterized in that, Including: An acquisition unit for acquiring the environmental information and operating parameter information of the target station. Among them, the environmental information includes rain and snow information, environmental temperature information, and humidity information, and the operating parameter information includes equipment temperature information, current information, and voltage information; A first determination unit for determining the inspection task information of each sub-region to be inspected based on the environmental information and the operating parameter information. Among them, the inspection task information includes inspection type information and inspection priority information; A second determination unit for performing task allocation based on the inspection task information to determine the inspection sub-task information of the inspection robot, the inspection sub-task information of the inspection drone, and the inspection sub-task information of the inspection personnel; A control unit for sending the inspection sub-task information of the inspection robot to the inspection robot, sending the inspection sub-task information of the inspection drone to the inspection drone, and sending the inspection sub-task information of the inspection personnel to the mobile terminal corresponding to the inspection personnel, so that the inspection robot, the inspection drone, and the inspection personnel execute the corresponding inspection sub-tasks.

Images