Intelligent photovoltaic self-cleaning method and device

Abstract

The application discloses an intelligent photovoltaic self-cleaning method and device, and relates to the technical field of photovoltaic modules, which comprises the following steps: automatically calculating the weekly power generation through a photovoltaic control platform, obtaining images monitored by a camera, constructing a prediction model, judging the dirt degree of each photovoltaic module, and calculating an optimal cleaning path; starting a first cleaning operation according to the optimal cleaning path, evaluating the cleaning effect, and judging whether to perform a second cleaning operation; resetting the cleaning operation according to the result of the second cleaning operation, and completing the intelligent photovoltaic self-cleaning operation. Through automatic data collection, prediction modeling and optimized cleaning path, the application realizes an efficient and accurate cleaning process, intelligently judges the dirt degree, executes automatic cleaning, dynamically evaluates the effect, decides whether secondary cleaning is needed, improves the cleaning efficiency and quality, optimizes resource utilization, and significantly improves the overall performance and reliability of the photovoltaic system.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic module technology, and in particular to an intelligent photovoltaic self-cleaning method and apparatus. Background Technology

[0002] Currently, there are three main methods for cleaning photovoltaic modules: manual cleaning, semi-automatic cleaning, and fully automatic cleaning. Manual cleaning relies on human labor for wiping and spraying water, which presents problems such as high labor costs, low work efficiency, and safety risks. Semi-automatic cleaning requires human operation of the water spraying equipment, has stringent requirements regarding the spacing and size of the photovoltaic modules, and can easily lead to operational inconvenience. While fully automatic cleaning delegates the entire cleaning process to equipment, its complexity and operational difficulties currently limit its widespread application.

[0003] In existing technologies, most photovoltaic cleaning equipment focuses on structural optimization, such as structural modifications to the cleaning brush, improvements to the nozzle and water spraying methods, and automatic control of the cleaning brush to determine whether the cleaning requirements have been met based on the degree of dirt on the photovoltaic modules. However, existing patents do not yet include corresponding programs to assist in the operation.

[0004] Long-term neglect of cleaning photovoltaic (PV) module surfaces can affect light transmittance, thereby reducing power generation efficiency. Currently, the operation and maintenance of distributed PV power stations mainly relies on manual cleaning, which is time-consuming and incurs high labor and transportation costs. To address these issues, this invention proposes an intelligent PV self-cleaning method and device. Through remote program calculations and camera comparison, it accurately locates the PV modules requiring cleaning and performs intelligent zoned cleaning, significantly saving cleaning costs, water resources, and electricity resources. Summary of the Invention

[0005] In view of the problems existing in the cleaning methods of photovoltaic modules, this invention is proposed.

[0006] Therefore, the problem to be solved by this invention is how to achieve efficient self-cleaning of photovoltaic modules through intelligent means, thereby reducing maintenance costs, improving power generation efficiency, and saving water and electricity resources.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] In a first aspect, embodiments of the present invention provide an intelligent photovoltaic self-cleaning method, comprising: automatically calculating weekly power generation through a photovoltaic control platform and acquiring images monitored by a camera; constructing a prediction model based on the weekly power generation and the images; determining the degree of dirtiness of each photovoltaic module according to the prediction results of the prediction model and calculating the optimal cleaning path; initiating a first cleaning operation according to the optimal cleaning path and evaluating the cleaning effect; determining whether to perform a second cleaning operation based on the evaluation results of the cleaning effect; and resetting the cleaning operation according to the result of the second cleaning operation to complete the intelligent photovoltaic self-cleaning operation.

[0009] As a preferred embodiment of the intelligent photovoltaic self-cleaning method of the present invention, the prediction model construction process includes the following steps: collecting power generation data, image data, and meteorological information of photovoltaic modules; extracting feature information from the image data using computer vision algorithms to quantify the degree of contamination; and establishing a prediction model based on the power generation data, extracted feature information, and meteorological information using a support vector machine algorithm to output the expected power generation efficiency η of each photovoltaic module. pred The expected power generation efficiency η pred The specific formula is as follows:

[0010]

[0011] Where, η pred For the expected power generation efficiency, E w E is the weekly power generation. w E is the weekly power generation. d Where W is the daily power generation, T is the meteorological factor, and S is the pollution score. The specific formula is as follows:

[0012] S=∑(w i *A i ) / A total ;

[0013] Among them, w i For the dirty weight of component i, A i Let A be the dirty area of ​​component i. total This represents the total area.

[0014] As a preferred embodiment of the intelligent photovoltaic self-cleaning method of the present invention, it further includes: for component i, if the dirt score S is greater than the dirt threshold S threshold If a component is identified as needing cleaning, a list of components requiring cleaning is created, recording the component ID and its corresponding dirt score. If the dirt score S is less than or equal to the dirt threshold S... thresholdThe component ID and current dirt score are recorded in the cleaning component list, and the dirt status of the component continues to be monitored. Components to be cleaned are sorted according to their priority scores, and the sorted component list is divided into several cleaning groups. A genetic algorithm is used to define the start and end points for each cleaning group, converting the component positions into nodes in a graph, calculating the distance matrix between nodes, and generating the optimal cleaning path. The optimal cleaning path is stored in the system database and transmitted to the cleaning operation. The specific formula for the priority score is as follows:

[0015] P=ω1S i +ω2L i +ω3T i ;

[0016] Where P is the priority score, ω1, ω2, and ω3 are all weight coefficients, and S i L is the dirtiness score for component i. i T is the position factor of component i. i Let i be the time since the last cleaning.

[0017] The expected power generation efficiency includes when the expected power generation efficiency η pred Less than the first cleaning threshold η threshold When this occurs, a cleaning operation is required; when the expected power generation efficiency η pred Greater than or equal to the first cleaning threshold η threshold If the condition is met, the ID of the component and its current expected power generation efficiency will be recorded in the list of normally operating components, and the power generation efficiency of the component will continue to be monitored.

[0018] As a preferred embodiment of the intelligent photovoltaic self-cleaning method of the present invention, the cleaning operation includes a first cleaning operation and a second cleaning operation; the first cleaning operation includes the following steps: starting the cleaning operation according to the optimal cleaning path; checking whether the liquid level of the storage tank level gauge and the power supply of each electrical device are normal, and whether all local switches are closed; if the liquid level of the storage tank level gauge and the power supply of each electrical device are normal, and all local switches are closed, then the cleaning operation begins; opening the cleaning function in the operation panel and checking whether there is an alarm on the panel; if there is an alarm on the panel, then immediately stopping all cleaning operation preparation processes and sending an alarm to the operator. Notification: If there is no alarm on the panel, open the pump inlet and outlet solenoid valves, and simultaneously open and check that the booster pump outlet pressure is normal; check that the cleaning permission light on the control station is on, start the cleaning program, and sequentially start all photovoltaic module inlet solenoid valves; simultaneously start this group of motors, driving the internal transmission device of the cleaning brush to rotate at the set speed, and the cleaning brush begins to brush the photovoltaic modules at a uniform speed; repeat this sequence for the entire photovoltaic module cleaning process in this area; take an image of the photovoltaic modules after the first cleaning operation is completed, transmit it to the prediction model to evaluate the cleaning effect, and check the dirt and water accumulation on the photovoltaic modules through the camera.

[0019] As a preferred embodiment of the intelligent photovoltaic self-cleaning method of the present invention, the evaluation of the cleaning effect includes: performing a second cleaning operation when the dirt on the photovoltaic module still exists or the water accumulated on the photovoltaic module has not been drained; performing a second cleaning operation when the photovoltaic power generation after cleaning is less than the preset photovoltaic power generation; if a second cleaning operation is performed, the operator is asked whether to confirm the execution of the second cleaning operation; if the operator rejects the second cleaning operation, the decision is recorded and the cleaning process ends; if the operator confirms the execution of the second cleaning operation, the second cleaning operation is performed, the photovoltaic module image is taken again and transmitted to the prediction model for evaluation; when the photovoltaic power generation after cleaning is greater than or equal to the preset photovoltaic power generation, the cleaning effect is determined to be good and no second cleaning operation is required.

[0020] As a preferred embodiment of the intelligent photovoltaic self-cleaning method of the present invention, the second cleaning operation includes the following steps: restarting the cleaning program and opening the cleaning operation again on the operation panel, checking whether there are any alarms on the panel and whether the power supply status of the tank level device is normal; turning on the booster pump and related solenoid valves, clicking the cleaning operation, starting the cleaning program, and cleaning the photovoltaic modules that need to be cleaned in detail; monitoring the cleaning effect in real time through the camera, checking whether the photovoltaic modules have been thoroughly cleaned, confirming that the accumulated water has been completely drained, and resetting the cleaning operation.

[0021] As a preferred embodiment of the intelligent photovoltaic self-cleaning method of the present invention, the reset includes the following steps: after completing all cleaning operations, sequentially close the solenoid valves of all photovoltaic modules and stop the booster pump; simultaneously control the cleaning brush motor to start, so that the cleaning brush returns to the stop position; confirm the correct position of the cleaning brush through the Hall limit switch, and stop the cleaning brush motor; after all mechanical parts have been reset, check whether all indicator lights on the control panel have returned to the standby state; if the control panel indicates normally, perform a safety check, send a notification to the operator that the cleaning operation is complete, and continue to monitor the power generation efficiency of the photovoltaic modules.

[0022] Secondly, embodiments of the present invention provide an intelligent photovoltaic self-cleaning device, comprising: an acquisition module for automatically calculating weekly power generation through a photovoltaic control platform and acquiring images monitored by a camera; a construction module for constructing a prediction model based on the weekly power generation and the images; a calculation module for determining the degree of dirtiness of each photovoltaic module based on the prediction results of the prediction model and calculating the optimal cleaning path; a cleaning module for initiating a first cleaning operation based on the optimal cleaning path and evaluating the cleaning effect; an evaluation module for determining whether to perform a second cleaning operation based on the evaluation results of the cleaning effect; and a control module for resetting the cleaning operation based on the result of the second cleaning operation to complete the intelligent photovoltaic self-cleaning operation.

[0023] Thirdly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program instructions, when executed by the processor, implement the steps of the intelligent photovoltaic self-cleaning method as described in the first aspect of the present invention.

[0024] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of the intelligent photovoltaic self-cleaning method as described in the first aspect of the present invention.

[0025] The beneficial effects of this invention are as follows: This invention achieves a highly efficient and precise cleaning process through automatic data acquisition, predictive modeling, and optimized cleaning path. It intelligently judges the degree of dirt, performs automated cleaning, and dynamically evaluates the effect to determine whether secondary cleaning is needed, thereby improving cleaning efficiency and quality, optimizing resource utilization, and significantly enhancing the overall performance and reliability of the photovoltaic system. Attached Figure Description

[0026] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0027] Figure 1 This is a flowchart of the intelligent photovoltaic self-cleaning method in Example 1.

[0028] Figure 2 This is a schematic diagram of the photovoltaic panel cleaning process in the intelligent photovoltaic self-cleaning method of Example 1.

[0029] Figure 3 This is an electronic device control diagram for the intelligent photovoltaic self-cleaning method in Example 2.

[0030] Figure 4 This is a diagram of the computer equipment used in the intelligent photovoltaic self-cleaning method of Example 3. Detailed Implementation

[0031] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0032] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0033] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0034] Example 1

[0035] Reference Figures 1-2 This is the first embodiment of the present invention, which provides an intelligent photovoltaic self-cleaning method, including,

[0036] S1: Automatically calculates weekly power generation through the photovoltaic control platform and acquires images monitored by cameras.

[0037] S2: Construct a prediction model based on the weekly power generation and the image.

[0038] Specifically, the process of building the prediction model includes the following steps: collecting power generation data, image data, and meteorological information from photovoltaic modules; power generation data is recorded hourly and weekly power generation is calculated; meteorological data includes temperature, humidity, wind speed, and solar radiation intensity; all data are standardized to remove outliers and feature scaling is performed to make data of different dimensions comparable.

[0039] Furthermore, computer vision algorithms are used to extract feature information from image data to quantify the degree of contamination. Feature information includes edge detection, color analysis, and texture analysis; edge detection identifies the contours of dust and stains, color analysis distinguishes different types of pollutants, and texture analysis assesses the degree of contamination. Based on power generation data, extracted feature information, and meteorological information, a prediction model is built using a support vector machine algorithm to output the expected power generation efficiency η of each photovoltaic module. pred .

[0040] Furthermore, the expected power generation efficiency η predThe specific formula is as follows:

[0041]

[0042] Where, η pred For the expected power generation efficiency, E w E is the weekly power generation. w E is the weekly power generation. d Where W is the daily power generation, T is the meteorological factor, and S is the pollution score. The specific formula is as follows:

[0043] S=∑(w i *A i ) / A total ;

[0044] Among them, w i For the dirty weight of component i, A i Let A be the dirty area of ​​component i. total This represents the total area.

[0045] S3: Based on the prediction results of the prediction model, determine the degree of dirtiness of each photovoltaic module and calculate the optimal cleaning path.

[0046] Specifically, for component i, if the dirtiness score S is greater than the dirtiness threshold S threshold If a component is identified as needing cleaning, a list of components requiring cleaning is created, recording the component ID and its corresponding dirt score. If the dirt score S is less than or equal to the dirt threshold S... threshold The component ID and current dirt score are recorded in the cleaning component list, and the dirt status of the component is monitored. The components to be cleaned are sorted according to their priority scores, and the sorted component list is divided into several cleaning groups. A genetic algorithm is used to define the start and end points of the cleaning groups, and the component positions are converted into nodes in the graph. The distance matrix between the nodes is calculated to generate the optimal cleaning path. The optimal cleaning path is stored in the system database and transmitted to the cleaning operation.

[0047] It should be noted that the dirt threshold S threshold This is a standard used to determine whether an individual photovoltaic module needs cleaning. When the dirt score of a module exceeds this threshold, the system marks the module as needing cleaning and adds it to the list of modules that need cleaning.

[0048] Furthermore, the specific formula for priority scoring is as follows:

[0049] P=ω1S i +ω2L i +ω3T i ;

[0050] Where P is the priority score, ω1, ω2, and ω3 are all weight coefficients, and S i L is the dirtiness score for component i. i T is the position factor of component i. i Let i be the time since the last cleaning.

[0051] Furthermore, the expected power generation efficiency includes, when the expected power generation efficiency η pred Less than the first cleaning threshold η threshold When this occurs, a cleaning operation is required; when the expected power generation efficiency η pred Greater than or equal to the first cleaning threshold η threshold If the condition is met, the ID of the component and its current expected power generation efficiency will be recorded in the list of normally operating components, and the power generation efficiency of the component will continue to be monitored.

[0052] It should be noted that the threshold η for the first cleaning... threshold It typically involves intervening promptly when the overall power generation efficiency begins to decline, based on the normal operating parameters and historical data of the photovoltaic system, to ensure that the photovoltaic system remains in optimal working condition.

[0053] S4: Based on the optimal cleaning path, initiate the first cleaning operation and evaluate the cleaning effect.

[0054] Specifically, the cleaning operation includes a first cleaning operation and a second cleaning operation; the first cleaning operation includes the following steps: start the cleaning operation according to the optimal cleaning path; check whether the liquid level of the storage tank level gauge and the power supply of each electrical device are normal, and whether the local switches are all closed.

[0055] Furthermore, if the tank level gauge and the power supply to all electrical equipment are normal, and all local switches are closed, then the cleaning operation can begin. Open the cleaning control panel and check for any alarms. If an alarm is detected, immediately stop all cleaning preparation procedures and send an alarm notification to the operator. If no alarm is detected, open the pump inlet and outlet solenoid valves and simultaneously open and check that the booster pump outlet pressure is normal.

[0056] Furthermore, the cleaning permit light at the control station is illuminated, the cleaning program is initiated, and all inlet solenoid valves of the photovoltaic modules are activated sequentially. Simultaneously, the motors in this group are started, driving the transmission device inside the cleaning brush to rotate at a set speed, and the cleaning brush begins to brush the photovoltaic modules at a uniform speed. This sequence is repeated for the entire photovoltaic module cleaning process in this area. After the first cleaning operation is completed, images of the photovoltaic modules are captured and transmitted to the predictive model to evaluate the cleaning effect. The dirt and water accumulation on the photovoltaic modules are also checked using a camera.

[0057] Specifically, the evaluation of the cleaning effect includes: if the photovoltaic modules are still dirty or the water in the photovoltaic modules has not been drained, a second cleaning operation is performed; if the photovoltaic power generation after cleaning is less than the preset photovoltaic power generation, a second cleaning operation is performed; if a second cleaning operation is performed, the operator is asked whether they confirm the execution of the second cleaning operation; if the operator rejects the second cleaning operation, the decision is recorded and the cleaning process ends; if the operator confirms the execution of the second cleaning operation, the second cleaning operation is performed, and the photovoltaic module image is taken again and transmitted to the prediction model for evaluation.

[0058] Furthermore, if the photovoltaic power generation after cleaning is greater than or equal to the preset photovoltaic power generation, the cleaning effect is considered good, and a second cleaning operation is not required.

[0059] S5: Based on the evaluation results of the cleaning effect, determine whether to perform a second cleaning operation.

[0060] Furthermore, the second cleaning operation includes the following steps: restart the cleaning program and open the cleaning operation again on the operation panel, check whether there are any alarms on the panel and whether the power supply status of the tank level device is normal; turn on the booster pump and related solenoid valves, click the cleaning operation to start the cleaning program, and clean the photovoltaic modules that need to be cleaned in detail; monitor the cleaning effect in real time through the camera, check whether the photovoltaic modules have been thoroughly cleaned, confirm that the accumulated water has been completely drained, and reset the cleaning operation.

[0061] S6: Based on the result of the second cleaning operation, reset the cleaning operation to complete the intelligent photovoltaic self-cleaning operation.

[0062] Specifically, the reset includes the following steps: After completing all cleaning operations, close the solenoid valves of all photovoltaic modules in sequence and stop the booster pump; at the same time, control the cleaning brush motor to start, so that the cleaning brush returns to the stop position; confirm the correct position of the cleaning brush through the Hall limit switch and stop the cleaning brush motor; after all mechanical parts have been reset, check whether all indicator lights on the control panel have returned to the standby state.

[0063] Furthermore, if the control panel indicates normal operation, a safety check is performed, and a notification is sent to the operator that the cleaning operation is complete, while the power generation efficiency of the photovoltaic modules continues to be monitored.

[0064] In summary, this invention achieves a highly efficient and precise cleaning process through automatic data acquisition, predictive modeling, and optimized cleaning paths. It intelligently judges the degree of dirt, performs automated cleaning, and dynamically evaluates the results to determine whether secondary cleaning is needed. This improves cleaning efficiency and quality, optimizes resource utilization, and significantly enhances the overall performance and reliability of photovoltaic systems.

[0065] Example 2

[0066] Reference Figure 3 According to one embodiment of the present invention, an intelligent photovoltaic self-cleaning device is provided, comprising: an electric booster pump, a solenoid valve, a motor, and a lower-level machine, wherein the lower-level machine mainly consists of a control chip, a communication unit, and a power supply; an acquisition module for automatically calculating weekly power generation through a photovoltaic control platform and acquiring images monitored by a camera; a construction module for constructing a prediction model based on the weekly power generation and the images; a calculation module for determining the degree of dirtiness of each photovoltaic module based on the prediction results of the prediction model and calculating the optimal cleaning path; a cleaning module for initiating a first cleaning operation according to the optimal cleaning path and evaluating the cleaning effect; an evaluation module for determining whether to perform a second cleaning operation based on the evaluation results of the cleaning effect; and a control module for resetting the cleaning operation according to the result of the second cleaning operation, thereby completing the intelligent photovoltaic self-cleaning operation.

[0067] Example 3

[0068] Reference Figure 4 This is one embodiment of the present invention, which differs from the previous embodiment in that:

[0069] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0070] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0071] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0072] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0073] Example 4

[0074] Referring to Table 1, the second embodiment of the present invention provides an intelligent photovoltaic self-cleaning method. To verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculations and simulation experiments.

[0075] Specifically, as shown in Table 1, by comparing test data, the intelligent photovoltaic self-cleaning method of the present invention shows significant advantages in several key indicators, where Group A: manual cleaning, Group B: fixed-cycle automatic cleaning, Group C: simple intelligent cleaning, and Group D: the method of the present invention. In terms of average daily power generation, Group D reaches 1750 kWh, which is 12.2% higher than the traditional manual cleaning method, 8.0% higher than the fixed-cycle automatic cleaning method, and 4.2% higher than the simple intelligent cleaning method. This significant increase in power generation directly reflects the superior performance of the method of the present invention in maintaining the cleanliness of photovoltaic panels and improving power generation efficiency.

[0076] Table 1 Comparison between the present invention and prior art

[0077] parameter Group A Group B Group C Group D Average daily power generation (kWh) 1560 1620 1680 1750 Cleaning frequency (times / month) 1 4 3 2.5 Water consumption per cleaning cycle (L / ㎡) 3.5 2.8 2.5 2.0 Cleaning labor cost (RMB / month) 2000 500 400 200 Equipment failure rate (%) 0 2.5 2.0 0.5 Average cleanliness score (0-100) 85 88 92 97

[0078] Furthermore, regarding cleaning frequency, Group D (2.5 times / month) was lower than Group B (4 times / month) and Group C (3 times / month), but higher than Group A (1 time / month); this indicates that the method of the present invention can more accurately determine cleaning needs, avoiding over-cleaning and under-cleaning. Simultaneously, Group D had the lowest water consumption per cleaning cycle, only 2.0L / ㎡, saving 42.9% of water compared to traditional manual methods, demonstrating the significant water-saving advantage of this method. In terms of cleaning labor costs, Group D only needed 200 yuan per month, significantly lower than other groups, especially saving 90% of labor costs compared to manual cleaning. This not only reduces operation and maintenance expenses but also simplifies personnel management.

[0079] Furthermore, equipment failure rate is a crucial indicator of system reliability. Group D's failure rate was only 0.5%, significantly lower than other automated solutions (Group B 2.5%, Group C 2.0%), approaching the reliability of manual cleaning. This demonstrates that the method of this invention not only achieves a high degree of automation but also maintains extremely high system stability. The average cleanliness score directly reflects the cleaning effect. Group D scored as high as 97 points, far exceeding other solutions. This excellent result is attributed to the invention's precise dirt detection, flexible cleaning strategies, and comprehensive, thorough cleaning design.

[0080] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent photovoltaic self-cleaning method, characterized in that: include, The photovoltaic control platform automatically calculates weekly power generation and acquires images from camera monitoring. Based on the weekly power generation and the image, a prediction model is constructed; Based on the prediction results of the prediction model, the degree of dirtiness of each photovoltaic module is determined, and the optimal cleaning path is calculated. Based on the optimal cleaning path, initiate the first cleaning operation and evaluate the cleaning effect; Based on the evaluation results of the cleaning effect, it is determined whether a second cleaning operation should be performed; Based on the result of the second cleaning operation, the cleaning operation is reset to complete the intelligent photovoltaic self-cleaning operation. The process of constructing the prediction model includes the following steps: Collect power generation data, image data, and meteorological information from photovoltaic modules; Computer vision algorithms are used to extract feature information from image data and quantify the degree of dirtiness. Based on power generation data, feature information, and meteorological information, a prediction model is established using the support vector machine algorithm to output the expected power generation efficiency of each photovoltaic module. ; The expected power generation efficiency The specific formula is as follows: ； in, For the expected power generation efficiency, Weekly power generation This represents the daily power generation on day i of a week. Meteorological factors, For temperature, The specific formula for scoring the level of dirtiness is as follows: ； in, The dirty weight of component i, Let i be the dirty area. The total area; For component i, if the dirtiness score Greater than the dirt threshold If a component is identified as needing cleaning, a list of components that need cleaning is created, recording the component ID and its corresponding dirt score. If the dirtiness score Less than or equal to the dirt threshold If so, the ID and current dirt score of this component are recorded in the cleaning component list, and the dirt status of this component is monitored. The components to be cleaned are sorted according to their priority scores, and the sorted list of components is divided into several cleaning groups. Using a genetic algorithm, a starting point and an ending point are defined for several cleaning groups, and the component positions are converted into nodes in a graph. The distance matrix between nodes is calculated to generate the optimal cleaning path. The optimal cleaning path is stored in the system database and then transmitted to the cleaning operation. The specific formula for the priority score is as follows: ； in, Score based on priority. , and All are weighting coefficients. The dirtiness score for component i. Let i be the position factor of component i. The time since component i was last cleaned; The expected power generation efficiency includes When the expected power generation efficiency Less than the first cleaning threshold At that time, a cleaning operation is required; When the expected power generation efficiency Greater than or equal to the first cleaning threshold If the component is in a normal operating state, its ID and current expected power generation efficiency are recorded in the list of normally operating components, and the power generation efficiency of the component is monitored. The cleaning operation includes a first cleaning operation and a second cleaning operation; the first cleaning operation includes the following steps: Start the cleaning operation according to the optimal cleaning path; Check the liquid level of the storage tank level gauge and the power supply of each electrical device to ensure they are normal, and that all local switches are turned off. If the tank level gauge reading and the power supply to all electrical equipment are normal, and all local switches are closed, then the cleaning operation can begin. Open the cleaning function in the operation panel and check for any alarms on the panel. If an alarm is detected on the control panel, immediately stop all cleaning preparation procedures and send an alarm notification to the operator. If there is no alarm on the panel, open the pump inlet and outlet solenoid valves, and at the same time open and check that the booster pump outlet pressure is normal. Check that the cleaning permission light on the control station is on, start the cleaning procedure, and sequentially activate all photovoltaic module inlet solenoid valves. At the same time, start this set of motors to drive the transmission device inside the cleaning brush to rotate at a set speed, and the cleaning brush begins to clean the photovoltaic module at a uniform speed. Perform the sequential cleaning process for the entire photovoltaic module group in this area in this order; Images of the photovoltaic modules after the first cleaning operation are captured and transmitted to a predictive model to evaluate the cleaning effect. The camera also checks the dirt and water accumulation on the photovoltaic modules.

2. The intelligent photovoltaic self-cleaning method as described in claim 1, characterized in that: The evaluation of cleaning effectiveness includes, If the photovoltaic modules are still dirty or the water in the photovoltaic modules has not been drained, a second cleaning operation will be performed. If the photovoltaic power generation after cleaning is less than the preset photovoltaic power generation, a second cleaning operation will be performed. If a second cleaning operation is to be performed, ask the operator if they confirm that the second cleaning operation is to be performed. If the operator rejects the second cleaning operation, the decision is recorded and the cleaning process ends. If the operator confirms the execution of the second cleaning operation, then the second cleaning operation will be performed, and images of the photovoltaic modules will be taken again and transmitted to the predictive model for evaluation. If the photovoltaic power generation after cleaning is greater than or equal to the preset photovoltaic power generation, the cleaning effect is considered good and no second cleaning operation is required.

3. The intelligent photovoltaic self-cleaning method as described in claim 2, characterized in that: The second cleaning operation includes the following steps: Restart the cleaning program and open the cleaning operation again on the operation panel. Check if there are any alarms on the panel and whether the power supply status of the tank level device is normal. Turn on the booster pump and related solenoid valves, click the cleaning operation to start the cleaning program, and clean the photovoltaic modules that need to be cleaned in detail. The cleaning effect is monitored in real time by a camera to check whether the photovoltaic modules have been thoroughly cleaned and to confirm that the accumulated water has been completely drained before resetting the cleaning operation.

4. The intelligent photovoltaic self-cleaning method as described in claim 3, characterized in that: The reset includes the following steps: After completing all cleaning operations, close the solenoid valves of all photovoltaic modules in sequence and stop the booster pump. At the same time, control the start of the cleaning brush motor so that the cleaning brush returns to the stop position; After confirming that the cleaning brush is in the correct position using the Hall limit switch, stop the cleaning brush motor. After all mechanical parts have been reset, check if all indicator lights on the control panel have returned to standby mode. If the control panel indicates normal operation, a safety check will be performed, and a notification will be sent to the operator indicating that the cleaning operation is complete. The power generation efficiency of the photovoltaic modules will continue to be monitored.

5. An apparatus employing an intelligent photovoltaic self-cleaning method as described in any one of claims 1 to 4, characterized in that, include: The acquisition module is used to automatically calculate weekly power generation through the photovoltaic control platform and acquire images monitored by the camera; The module constructs a prediction model based on the weekly power generation and the image; The calculation module is used to determine the degree of dirtiness of each photovoltaic module based on the prediction results of the prediction model, and to calculate the optimal cleaning path. The cleaning module is used to initiate the first cleaning operation based on the optimal cleaning path and evaluate the cleaning effect. The evaluation module, based on the evaluation results of the cleaning effect, determines whether to perform a second cleaning operation; The control module is used to reset the cleaning operation based on the result of the second cleaning operation, thereby completing the intelligent photovoltaic self-cleaning operation.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the intelligent photovoltaic self-cleaning method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the intelligent photovoltaic self-cleaning method according to any one of claims 1 to 4.

Images