Intelligent evaluation system for photovoltaic power station

Abstract

The invention provides an intelligent evaluation system for a photovoltaic power station. The system comprises a basic information management module, a detection evaluation task management module, a power station data evaluation module, a power station quality evaluation module, an intelligent identification module and a power station performance detection module. The intelligent identification module is used for analyzing visible light and infrared thermal imaging images and outputting defects of the photovoltaic string; and the power station performance detection module is used for discovering the defects of the photovoltaic string according to the detection item associated with the inspection task and the model algorithm associated with the detection item for the acquired image by using the intelligent identification module. The method has the advantages that the evaluation cost is reduced, the evaluation period is shortened, errors caused by manual evaluation are avoided, the evaluation quality and accuracy are improved, and a guidance basis is provided for quality improvement and efficiency improvement of the photovoltaic power station.

Description

Technical Field

[0001] The present application belongs to the field of image detection, and specifically relates to an intelligent evaluation system for photovoltaic power stations. Background Art

[0002] The domestic photovoltaic power station smart management system started late, mainly developed from the monitoring control and data acquisition system (SCADA), and gradually moved from the laboratory to commercial applications. With the substantial increase in the amount of photovoltaic power station data and the continuous improvement of analysis needs in the future, data analysis technology will also be expanded and enhanced in multiple dimensions. Due to the substantial increase in data volume, the existing processing and analysis technology can no longer meet the needs of timely data analysis.

[0003] The existing photovoltaic power station management system has the following defects:

[0004] Difficulty in fault location: There are a huge number of components and nodes in photovoltaic power stations, and there is a lack of effective fault location methods. Fault detection mainly relies on manual site visits and manual measurement and comparison with multimeters. The fault handling cycle is long, which affects power generation output, has low maintenance efficiency, and requires a lot of manpower.

[0005] Power station management lacks digital means: monitoring information is simply collected but not presented, a large number of data reports are manually processed through Excel, the ability to conduct comprehensive data analysis is poor, power station operation analysis and improvement lack quantitative means, and multi-power station management cannot be achieved.

[0006] Low prediction accuracy: Problems with the optical power prediction system include low prediction accuracy and precision, which are severely subject to the implementation rules of the power dispatching department; affecting the release of power plant capacity; meteorological equipment is not calibrated for a long time, affecting ultra-short-term predictions and power prediction corrections; third-party meteorological data is not accurate; the optical power prediction system model is established without combining the actual situation of the power plant and the design plan for correction, and the power prediction algorithm has loopholes.

[0007] Equipment failure and energy loss: Various defects may occur in the operation of photovoltaic power plants, such as equipment failure, energy loss, etc. If these defects are not discovered in time, they will not only reduce the power generation efficiency of the power plant, but also have a negative impact on the stability and life of the power plant equipment.

[0008] Monitoring data is unreliable and untrustworthy: There is no string monitoring or only simple string data collection, the monitoring measurement accuracy is not high, and the measurement data is inaccurate; it is difficult to upload communication data. Monitoring data is uploaded through the RS-485 bus, the transmission rate is low, communication failures are frequent, and false alarms and missed reports are serious. Summary of the invention

[0009] In order to overcome the above-mentioned defects of the prior art, the present application proposes a photovoltaic power station intelligent evaluation system, the system comprising:

[0010] Basic information management module, used to manage and maintain the basic data of photovoltaic power plants, detection library information and power plant information;

[0011] The detection and evaluation task management module is used to create and maintain detection and evaluation tasks, and determine the detection items and evaluation standards included in the detection and evaluation tasks;

[0012] The power plant data evaluation module is used to maintain the power plant data, maintain the technical parameters of each equipment in the power plant, set inspection rules to evaluate the power plant data, or record the manual evaluation results of the data;

[0013] The power plant quality assessment module is used to record and maintain the user's quality assessment results of the power plant related parameters;

[0014] Intelligent recognition module, used to analyze visible light and infrared thermal imaging images and output defects of photovoltaic strings;

[0015] The power station performance detection module is used to use the intelligent recognition module to detect defects in photovoltaic strings based on the collected images according to the detection items associated with the inspection tasks and the model algorithms associated with the detection items.

[0016] As an improvement of the above system, the data managed by the basic information management module includes: station data, station climate condition data, detection library data, system management data, evaluation sampling data and template data.

[0017] As an improvement of the above system, the functions of the detection and evaluation task management module include: creating detection and evaluation tasks, generating detection items, adjusting detection items and modifying detection project status.

[0018] As an improvement of the above system, the detection item status includes:

[0019] Draft status: This status is the initial status of the detection project creation, and the user can adjust the content;

[0020] Testing status: in this status, the tester is allowed to add new test items, conduct project tests and enter test data;

[0021] Status in the report, which is the status when all test items are completed, allowing adjustment of test items or output of test reports;

[0022] Completed status: When the inspection task is reviewed, it is in completed status.

[0023] As an improvement of the above system, the power plant data evaluation module supports classified uploading of file data of the power plant, supports online preview of corresponding electronic file data, supports setting threshold rules for power plant data evaluation, supports evaluating power plant data according to the set rules and outputting corresponding evaluation results.

[0024] As an improvement of the above system, the quality assessment content of the power station quality assessment module includes: actual installed power of the power station, visual quality of photovoltaic modules, form and quality of photovoltaic array brackets, form and quality of photovoltaic array foundations, photovoltaic module / array arrangement and installation quality, DC cable model and quality, cable laying quality, junction box function and quality, electrical clearance / creepage distance in the junction box, photovoltaic and inverter capacity ratio, inverter concentration / location and machine room quality, transformer installation method / distance, lightning protection grounding and construction quality, power station fence and quality and photovoltaic array cleaning plan / water consumption.

[0025] As an improvement of the above system, the intelligent recognition module includes an algorithm warehouse;

[0026] The algorithm warehouse is a fault detection model for intelligently identifying various photovoltaic string defects;

[0027] The photovoltaic string defect types include: hot spots, diode breakdown, component failure, component open circuit, string short circuit, string open circuit, string inefficient operation and abnormal component appearance.

[0028] As an improvement to the above system, the training process of the fault detection model includes:

[0029] First, data collection specifications are formulated to collect images of photovoltaic modules taken by drones to form original data sets;

[0030] Secondly, formulate data annotation specifications, annotate the original data set, and form a standard data set;

[0031] Finally, the yolov8 algorithm model is used to iteratively train the standard data set to eventually form a fault detection model.

[0032] As an improvement of the above system, the implementation process of the power station performance detection module includes:

[0033] Input the image taken by the drone, call the intelligent recognition module, and obtain the fault type and fault pixel coordinates of the photovoltaic string;

[0034] According to the longitude and latitude of the center point of the UAV image, the true longitude and latitude of the center of the UAV image is obtained using the quaternion algorithm and the aircraft space position inversion algorithm;

[0035] According to the fault pixel coordinates and the true center longitude and latitude coordinates of the UAV image, the pixel positioning algorithm is used to match the longitude and latitude codes of the photovoltaic module to obtain the code of the faulty photovoltaic module and achieve accurate fault positioning.

[0036] As an improvement of the above system, the power station performance detection module is also used to input the station data into the detection item formula for calculation, compare the calculation result data with the standard, and determine whether the station data meets the standard.

[0037] Compared with the prior art, the advantages of this application are:

[0038] 1. By strengthening data mining and deep learning functions, using AI technology, exploring the establishment of a power station evaluation algorithm library, collecting operating data and offline detection data, the system automatically processes data, and selects algorithms for corresponding evaluation items in the algorithm library. Combined with the technical standards and specifications for implementation, the operating status of each device and component of the power station and the existing safety hazards are obtained, and the corresponding charts are automatically calculated to achieve intelligent evaluation of the performance of photovoltaic power stations.

[0039] 2. It reduces the evaluation cost, shortens the evaluation cycle, avoids the errors caused by manual evaluation, improves the quality and accuracy of the evaluation, and provides guidance for improving the quality and efficiency of photovoltaic power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The figure shows the structure diagram of the photovoltaic power station intelligent evaluation system;

[0041] Figure 2 Shown is a schematic diagram of hot spot identification of photovoltaic strings;

[0042] Figure 3 The figure shows the schematic diagram of photovoltaic string diode breakdown identification;

[0043] Figure 4 Shown is a schematic diagram of photovoltaic string component failure identification;

[0044] Figure 5 The figure shows a schematic diagram of identifying missing photovoltaic string components;

[0045] Figure 6 The figure shows the schematic diagram of photovoltaic string component circuit break identification;

[0046] Figure 7 The figure shows the schematic diagram of photovoltaic string short circuit identification;

[0047] Figure 8 The figure shows the schematic diagram of photovoltaic string open circuit identification;

[0048] Fig. 9 The figure shows the schematic diagram of identifying the inefficient operation of photovoltaic strings;

[0049] Fig.10 The figure shows a schematic diagram for identifying abnormal appearance of a part;

[0050] Fig.11 Shown is a flow chart of the implementation process of the power station performance detection module. DETAILED DESCRIPTION

[0051] The technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0052] This application provides a photovoltaic power station intelligent evaluation system that integrates the quality inspection, performance evaluation and data analysis functions of photovoltaic power stations. Quality inspection focuses on appearance and compliance to achieve multi-dimensional and intelligent evaluation. Performance evaluation, through machine vision image recognition algorithms and AI algorithm libraries, combined with offline imported power station operation, meteorological environment and equipment status data, conducts cross-analysis from multiple aspects such as photovoltaic modules, photovoltaic strings, inverters, etc., to scientifically evaluate the current status of the power station. Through digital empowerment of testing and certification innovation and development, we actively promote the intelligent and precise analysis of digitalization and intelligence in testing and certification work, and find out the problems that affect the power generation efficiency of photovoltaic power stations faster and more accurately.

[0053] Photovoltaic power station intelligent evaluation system, including:

[0054] 1. Basic information management module, used to manage and maintain the basic data of photovoltaic power stations, detection library information, power station information and other information.

[0055] The data managed by the basic information management module includes:

[0056] Site data: mainly includes necessary basic data, such as site name, code, geographical location, installed capacity and other basic information, inverter model - inverter number - photovoltaic rated power connected to the inverter - power generation - time period, inverter, string, combiner box, each module number, power + each inverter number, power, etc.;

[0057] Site climate condition data: including annual average total horizontal radiation, annual average sunshine hours, annual average temperature, extreme maximum temperature, extreme minimum temperature, etc.

[0058] Test library data: including test library information, test item information, test standard information, test algorithm association information, test method description, etc.;

[0059] System management data: including user information, role information, permission information, data dictionary, operation log, etc.;

[0060] Evaluation sampling data: including power plant data evaluation data, operation data, sampling setting data, sampling data, sample backup data, etc.;

[0061] Data entry and template data: The data entry templates for various tests and performance evaluations can be managed through the system background. It supports uploading data from standard templates and automatically importing them into the database. The system can automatically match the data format exported by the power station.

[0062] 2. The detection and evaluation task management module is used to create and maintain detection and evaluation tasks, determine the detection items included in the detection and evaluation tasks, and the evaluation standards.

[0063] The functions of the detection and evaluation task management module include:

[0064] Create inspection and assessment tasks: The main maintenance information includes: site number, assessment category, assessment year, project status, creation time, creator, etc.

[0065] Generate detection items: Based on the selected evaluation category, randomly extract detection items from the corresponding detection library according to the rules.

[0066] Adjust detection items: Users can add or delete detection items as needed.

[0067] Modify the status of the test item: The status of the test item includes:

[0068] Test project status: draft, under test, in report, completed;

[0069] Draft: The initial state of the project creation, where users can adjust the content;

[0070] During testing: Testers can add new test items, conduct project tests and enter test data;

[0071] In the report: After all test items are completed, the test items can be adjusted or the test report can be output;

[0072] Completed: The task status is completed after the inspection is completed.

[0073] 3. The power plant data evaluation module is used to maintain the power plant data, maintain the technical parameters of each equipment in the power plant, set inspection rules to evaluate the power plant data, or record the results of manual data evaluation.

[0074] The power station data collection and evaluation work is the basic work content link of the test team before sampling and selecting photovoltaic power stations. This functional module can support the classification and uploading of power station file data, support online preview of corresponding electronic file data, support setting threshold rules for power station data evaluation, support automatic evaluation of power station data according to the set rules and output of corresponding evaluation results.

[0075] Before evaluating the power station data, it is necessary to sample the power station data. To ensure that the system can effectively save manpower and time costs, this function item can support the use of detection items in the use scenario. After setting the query conditions (such as data time, area, sub-array, string range, etc.), the system automatically completes the data sampling operation in the system according to the sampling rules and automatically collects the required data.

[0076] The rules for power station data sampling can be set flexibly, and the system can output the corresponding sampling test data results by automatically tracking the data of the "good, medium and poor" inverter units divided by the test. It can automatically screen out 6 groups of strings (2 good, 2 medium and 2 poor) from the data test results of different units according to the principle of "selecting the best from the good and the worst from the worst" for the tracked test data, and provide the final string data to be tested for a single test task for subsequent on-site testing, providing pre-data support for subsequent on-site testing.

[0077] 4. Power plant quality assessment module, used to record and maintain the user's quality assessment results on power plant related parameters.

[0078] The quality assessment contents include: actual installed power of the power station, visual quality of PV modules, form and quality of PV array brackets, form and quality of PV array foundations, arrangement and installation quality of PV modules / arrays, DC cable model and quality, cable laying quality, combiner box function and quality, electrical clearance / creepage distance in the combiner box, PV and inverter capacity ratio, inverter concentration / location and machine room quality, transformer installation method / distance, lightning protection grounding and construction quality, power station fence and quality, PV array cleaning plan / water consumption, environmental assessment, equipment identification and other monitoring contents. The system only manages the above quality inspection contents.

[0079] 5. Intelligent recognition module, which is used to analyze visible light and infrared thermal imaging images and discover various defects of photovoltaic strings. It includes algorithm warehouse, self-service modeling, external model integration, data standard management, data set management, model library management, model scheduling management and other functions, and supports docking with the enterprise's infrastructure platform and operation platform to achieve rapid docking of data required by the algorithm and supporting applications.

[0080] The algorithm warehouse of the intelligent identification module includes a fault detection model for intelligent identification of various photovoltaic string defects. The photovoltaic string defects that can be identified include:

[0081] Hot spot identification: Under certain conditions, defective areas (such as obstruction, cracks, bubbles, delamination, dirt, internal connection failure, etc.) in photovoltaic modules are treated as loads and consume the energy generated by other areas, resulting in local overheating. This phenomenon is called the "hot spot effect" of photovoltaic modules. This failure is one of the important reasons that affect the output power and service life of photovoltaic modules. Figure 2 shown.

[0082] Diode breakdown identification: A certain number of photovoltaic cells in a photovoltaic module are connected in parallel with bypass diodes. When certain faults occur in the cell, the diode acts as a bypass and the current flows through the diode. The bypass diode can reduce the power loss caused by the fault and effectively reduce the impact of higher reverse bias voltage on the cell. When power dissipation occurs in the diode, it may cause the diode to overload and short-circuit. In addition, the diode will cause a short circuit under faults such as lightning strikes. The diode breakdown identification diagram is as follows Figure 3 shown.

[0083] Module failure identification: In the infrared image, the entire module battery is in an open circuit state, which is manifested as a single module being too bright. The temperature of this circuit battery is higher than that of a normal working battery. Module failure identification is shown as follows: Figure 4 shown.

[0084] Missing Component Identification: Due to the influence of bracket installation and wind, the photovoltaic panel components fall or are blown away, leaving only the bracket. Figure 5 shown.

[0085] Component disconnection identification: A disconnection fault is a disconnection or breakage of the photovoltaic power generation unit including the components, strings, cables and junction boxes, which will cause the photovoltaic array to be unable to output power. The main causes of disconnection faults are severe component breakage caused by external forces, loose string connection cables, and disconnection of junction cables. Component disconnection identification diagram is as follows Figure 6 shown.

[0086] String short circuit identification: When a short circuit occurs in a component of a photovoltaic string, the bypass diode inside the photovoltaic component is turned on, and the parallel resistance of the photovoltaic component is close to zero. There is also a form of abnormal string shape that looks like hot spots are gathered all over the string. Usually the surface temperature of the component can reach 80 degrees. The severity level is high and needs to be handled as soon as possible. The string short circuit identification diagram is as follows Figure 7 shown.

[0087] String open circuit identification: In the infrared image, the temperature of the string where the entire string connection is lost is obviously different from the temperature of other strings. Due to the different wiring methods when the branches are physically connected, the fault manifestation is also different. String open circuit identification is shown as follows Figure 8 shown.

[0088] Identification of inefficient string operation: There are many reasons for inefficient operation, such as hidden cracks, hot spots, dust, and internal connection failure of strings. Hidden cracks are cracks caused by stress or pressure in the silicon solar cells of photovoltaic modules. These cracks are usually not visible to the naked eye and can only be seen through specific testing methods. Hidden cracks are the main cause of malfunction or failure of photovoltaic modules. Hot spots are unstable and may eventually cause damage to the module. When the area of ​​hot spots or highlighted areas accounts for a high ratio of the entire string area, it can be determined that the string is inefficient. The following is a diagram of identifying inefficient string operation: Fig. 9 shown.

[0089] Identification of abnormal appearance of components: Abnormal appearance mainly includes foreign objects (shade, weeds, bird droppings, dirt, local dust). Fig.10 shown.

[0090] The training process of the algorithm model includes: first, formulating data collection specifications, and collecting images of photovoltaic components taken by drones to form an original data set; second, formulating data annotation specifications (component hot spots, diode failures, component failures, component missing, component open circuits, string short circuits, string open circuits, abnormal component appearance, inefficient operation, etc.), annotating the original data set, and forming a standard data set; finally, using the yolov8 algorithm model to iteratively train the standard data set to ultimately form a fault detection model.

[0091] 6. The power station performance detection module is used to use the algorithm warehouse of the intelligent recognition module to automatically perform AI analysis on the collected images according to the detection items associated with the inspection tasks and the model algorithms associated with the detection items, including fault analysis such as dust, dirt, cracks, occlusions, diode damage, hot spots, etc., and generate intelligent analysis results. It can support automatic rapid positioning of abnormal components on the map and display the position code of abnormal components. It supports correction of the identified result data, and the corrected result data can also feed back the deep learning of the model algorithm. In the data detection process, it supports data comparison between visible light and infrared thermal imaging, and supports three-light data comparison of visible light, thermal infrared thermal imaging and imported EL detection data, etc., to further improve the accuracy of recognition and correction.

[0092] The power station performance detection module is also used to calculate by inputting the site data into the detection item formula, and the calculation result data is compared with the standard to determine whether it meets the standard. The calculated detection items mainly include: PV module power attenuation, PV array temperature rise loss, PV system stain and dust shielding loss, PV array shielding loss, PV module series mismatch loss in the string, parallel mismatch from each string to the string inverter, PV array insulation, inverter weighted efficiency, inverter maximum efficiency, grounding continuity detection, PV string to combiner box DC line loss, combiner box to inverter DC line loss, PV string to string inverter DC line loss, centralized inverter AC line loss, string inverter AC line loss, etc.

[0093] like Fig.11 As shown in the figure, the implementation process of the power station performance detection module includes:

[0094] (1) Input the image taken by the drone, call the algorithm warehouse of the intelligent recognition module, and obtain the fault type and fault pixel coordinates of the photovoltaic string;

[0095] (2) Based on the longitude and latitude of the center point of the UAV image, the true longitude and latitude of the center of the UAV image is obtained using the quaternion algorithm and the aircraft spatial position back-calculation algorithm;

[0096] (3) According to the fault pixel coordinates and the true center longitude and latitude coordinates of the UAV image, the pixel positioning algorithm is used to match the basic data in the database (longitude and latitude coding of the photovoltaic module) to obtain the coding of the faulty photovoltaic module and achieve accurate fault positioning.

[0097] The present application may also provide a computer device, comprising: at least one processor, a memory, at least one network interface and a user interface. The various components in the device are coupled together through a bus system. It is understood that the bus system is used to achieve connection and communication between these components. In addition to the data bus, the bus system also includes a power bus, a control bus and a status signal bus.

[0098] The user interface may include a display, a keyboard or a pointing device, such as a mouse, a trackball, a touch pad or a touch screen.

[0099] It is understood that the memory in the embodiments disclosed in the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DRRAM). The memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.

[0100] In some embodiments, the memory stores the following elements, executable modules or data structures, or a subset thereof, or an extended set thereof: an operating system and applications.

[0101] The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application includes various application programs, such as a media player (Media Player), a browser (Browser), etc., which are used to implement various application services. The program for implementing the method of the embodiment of the present disclosure can be included in the application.

[0102] In the above embodiment, the processor may also call a program or instruction stored in the memory, specifically, a program or instruction stored in an application program, and is used to:

[0103] Execute the steps of the above method.

[0104] The above method can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor may be a general processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The above-disclosed methods, steps and logic block diagrams can be implemented or executed. The general processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the above-disclosed method can be directly embodied as a hardware decoding processor to execute, or the hardware and software modules in the decoding processor are combined to execute. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0105] It is understood that the embodiments described in the present application can be implemented by hardware, software, firmware, middleware, microcode or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described in the present application or a combination thereof.

[0106] For software implementation, the technology of the present application can be implemented by executing the functional modules (such as procedures, functions, etc.) of the present application. The software code can be stored in a memory and executed by a processor. The memory can be implemented in the processor or outside the processor.

[0107] The present application may also provide a non-volatile storage medium for storing a computer program. When the computer program is executed by a processor, each step in the above method embodiment can be implemented.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application is described in detail with reference to the embodiments, a person skilled in the art should understand that any modification or equivalent replacement of the technical solution of the present application does not depart from the spirit and scope of the technical solution of the present application and should be included in the scope of the claims of the present application.

Claims

1. A photovoltaic power station intelligent evaluation system, characterized in that: The system comprises: Basic information management module, used to manage and maintain the basic data of photovoltaic power plants, detection library information and power plant information; The detection and evaluation task management module is used to create and maintain detection and evaluation tasks, and determine the detection items and evaluation standards included in the detection and evaluation tasks; The power plant data evaluation module is used to maintain the power plant data, maintain the technical parameters of each equipment in the power plant, set inspection rules to evaluate the power plant data, or record the manual evaluation results of the data; The power plant quality assessment module is used to record and maintain the user's quality assessment results of the power plant related parameters; Intelligent recognition module, used to analyze visible light and infrared thermal imaging images and output defects of photovoltaic strings; and The power station performance detection module is used to use the intelligent recognition module to detect defects in photovoltaic strings based on the collected images according to the detection items associated with the inspection tasks and the model algorithms associated with the detection items.

2. The photovoltaic power station intelligent evaluation system according to claim 1, characterized in that: The data managed by the basic information management module include: station data, station climate condition data, detection library data, system management data, evaluation sampling data and template data.

3. The photovoltaic power station intelligent evaluation system according to claim 1, characterized in that: The functions of the detection and evaluation task management module include: creating detection and evaluation tasks, generating detection items, adjusting detection items and modifying detection project status.

4. The photovoltaic power station intelligent evaluation system according to claim 3, characterized in that: The detection item status includes: Draft status: This status is the initial status of the detection project creation, and the user can adjust the content; Testing status: in this status, the tester is allowed to add new test items, conduct project tests and enter test data; Status in the report, which is the status when all test items are completed, allowing adjustment of test items or output of test reports; Completed status: When the inspection task is reviewed, it is in completed status.

5. The photovoltaic power station intelligent evaluation system according to claim 1, characterized in that: The power plant data evaluation module supports the classified uploading of power plant file data, supports online preview of corresponding electronic file data, supports setting threshold rules for power plant data evaluation, supports evaluating power plant data according to the set rules and outputting corresponding evaluation results.

6. The photovoltaic power station intelligent evaluation system according to claim 1, characterized in that: The quality assessment content of the power station quality assessment module includes: actual installed power of the power station, visual quality of PV modules, form and quality of PV array brackets, form and quality of PV array foundations, arrangement and installation quality of PV modules / arrays, DC cable model and quality, cable laying quality, combiner box function and quality, electrical clearance / creepage distance in the combiner box, PV and inverter capacity ratio, inverter concentration / location and machine room quality, transformer installation method / distance, lightning protection grounding and construction quality, power station fence and quality and PV array cleaning plan / water consumption.

7. The photovoltaic power station intelligent evaluation system according to claim 1, characterized in that: The intelligent recognition module includes an algorithm warehouse; The algorithm warehouse is a fault detection model for intelligently identifying various photovoltaic string defects; The photovoltaic string defect types include: hot spots, diode breakdown, component failure, component open circuit, string short circuit, string open circuit, string inefficient operation and abnormal component appearance.

8. The photovoltaic power station intelligent evaluation system according to claim 7, characterized in that: The training process of the fault detection model includes: First, data collection specifications are formulated to collect images of photovoltaic modules taken by drones to form original data sets; Secondly, formulate data annotation specifications, annotate the original data set, and form a standard data set; Finally, the yolov8 algorithm model is used to iteratively train the standard data set to eventually form a fault detection model.

9. The photovoltaic power station intelligent evaluation system according to claim 1, characterized in that: The implementation process of the power station performance detection module includes: Input the image taken by the drone, call the intelligent recognition module, and obtain the fault type and fault pixel coordinates of the photovoltaic string; According to the longitude and latitude of the center point of the UAV image, the true longitude and latitude of the center of the UAV image is obtained using the quaternion algorithm and the aircraft space position inversion algorithm; According to the fault pixel coordinates and the true center longitude and latitude coordinates of the UAV image, the pixel positioning algorithm is used to match the longitude and latitude coding of the photovoltaic module to obtain the coding of the faulty photovoltaic module and achieve accurate fault positioning.

10. The photovoltaic power station intelligent evaluation system according to claim 1, characterized in that: The power station performance detection module is also used to input the station data into the detection item formula for calculation, compare the calculation result data with the standard, and determine whether the station data meets the standard.

Images