Regional renewable energy power generation equipment performance evaluation method

Abstract

The invention discloses a regional renewable energy power generation equipment performance evaluation method and system, and the method comprises the steps: enabling a power dispatching center to receive or generate a renewable energy equipment performance evaluation demand, and planning a space region; according to the meteorological information acquisition range of the unmanned aerial vehicle, dividing a space region into a plurality of acquisition blocks; the unmanned aerial vehicles independently complete a meteorological data acquisition task and a renewable energy equipment detection task or multiple unmanned aerial vehicles cooperatively complete a meteorological data acquisition task and a renewable energy equipment detection task; after the task is completed, the unmanned aerial vehicle returns to the unmanned aerial vehicle base station and transmits the collected data to the power dispatching center; and the power dispatching center predicts the future power generation condition in the spatial region according to the collected meteorological data and the renewable energy equipment detection data. The method has the advantages of quicker data acquisition, lower communication requirements of the distributed execution algorithm, better collaboration flexibility of the ad hoc network and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of renewable energy power generation equipment detection and power generation performance evaluation of a new power system, and specifically to a regional renewable energy power generation equipment performance evaluation method based on meteorological factors and multi-UAV collaboration. Background Art

[0002] In the new power system, renewable energy power generation equipment plays an increasingly important role. Its normal operation is directly related to the supply of power generation in the region and the accurate prediction of power load. In order to test the performance of renewable energy power generation equipment and then evaluate and predict the power generation in the region, it is mainly necessary to start from the following two aspects: On the one hand, collecting meteorological conditions around equipment is a basic task for evaluating the efficiency of renewable energy utilization and predicting future power generation. However, the current collection of meteorological data mainly relies on fixed meteorological observation stations and remote sensing satellites, which limits the accurate assessment of meteorological conditions around equipment.

[0003] On the other hand, it is also necessary to inspect the renewable energy power generation equipment itself. Traditional inspection methods usually rely on manual labor to reach fixed inspection nodes, but when faced with complex and widely distributed renewable energy power generation equipment, a large number of experienced maintenance personnel are required. In addition, many devices are located on the top of houses or high towers, and the risks and costs of manual maintenance are extremely high.

[0004] In summary, the power generation capacity of renewable energy is highly dependent on meteorological conditions, such as wind speed, light intensity, precipitation, and temperature. These factors not only increase the volatility of power generation capacity, but also pose challenges to the safety of equipment. With the continuous increase in the proportion of renewable energy and the increasing dispersion of distributed power generation equipment, the requirements for equipment performance testing are becoming more urgent. How to combine the meteorological conditions around renewable energy power generation equipment to achieve performance evaluation of renewable energy power generation equipment, and then predict future power generation and regional power load, is of great significance to the power dispatch of new power systems. Summary of the invention

[0005] The present invention aims to solve one of the technical problems existing in the related art at least to a certain extent.

[0006] The purpose of the present invention is to provide a method for evaluating the performance of regional renewable energy power generation equipment, which combines drone collaborative technology to evaluate the performance of regional renewable energy equipment, so as to provide a guarantee for scientifically evaluating the performance of renewable energy power generation equipment, predicting future power generation capacity and pre-dispatching of regional power loads.

[0007] In order to achieve the above-mentioned purpose, the present invention first provides a method for evaluating the performance of regional renewable energy power generation equipment, comprising: S100, the power dispatching center accepts or generates a demand for renewable energy equipment performance evaluation, and plans a spatial area that needs to collect meteorological data and includes all renewable energy equipment to be evaluated; S200, the power dispatching center divides the spatial area into several collection blocks according to the meteorological information collection range of the UAV, the center of each collection block is the corresponding collection point, the collection point corresponding to the collection block containing the renewable energy equipment to be evaluated is the interest collection point, and the rest are common collection points; S300, the drone dynamically obtains the information of the current point to be collected through the self-organizing network, and then plans the flight path according to the information of the point to be collected and its own power reserve, and independently completes or multiple drones cooperate to complete the meteorological data collection task of the collection point, as well as the renewable energy equipment detection task of the interest point; after completing all the interest point detection tasks and the collection tasks of the collection point, the drone returns to the drone base station and transmits the collected data to the power dispatching center; S400. The power dispatching center evaluates the operating performance of each renewable energy device based on the collected meteorological data and renewable energy device detection data, and then predicts the future power generation situation in the spatial area to guide the pre-dispatching of power loads.

[0008] A further preferred technical solution of the present invention is that step S200 includes: S210, the power dispatching center dispatches a number of drones, and obtains the current base station locations of all drones, and the effective detection range of the sensors used by each drone to collect meteorological data; S220, the minimum effective detection range among all drone sensors As a standard, the spatial area is divided into several acquisition blocks; S230. Taking the center of each collection block as the collection point, the power dispatching center sets the collection points corresponding to the collection blocks containing the renewable energy equipment to be evaluated as the collection points of interest according to the distribution of renewable energy equipment in the collection blocks, and the rest as ordinary collection points. The task type of each collection block is set accordingly; and according to the task type, the type of renewable energy equipment in the collection block, and the distance between the renewable energy equipment and the collection point, the additional task reward and the shortest collection time of the corresponding collection block are set.

[0009] Preferably, in step S220, the minimum effective detection range of all the drone sensors is used as a standard to divide the space area into a number of acquisition blocks, and the specific method is as follows: A sphere is constructed with the minimum effective detection range of all drone sensors, and the largest rhombic dodecahedron inscribed in the sphere is determined. The rhombic dodecahedron is used to fill the spatial area planned in step S100 to obtain a spatial grid evenly distributed in the spatial area, and each spatial network is used as a collection block.

[0010] Preferably, step S300 includes: S310, the power dispatching center initializes the map information, determines the information of each collection block, and synchronizes the mission map with the UAV base station. The UAV takes off from the base station and goes to the collection point to perform the mission; S320, the UAV performs the meteorological data collection task and the renewable energy equipment detection task according to the task type of the collection point and its own power reserve. After completing the task of the current collection point, the UAV updates the task map and marks the collection point that has completed the task; S330, when the drone is performing a task or heading to the next collection point, the drone attempts to establish communication with the base station or other drones at regular intervals. If communication is established, the drone and the base station or other drones are networked with each other, and the drone exchanges its own task map with the base station or other drones, and informs its own location information and the collection point it is working on; S340: The drone plans the next action route according to the updated mission map, and loops through steps S320 and S330.

[0011] Preferably, in step S310, the grid point information in the task map includes the following contents: Grid collection points Location ; Collection identification , used to mark whether the point has been collected. If it has been collected, it is 0, if it has not been collected, it is 1; interest identifier , used to mark whether the block is an interest block, if it is not an interest block, it is 0, if it is an interest block, it is 1; the minimum acquisition time at the acquisition point ; Location table of renewable energy equipment within the grid ,in, It is The three-dimensional coordinates of renewable energy power generation equipment, is the three-dimensional coordinate of the last renewable energy power generation device, is the number of renewable energy generation equipment in the grid; the shortest inspection schedule of the equipment in the grid ,in It is The shortest inspection time for renewable energy power generation equipment, is the shortest inspection time of the last renewable energy power generation equipment; timestamp Used to save the last time the point status was updated; meteorological data matrix Used to store meteorological data and equipment data matrix within the grid Used to store the detection data of each device; grid reward value Defined as:

[0012] in, Reward yourself, For derived rewards, is the charging incentive factor, It is the basic reward for all collection points. is the reward attached to the block, yes The number of surrounding collection points, It is the collection mark, which is used to identify whether the block has been collected; It is around The block's own reward, It is around According to the division method in step S230, Usually 12, It will be different at the border; At the same time, the drone base station that can be used to replace the drone battery ,in , is the number of drone base stations, marking the locations of the base stations , charging reward value Defined as:

[0013] in, Reward yourself, For derived rewards, is the charging incentive factor, It is the basic reward for replacing the battery. From the current position of the drone to the base station The distance For the i Drone at the moment t location, yes location, yes The number of collection points around the block.

[0014] Preferably, in step S340, when the drone plans the next action route, it determines to select one of the adjacent collection points at the current location or return to the base station to replenish power according to its own power reserve; and designs a reward function to encourage the drone to fully and economically use the power reserve while ensuring the minimum safe power reserve; Specifically design the following reward function to constrain the behavior of the drone:

[0015] in, is the residual energy excitation factor, It's the drone. The remaining power at the moment, It is the minimum power limit of the drone;

[0016] in, is the battery surplus penalty factor, is the minimum distance between the drone and the battery replacement base station, is the maximum flight speed of the drone, is the maximum battery capacity of the drone, is the energy consumption per unit time of flight;

[0017] in, is the power shortage penalty factor, At the collection point The shortest collection time, is the energy consumption per unit time of hover collection;

[0018] in, is the flight stability excitation factor, It's the drone. the height of the moment; In summary, the reward function for the drone to select the next action block is:

[0019]

[0020] The constraints are: Power constraints, each drone At any time The remaining energy Cannot fall below the minimum threshold ; Position constraint: the position of each drone must be continuous. Departure, arrival Finish; Boundary constraints: drones should not exceed the the extent of the region; No interruption constraints. The tasks of each collection point must be completed at one time without interruption. There is no duplication constraint. The reward of each collection point in a grid can only be obtained once and cannot be obtained repeatedly by subsequent drones that visit the grid. The shortest detection time constraint: for each detection point, the detection time must not be less than ; ( For collection point The shortest collection time).

[0021] Preferably, in step S320, when the current collection point is a common collection point, the drone performs the meteorological data collection task, and the specific method is as follows: After the drone arrives at the collection point according to the planned route, the sensors installed on the drone collect the meteorological data of the collection area; When the current collection point is the collection point of interest, the drone performs the meteorological data collection task and the renewable energy equipment detection task. The specific method is as follows: When the drone arrives at the collection point, it first performs the meteorological data collection task. During the execution of the meteorological data collection task, it plans the route between the collection point and the renewable energy equipment. After completing the meteorological data collection task, it goes to the vicinity of the renewable energy equipment and uses its own imaging equipment to detect the operating status of the renewable energy equipment. The detection time is not less than the shortest collection time.

[0022] As a preference, before the UAV performs the renewable energy equipment detection task, during the process of planning the action route between the collection point and the renewable energy equipment, it evaluates whether it has the ability to independently complete the detection task within the power reserve limit. The minimum power requirement of the UAV for this detection sequence is:

[0023] in, It's the drone in the block The total flight time of the planned path, is the energy consumption per unit time of flight, It is for The minimum detection time of a device, is the number of renewable energy generation equipment in the current grid, It is The shortest inspection time for renewable energy equipment is the energy consumption per unit time of hover collection, From the current position of the drone to the base station The distance For the i Drone at the moment t location, yes location, is the maximum flight speed of the drone; If satisfied The drone will detect the operation of renewable energy power generation equipment one by one according to this route, and the detection time for each device is not less than ; It's the drone. The remaining power at the moment, is the minimum power limit of the drone; after completing a series of detection tasks, the drone will continue to the next action point; if , then it issues a collaboration request to other drones within the direct communication range, and tries to dispatch the detection task of renewable energy equipment in this collection block to other drones to perform joint detection.

[0024] As a preferred method, if the UAV does not have the ability to independently complete the detection task within the power reserve limit, the specific method of performing joint detection is: The drone in this collection block sends a collaboration request to other drones within the direct communication range, and other drones in the ad hoc network return to their current locations. , the position of the next target point , remaining power and their ability to cooperate , It is expressed as:

[0025] in, The participation in other detection tasks. If the user is participating in other detection tasks or is currently performing data collection, the task completion status will be used to select the If the value is 1, it means that the acquisition of a certain point has just been completed, or the collaboration has just ended, or the flight is in progress and there is no detection task to be completed; It's a drone With the collection point in the block to be coordinated distance; is a ratio used to adjust the influence of distance factor and residual power factor on cooperation ability. The current location of the drone and the block to be cooperated Location The distance between The drone in this collection block determines the best cooperative drone based on the current position of other drones, the position of the next target point, the remaining power reserve, and the cooperation ability, and sends the required detection points to the best cooperative drone through the self-organizing network. The specific method is as follows: according to Sort from high to low, and use swarm intelligence optimization algorithm to join the drone The route of the subsequent detection points is planned, where each drone The path should be from the current location Set off until the next destination When the power is insufficient, the drone should be planned to return to the nearby base station in time to replace the battery; the objective function of the algorithm is:

[0026] in, , are the weights of the UAV flight time and altitude change in the scheme respectively; For drones Flight time; For drones The height change value is used to encourage the drone to fly smoothly; is the number of drones included in the local path planning; the constraints are: Power constraints, each drone At any time The remaining energy Cannot fall below the minimum threshold , and upon arrival , that is, the power consumption at the next original destination shall not be less than ; Position constraint: the position of each drone must be continuous. Departure, arrival Finish; No interruption constraints. The tasks of each detection point must be completed at one time without interruption. There is no duplication constraint, each detection point can only be detected once by a drone; Full coverage constraint: all detection points in the block must be fully detected in the local path planning; The shortest detection time constraint: for each detection point, the detection time must not be less than ; Get the target value of the solution after adding the new drone , if it satisfies:

[0027] The drone Add to the scheduling plan, where is the planning target value before adding the drone, is the importance ratio of time optimization and collaboration ability; repeat this step until the equation is not satisfied, and use the last planned The corresponding scheme is used, and the detection points required for each drone are sent to the corresponding drone through the self-organizing network; After completing the current collection point task, the corresponding drone goes to the received detection point location to perform the detection task.

[0028] Preferably, step S400 includes: S410, the power dispatching center calls the expected power generation model, calculates the energy efficiency and equipment capacity factor based on the renewable energy equipment detection data collected by the drone, evaluates the equipment integrity and equipment cleanliness, and performs a fuzzy comprehensive evaluation on each renewable energy power generation equipment; S420. Use the collected meteorological data to make a weather forecast for the spatial area planned in step S100. The power dispatching center predicts the total power generation capacity and the future power load in the spatial area through a neural network model based on the meteorological forecast results and the evaluation results obtained in step S410, thereby realizing power load pre-dispatching.

[0029] In another aspect, the present invention provides a regional renewable energy power generation equipment performance evaluation system, comprising: Power dispatching center, multiple drone base stations, and several drones deployed at the drone base stations; The renewable energy power generation equipment is one or more of photovoltaic power generation equipment, wind power generation equipment and hydropower generation equipment; The power dispatching center accepts or generates the demand for renewable energy equipment performance evaluation, plans the spatial area where meteorological data needs to be collected and contains all the renewable energy equipment to be evaluated; and divides the spatial area into several collection blocks according to the meteorological information collection range of the drone. The center of each collection block is the corresponding collection point, among which the collection point corresponding to the collection block containing the renewable energy equipment to be evaluated is the collection point of interest, and the rest are ordinary collection points; The drone dynamically obtains the information of the current collection point through the self-organizing network, and then plans the flight path according to the information of the collection point and its own power reserve, and independently completes or multiple drones work together to complete the meteorological data collection task of the collection point, as well as the renewable energy equipment detection task of the interest point; after completing all the interest point detection tasks and the collection tasks of the collection point, the drone returns to the drone base station and transmits the collected data to the power dispatching center; The power dispatching center evaluates the operating performance of each renewable energy device based on the collected meteorological data and renewable energy equipment detection data, and then predicts the future power generation situation in the spatial area to guide the pre-dispatching of power loads.

[0030] Beneficial effects: The present invention provides a regional renewable energy equipment performance evaluation method combined with UAV collaborative technology, which has the advantages of faster data collection, lower communication requirements for distributed execution algorithms, and better flexibility in self-organizing network collaboration. It solves the problems of traditional renewable energy equipment performance evaluation, such as a single data collection method, poor adaptability to multi-model distributed systems, and insufficient flexibility in traditional multi-UAV detection and scheduling responses.

[0031] The present invention adopts a multi-UAV collaboration method that simultaneously completes meteorological data collection and renewable energy power generation equipment detection. It can collect performance data of renewable energy power generation equipment in a targeted manner while taking into account meteorological data collection in the region, achieving a good balance between energy conservation and comprehensive collection, and can provide scientific guarantees for performance evaluation of renewable energy power generation equipment and pre-dispatching of regional power loads.

[0032] The present invention proposes a collection point formulation strategy that divides the spatial area according to the rhombic dodecahedron filling theory. This strategy can not only fully utilize the collection range of the drone, but also ensure that the distribution of the collection points has a high regularity. Compared with the typical cube division, when the effective collection radius of the drone is the same, the number of blocks required can be reduced by about 23%.

[0033] Through carefully designed reward functions and constraints, the present invention allows UAVs to conduct self-organized networking communications, allowing UAVs to complete meteorological field data collection under conditions of poor communication and high interference, and to conduct collaborative detection of renewable energy power generation equipment through local swarm intelligence optimization algorithms, greatly improving work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A flowchart of a method for evaluating performance of regional renewable energy power generation equipment according to Embodiment 1 of the present invention; Figure 2 A schematic diagram of a space region filling method of the present invention; Figure 3 This is a schematic diagram of the framework of a regional renewable energy power generation equipment performance evaluation system according to Example 2 of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments, and they should not be understood as limitations on the present invention. 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. In the description of the present invention, it should be understood that the terms used are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0036] The present invention relates to regional renewable energy power generation equipment performance detection based on meteorological factors and multi-UAV collaboration, aiming to solve complex renewable energy power generation equipment performance detection problems in new power systems.

[0037] Combined with the meteorological conditions around renewable energy power generation equipment, the performance of renewable energy power generation equipment is evaluated, and the future power generation and regional power load are predicted, which is of great significance to the power dispatch of the new power system. The implementation of this work requires the collection of a large amount of data. The particularity of the location of renewable energy power generation equipment and its sensitivity to meteorological conditions make drones an ideal choice for this data collection work, and they can effectively collect meteorological information. In addition, many devices are located on the top of houses or high towers, and using drones for routine inspections (such as maintenance node information extraction, cleanliness detection, thermal imaging scanning, etc.) is obviously a more effective solution.

[0038] At present, the solutions for multi-UAV meteorological data collection and multi-UAV collaborative fault detection of renewable energy power generation equipment are mostly separate. The former tends to collect meteorological data for the entire area, which is a supplement to the work of meteorological stations and remote sensing satellites; the latter focuses on maintenance after a fault occurs.

[0039] The present invention is aimed at the problems of traditional renewable energy equipment performance evaluation, such as single data collection mode, poor adaptability to multi-model distributed systems, and inflexible response to traditional multi-UAV detection and scheduling. The proposed regional renewable energy power generation equipment performance evaluation method and system, through the collaborative path planning of multiple UAVs, can efficiently collect the meteorological field around the renewable energy power generation equipment during a certain period of time, and conduct routine detection of the renewable energy power generation equipment itself, so as to achieve efficient inspection of renewable energy power generation equipment in the region. It provides a guarantee for scientifically evaluating the performance of renewable energy power generation equipment, predicting future power generation capacity and regional power load pre-scheduling, and can also effectively respond to the challenges brought by emergencies.

[0040] Combine the following Figures 1 to 3 The present invention describes a method and system for evaluating the performance of regional renewable energy power generation equipment.

[0041] Embodiment 1: This embodiment provides a method for evaluating the performance of regional renewable energy power generation equipment.

[0042] Next, the evaluation method of the present invention is described in detail. Figure 1 As shown, the specific steps are: S100, the power dispatching center accepts or generates renewable energy equipment performance evaluation requirements and plans the spatial areas where meteorological data needs to be collected . Set the space area The division standard is the power dispatching area of ​​the new power system.

[0043] S200, the power dispatching center determines the size of the area to be collected meteorological information and the meteorological information collection range of the drone. ,like Figure 2 As shown in the figure, the space is filled by rhombic dodecahedron filling, and the collection points are planned accordingly. The specific process includes: S210, the power dispatching center dispatches from two base stations Get a drone Current base station location ,in, ; ,in, . Collect all drones Sensor parameters: In order to collect meteorological information, the drone should carry at least a lidar, a temperature and humidity sensor, a wind speed and direction sensor, an air pressure sensor, and a radiation sensor to measure the necessary meteorological data. The minimum value of the effective detection range of these sensors is specified as .

[0044] S220, For a sphere with radius , determine the largest rhombic dodecahedron inscribed in the sphere. The side length of this rhombic dodecahedron is . Drones can effectively collect detailed meteorological data inside this rhombic dodecahedron.

[0045] S230, using the rhombic dodecahedron to By stacking and filling, a series of evenly distributed spatial grids can be obtained, and the center of each grid is the collection point. The coordinates are , then the coordinates of the twelve collection points around it (if they are within the boundary) are:

[0046]

[0047]

[0048]

[0049] Therefore, the actual distance between each two collection points is m, through this principle, the task space is divided into a series of evenly distributed collection blocks. In this embodiment, it is assumed that a collection point is located at , there are 231 collection points in this space.

[0050] S240. The power dispatching center selects the collection block where the renewable energy power generation equipment is located and its corresponding collection point according to the regional characteristics and the distribution of renewable energy equipment in the region. As the interest blocks and points of interest for this task. Each collection point should contain its three-dimensional coordinates , the task type to which the point belongs (interest collection point, ordinary collection point), as an additional task reward for the point of interest , and the shortest acquisition time at each point is .

[0051] In this embodiment, the distribution details of the renewable energy power generation equipment in the region are shown in Table 1: Table 1 Distribution information of renewable energy power generation equipment

[0052] Additional quest rewards for points of interest and the shortest collection time This collection point can be The device type, the distance between the device and the collection point, and the minimum detection time are used to determine the detection time. For example, for a collection point with coordinates at (-400, 2000, 100), It can be calculated by unit basic reward * minimum detection time + d (detection point, collection point), that is, ; It can be calculated by basic collection time + device type parameter * device minimum detection time, that is, .

[0053] S300, the drone dynamically obtains information about the points to be collected through the self-organizing network, collects meteorological data through the designed algorithm, and returns to the base station to replace the battery in a timely manner according to its own power reserve. After completing the perception and collection of meteorological data in all interest blocks, the data is sent to the base station and then transmitted to the power dispatching center. If the collection task is completed, it returns to its own base station. If there is a device to be detected in the interest block, the device to be detected will be tested. If the capacity is insufficient, multi-drone collaborative detection will be carried out autonomously. The specific process includes: S310, the power dispatching center initializes the map information and determines the information of each grid. Then the power dispatching center connects to the UAV base station to synchronize the mission map, and the UAV takes off from the base station to perform the meteorological data collection mission.

[0054] The grid point information in the mission map includes the following: Grid collection points Location ; Collection identification Used to mark whether the point has been collected (if it has been collected, it is 0, if it has not been collected, it is 1); interest identifier Used to mark whether the block is an interest block (if it is not an interest block, it is 0, if it is an interest block, it is 1); the minimum acquisition time at the acquisition point ; Location table of renewable energy equipment within the grid ,in, It is The three-dimensional coordinates of renewable energy power generation equipment, is the three-dimensional coordinate of the last renewable energy power generation device, is the number of renewable energy generation equipment in the grid; the shortest inspection schedule of the equipment in the grid ,in It is The shortest inspection time for renewable energy power generation equipment, is the shortest inspection time of the last renewable energy power generation equipment; timestamp Used to save the last time the point status was updated; meteorological data matrix Used to store meteorological data and equipment data matrix within the grid It is used to store the detection data of each device; and considering the spatial continuity of meteorological data, the blocks around each interest block should also increase a certain reward. Therefore, the grid reward value Depends on the basic reward, the additional reward as a point of interest, and the combination of its own rewards And the derived rewards brought by the surrounding collection points , defined as:

[0055] in, Reward yourself, For derived rewards, is the charging incentive factor, It is the basic reward for all collection points. is the reward attached to the block, yes The number of surrounding collection points, is a collection mark, used to identify whether the block is collected. According to the division method in step S230, Usually 12, This derived reward can effectively attract drones to blocks around the block of interest and encourage drones to form collaboration through self-organizing networks.

[0056] At the same time, the drone base station that can be used to replace the drone battery ,in , is the number of drone base stations, marking the locations of the base stations , charging reward value Depends on the distance the drone needs to fly to replace the battery here, and the derived rewards brought by the collection points around the base station , defined as:

[0057] in, For charging rewards, For derived rewards, is the charging incentive factor, It is the basic reward for replacing the battery. From the current position of the drone to the base station The distance yes The number of collection points around the block. This formula is designed to encourage drones to find base stations that are closer to themselves and have higher value of collection points around them for battery replacement when the power is low.

[0058] S320: Every once in a while, the drone attempts to establish a self-network with the base station or other drones to exchange map information and collaboration information. If there are no other drones or ground base stations within the communication range, then directly proceed to step S330. If communication is established, after the networking is completed, the drone will update the collection point in its map exchanged with other drones and ground base stations. of , and inform your location information And the collection points in operation Finally, update all nodes in the map based on the communication results. .

[0059] In order to prevent two drones from going to the same collection point or even colliding, the communication interval requirements are as follows:

[0060] in, is the maximum flight speed of the drone. is the maximum communication range of the drone. That is, the communication interval does not exceed It can ensure that two drones will not go to the same node, and the communication interval does not exceed It can be ensured that the two drones will not collide. , ,but In actual use, this value should be about half of 14s.

[0061] In this step, the power dispatching center can send new environmental data to the drone through the base station to adjust map information such as points of interest.

[0062] S330, drone based on updated local map The distributed execution mode of the multi-agent reinforcement learning algorithm is used to generate the next action. In each round of decision making, the multi-agent reinforcement learning algorithm will Output the next action. The action space of the next action is: you can select one of the adjacent collection points from the current position, or return to the base station to replace the battery. After the execution is completed, the updated state is , iteratively generate the next action until all of ,That are all updated to 0, that is, until all the blocks of interest are collected.

[0063] In order to encourage the drone to save energy when completing the collection, the following reward function is designed to constrain the drone's behavior:

[0064] in, is the residual energy excitation factor, The drone is in state The remaining energy when is the minimum power limit of the drone. This reward function is designed to encourage the drone to continue the mission when it has sufficient power.

[0065]

[0066] in, is the battery surplus penalty factor, is the minimum distance between the drone and the battery replacement base station, is the maximum battery capacity of the drone, It is the energy consumption per unit time of flight. This reward is designed to encourage drones to make full use of the battery and not return to the base station for battery replacement too early.

[0067]

[0068] in, is the power shortage penalty factor, is the energy consumption per unit time of hovering collection. This reward is designed to encourage the drone to ensure that it has enough power for resupply. Should be much greater than , and with increases with the increase of .

[0069]

[0070] in, is the flight stability excitation factor, It's the drone. The altitude at the moment is determined by the distance between the two adjacent collection points, because although the straight-line distance between the two adjacent collection points is the same, the change in altitude will cause the drone to consume extra power. This reward is designed to encourage drones to reduce altitude changes and fly smoothly. The formulation should not be too large.

[0071] In summary, the reward function for the drone to select the next action block is:

[0072]

[0073] The constraints are: · Power constraints. Each drone At any time The remaining energy Cannot fall below the minimum threshold .

[0074] Position constraints. The position of each drone must be continuous, starting from Departure, arrival Finish.

[0075] Boundary constraints. The drone should not exceed the The scope of the area.

[0076] No interruption constraints. The task of each collection point needs to be completed at one time, and no interruption is allowed.

[0077] No duplication constraints. The rewards of each collection point in a grid can only be obtained once and cannot be obtained repeatedly by subsequent drones that visit the grid.

[0078] · Minimum detection time constraint. For each detection point, the detection time must not be less than .

[0079] S340, after the drone determines the action, it will fly to the target point according to the plan. If it is a collection point, it will use the temperature and humidity sensor, wind speed and direction sensor, air pressure sensor, radiation sensor and precipitation sensor to collect meteorological data after arriving at the collection point. During the collection process, if the current block If it is not empty, execute step S350.

[0080] S350, the drone obtains the next action point according to the multi-agent reinforcement learning algorithm, and uses the current position as the starting point and the next action point as the end point, and calls the heuristic algorithm to plan a detection sequence. According to the detection sequence, the drone needs to evaluate whether it has the ability to independently complete the detection sequence within the power limit. The minimum power requirement of the detection sequence is:

[0081] in, It's the drone in the block The total flight time of the planned path, It is for The minimum detection time of a device. If it meets The drone will use high-definition cameras, thermal imagers and other equipment to detect the operation of renewable energy power generation equipment one by one according to this route (such as structural damage of wind turbines, cleanliness of solar panels, thermal conditions of components, etc.), and the detection time for each device is not less than After completing a series of inspection tasks, the drone will continue to the next action point. , then execute step S360.

[0082] S360: The drone issues a collaboration request to other drones within the direct communication range, attempts to distribute the detection points to be detected in this block to other drones, and then performs joint detection. The specific process is as follows: S361: The drone in this block issues a collaboration request to other drones within the direct communication range. Returns the current location , the position of the next target point , remaining power and their ability to cooperate , It can be expressed as:

[0083] in, The participation in other detection tasks. If the user is participating in other detection tasks or is currently performing data collection, the task completion status will be used to select the If the value is 1, it means that the acquisition of a certain point has just been completed, or the collaboration has just ended, or the flight is in progress and there is no detection task to be completed; It's a drone With the collection point in the block to be coordinated distance; is a ratio used to adjust the influence of distance factor and remaining power factor on cooperation ability. If there is a drone base station in the network, the drone base station executes step S362; if not, the drone in this block executes step S362.

[0084] S362, according to Sort from high to low, and use swarm intelligence optimization algorithm to join the drone The route of the subsequent detection points is planned, where each drone The path should be from the current location Set off until the next destination When the power is insufficient, the drone should be planned to return to the nearby base station in time to replace the battery. The objective function of the algorithm is:

[0085] in, , are the weights of the UAV flight time and altitude change in the scheme respectively; For drones Flight time; For drones The height change value is used to encourage the drone to fly smoothly; is the number of drones included in the local path planning. The constraints are: · Power constraints. Each drone At any time The remaining energy Cannot fall below the minimum threshold , and upon arrival The power shall not be less than .

[0086] Position constraints. The position of each drone must be continuous, starting from Departure, arrival Finish.

[0087] No interruption constraints. The task of each detection point needs to be completed in one go, and no interruption is allowed.

[0088] No duplication constraints. Each detection point can only be detected once by a drone.

[0089] Full coverage constraint. All detection points in the block must be fully detected in the local path planning.

[0090] · Minimum detection time constraint. For each detection point, the detection time must not be less than .

[0091] Get the target value of the solution after adding the new drone , if it satisfies:

[0092] The drone Add to the scheduling plan. Among them, is the planned target value before adding the drone, is the ratio of the importance of time optimization to collaboration. Repeat this step until the equation is no longer satisfied, and use the last planned The corresponding solution is used, and the detection points required for each drone to be detected are sent to the corresponding drone through the self-organizing network.

[0093] S363. The drone that receives the path point departs from its own position. If it is in the collection state or the detection state, it departs after completing the current work. After the drone arrives at the detection point according to the received path, it uses high-definition cameras, thermal imagers and other detection equipment to detect the operation of renewable energy power generation equipment one by one. According to this path planning, the drone will complete the task and successfully reach the next original destination. .

[0094] S400, the power dispatching center completes the collection of regional meteorological data and observation data of renewable energy power generation equipment in the region through the drone base station, evaluates the operating performance of each device, and then The future power generation situation can be predicted to guide the pre-dispatch of power load.

[0095] S410, the power dispatch center calls the expected power generation model and calculates the power generation equipment based on the information collected by the drone Expected power generation during the data collection period , and retrieve the actual power generation of the device , calculate the energy efficiency ; According to the installed capacity of the equipment Calculating Capacity Factor ; Evaluate equipment integrity based on drone inspection data (0 means the equipment is completely damaged, 1 means the equipment is intact) and equipment cleanliness (0 means the equipment is seriously contaminated, 1 means the equipment is clean and bright), and a fuzzy comprehensive evaluation is conducted on each renewable energy power generation equipment based on these factors to scientifically evaluate the operating status of each equipment; S420: The power dispatching center obtains the regional weather forecast from the weather station and, combined with the evaluation result of step S410, estimates the total power generation capacity in the region in the future. 、Future regional power load Predictions are made through neural network models to guide power load pre-dispatching.

[0096] The evaluation method of this embodiment adopts multi-UAV collaboration to collect targeted meteorological data and conduct efficient and accurate detection of renewable energy power generation equipment. The block division is carried out using the rhombic dodecahedron filling principle to make full use of the meteorological data collection range of the UAV. The distributed execution of the multi-agent reinforcement learning strategy allows the UAV to effectively and dynamically plan the action route under poor communication conditions. A UAV self-organizing network mechanism is established, and a group intelligence optimization algorithm is introduced. The decision-making and task allocation capabilities of the UAV are enhanced through the networking combined with timing and broadcast requests, so that multiple UAVs can achieve effective coordination of renewable energy power generation equipment detection and meet the detection needs of regional renewable energy power generation equipment under complex conditions. Combined with multiple factors of the detection results, the performance of the power generation equipment is evaluated using fuzzy comprehensive evaluation. At the same time, the total power generation capacity in the future region and the future power load in the region are predicted through a neural network model to guide the pre-dispatch of power load. This method provides an efficient and reliable solution for the detection of renewable energy power generation equipment and the pre-dispatch of power load in the region, effectively improving the detection efficiency of renewable energy power generation equipment in the region and the accuracy of power load pre-dispatch.

[0097] Embodiment 2: This embodiment provides a regional renewable energy power generation equipment performance evaluation system, such as Figure 3 As shown, including: Power dispatching center, multiple drone base stations, and several drones deployed at the drone base stations; Renewable energy power generation equipment includes photovoltaic power generation equipment and wind power generation equipment.

[0098] The power dispatching center accepts or generates the demand for renewable energy equipment performance evaluation, plans the spatial area where meteorological data needs to be collected and contains all the renewable energy equipment to be evaluated; and divides the spatial area into several collection blocks according to the meteorological information collection range of the drone. The center of each collection block is the corresponding collection point, among which the collection point corresponding to the collection block containing the renewable energy equipment to be evaluated is the collection point of interest, and the rest are ordinary collection points; The drone dynamically obtains the information of the current collection point through the self-organizing network, and then plans the flight path according to the information of the collection point and its own power reserve, and independently completes or multiple drones work together to complete the meteorological data collection task of the collection point, as well as the renewable energy equipment detection task of the interest point; after completing all the interest point detection tasks and the collection tasks of the collection point, the drone returns to the drone base station and transmits the collected data to the power dispatching center; The power dispatching center evaluates the operating performance of each renewable energy device based on the collected meteorological data and renewable energy equipment detection data, and then predicts the future power generation situation in the spatial area to guide the pre-dispatching of power loads.

[0099] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating the performance of regional renewable energy power generation equipment, characterized in that: include: S100, the power dispatching center accepts or generates a demand for renewable energy equipment performance evaluation, and plans a spatial area that needs to collect meteorological data and includes all renewable energy equipment to be evaluated; S200, the power dispatching center divides the spatial area into several collection blocks according to the meteorological information collection range of the UAV, the center of each collection block is the corresponding collection point, the collection point corresponding to the collection block containing the renewable energy equipment to be evaluated is the interest collection point, and the rest are common collection points; S300, the drone dynamically obtains the information of the current point to be collected through the self-organizing network, and then plans the flight path according to the information of the point to be collected and its own power reserve, and independently completes or multiple drones cooperate to complete the meteorological data collection task of the collection point, as well as the renewable energy equipment detection task of the interest point; after completing all the interest point detection tasks and the collection tasks of the collection point, the drone returns to the drone base station and transmits the collected data to the power dispatching center; S400. The power dispatching center evaluates the operating performance of each renewable energy device based on the collected meteorological data and renewable energy device detection data, and then predicts the future power generation situation in the spatial area to guide the pre-dispatching of power loads.

2. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 1, characterized in that: Step S200 includes: S210, the power dispatching center dispatches several drones, and obtains the current base station locations of all drones, and the effective detection range of the sensors used by each drone to collect meteorological data; S220, the minimum effective detection range among all drone sensors As a standard, the spatial area is divided into several acquisition blocks; S230. Taking the center of each collection block as the collection point, the power dispatching center sets the collection points corresponding to the collection blocks containing the renewable energy equipment to be evaluated as the collection points of interest according to the distribution of renewable energy equipment in the collection blocks, and the rest as ordinary collection points. The task type of each collection block is set accordingly; and according to the task type, the type of renewable energy equipment in the collection block, and the distance between the renewable energy equipment and the collection point, the additional task reward and the shortest collection time of the corresponding collection block are set.

3. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 2, characterized in that: In step S220, the minimum effective detection range of all drone sensors is used as the standard to divide the space area into several acquisition blocks. The specific method is as follows: A sphere is constructed with the minimum effective detection range of all drone sensors, and the largest rhombic dodecahedron inscribed in the sphere is determined. The rhombic dodecahedron is used to fill the spatial area planned in step S100 to obtain a spatial grid evenly distributed in the spatial area, and each spatial network is used as a collection block.

4. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 2, characterized in that: Step S300 includes: S310, the power dispatching center initializes the map information, determines the information of each collection block, and synchronizes the mission map with the UAV base station. The UAV takes off from the base station and goes to the collection point to perform the mission; S320, the UAV performs the meteorological data collection task and the renewable energy equipment detection task according to the task type of the collection point and its own power reserve. After completing the task of the current collection point, the UAV updates the task map and marks the collection point that has completed the task; S330, when the drone is performing a task or heading to the next collection point, the drone attempts to establish communication with the base station or other drones at regular intervals. If communication is established, the drone and the base station or other drones are networked with each other, and the drone exchanges its own task map with the base station or other drones, and informs its own location information and the collection point it is working on; S340: The drone plans the next action route according to the updated mission map, and loops through steps S320 and S330.

5. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 4, characterized in that: In step S310, the grid point information in the task map includes the following contents: Grid collection points Location ; Collection identification , used to mark whether the point has been collected. If it has been collected, it is 0, if it has not been collected, it is 1; interest identifier , used to mark whether the block is an interest block, if it is not an interest block, it is 0, if it is an interest block, it is 1; the minimum acquisition time at the acquisition point ; Location table of renewable energy equipment within the grid ,in, It is The three-dimensional coordinates of renewable energy power generation equipment, is the three-dimensional coordinate of the last renewable energy power generation device, is the number of renewable energy generation equipment in the grid; the shortest inspection schedule of the equipment in the grid ,in It is The shortest inspection time for renewable energy power generation equipment, is the shortest inspection time of the last renewable energy power generation equipment; timestamp Used to save the last time the point status was updated; meteorological data matrix Used to store meteorological data and equipment data matrix within the grid Used to store the detection data of each device; grid reward value Defined as: ； in, Reward yourself, For derived rewards, is the charging incentive factor, It is the basic reward for all collection points. is the reward attached to the block, yes The number of surrounding collection points, It is the collection mark, which is used to identify whether the block has been collected; It is around The block's own reward, It is around According to the division method in step S230, Usually 12, It will be different at the border; At the same time, the drone base station that can be used to replace the drone battery ,in , is the number of drone base stations, marking the locations of the base stations , charging reward value Defined as: ； in, For charging rewards, For derived rewards, is the charging incentive factor, It is the basic reward for replacing the battery. From the current position of the drone to the base station The distance For the i Drone at the moment t location, yes location, yes The number of collection points around the block.

6. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 4, characterized in that: In step S340, when the drone plans the next action route, it determines to select one of the collection points adjacent to the current position according to its own power reserve, or returns to the base station to replenish power; And design a reward function to encourage drones to fully and economically use power reserves while ensuring a minimum safe power reserve; Specifically design the following reward function to constrain the behavior of the drone: ； in, is the residual energy excitation factor, It's the drone. The remaining power at the moment, It is the minimum power limit of the drone; ； in, is the battery surplus penalty factor, is the minimum distance between the drone and the battery replacement base station, is the maximum flight speed of the drone, is the maximum battery capacity of the drone, is the energy consumption per unit time of flight; ； in, is the power shortage penalty factor, At the collection point The shortest collection time, is the energy consumption per unit time of hover collection; ； in, is the flight stability excitation factor, It's the drone. the height of the moment; In summary, the reward function for the drone to select the next action block is: ； ； The constraints are: Power constraints, each drone At any time The remaining energy Cannot fall below the minimum threshold ; Position constraint: the position of each drone must be continuous. Departure, arrival Finish; Boundary constraints: drones should not exceed the the extent of the region; No interruption constraints. The tasks of each collection point must be completed at one time without interruption. There is no duplication constraint. The reward of each collection point in a grid can only be obtained once and cannot be obtained repeatedly by subsequent drones that visit the grid. The shortest detection time constraint: for each detection point, the detection time must not be less than ; ( For collection point The shortest collection time).

7. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 4, characterized in that: In step S320, when the current collection point is a common collection point, the UAV performs the meteorological data collection task, and the specific method is as follows: After the drone arrives at the collection point according to the planned route, the sensors installed on the drone collect the meteorological data of the collection area; When the current collection point is the collection point of interest, the drone performs the meteorological data collection task and the renewable energy equipment detection task. The specific method is as follows: When the drone arrives at the collection point, it first performs the meteorological data collection task. During the execution of the meteorological data collection task, it plans the route between the collection point and the renewable energy equipment. After completing the meteorological data collection task, it goes to the vicinity of the renewable energy equipment and uses its own imaging equipment to detect the operating status of the renewable energy equipment. The detection time is not less than the shortest collection time.

8. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 7, characterized in that: Before the UAV performs the renewable energy equipment inspection task, it plans the route between the collection point and the renewable energy equipment and evaluates whether it has the ability to independently complete the inspection task within the power reserve limit. The minimum power requirement of the UAV for this inspection sequence is: ； in, It's the drone in the block The total flight time of the planned path, is the energy consumption per unit time of flight, It is for The minimum detection time of a device, is the number of renewable energy generation equipment in the current grid, It is The shortest inspection time for renewable energy equipment is the energy consumption per unit time of hover collection, From the current position of the drone to the base station The distance For the i Drone at the moment t location, yes location, is the maximum flight speed of the drone; If satisfied The drone will detect the operation of renewable energy power generation equipment one by one according to this route, and the detection time for each device is not less than ; It's the drone. The remaining power at the moment, is the minimum power limit of the drone; after completing a series of detection tasks, the drone will continue to the next action point; if , then it issues a collaboration request to other drones within the direct communication range, and tries to dispatch the detection task of renewable energy equipment in this collection block to other drones to perform joint detection.

9. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 8, characterized in that: If the UAV does not have the ability to independently complete the inspection task within the power reserve limit, the specific method for performing joint inspection is as follows: The drone in this collection block sends a collaboration request to other drones within the direct communication range, and other drones in the ad hoc network return to their current locations. , the position of the next target point , remaining power and their ability to cooperate , It is expressed as: ； in, The participation in other detection tasks. If the user is participating in other detection tasks or is currently performing data collection, the task completion status will be used to select the If the value is 1, it means that the acquisition of a certain point has just been completed, or the collaboration has just ended, or the flight is in progress and there is no detection task to be completed; It's a drone With the collection point in the block to be coordinated distance; is a ratio used to adjust the influence of distance factor and residual power factor on cooperation ability. The current location of the drone and the block to be cooperated Location The distance between The drone in this collection block determines the best cooperative drone based on the current position of other drones, the position of the next target point, the remaining power reserve, and the cooperation ability, and sends the required detection points to the best cooperative drone through the self-organizing network. The specific method is as follows: according to Sort from high to low, and use swarm intelligence optimization algorithm to join the drone The route of the subsequent detection points is planned, where each drone The path should be from the current location Set off until the next destination When the power is insufficient, the drone should be planned to return to the nearby base station in time to replace the battery; the objective function of the algorithm is: ； in, , are the weights of the UAV flight time and altitude change in the scheme respectively; For drones Flight time; For drones The height change value is used to encourage the drone to fly smoothly; is the number of drones included in the local path planning; the constraints are: Power constraints, each drone At any time The remaining energy Cannot fall below the minimum threshold , and upon arrival , that is, the power consumption at the next original destination shall not be less than ; Position constraint: the position of each drone must be continuous. Departure, arrival Finish; No interruption constraints. The tasks of each detection point must be completed at one time without interruption. There is no duplication constraint, each detection point can only be detected once by a drone; Full coverage constraint: all detection points in the block must be fully detected in the local path planning; The shortest detection time constraint: for each detection point, the detection time must not be less than ; Get the target value of the solution after adding the new drone , if it satisfies: ； The drone Add to the scheduling plan, where is the planned target value before adding the drone, is the importance ratio of time optimization to collaboration ability; repeat this step until the equation is not satisfied, and use the last planned The corresponding scheme is used, and the detection points required for each drone are sent to the corresponding drone through the self-organizing network; After completing the current collection point task, the corresponding drone goes to the received detection point location to perform the detection task.

10. The method for evaluating the performance of regional renewable energy power generation equipment according to claim 4, characterized in that: Step S400 includes: S410, the power dispatching center calls the expected power generation model, calculates the energy efficiency and equipment capacity factor based on the renewable energy equipment detection data collected by the drone, evaluates the equipment integrity and equipment cleanliness, and performs a fuzzy comprehensive evaluation on each renewable energy power generation equipment; S420. Use the collected meteorological data to make a weather forecast for the spatial area planned in step S100. The power dispatching center predicts the total power generation capacity and the future power load in the spatial area through a neural network model based on the meteorological forecast results and the evaluation results obtained in step S410, thereby realizing power load pre-dispatching.

Images