Photovoltaic panel surface intelligent maintenance method and device, electronic equipment and storage medium

Abstract

The application discloses a kind of photovoltaic panel face intelligent maintenance method, device, electronic equipment and storage medium, involve photovoltaic power station operation and maintenance management technical field.The method comprises: obtaining the operation data of each photovoltaic panel management partition, environmental data and historical maintenance data, determine the fouling state index indicating current pollution degree;Based on the fouling state index and weather forecast data, predict the fouling state index in future time window, and calculate the risk index indicating comprehensive maintenance risk;According to fouling state index, risk index and maintenance related data, generate the maintenance scheme containing operation sequence, path planning and task schedule.The application predicts the fouling state and comprehensive risk by foresight, realizes the automation and fine management of photovoltaic panel face maintenance, improves the economy and adaptability of power station operation and maintenance.

Description

Technical Field

[0001] This application relates to the field of photovoltaic power plant operation and maintenance management technology, and in particular to intelligent maintenance methods, devices, electronic equipment and storage media for photovoltaic panels. Background Technology

[0002] During operation, photovoltaic power plants are susceptible to contaminants such as dust, bird droppings, and fallen leaves adhering to the surface of photovoltaic panels, leading to decreased light transmittance and reduced power generation efficiency. This problem is particularly pronounced in arid, windy, and low-rainfall areas.

[0003] Currently, photovoltaic panel maintenance mainly employs methods such as manual inspection and fixed-period cleaning, automatic cleaning triggered by output power or dust sensor thresholds, and simple rule-based decision-making combining drone inspection and image recognition. However, manual inspection and fixed-period cleaning methods are prone to problems such as untimely or excessive cleaning; furthermore, the working environment of photovoltaic panels is variable, and relying solely on a single threshold or simple rules cannot achieve accurate decision-making for maintenance tasks, thus affecting power generation revenue and the rational allocation of operation and maintenance resources.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide an intelligent maintenance method for photovoltaic panels, which aims to solve the technical problem that the existing technology cannot provide a method that can proactively predict the evolution trend of pollution and, on this basis, perform multi-objective dynamic optimization and rolling replanning of maintenance tasks, resulting in insufficient accuracy of maintenance decisions.

[0006] To achieve the above objectives, this application proposes a method for intelligent maintenance of photovoltaic panels, which is applied to photovoltaic power plants. The method includes: Obtain operational data, environmental data, and historical maintenance data for each photovoltaic panel management zone of the photovoltaic power plant; Based on the operational data, environmental data, and historical maintenance data of each photovoltaic panel management zone, the pollution status index of each photovoltaic panel management zone is determined, wherein the pollution status index is used to characterize the current pollution level of the photovoltaic panel management zone; Based on the fouling status indicators of each photovoltaic panel management zone and the weather forecast data of the area where the photovoltaic power station is located, the fouling status indicators of each photovoltaic panel management zone in the future time window are predicted, and based on the fouling status indicators of each photovoltaic panel management zone in the future time window, the risk indicators of each photovoltaic panel management zone in the future time window are calculated, wherein the risk indicators are used to characterize the comprehensive maintenance risks caused by the photovoltaic panel management zone in the future time window without cleaning; Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station is generated, which includes the operation sequence, path planning and task schedule. The maintenance-related data includes at least expected power generation loss, expected water consumption and spatial location information.

[0007] In one embodiment, the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, includes: Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone, the maintenance urgency indicators of each photovoltaic panel management zone are determined. Based on the maintenance urgency index of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, a maintenance plan including the work sequence, path planning and task schedule is generated for the photovoltaic power station.

[0008] In one embodiment, the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the maintenance urgency index of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station includes: In each of the photovoltaic panel management zones, the photovoltaic panel management zone with a maintenance urgency index greater than a first preset threshold is classified as a non-delay zone. In each of the photovoltaic panel management zones, the photovoltaic panel management zone whose maintenance urgency index is greater than the second preset threshold and not greater than the first preset threshold is classified as a deferred zone. Among the photovoltaic panel management zones, those whose maintenance urgency index is less than the second preset threshold are designated as candidate zones; For any of the candidate partitions, the minimum spatial distance between the candidate partition and the set consisting of the non-delayable partition and the delayable partition is calculated based on the maintenance-related data of the photovoltaic power station, and the marginal cleaning cost of the candidate partition is determined. When the minimum spatial distance is less than the third preset threshold and the marginal cleaning cost is less than the fourth preset threshold, the candidate partition is divided into a flexible supplementary partition. Based on the non-delayable partition, the delayable partition, the flexible supplementary partition, and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station, including the work sequence, path planning, and task schedule, is generated.

[0009] In one embodiment, the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the non-delayable partition, the delayable partition, the flexible supplementary partition, and the maintenance-related data of the photovoltaic power station, includes: Add each of the aforementioned non-delayable partitions to the set of tasks to be executed in the current maintenance cycle; Each of the deferable partitions is added to the set of tasks to be executed, and the deferable partitions that cannot be executed in the current maintenance cycle due to preset resource constraints are deferred to the next maintenance cycle. Based on the remaining resources in the current maintenance cycle, the elastic supplementation partition is added to the set of tasks to be executed; Based on the set of tasks to be executed and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station, including the work sequence, path planning and task schedule, is generated.

[0010] In one embodiment, after the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, the method further includes: After monitoring that the photovoltaic power station is performing maintenance tasks according to the maintenance plan, the task execution feedback data of each photovoltaic panel management zone is obtained; The historical maintenance data is updated based on the task execution feedback data of each photovoltaic panel management zone.

[0011] In one embodiment, after the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, the method further includes: Upon receiving a replanning instruction, the process returns to the step of obtaining the operating data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station, in order to regenerate the maintenance plan for the photovoltaic power station in the current maintenance cycle.

[0012] In one embodiment, after the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, the method further includes: Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone, the maintenance urgency indicators of each photovoltaic panel management zone are determined. Calculate the sum of the maintenance urgency indices for each of the photovoltaic panel management zones; When the sum exceeds the preset global urgency threshold, return to the step of obtaining the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station, so as to regenerate the maintenance plan of the photovoltaic power station in the current maintenance cycle.

[0013] Furthermore, to achieve the above objectives, this application also proposes a photovoltaic panel intelligent maintenance device, which includes: The data acquisition module is used to acquire operational data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station; The status assessment module is used to determine the pollution status index of each photovoltaic panel management zone based on the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone, wherein the pollution status index is used to characterize the current pollution level of the photovoltaic panel management zone; The risk assessment module is used to predict the contamination status index of each photovoltaic panel management zone in a future time window based on the contamination status index of each photovoltaic panel management zone and the weather forecast data of the area where the photovoltaic power station is located, and to calculate the risk index of each photovoltaic panel management zone in the future time window based on the contamination status index of each photovoltaic panel management zone in the future time window. The risk index is used to characterize the comprehensive maintenance risk caused by the photovoltaic panel management zone in the future time window without cleaning. The scheme generation module is used to generate a maintenance scheme for the photovoltaic power station, which includes the operation sequence, path planning and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station. The maintenance-related data includes at least expected power generation loss, expected water consumption and spatial location information.

[0014] In addition, to achieve the above objectives, this application also proposes an electronic device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the photovoltaic panel intelligent maintenance method as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the photovoltaic panel intelligent maintenance method described above.

[0016] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the photovoltaic panel intelligent maintenance method described above.

[0017] The one or more technical solutions proposed in this application have at least the following technical effects: First, by acquiring operational data, environmental data, and historical maintenance data of each photovoltaic panel management zone, a fouling status index characterizing the current degree of fouling is determined; based on the fouling status index and weather forecast data, the fouling status index of the photovoltaic panel management zone in future time windows is predicted, and a risk index characterizing the comprehensive maintenance risks that may occur in future time windows without cleaning is calculated based on the prediction results. This method does not passively rely on the current degree of pollution, but is based on a forward-looking prediction of the future evolution trend of the degree of pollution and the risk index calculated therefrom. It can identify areas that are about to deteriorate before fouling causes significant losses, avoiding missing the best cleaning opportunity due to delayed decision-making, thereby effectively improving the timeliness of maintenance decisions and reducing power generation losses caused by fouling accumulation.

[0018] Meanwhile, this application generates a maintenance plan that includes work sequence, path planning and task schedule based on the pollution status indicators, risk indicators and maintenance-related data of each photovoltaic panel management zone. It comprehensively considers multi-dimensional factors such as expected power generation loss, expected water consumption and spatial location information, so as to achieve a reasonable allocation of cleaning resources at the global level, and take into account both the timeliness and economy of maintenance, thereby improving the accuracy and intelligence level of photovoltaic power plant operation and maintenance. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating an embodiment of the intelligent maintenance method for photovoltaic panels in this application. Figure 2This is a detailed flowchart of a first embodiment of the intelligent maintenance method for photovoltaic panels in this application; Figure 3 This is a flowchart illustrating Embodiment 2 of the intelligent maintenance method for photovoltaic panels in this application. Figure 4 This is a schematic diagram of the module structure of the intelligent maintenance device for photovoltaic panels according to an embodiment of this application; Figure 5 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the intelligent maintenance method for photovoltaic panels in this application embodiment.

[0022] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0023] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of this application and are not intended to limit this application.

[0024] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0025] It should be noted that the execution subject in this embodiment can be an electronic device with data processing, network communication and program execution functions, such as a central monitoring server of a photovoltaic power station, an edge computing gateway or a cloud platform.

[0026] The following uses a photovoltaic power plant operation and maintenance management platform as an example to illustrate this embodiment and the following embodiments.

[0027] Example 1 During the operation of a photovoltaic power station, the degree of surface fouling is affected by multiple environmental factors, including irradiance, temperature, wind speed, rainfall, and dust, and its accumulation and reduction processes exhibit nonlinear and time-varying characteristics. In actual operation and maintenance, maintenance decisions need to answer three progressive questions: Is cleaning necessary? When is cleaning most economical? How should cleaning tasks be scheduled?

[0028] Currently widely adopted technical approaches can be broadly categorized into three types. The first type is preventative cleaning based on fixed cycles, such as monthly or quarterly cleaning. This method is simple to implement but completely ignores the differences in fouling rates across different areas and seasons, easily leading to untimely or excessive cleaning. The second type is trigger-based cleaning based on a single threshold, such as initiating cleaning when the component's output power falls below a certain percentage of its theoretical value. This method can only determine "whether cleaning is needed at the current moment," unable to predict the rapid deterioration of fouling in the next few hours, nor assess "how much damage will occur if cleaning is not performed now." The third type combines drone inspection, image recognition, and simple rules. While this improves the accuracy of fouling detection, task scheduling still relies primarily on preset rules or manual scheduling. When equipment malfunctions, weather changes, or task backlogs occur, it lacks the ability to dynamically rearrange the order, path, and resources of operations.

[0029] The aforementioned technical approaches share a common limitation: the contradiction between the static nature of the decision-making logic and the dynamic nature of the operational environment. Specifically, existing solutions all use monitoring data from the "current moment" as the input for decision-making, while the evolution of panel fouling is a continuously changing process over time; a good current state does not guarantee low future risk. Furthermore, cleaning resources (equipment, personnel, water) are limited, and different management zones vary in spatial distribution, criticality level, and urgency of fouling. Existing solutions struggle to incorporate these multi-dimensional, time-varying factors into a unified optimization framework for global solutions. When environmental or resource conditions change abruptly, preset solutions often fail to adapt adaptively, leading to decreased efficiency and cost-effectiveness in maintenance plan execution.

[0030] Based on this, the embodiments of this application provide a method for intelligent maintenance of photovoltaic panels, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the intelligent maintenance method for photovoltaic panels according to this application.

[0031] In this embodiment, the intelligent maintenance method for photovoltaic panels includes steps S10 to S40: Step S10: Obtain the operation data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station; It should be noted that a photovoltaic (PV) panel management zone refers to a pre-defined maintenance area based on a power plant array, string, or geographical region as the smallest management unit. Each PV panel management zone has a unique zone identifier (zone ID) to facilitate data association and traceability.

[0032] Operational data includes, but is not limited to, real-time operating parameters such as inverter output power, string current / voltage, and module temperature. This data can be acquired through the power plant's existing sampling system or data acquisition unit. The sampling frequency can be set according to actual needs, for example, once every 5 minutes.

[0033] Environmental data includes, but is not limited to, measured meteorological data such as irradiance, ambient temperature and humidity, wind speed, and rainfall. This data can be obtained through environmental monitoring instruments installed at the power station or from nearby weather stations. When conditions permit, visible light and infrared images obtained from fixed cameras or drone inspections can also be used to assist in assessing the surface contamination status.

[0034] Historical maintenance data includes, but is not limited to: historical cleaning time, cleaning area, cleaning method (manual or robotic cleaning), post-cleaning power generation recovery curve characteristics, equipment availability status, and operator or robot scheduling information. This historical maintenance data records the actual execution and effectiveness of each cleaning task and serves as a crucial basis for subsequent parameter optimization.

[0035] All the aforementioned data types are written to a time-series database with a unified timestamp, and a three-level association relationship of "partition ID - device ID - time slice" is established to form a data foundation that can be used for online decision-making. Before the data is written, the system can also perform preprocessing operations on the raw data, such as missing data completion, anomaly removal, smoothing and noise reduction, time alignment, and standardization, to ensure the accuracy of subsequent calculations.

[0036] Step S20: Determine the pollution status index of each photovoltaic panel management zone based on the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone. The pollution status index is used to characterize the current pollution level of the photovoltaic panel management zone. It should be noted that the Contamination Status Index (DI) is a normalized comprehensive quantitative index with a value range of [0, 1]. The larger the value, the more severe the contamination of the board surface.

[0037] In one feasible implementation, the fouling state index DI_i(t) is calculated using the following formula:

[0038] Where w1, w2, w3, and w4 are weighting coefficients, satisfying w1 + w2 + w3 + w4 = 1. Specifically, w1 is the weighting coefficient corresponding to η_i(t), w2 is the weighting coefficient corresponding to σ_i(t), w3 is the weighting coefficient corresponding to C_i(t), and w4 is the weighting coefficient corresponding to D_i(t). Each weighting coefficient can be determined based on historical data regression analysis or configured by operation and maintenance experts according to the actual situation of the power plant. For example, in arid and dusty areas, the weights of w1 and w4 can be appropriately increased; in power plants with significant differences in string configuration, the weight of w2 can be appropriately increased.

[0039] η_i(t) is the normalized efficiency loss, used to characterize the degree of power generation efficiency degradation caused by pollution. Its calculation formula is:

[0040] Where P_act_i(t) is the actual output power of partition i at time t, and P_clean_i(t) is the clean reference power, which is the theoretical output power under the same irradiance and temperature conditions assuming the panel is completely clean. P_clean_i(t) can be estimated by the clean reference model, which is calibrated based on the photovoltaic module's factory parameters and measured data under historical clean conditions.

[0041] σ_i(t) represents the string dispersion, which reflects the degree of difference in current or power between different strings within the same partition. When local contamination occurs on the board surface (such as bird droppings or fallen leaves blocking the path), the output of the blocked strings will decrease significantly, leading to an increase in the dispersion between strings. String dispersion can be represented by the standard deviation or coefficient of variation of the string current.

[0042] C_i(t) represents the pollution coverage rate based on image recognition. In power plants equipped with cameras or drones for inspection, image processing algorithms can be used to identify polluted areas on the panels and calculate the percentage of the polluted area relative to the total area of ​​the partition. Image recognition can be implemented using a semantic segmentation model based on deep learning.

[0043] D_i(t) is the continuous rainless accumulation factor, used to characterize the cleaning effect of natural rainfall on the surface. The longer the continuous rainless period, the more severe the accumulation of dirt. D_i(t) can be normalized based on the number of days since the last effective rainfall (rainfall exceeding a preset threshold).

[0044] In this embodiment, the pollution level of the photovoltaic panel is comprehensively characterized by the weighted fusion of multi-dimensional features. It considers the direct reflection of electrical parameters, integrates the intuitive information of image recognition, and also incorporates the influence of environmental factors such as rainfall, thereby improving the accuracy and robustness of pollution level assessment.

[0045] In another feasible implementation, for power plants that are not equipped with image acquisition equipment, the C_i(t) term can be omitted, and the weights w1, w2, and w4 can be redistributed. The pollution level can still be effectively assessed through electrical parameters and rainfall information.

[0046] It is understood that the two implementation methods described above can be flexibly selected based on the hardware configuration of the power plant. The first implementation method has higher evaluation accuracy, but requires additional investment in image acquisition equipment; the second implementation method has lower implementation costs and is suitable for the intelligent transformation of existing power plants. The above are only two feasible implementation methods for step S20 provided in this embodiment, and this embodiment does not specifically limit the specific implementation method of step S20.

[0047] Step S30: Based on the pollution status indicators of each photovoltaic panel management zone and the weather forecast data of the area where the photovoltaic power station is located, predict the pollution status indicators of each photovoltaic panel management zone in the future time window, and calculate the risk indicators of each photovoltaic panel management zone in the future time window based on the pollution status indicators of each photovoltaic panel management zone in the future time window. The risk indicators are used to characterize the comprehensive maintenance risks caused by the photovoltaic panel management zone in the future time window without cleaning. It should be noted that the future time window can be set according to the operation and maintenance scheduling cycle, such as setting it to the next 24 hours or 48 hours. Weather forecast data includes information such as wind speed, wind direction, rainfall, humidity, temperature, and dust forecast for each time point within the future time window, which can be obtained through the meteorological service interface.

[0048] It should be noted that the comprehensive maintenance risk is a quantification of the complex negative impacts that the decision to "not carry out cleaning for the time being" will cause in the future, and it includes risk dimensions such as the estimated cumulative power generation loss and the rate of pollution deterioration.

[0049] In one feasible implementation, the fouling status index within the future time window is predicted iteratively hourly using the following fouling evolution model:

[0050] Wherein, DI_i(t) is the fouling status index at time t; X_i(t) is the meteorological and operational feature vector including wind speed, rainfall, humidity, and temperature; β_i^T is the transpose of the corresponding coefficient; u_i(t) is the cleaning decision variable (takes 1 if cleaning is performed at time t, otherwise takes 0); α_i is the fouling self-attenuation / accumulation coefficient, reflecting the natural change trend of fouling without external intervention; β_i is the meteorological influence coefficient vector, reflecting the degree of influence of different meteorological conditions on fouling accumulation or reduction; γ_i is the cleaning recovery coefficient, reflecting the degree of recovery of the degree of pollution after cleaning; f(·) is the amplitude limiting function, which restricts the prediction results to the interval [0, 1].

[0051] Model parameters α_i, β_i, and γ_i can be determined through regression analysis of historical data for each partition. For example, during dusty weather, the coefficients in β_i corresponding to wind speed and particulate matter concentration will cause DI to increase; during rainy weather, the coefficient corresponding to rainfall will cause DI to decrease.

[0052] Based on the predicted pollution state index sequence DI_i(t+h) for each time point within the future time window, the corresponding power generation can be calculated using a power mapping model:

[0053] Wherein, P_clean_i(t+h) is the theoretical power under clean conditions estimated based on irradiance and temperature forecasts, k_i is the fouling attenuation coefficient, and p_i is the nonlinear exponent, reflecting the nonlinear relationship between the degree of fouling and power attenuation.

[0054] The power generation loss at each moment is calculated in the following way:

[0055] The estimated cumulative power generation loss is:

[0056] Where p_i(t+h) is the time-of-use electricity price or generation weight coefficient (higher weight during peak hours and lower weight during off-peak hours), and Δt is the time interval.

[0057] Based on this, the risk indicator RI_i is calculated in the following way:

[0058] Wherein, λ1 is the weighting coefficient corresponding to Li(H), λ2 is the weighting coefficient corresponding to max_h d(ΔP_i) / dh, λ3 is the weighting coefficient corresponding to W_i, λ4 is the weighting coefficient corresponding to C_i, Norm(·) is the normalization function, and Li(H) is the estimated cumulative power generation loss within the future time window H, represented by the normalized estimated cumulative power generation loss. For example, if partition A is estimated to lose 200 kWh in the next 24 hours and partition B is estimated to lose 50 kWh, then under the same conditions, the value of this component of partition A will be higher. max_h d(ΔP_i) / dh is the maximum rate of change of power generation attenuation within the future time window H, used to characterize the drastic degree of pollution deterioration. When the weather forecast indicates that a sandstorm or sustained high temperature and dry weather will occur in the future, the slope of the power attenuation curve will increase sharply, and this component value will increase significantly. For example, during the sandstorm weather warning period, even if the current pollution is still light, this component will drive the risk index to rise, triggering preventive cleaning. W_i represents the penalty for the unwashed time window. The longer the time since the last wash, the larger this penalty becomes, reflecting the operational experience that "the longer the wash is not performed, the greater the potential risk." C_i is the criticality level coefficient of the partition, which can be set according to the partition's importance in the power plant (such as the proportion of installed capacity, grid connection level, etc.). C_i values ​​are larger for critical partitions, giving high-risk, high-value areas higher maintenance priority.

[0059] Step S40: Based on the pollution status indicators, risk indicators, and maintenance-related data of each photovoltaic panel management zone, generate a maintenance plan for the photovoltaic power station that includes the work sequence, path planning, and task schedule. The maintenance-related data includes at least the expected power generation loss, expected water consumption, and spatial location information.

[0060] It should be noted that the core of this step is to comprehensively process the pollution status assessment results (pollution status indicators) and future risk prediction results (risk indicators) of each photovoltaic panel management zone with the resource data (maintenance-related data) required for the cleaning operation, and generate a directly executable maintenance plan through optimization solution.

[0061] It should be noted that maintenance-related data refers to parameter information directly related to the execution of cleaning tasks in each photovoltaic panel management zone, which is used as the basis for generating the maintenance plan. Specifically: Expected power generation loss refers to the estimated power generation loss that would result if the photovoltaic panel management zone were not cleaned within a preset future time window, based on the current state of contamination and predictions of future contamination evolution. This value can be expressed in kilowatt-hours or converted into a corresponding economic loss amount. When generating maintenance plans, expected power generation loss serves as a core indicator for measuring the urgency and economic benefits of cleaning each zone; zones with higher expected power generation loss are more likely to be prioritized for cleaning in the plan.

[0062] The estimated water consumption refers to the amount of water resources required to perform a complete cleaning operation on each photovoltaic panel management zone. This estimate is determined by considering the following factors: the total area of ​​the photovoltaic panels in the zone, the severity of the current fouling status (more severe fouling usually requires more water), the rated water consumption per unit area of ​​the cleaning equipment, and the statistical values ​​of actual water consumption for similar zones with the same degree of fouling in historical maintenance data. When generating a maintenance plan, the estimated water consumption is used to determine whether the cleaning equipment has sufficient water reserves to complete the cleaning task for that zone in a single operation, thus affecting task allocation and equipment scheduling decisions.

[0063] Spatial location information refers to the geographical location data of each photovoltaic panel management zone within the photovoltaic power plant area, such as latitude and longitude coordinates, zone center point coordinates, and zone boundary vertex coordinates. Based on the spatial location information of each zone, the travel distance or travel time between any two zones can be calculated. When generating maintenance plans, spatial location information is mainly used to plan the movement path of cleaning equipment between different zones. By optimizing the access sequence between zones, the idle mileage and ineffective maneuvering time of the equipment can be reduced, thereby reducing overall operating costs and improving operating efficiency.

[0064] It should be noted that the maintenance-related data is not limited to the items listed above. In practical applications, it can be further included, but not limited to, the following, based on the power plant's operation and maintenance management needs: historical cleaning frequency of each photovoltaic panel management zone, available quantity and real-time status of cleaning equipment, shift schedule information of operators, travel speed of cleaning equipment under different road conditions, location distribution of water replenishment points or charging piles, and time window restrictions for cleaning operations in each zone. This embodiment does not impose specific limitations on this, as long as the data can assist in generating or optimizing the maintenance plan.

[0065] It should be noted that the operation sequence is a list of photovoltaic panel management zones that each cleaning device will visit in turn, clearly defining the order in which each zone will be cleaned. The path planning is a suggested route for each cleaning device between different zones, usually represented by a sequence of lines connecting the zones. The task schedule includes the planned start and estimated completion times for each cleaning task, as well as the times when each device returns to the water replenishment point or ends its shift.

[0066] During the maintenance plan generation process, contamination status indicators, risk indicators, and maintenance-related data for each zone can be input into a preset optimization model. This optimization model can comprehensively consider one or more of the following optimization objectives: (1) Minimize expected power generation loss: Prioritize the execution of the partition with greater expected power generation loss in the earlier period of the current maintenance cycle, so as to recover potential power generation loss as early as possible; (2) Minimize cleaning execution costs: Reduce direct cleaning costs such as manpower, equipment wear and tear and energy consumption by reasonably allocating tasks and planning routes; (3) Minimize water consumption: Under the premise of meeting the cleaning effect requirements, control the overall water consumption by selecting reasonable cleaning methods and scheduling sequence; (4) Minimize the distance of the operation path and the maneuver time: Based on the spatial location information of each partition, plan the partition access order with the shortest total travel distance or the least total maneuver time; (5) Equipment utilization and task balance: The cleaning tasks are reasonably distributed among the available cleaning equipment to avoid some equipment from working continuously for a long time while other equipment is idle, so that the workload of each equipment tends to be balanced. (6) Time window violation penalty: For some zones with specific workable periods (such as avoiding the midday high temperature period or the power plant power restriction period), if the start time of the scheduled task exceeds its allowed time window, the corresponding penalty item will be applied in the optimization objective.

[0067] The optimized solution must also satisfy one or more of the following constraints to ensure the feasibility of the generated maintenance plan in actual implementation: (a) Equipment capacity constraints: Each cleaning equipment has capacity limitations such as maximum operating range, maximum water capacity, maximum continuous operation time and rated operating speed. The cumulative consumption of tasks assigned to the equipment shall not exceed its capacity limit. (b) Operation window constraints: Photovoltaic power plants may stipulate that cleaning operations are prohibited during certain periods (such as at night or under severe weather conditions), and task scheduling needs to avoid these prohibited periods; (c) Weather conditions: When the weather forecast indicates that the wind speed will exceed the safe operating threshold of the cleaning equipment during certain periods within the future time window, or when rainfall is forecast and cleaning is not advisable, the task shall not be scheduled during these periods; (d) Water replenishment / endurance constraints: For water-washing cleaning equipment, when the estimated cumulative water consumption is about to exceed the equipment's water capacity, the plan should reserve the travel time and space to the water replenishment point; for electric equipment, the time and path for charging or battery swapping should be reserved. (e) Constraint on the number of concurrent tasks: The number of devices that can perform cleaning tasks at the same time is limited by the power plant's maintenance personnel configuration or on-site management capabilities.

[0068] In practical implementation, mixed-integer programming can be used to construct the above optimization objectives and constraints into a mathematical model for accurate solution. Alternatively, heuristic algorithms such as genetic algorithms and particle swarm optimization can be used to quickly obtain approximate optimal solutions for large-scale problems. During the solution process, the model treats each photovoltaic panel management zone as a task node to be processed. Each task node is associated with its corresponding pollution status indicators, risk indicators, and maintenance-related data. Through iterative search or evolutionary calculation, the task allocation and sorting scheme that optimizes the comprehensive objective function is found.

[0069] In this embodiment, a data foundation for multi-source information fusion is established by acquiring operational data, environmental data, and historical maintenance data from each photovoltaic panel management zone. Based on this, a pollution status index characterizing the current pollution level is determined, achieving accurate quantification of the panel surface pollution status. By combining weather forecast data to predict the pollution status index within a future time window and calculating a comprehensive maintenance risk index, maintenance needs can be identified in advance before pollution causes significant losses, avoiding delayed decision-making. Finally, a maintenance plan including work sequence, path planning, and task schedule is generated by integrating the pollution status index, risk index, and maintenance-related data, ensuring reasonable allocation of cleaning resources, balancing maintenance timeliness and economy, and improving the intelligent level of photovoltaic power plant operation and maintenance.

[0070] In one feasible implementation, step S40 may include steps S41-S42: Step S41: Determine the maintenance urgency index for each photovoltaic panel management zone based on the pollution status index and risk index of each photovoltaic panel management zone; It should be noted that the maintenance urgency index is used to quantify the urgency of cleaning tasks in each photovoltaic panel management zone, so as to differentiate the processing priority of different zones when generating subsequent maintenance plans. The contamination status index reflects the actual degree of contamination on the panel surface at the current moment, while the risk index reflects the comprehensive maintenance risks that may be faced in the future time window without cleaning. Combining the two allows for a more comprehensive assessment of the maintenance urgency of a zone.

[0071] For example, the maintenance urgency index U_i can be calculated using a weighted summation:

[0072] Where DI_i(t) is the current pollution status index of partition i, RI_i is the risk index of partition i within a future time window, μ1 is the weight coefficient corresponding to DI_i(t), and μ2 is the weight coefficient corresponding to RI_i, satisfying μ1+μ2=1. The weight coefficients can be configured according to the power plant operation and maintenance strategy. For example, if the power plant focuses more on the immediate treatment of the current pollution level, the weight of μ1 can be appropriately increased; if the power plant focuses more on the early prevention of future losses, the weight of μ2 can be appropriately increased. A typical configuration is μ1=μ2=0.5, that is, the current status and future risk are equally important.

[0073] For example, maintenance urgency indicators can also be combined in a non-linear manner, such as:

[0074] The greater of the current contamination status indicator and the risk indicator is taken as the urgency indicator. This method emphasizes the "weakest link effect," meaning that whichever indicator is too high should be addressed first.

[0075] Step S42: Based on the maintenance urgency index of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, generate a maintenance plan for the photovoltaic power station that includes the work sequence, path planning and task schedule.

[0076] It should be noted that, based on the maintenance urgency index determined in step S41, this step, combined with relevant maintenance data of the photovoltaic power station, uniformly arranges the cleaning tasks for each photovoltaic panel management zone. The maintenance urgency index serves as a quantitative basis for measuring the urgency of the cleaning tasks in each zone, and is used to guide the determination of task priorities when generating the maintenance plan—the higher the maintenance urgency index of a zone, the more likely its corresponding cleaning tasks will be prioritized in the plan.

[0077] In this process, the maintenance urgency index achieves a comprehensive quantitative assessment of zonal maintenance needs by integrating pollution status and risk indicators. Compared to single-dimensional judgments relying solely on pollution status or risk indicators, the maintenance urgency index simultaneously considers both the "current actual pollution status" and the "future risk evolution trend." For example, for a zonal with relatively light pollution but a high risk indicator in the future time window, decision-making based solely on the pollution status index might lead to incorrectly delayed processing, resulting in avoidable power generation losses. Conversely, for a zonal with severe pollution but naturally improving weather conditions (such as predicted rainfall), decision-making based solely on the risk indicator might lead to over-priority processing, wasting cleaning resources. By combining both into a unified maintenance urgency index, a reasonable balance can be achieved between the current state and future risks, providing a more scientific prioritization reference for subsequent solution generation.

[0078] In practical implementation, the maintenance urgency index of each photovoltaic panel management zone, along with maintenance-related data such as expected power generation loss, expected water consumption, and spatial location information, can be input into the optimization model. Under the premise of satisfying equipment capacity constraints, operation window constraints, and weather condition constraints, the optimization model uses the maintenance urgency index as the main ranking criterion and combines it with maintenance-related data to perform a comprehensive solution, ultimately generating a maintenance plan that includes operation sequence, path planning, and task schedule.

[0079] In this embodiment, steps S41-S42 integrate the contamination status index and risk index into a unified maintenance urgency index, thereby achieving a comprehensive quantitative assessment and ranking of the maintenance needs of each zone. This avoids the one-sidedness of relying solely on the current contamination status or future risks, enabling subsequent solution generation to achieve a reasonable balance between the current status and future risks. Based on this, maintenance solutions are generated by combining maintenance-related data, ensuring that the allocation of cleaning resources responds to the actual comprehensive urgency of each zone while also taking into account operating costs and resource constraints, further improving the scientificity and rationality of the maintenance solutions.

[0080] In one feasible implementation, step S40 may further include steps S43 to S48: Step S43: In each photovoltaic panel management zone, the photovoltaic panel management zone with a maintenance urgency index greater than the first preset threshold is classified as a non-delay zone; It should be noted that the first preset threshold is used to define the highest level of maintenance urgency. Undelayable zones refer to zones whose maintenance urgency index has exceeded the first preset threshold. These zones are severely contaminated or pose extremely high future risks. If not addressed within the current maintenance cycle, they will result in significant power generation losses or operational risks. Therefore, cleaning tasks need to be prioritized in the current maintenance plan.

[0081] Step S44: In each photovoltaic panel management zone, the photovoltaic panel management zone with a maintenance urgency index greater than the second preset threshold and not greater than the first preset threshold is divided into a delayable zone. It should be noted that the second preset threshold is lower than the first preset threshold. Delayable partitions refer to partitions whose maintenance urgency index falls between the second and first preset thresholds. While these partitions have reached the point where cleaning is necessary, they have a certain time tolerance—cleaning them within the current maintenance cycle can effectively mitigate losses; if cleaning cannot be scheduled due to resource constraints, postponing it to the next maintenance cycle will not cause overly serious consequences. The division of delayable partitions provides flexibility for subsequent scheduling adjustments under resource constraints.

[0082] Step S45: In each photovoltaic panel management zone, the photovoltaic panel management zone with a maintenance urgency index less than the second preset threshold is classified as a candidate zone; It should be noted that the alternative zones refer to zones whose maintenance urgency index is below the second preset threshold. These zones have relatively minor current contamination and low future risk, and according to conventional decision-making logic, they do not need to be included in the current maintenance plan. However, considering that the photovoltaic power plant cleaning equipment may generate surplus resources in actual operation (such as remaining water volume or remaining operation time after completing the main tasks), the existence of alternative zones provides a potential source of tasks for opportunistic scheduling such as "cleaning along the way".

[0083] There are several ways to set the first and second preset thresholds. In one specific implementation, the thresholds can be dynamically adjusted based on the total amount of cleaning resources available during the shift. For example: First preset threshold:

[0084] Second preset threshold:

[0085] Where Th_base_high and Th_base_low are baseline thresholds, C_available is the number of currently available cleaning devices, C_total is the total number of cleaning devices, α is the adjustment coefficient for the first preset threshold, and β is the adjustment coefficient for the second preset threshold. When available resources are sufficient, the first preset threshold is appropriately lowered to classify more partitions as non-delayable, making full use of idle resources; when resources are scarce, the first preset threshold is appropriately raised, retaining only the most urgent partitions as non-delayable, and classifying the rest as delayable.

[0086] In another specific implementation, the threshold can also be a fixed value, preset by operations and maintenance personnel based on historical experience. For example, the urgency index range [0, 1] can be divided into three intervals: [0, 0.3] for low urgency, (0.3, 0.7] for medium urgency, and (0.7, 1] for high urgency. The first preset threshold is set to 0.7, and the second preset threshold is set to 0.3. Partitions with urgency greater than 0.7 are non-delayable partitions, partitions with urgency between 0.3 and 0.7 are delayable partitions, and partitions with urgency less than 0.3 are candidate partitions.

[0087] Step S46: For any candidate partition among the candidate partitions, calculate the minimum spatial distance between the candidate partition and the set consisting of the non-delayable partition and the delayable partition based on the maintenance-related data of the photovoltaic power station, and determine the marginal cleaning cost of the candidate partition. It should be noted that the minimum spatial distance refers to the shortest spatial distance between the candidate partition and the higher-priority partitions (i.e., non-delayable and delayable partitions) that require cleaning tasks. This distance can be calculated based on the spatial location information of each partition, such as using Euclidean distance or the actual driving distance based on the internal road topology network of the power plant. The calculation range can be limited to the distance between the candidate partition and each partition in the above set. To reduce the computational load, a maximum search radius can be set, and areas exceeding this radius will not be calculated.

[0088] Marginal cleaning cost refers to the additional cost incurred by cleaning equipment when cleaning an alternative zone beyond the completion of existing tasks. Marginal cleaning cost can be comprehensively considered from the following factors: path-related costs (the additional travel distance and time required to move from the existing task point to the alternative zone and back along the original path), operation-related costs (the water, electricity, and operation time consumed in cleaning the alternative zone), and time window impact (the penalty for exceeding the time window of subsequent tasks if adding the alternative zone causes it).

[0089] In one specific implementation, the marginal cleaning cost MC can be expressed as:

[0090] Where Δd is the extra driving distance, Δt is the extra time spent, Δw is the extra water consumption, c1 is the cost weight coefficient corresponding to the extra driving distance, c2 is the cost weight coefficient corresponding to the extra time spent, c3 is the cost weight coefficient corresponding to the extra water consumption, and P_window is the time window penalty item.

[0091] Step S47: When the minimum spatial distance is less than the third preset threshold and the marginal cleaning cost is less than the fourth preset threshold, the candidate partition is divided into an elastic supplementary partition. It should be noted that the third preset threshold is used to define the scope of "spatial proximity". For example, for a small distributed power station, the third threshold can be set to 100 meters; for a large ground power station, the third threshold can be set to 500 meters or more. If the minimum spatial distance is less than the third threshold, it means that the candidate zone is located on the necessary route or in the vicinity of the cleaning equipment to the existing task zone, and has the spatial conditions for "cleaning along the way".

[0092] The fourth preset threshold is used to define the standard for "sufficiently low marginal cost". For example, the fourth threshold can be set to 30% of the cost required to clean the partition once normally. If the marginal cleaning cost is less than the fourth threshold, it means that "on-the-way cleaning" is significantly more economical than future individual scheduled cleaning.

[0093] Elastic supplementary partitions are those partitions that, while not urgently requiring cleaning, can be cleaned "on the way" when remaining resources allow because they are adjacent to partitions already designated for cleaning and have sufficiently low marginal costs. Identifying and marking elastic supplementary partitions provides a low-cost source of tasks for opportunistic scheduling when resources are idle.

[0094] Step S48: Based on the non-delayable partition, delayable partition, flexible supplementary partition, and maintenance-related data of the photovoltaic power station, generate a maintenance plan for the photovoltaic power station that includes the work sequence, path planning, and task schedule.

[0095] It should be noted that this step uses the three types of partitions identified in the previous steps—non-delayable partitions, delayable partitions, and flexible supplementary partitions—as the source of tasks for generating maintenance plans. When generating plans, non-delayable partitions have the highest processing priority, followed by delayable partitions, with flexible supplementary partitions serving as a supplementary option when resources are idle. Combining the maintenance-related data corresponding to each partition (expected power generation loss, expected water consumption, spatial location information, etc.), an optimization solution is used to generate a maintenance plan that includes the work sequence, path planning, and task schedule.

[0096] In this embodiment, by dividing each photovoltaic panel management zone into three levels—non-delayable zone, delayable zone, and alternative zone—based on maintenance urgency, a refined classification of maintenance tasks is achieved. This ensures that high-urgency tasks are prioritized, medium-urgency tasks have flexible processing space, and low-urgency tasks retain the possibility of utilization. Based on this, for the alternative zones, flexible supplementary zones are further identified by combining minimum spatial distance and marginal cleaning cost. Zones that are spatially adjacent and have sufficiently low marginal cleaning costs are marked as supplementary tasks that can be executed along the route. This provides a structured task classification basis for improving equipment utilization at extremely low cost when resources permit, thereby ensuring both timely maintenance and operational economy.

[0097] For example, please refer to Figure 2 , Figure 2 A detailed flowchart illustrating the maintenance scheme generation steps in a smart maintenance method for photovoltaic panels is provided. First, after obtaining the contamination status index Di and risk index Ri for each photovoltaic panel management zone (step 101), the maintenance urgency index for each photovoltaic panel management zone is determined based on Di and Ri (step 102). Then, it is determined whether the maintenance urgency index is greater than a first preset threshold (step 103). If it is greater than the first preset threshold, the zone is classified as a non-delayable zone (step 105). If it is less than the first preset threshold, it is determined whether the maintenance urgency index is greater than a second preset threshold (step 104). If it is greater than the second preset threshold, the zone is classified as a delayable zone (step 106). If it is not greater than the second preset threshold, the zone is classified as a candidate zone (step 108).

[0098] After completing the hierarchical division of each photovoltaic panel management zone, the non-delayable zone and the delayable zone can be merged into a set (step 107). For each candidate zone, the minimum spatial distance between it and the set formed by the non-delayable zone and the delayable zone is calculated (step 109), and the marginal cleaning cost of the candidate zone is determined. It is then determined whether the minimum spatial distance is less than the third preset threshold and whether the marginal cleaning cost is less than the fourth preset threshold (step 110). If the conditions are met, the candidate zone is divided into a flexible supplementary zone (step 111).

[0099] In one feasible implementation, step S40 may further include steps S401 to S404: Step S401: Add each non-delayable partition to the set of tasks to be executed in the current maintenance cycle; It should be noted that the "current maintenance cycle" refers to the time range targeted by this scheduling optimization, such as a work shift (8 hours) or a scheduling window of the next 24 hours. The "set of tasks to be executed" refers to the set of cleaning tasks planned to be executed within the current maintenance cycle.

[0100] Non-delayable partitions are those whose maintenance urgency index, as determined in step S43, exceeds a first preset threshold. These partitions are severely contaminated or pose extremely high future risks. If they are not addressed within the current maintenance cycle, they will result in significant power generation losses or operational risks. Therefore, when constructing the set of tasks to be executed, each non-delayable partition is unconditionally prioritized to ensure that the most urgent maintenance tasks are guaranteed within the current maintenance cycle.

[0101] Step S402: Add each deferable partition to the set of tasks to be executed, and defer the deferable partitions that cannot be executed in the current maintenance cycle due to preset resource constraints to the next maintenance cycle. It should be noted that the deferable partitions are those whose maintenance urgency index, determined in step S44, falls between the second preset threshold and the first preset threshold. Although these partitions have reached the point where they need to be cleaned, they have a certain time tolerance—cleaning them within the current maintenance cycle can effectively mitigate losses, and if they cannot be scheduled due to resource constraints, postponing them to the next maintenance cycle will not cause too serious consequences.

[0102] When constructing the set of tasks to be executed, this step first attempts to include each deferred partition in the set, making it a candidate scheduling object for the current maintenance cycle. However, the available cleaning resources (such as the number of devices, available operating time, remaining water volume, and operating range) within the current maintenance cycle are limited. During the optimization process, the system needs to allocate each task in the set of tasks to be executed to available cleaning devices and arrange them in a time-based manner, while satisfying preset resource constraints.

[0103] The preset resource constraints may include one or more of the following: (1) Equipment quantity constraint: The total number of available cleaning equipment is limited, and each piece of equipment can only perform one cleaning task at a time; (2) Operation time constraint: Each piece of equipment has a maximum working time within the current maintenance cycle, and the total time spent on all tasks assigned to that equipment shall not exceed that time. (3) Range constraint: The cleaning equipment has a limited maximum range after a single recharge. If going to a certain zone and returning will cause the total range to exceed the limit, then the task cannot be assigned to the equipment. (4) Water capacity constraint: For water washing type cleaning equipment, the maximum water capacity after a single filling is limited. If the estimated cleaning water consumption of a certain zone exceeds the remaining water volume, the location of the water replenishment point and the water replenishment time need to be considered. (5) Time window constraint: Some partitions may have specific workable time periods set. If it cannot be arranged within the time limit, it cannot be forcibly allocated.

[0104] When the optimization model detects that, under the above resource constraints, it is impossible to schedule all cleaning tasks corresponding to all deferred partitions within the current maintenance cycle, the model will mark the deferred partitions that cannot be scheduled as "deferred" and postpone them to the next maintenance cycle. That is, these deferred partitions will not be used as scheduling objects in the current maintenance cycle and will be left to participate in the scheduling decision again in the next maintenance cycle.

[0105] Step S403: Based on the remaining resources in the current maintenance cycle, add the elastic supplementation partition to the set of tasks to be executed; It should be noted that the flexible supplementary partition is the candidate partition identified in step S47 that meets the spatial proximity condition and has a sufficiently low marginal cleaning cost. Although such partitions are currently less contaminated and have a lower maintenance urgency, they are economically feasible for "on-the-way cleaning" because they are spatially close to the partitions that have been identified for cleaning and the marginal cost of additional cleaning is low.

[0106] After adding non-delayable partitions to the task set in step S401 and adding delayable partitions to the task set and delaying delayable partitions that cannot be executed in step S402, residual resources may remain in the scheduling scheme of the current maintenance cycle. For example, after a cleaning machine completes all assigned tasks in the task set, it may still have remaining operating time before the end of its shift, and its remaining water volume and range are sufficient; or, a machine may have underutilized mileage resources on its planned route. If these residual resources are not utilized, it will lead to a decrease in equipment utilization and a waste of maintenance resources.

[0107] This step is precisely to opportunistically supplement the remaining resources mentioned above. Specifically, based on the remaining resources of the current maintenance cycle (such as remaining operation time, remaining water volume, remaining driving range, etc.), the system selectively adds some elastic supplementation zones from the identified elastic supplementation zones to the task set to be executed, according to their spatial proximity and marginal cleaning cost.

[0108] When adding elastic supplementary partitions to the set of tasks to be executed, the spatial distance and marginal cleaning cost can be considered. The elastic supplementary partitions can be added to the set of tasks to be executed one by one or in batches. After each addition, the remaining resources should be reassessed until the remaining resources can no longer meet the execution needs of any elastic supplementary partition.

[0109] In a more preferred embodiment, proactive task replacement can be performed before or simultaneously with adding the elastic supplementation partition to the set of tasks to be executed, based on a comparison of resource efficiency. Specifically, this embodiment further includes steps S4031-S4032: Step S4031: For each delayed partition and each elastic supplement partition, based on the expected power generation loss and expected water consumption in the maintenance-related data, determine the expected power generation loss recovery amount corresponding to its unit resource consumption. It should be noted that the expected power generation loss recovery per unit of resource consumption refers to the power generation loss that can be recovered for each unit of resource consumed by the cleaning task corresponding to that partition. For example, the unit resource recovery can be calculated by dividing the expected power generation loss of that partition by its expected water consumption, i.e., the power generation loss that can be recovered per unit of water consumption. Alternatively, a weighted calculation can be performed considering multiple resource dimensions such as water consumption and operation time; this embodiment does not specifically limit this approach.

[0110] Step S4032: When the unit resource recovery amount of a certain elastic supplement partition is higher than the unit resource recovery amount of a certain deferred partition in the current task set to be executed, the deferred partition is removed from the task set to be executed and deferred to the next maintenance cycle, and the elastic supplement partition is added to the task set to be executed.

[0111] It's important to note that the core of this step lies in considering not only the feasibility of "whether it can be included" but also the efficiency optimization of "which one to include." When the unit resource recovery of the elastic supplementary partition is higher than that of a certain deferred partition, it indicates that using the limited cleaning resources to execute the cleaning task corresponding to that elastic supplementary partition can generate higher power generation revenue. In this case, even if there are enough remaining resources in the current maintenance cycle to accommodate some deferred partitions, the system can proactively postpone the deferred partitions with lower resource efficiency and prioritize the execution of the elastic supplementary partitions with higher resource efficiency.

[0112] In this implementation, by introducing a comparison mechanism for unit resource recovery, the scheduling logic of elastic supplementation partitions is upgraded from "passively filling remaining resources" to "actively replacing inefficient tasks." This mechanism enables the system to continuously optimize the composition of the task set to be executed under resource constraints, ensuring that limited cleaning resources are always used for tasks with the highest unit resource return, further improving the economic efficiency of the maintenance scheme.

[0113] Step S404: Based on the set of tasks to be executed and the maintenance-related data of the photovoltaic power station, generate a maintenance plan for the photovoltaic power station that includes the work sequence, path planning and task schedule.

[0114] It should be noted that, after processing steps S401 to S403, the set of tasks to be executed includes the following tasks: cleaning tasks corresponding to all non-delayable partitions, cleaning tasks corresponding to delayable partitions that can be executed within the current maintenance cycle, and cleaning tasks corresponding to elastic supplementary partitions that will be added as needed based on remaining resources. This set of tasks to be executed is the final set of cleaning tasks to be executed in the current maintenance cycle.

[0115] In this step, the system uses the maintenance-related data corresponding to each task in the task set to be executed as input, and generates a specific maintenance plan through optimization. The specific content of the maintenance-related data, the target factors and constraints on which the optimization is based, and the specific composition of the solution output can all be referred to the relevant description in step S40, and will not be repeated here.

[0116] Compared to step S40, this step is characterized by the fact that the tasks used to generate the maintenance plan are not directly selected from each photovoltaic panel management zone, but rather formed into a set of tasks to be executed through a three-level construction mechanism: mandatory selection of non-delayed zones, flexible adjustment of delayed zones, and opportunity supplementation of flexible supplementary zones. This mechanism ensures that the final generated maintenance plan prioritizes the execution of high-urgency tasks while achieving dynamic task balance and efficient resource utilization under resource constraints.

[0117] In this embodiment, a three-tiered task construction mechanism of "mandatory-flexible-supplementary" is formed by unconditionally including non-delayable partitions in the task set to be executed, attempting to include delayable partitions and deferring the parts that cannot be executed due to resource constraints to the next maintenance cycle, and adding elastic supplementary partitions as appropriate based on remaining resources. On this basis, a maintenance plan is generated based on the task set to be executed and maintenance-related data, so that the final maintenance plan not only ensures the priority execution of high-urgency partitions, but also avoids the infeasibility of the plan caused by rigid scheduling through the flexible yielding of delayable partitions. Whether it is necessary to clean up and fill the gap by taking advantage of the opportunity of elastic supplementary partitions, the equipment utilization rate and task saturation are effectively improved, and a dynamic balance between maintenance timeliness and operation and maintenance economy under resource constraints is achieved with low marginal cost.

[0118] Example 2 Based on the first embodiment of this application, in another embodiment of this application, the same or similar content as in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 After the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, the method further includes steps A10 to A20: Step A10: After monitoring that the photovoltaic power station is performing maintenance tasks according to the maintenance plan, obtain the task execution feedback data of each photovoltaic panel management zone; It should be noted that this step involves data collection after the maintenance plan has been fully implemented. The task execution feedback data refers to relevant data generated during and after the execution of this maintenance task, which can be used to evaluate the cleaning effect and optimize subsequent decisions.

[0119] For example, task execution feedback data may include one or more of the following: (1) Actual power generation recovery: After the cleaning task is performed on each photovoltaic panel management zone, the system continuously collects the power generation data of that zone, records the power generation change curve from the polluted state to the clean state, and calculates the actual power generation recovered. This data is the core basis for evaluating the actual effect of this cleaning and calibrating the polluted state assessment model; (2) Actual water consumption: The actual amount of water consumed by the cleaning equipment during the execution of the task can be compared with the estimated water consumption to optimize the subsequent water consumption prediction model; (3) Actual operation time: The actual execution time of each cleaning task, including the actual cleaning operation time and equipment maneuver time, which can be used to calibrate the operation time prediction model; (4) Actual travel path: The actual routes and mileage of the cleaning equipment between different zones can be used to optimize the path planning model; (5) Abnormal event record: Abnormal situations that occur during task execution, such as equipment failure, sudden weather changes, road blockage, etc., can be used to optimize the event trigger threshold of subsequent dynamic replanning strategies.

[0120] Step A20: Update historical maintenance data based on task execution feedback data for each photovoltaic panel management zone.

[0121] It should be noted that in this step, the task execution feedback data obtained in step A10 is written to the historical maintenance database with a unified timestamp, and merged or appended with the historical maintenance data obtained in step S10, thereby expanding the sample size of historical data. The updated historical maintenance data will be obtained in the next round of execution of step S10 and used for parameter calibration when determining the contamination status indicators in step S20.

[0122] Specifically, updating historical maintenance data based on task execution feedback data can achieve one or more of the following optimizations during the generation of subsequent maintenance plans: (1) Calibration of the clean reference model: Using the actual power generation recovery data collected after each cleaning, the actual output power value of the partition under the clean state is extracted, and the component aging attenuation coefficient and temperature coefficient in the clean reference model are updated by regression, so that the estimation of theoretical power under the clean state is more accurate. (2) Optimization of weight coefficients for pollution status indicators: Based on the correlation between the actual power generation recovery and each characteristic component, the weight coefficients in the calculation formula of pollution status indicators are adjusted. (3) Adaptive update of trigger thresholds: The actual recovery effect of each cleaning of the partition is correlated with the current fouling status indicators and risk indicators, and the various trigger thresholds of the partition are dynamically adjusted in different seasons and time periods. (4) Calibration of the parameters of the fouling evolution model: The actual fouling state change trajectory in the period before cleaning is compared with the model prediction trajectory, the prediction error is calculated and the evolution model parameters are updated.

[0123] In this embodiment, by introducing a feedback mechanism after task execution, this application forms a complete closed-loop mechanism of "prediction-execution-feedback-optimization". After each cleaning task is completed, the system feeds back the actual execution effect to the historical maintenance database, enabling the continuous accumulation and updating of historical maintenance data. In the next round of maintenance decisions, the model parameters relied upon by steps S20 and S30 can be calibrated based on richer and more accurate historical data, thereby making the assessment of contamination status indicators, the prediction of future contamination status, and the calculation of risk indicators gradually more accurate. This achieves the continuous evolution of the system's assessment and prediction capabilities, further improving the economy and accuracy of photovoltaic power plant operation and maintenance decisions.

[0124] In one possible implementation, step B10 is included after step S40: Step B10: Upon receiving the replanning instruction, return to the step of obtaining the operation data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station in order to regenerate the maintenance plan for the photovoltaic power station in the current maintenance cycle.

[0125] It should be noted that the replanning instruction refers to the instruction signal that triggers the system to replan tasks that have not yet been executed in the current maintenance plan. This replanning instruction can originate from various triggering scenarios, such as: (1) Manual triggering by maintenance personnel: Maintenance personnel can manually issue a replanning instruction when they discover changes in the field situation through the monitoring terminal (such as sudden weather changes, sudden equipment failures, etc.); (2) External system linkage trigger: When the energy management system or meteorological early warning system of the photovoltaic power station detects preset conditions that affect the cleaning operation (such as sandstorm warning, power curtailment dispatch instructions, etc.), it automatically sends a replanning instruction to this system; (3) Timed triggering: The system automatically generates replanning instructions at preset fixed time intervals to periodically update the maintenance plan.

[0126] Upon receiving the rescheduling instruction, this step returns the program execution flow to step S10, re-acquires the current operating data, environmental data, and historical maintenance data for each photovoltaic panel management zone, and then executes steps S20 to S40 sequentially. Since the current maintenance cycle may have been partially executed at this point, the regenerated maintenance plan will only schedule the remaining tasks that have not yet been executed, and the completed tasks will not be included in the new scheduling scope.

[0127] In this embodiment, by incorporating a replanning mechanism, the system gains the ability to replan on demand in response to external commands. The triggering method based on replanning commands provides maintenance personnel or related systems with greater control flexibility, enabling timely adjustments to the plan when sudden anomalies or significant changes in the operating environment are detected. This avoids response delays caused by waiting for the next automatic trigger cycle, thereby further enhancing the maintenance plan's real-time adaptability to dynamic operating environments.

[0128] In one feasible implementation, after step S40, steps C10 to C30 are further included: Step C10: Determine the maintenance urgency index for each photovoltaic panel management zone based on the pollution status index and risk index of each photovoltaic panel management zone; It should be noted that this step is performed after the initial maintenance plan is generated, and its purpose is to provide a quantitative basis for subsequent judgment on whether a global replanning needs to be triggered. The method for determining the maintenance urgency index is the same as that described in step S41, that is, by integrating the contamination status index and risk index of each photovoltaic panel management zone, a comprehensive index for quantifying the urgency of cleaning in each zone is obtained.

[0129] For example, the maintenance urgency index can be calculated using a weighted summation method, or determined by taking the maximum value or a piecewise mapping method. This embodiment does not limit the specific calculation method of the maintenance urgency index, as long as it can comprehensively reflect the current contamination status and future risk level of each photovoltaic panel management zone.

[0130] By recalculating the maintenance urgency index of each photovoltaic panel management zone in this step, the system can obtain the distribution of maintenance needs of each zone in the entire station at the current moment, providing basic data for step C20 to calculate the total maintenance urgency index of the entire station and then determine whether a global replanning needs to be triggered.

[0131] Step C20: Calculate the sum of the maintenance urgency indices for each photovoltaic panel management zone; It should be noted that this step is performed after the maintenance urgency index of each photovoltaic panel management zone is determined in step C10. By summing the maintenance urgency indices of all photovoltaic panel management zones in the entire station, a scalar value reflecting the cumulative degree of overall maintenance needs of the entire station is obtained.

[0132] The sum of the maintenance urgency indicators has the following technical meaning: When the soiling status of most zones in the entire station is relatively light and the risk index is low, the maintenance urgency indicators of each zone are generally at a low level, and their sum is also relatively small, indicating that the current overall maintenance demand of the entire station is relatively mild, and the initially generated maintenance plan can still adapt well to the current state; however, when multiple zones in the entire station experience a general increase in maintenance urgency indicators due to environmental factors (such as continuous lack of rain leading to widespread soiling accumulation, or sandstorm weather leading to large-scale pollution aggravation) or future risks (such as weather forecasts indicating that unfavorable weather conditions for natural cleaning are about to occur), the maintenance urgency indicators of each zone will generally increase, and their sum will also increase significantly, indicating that the overall maintenance demand of the entire station has changed significantly, and the initial maintenance plan may no longer be applicable to the current global state.

[0133] The sum of maintenance urgency indicators calculated in this step provides a quantitative basis for comparison with the preset global urgency threshold in step C30, and for determining whether a global replanning needs to be triggered.

[0134] Step C30: When the total exceeds the preset global urgency threshold, return to the step of obtaining the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station in order to regenerate the maintenance plan of the photovoltaic power station in the current maintenance cycle.

[0135] It should be noted that the global urgency threshold is a criterion used to determine whether the overall maintenance needs of the entire station have reached the level of "requiring a review and adjustment of the current maintenance plan." When the sum of the maintenance urgency indicators calculated in step C20 exceeds this threshold, it indicates that since the initial maintenance plan was generated, the contamination status or risk status of each photovoltaic panel management zone in the entire station has significantly accumulated or deteriorated. Continuing to implement the original maintenance plan may result in some high-urgency zones not being cleaned in a timely manner, thereby causing increased power generation losses or increased operation and maintenance risks.

[0136] There are several ways to set the global urgency threshold. In one specific implementation, it can be determined based on historical data statistics, for example, by using the upper quantile (e.g., the 85th percentile) of the sum of maintenance urgency indicators in historical operational data as the global urgency threshold. In another specific implementation, the global urgency threshold can be correlated with the current amount of available cleaning resources to achieve dynamic adjustment—when there are many available devices and sufficient resources, the threshold can be appropriately lowered to make the system more sensitive to changes in maintenance needs; when there are few available devices and resources are scarce, the threshold can be appropriately raised to avoid frequent triggering of invalid replanning due to insufficient resources.

[0137] Once the replanning conditions for this step are triggered, the system returns to step S10 to reacquire the current operating data, environmental data, and historical maintenance data for each photovoltaic panel management zone. Steps S20 to S40 are then executed sequentially to regenerate a maintenance plan suitable for the current maintenance cycle based on the latest data. The newly generated maintenance plan will replace the unexecuted portions of the original plan and will guide cleaning operations within the remaining time window.

[0138] In this embodiment, through the global urgency threshold triggering mechanism described in steps C10 to C30, the system can perceive global state changes caused by the slow accumulation of environmental factors from a macroscopic perspective of the overall maintenance needs of the entire station—for example, a situation where continuous days without rain lead to a general increase in fouling throughout the station. Such gradual global changes are often difficult to capture in a timely manner through a single event triggering mechanism, while the global urgency threshold triggering mechanism fills this response gap. It complements the periodic triggering mechanism and the event triggering mechanism, together forming a more complete dynamic replanning strategy system, further enhancing the adaptability of the maintenance scheme to changes in the operating state of the power station.

[0139] This application also provides an intelligent maintenance device for photovoltaic panels; please refer to... Figure 4 The intelligent maintenance device for photovoltaic panels includes: The data acquisition module 10 is used to acquire the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station; The status assessment module 20 is used to determine the pollution status indicators of each photovoltaic panel management zone based on the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone. The pollution status indicators are used to characterize the current pollution level of the photovoltaic panel management zone. The risk assessment module 30 is used to predict the pollution status index of each photovoltaic panel management zone in the future time window based on the pollution status index of each photovoltaic panel management zone and the weather forecast data of the area where the photovoltaic power station is located, and to calculate the risk index of each photovoltaic panel management zone in the future time window based on the pollution status index of each photovoltaic panel management zone in the future time window. The risk index is used to characterize the comprehensive maintenance risk of the photovoltaic panel management zone in the future time window if it is not cleaned. The scheme generation module 40 is used to generate a maintenance scheme for the photovoltaic power station, which includes the operation sequence, path planning and task schedule, based on the pollution status indicators, risk indicators and maintenance-related data of each photovoltaic panel management zone. The maintenance-related data includes at least expected power generation loss, expected water consumption and spatial location information.

[0140] In some embodiments, the solution generation module 40 is further configured to: Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone, the maintenance urgency indicators of each photovoltaic panel management zone are determined. Based on the maintenance urgency index of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, a maintenance plan including the work sequence, path planning and task schedule is generated for the photovoltaic power station.

[0141] In some embodiments, the solution generation module 40 is further configured to: In each of the photovoltaic panel management zones, the photovoltaic panel management zone with a maintenance urgency index greater than a first preset threshold is classified as a non-delay zone. In each of the photovoltaic panel management zones, the photovoltaic panel management zone whose maintenance urgency index is greater than the second preset threshold and not greater than the first preset threshold is classified as a deferred zone. Among the photovoltaic panel management zones, those whose maintenance urgency index is less than the second preset threshold are designated as candidate zones; For any of the candidate partitions, the minimum spatial distance between the candidate partition and the set consisting of the non-delayable partition and the delayable partition is calculated based on the maintenance-related data of the photovoltaic power station, and the marginal cleaning cost of the candidate partition is determined. When the minimum spatial distance is less than the third preset threshold and the marginal cleaning cost is less than the fourth preset threshold, the candidate partition is divided into a flexible supplementary partition. Based on the non-delayable partition, the delayable partition, the flexible supplementary partition, and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station, including the work sequence, path planning, and task schedule, is generated.

[0142] In some embodiments, the solution generation module 40 is further configured to: Add each of the aforementioned non-delayable partitions to the set of tasks to be executed in the current maintenance cycle; Each of the deferable partitions is added to the set of tasks to be executed, and the deferable partitions that cannot be executed in the current maintenance cycle due to preset resource constraints are deferred to the next maintenance cycle. Based on the remaining resources in the current maintenance cycle, the elastic supplementation partition is added to the set of tasks to be executed; Based on the set of tasks to be executed and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station, including the work sequence, path planning and task schedule, is generated.

[0143] In some embodiments, the solution generation module 40 is further configured to: After monitoring that the photovoltaic power station is performing maintenance tasks according to the maintenance plan, the task execution feedback data of each photovoltaic panel management zone is obtained; The historical maintenance data is updated based on the task execution feedback data of each photovoltaic panel management zone.

[0144] In some embodiments, the solution generation module 40 is further configured to: Upon receiving a replanning instruction, the process returns to the step of obtaining the operating data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station, in order to regenerate the maintenance plan for the photovoltaic power station in the current maintenance cycle.

[0145] In some embodiments, the solution generation module 40 is further configured to: Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone, the maintenance urgency indicators of each photovoltaic panel management zone are determined. Calculate the sum of the maintenance urgency indices for each of the photovoltaic panel management zones; When the sum exceeds the preset global urgency threshold, return to the step of obtaining the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station, so as to regenerate the maintenance plan of the photovoltaic power station in the current maintenance cycle.

[0146] The intelligent photovoltaic panel maintenance device provided in this application, employing the intelligent photovoltaic panel maintenance method described in the above embodiments, can solve the technical problem of how to proactively predict and optimize maintenance decision-making schemes. Compared with the prior art, the beneficial effects of the intelligent photovoltaic panel maintenance device provided in this application are the same as those of the intelligent photovoltaic panel maintenance method provided in the above embodiments, and other technical features in the intelligent photovoltaic panel maintenance device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0147] This application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the photovoltaic panel intelligent maintenance method in Embodiment 1 above.

[0148] The following is for reference. Figure 5The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of this application. The electronic devices in these embodiments may include, but are not limited to, mobile terminals such as mobile phones, laptops, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0149] like Figure 5 As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. While electronic devices with various systems are shown in the figures, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0150] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0151] The electronic device provided in this application, employing the intelligent photovoltaic panel maintenance method described in the above embodiments, can solve the technical problem of how to proactively predict and optimize maintenance decision-making schemes. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the intelligent photovoltaic panel maintenance method provided in the above embodiments, and other technical features of this electronic device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0152] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0153] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0154] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the intelligent maintenance method for photovoltaic panels in the above embodiments.

[0155] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0156] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0157] The aforementioned computer-readable storage medium carries one or more programs that, when executed by an electronic device, cause the electronic device to: acquire operational data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station; Based on the operational data, environmental data, and historical maintenance data of each photovoltaic panel management zone, the pollution status index of each photovoltaic panel management zone is determined, wherein the pollution status index is used to characterize the current pollution level of the photovoltaic panel management zone; Based on the fouling status indicators of each photovoltaic panel management zone and the weather forecast data of the area where the photovoltaic power station is located, the fouling status indicators of each photovoltaic panel management zone in the future time window are predicted, and based on the fouling status indicators of each photovoltaic panel management zone in the future time window, the risk indicators of each photovoltaic panel management zone in the future time window are calculated, wherein the risk indicators are used to characterize the comprehensive maintenance risks caused by the photovoltaic panel management zone in the future time window without cleaning; Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station is generated, which includes the operation sequence, path planning and task schedule. The maintenance-related data includes at least expected power generation loss, expected water consumption and spatial location information.

[0158] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0159] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0160] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0161] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., computer programs) for executing the above-described intelligent maintenance method for photovoltaic panels, and can solve the technical problem of how to proactively predict and optimize maintenance decision-making schemes. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the intelligent maintenance method for photovoltaic panels provided in the above embodiments, and will not be repeated here.

[0162] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described intelligent maintenance method for photovoltaic panels.

[0163] The computer program product provided in this application can solve the technical problem of how to proactively predict and optimize maintenance decision-making schemes. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the intelligent maintenance method for photovoltaic panels provided in the above embodiments, and will not be repeated here.

[0164] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for intelligent maintenance of photovoltaic panels, characterized in that, The method includes: Obtain operational data, environmental data, and historical maintenance data for each photovoltaic panel management zone of the photovoltaic power plant; Based on the operational data, environmental data, and historical maintenance data of each photovoltaic panel management zone, the pollution status index of each photovoltaic panel management zone is determined, wherein the pollution status index is used to characterize the current pollution level of the photovoltaic panel management zone; Based on the fouling status indicators of each photovoltaic panel management zone and the weather forecast data of the area where the photovoltaic power station is located, the fouling status indicators of each photovoltaic panel management zone in the future time window are predicted, and based on the fouling status indicators of each photovoltaic panel management zone in the future time window, the risk indicators of each photovoltaic panel management zone in the future time window are calculated, wherein the risk indicators are used to characterize the comprehensive maintenance risks caused by the photovoltaic panel management zone in the future time window without cleaning; Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station is generated, which includes the operation sequence, path planning and task schedule. The maintenance-related data includes at least expected power generation loss, expected water consumption and spatial location information.

2. The method according to claim 1, characterized in that, The step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, includes: Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone, the maintenance urgency indicators of each photovoltaic panel management zone are determined. Based on the maintenance urgency index of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, a maintenance plan including the work sequence, path planning and task schedule is generated for the photovoltaic power station.

3. The method according to claim 2, characterized in that, The step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the maintenance urgency index of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station includes: In each of the photovoltaic panel management zones, the photovoltaic panel management zone with a maintenance urgency index greater than a first preset threshold is classified as a non-delay zone. In each of the photovoltaic panel management zones, the photovoltaic panel management zone whose maintenance urgency index is greater than the second preset threshold and not greater than the first preset threshold is classified as a deferred zone. Among the photovoltaic panel management zones, those whose maintenance urgency index is less than the second preset threshold are designated as candidate zones; For any of the candidate partitions, the minimum spatial distance between the candidate partition and the set consisting of the non-delayable partition and the delayable partition is calculated based on the maintenance-related data of the photovoltaic power station, and the marginal cleaning cost of the candidate partition is determined. When the minimum spatial distance is less than the third preset threshold and the marginal cleaning cost is less than the fourth preset threshold, the candidate partition is divided into a flexible supplementary partition. Based on the non-delayable partition, the delayable partition, the flexible supplementary partition, and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station, including the work sequence, path planning, and task schedule, is generated.

4. The method according to claim 3, characterized in that, The step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the non-delayable partition, the delayable partition, the flexible supplementary partition, and the maintenance-related data of the photovoltaic power station includes: Add each of the aforementioned non-delayable partitions to the set of tasks to be executed in the current maintenance cycle; Each of the deferable partitions is added to the set of tasks to be executed, and the deferable partitions that cannot be executed in the current maintenance cycle due to preset resource constraints are deferred to the next maintenance cycle. Based on the remaining resources in the current maintenance cycle, the elastic supplementation partition is added to the set of tasks to be executed; Based on the set of tasks to be executed and the maintenance-related data of the photovoltaic power station, a maintenance plan for the photovoltaic power station, including the work sequence, path planning and task schedule, is generated.

5. The method according to claim 1, characterized in that, After the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, the method further includes: After monitoring that the photovoltaic power station is performing maintenance tasks according to the maintenance plan, the task execution feedback data of each photovoltaic panel management zone is obtained; The historical maintenance data is updated based on the task execution feedback data of each photovoltaic panel management zone.

6. The method according to claim 1, characterized in that, After the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, the method further includes: Upon receiving a replanning instruction, the process returns to the step of obtaining the operating data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station, in order to regenerate the maintenance plan for the photovoltaic power station in the current maintenance cycle.

7. The method according to any one of claims 1 to 6, characterized in that, After the step of generating a maintenance plan for the photovoltaic power station, including work sequence, path planning, and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station, the method further includes: Based on the pollution status indicators and risk indicators of each photovoltaic panel management zone, the maintenance urgency indicators of each photovoltaic panel management zone are determined. Calculate the sum of the maintenance urgency indices for each of the photovoltaic panel management zones; When the sum exceeds the preset global urgency threshold, return to the step of obtaining the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station, so as to regenerate the maintenance plan of the photovoltaic power station in the current maintenance cycle.

8. A photovoltaic panel intelligent maintenance device, characterized in that, The intelligent maintenance device for photovoltaic panels includes: The data acquisition module is used to acquire operational data, environmental data, and historical maintenance data of each photovoltaic panel management zone of the photovoltaic power station; The status assessment module is used to determine the pollution status index of each photovoltaic panel management zone based on the operation data, environmental data and historical maintenance data of each photovoltaic panel management zone, wherein the pollution status index is used to characterize the current pollution level of the photovoltaic panel management zone; The risk assessment module is used to predict the contamination status index of each photovoltaic panel management zone in a future time window based on the contamination status index of each photovoltaic panel management zone and the weather forecast data of the area where the photovoltaic power station is located, and to calculate the risk index of each photovoltaic panel management zone in the future time window based on the contamination status index of each photovoltaic panel management zone in the future time window. The risk index is used to characterize the comprehensive maintenance risk caused by the photovoltaic panel management zone in the future time window without cleaning. The scheme generation module is used to generate a maintenance scheme for the photovoltaic power station, which includes the operation sequence, path planning and task schedule, based on the pollution status indicators and risk indicators of each photovoltaic panel management zone and the maintenance-related data of the photovoltaic power station. The maintenance-related data includes at least expected power generation loss, expected water consumption and spatial location information.

9. An electronic device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the intelligent maintenance method for photovoltaic panels as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the photovoltaic panel intelligent maintenance method as described in any one of claims 1 to 7.

Images