CN-121980325-A - Unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system

CN121980325ACN 121980325 ACN121980325 ACN 121980325ACN-121980325-A

Abstract

The application relates to the technical field of operation and maintenance of photovoltaic power stations, in particular to a photovoltaic power station hot spot fault diagnosis system based on an unmanned aerial vehicle. The system comprises an unmanned aerial vehicle acquisition unit, an onboard calculation unit, a path planning unit, a diagnosis and visualization unit and a decision and visualization unit, wherein the unmanned aerial vehicle acquisition unit is used for acquiring multi-mode data of a photovoltaic power station, the onboard calculation unit is used for carrying out real-time analysis on the multi-mode data, identifying suspected hot spot faults and generating corresponding rechecking instructions, the path planning unit is used for dynamically adjusting a preset flight path of the unmanned aerial vehicle according to the rechecking instructions, generating a targeted inspection path pointing to the suspected hot spot faults, carrying out rechecking data acquisition on a suspected hot spot fault area and generating rechecking data, the diagnosis and analysis unit is used for receiving the rechecking data, carrying out accurate judgment and geographic positioning on the hot spot faults, and generating a diagnosis report according to the accurate judgment and geographic positioning on the hot spot faults. Autonomous, accurate and efficient diagnosis of the hot spot fault of the photovoltaic power station is realized.

Inventors

ZHU RONGKANG
Feng Xuhao
ZHAO JUNLIN
HU YONG
Cheng Zhongyue
LI DONGRU

Assignees

西安沣东华能热力有限公司

Dates

Publication Date: 20260505
Application Date: 20251128

Claims (10)

1. Unmanned aerial vehicle-based photovoltaic power plant hot spot fault diagnosis system, characterized by comprising: the unmanned aerial vehicle acquisition unit is used for acquiring multi-mode data of the photovoltaic power station; The airborne computing unit is used for analyzing the multi-mode data in real time, identifying suspected hot spot faults and generating corresponding rechecking instructions; The path planning unit is used for dynamically adjusting a preset flight path of the unmanned aerial vehicle according to the rechecking instruction to obtain a targeted inspection path of the suspected hot spot fault, and collecting rechecking data of the suspected hot spot fault area to generate rechecking data; The diagnosis analysis unit is used for receiving the re-detection data and carrying out accurate judgment and geographic positioning on the hot spot fault; And the decision and visualization unit is used for generating a diagnosis report according to the accurate decision and the geographic positioning of the hot spot fault.
2. The unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system according to claim 1, wherein the unmanned aerial vehicle acquisition unit is configured to acquire multi-modal data of the photovoltaic power station, and comprises: the first processing module is used for acquiring the space pose data of the unmanned aerial vehicle platform in real time under the navigation coordinate system of the photovoltaic power station; And the second processing module is used for synchronously acquiring an infrared thermal imaging sequence and a visible light image sequence of the photovoltaic module.
3. The unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system according to claim 2, wherein the on-board computing unit is configured to analyze the multi-mode data in real time, and when a suspected hot spot fault is identified and a corresponding recheck instruction is generated, the system comprises: the third processing module is used for preprocessing the infrared thermal imaging sequence, and carrying out preliminary screening on the hot spots to generate a preliminary screening result; and the fourth processing module is used for generating a rechecking instruction aiming at the suspected fault area according to the preliminary screening result and the space pose data.
4. The photovoltaic power station hot spot fault diagnosis system based on the unmanned aerial vehicle according to claim 3, wherein the path planning unit is configured to dynamically adjust a preset flight path of the unmanned aerial vehicle according to the rechecking instruction, generate a targeted inspection path pointing to a suspected hot spot fault, and perform rechecking data acquisition on the suspected hot spot fault area, and when generating rechecking data, the path planning unit includes: The fifth processing module is used for planning a global cruising path of the unmanned aerial vehicle based on a digital earth surface model of the photovoltaic power station, wherein the global cruising path is integrated with fixed obstacle information in the range of the power station in advance; And the sixth processing module is used for receiving and executing the rechecking instruction on the basis of the global cruising path, dynamically generating the targeted routing inspection path and driving the unmanned aerial vehicle to closely observe the suspected fault point.
5. The photovoltaic power plant hot spot fault diagnosis system according to claim 4, wherein the fifth processing module is configured to, when planning a global cruising path of the unmanned aerial vehicle based on a digital surface model of the photovoltaic power plant, include: global path planning is achieved by solving the following functions: ; Constraint conditions: ; Wherein, the Is the overall cost function of the global path planning, For the length of the path of the i-th segment, For the cruising speed of the unmanned aerial vehicle, In part at the expense of time, To fly over the pre-energy expenditure factor of the terrain, For the energy cost portion, α is the weight coefficient of the time cost portion, β is the weight coefficient of the energy cost portion, Is a preset safety distance from a known static obstacle.
6. The unmanned aerial vehicle-based photovoltaic power plant hot spot fault diagnosis system according to claim 5, wherein the sixth processing module further comprises: Executing real-time dynamic obstacle avoidance, which specifically comprises the following steps: based on the environmental information perceived by the unmanned plane in real time, an attractive force is generated for the rechecking target point And generating a repulsive force for the detected obstacle ; The flight direction of the unmanned aerial vehicle is controlled in real time by resolving the resultant force of the attractive force and the repulsive force, and a local obstacle avoidance path is generated; Resultant force of the attractive force and the repulsive force The method comprises the following steps: ; Wherein Fr is the resultant force received by the re-inspection target point, For the review object point attractive force, Is the repulsive force from the nth obstacle.
7. The unmanned aerial vehicle-based photovoltaic power plant hot spot fault diagnosis system according to claim 6, wherein the sixth processing module further comprises: evaluating the value of each unmanned aerial vehicle executing task for each suspected fault point to be rechecked, and generating a value evaluation result; the value comprehensively considers the efficiency of the unmanned aerial vehicle for reaching the task point and the cost required for executing the task; Based on the value evaluation results of all unmanned aerial vehicles on all tasks, a global task allocation scheme is constructed, and the scheme needs to meet the following constraint: each task to be rechecked is distributed to one unmanned aerial vehicle at most; The total number of tasks assigned to each unmanned aerial vehicle does not exceed the maximum task capacity.
8. The unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system according to claim 7, wherein the evaluating the value of each unmanned aerial vehicle performing a task for each suspected fault point to be rechecked comprises: ; Wherein, the Represents the value of assigning unmanned aerial vehicles to perform tasks, Representing the priority weights of the tasks, Representing predicted time of flight for unmanned aerial vehicle flight mission points Represents the estimated energy consumption of the unmanned aerial vehicle to execute the task, In order to be a time-cost factor, Is an energy cost coefficient.
9. The unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system according to claim 8, wherein the diagnosis and analysis unit is configured to receive the recheck data, and when performing accurate determination and geolocation of a hot spot fault, the diagnosis and analysis unit includes: The seventh processing module is used for accurately identifying hot spots and classifying faults of the returned infrared thermal imaging data based on the deep learning model; and the eighth processing module is used for fusing the positioning and attitude determination data with the image recognition result and calculating the accurate geographic coordinates of the fault photovoltaic module.
10. The unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system according to claim 9, wherein the decision and visualization unit is configured to generate a visual hot spot distribution map and a diagnosis report according to the accurate determination and the geographic positioning of the hot spot fault, and the method comprises: The ninth processing module is used for visually marking all fault points on the digital map of the power station and generating a hot spot distribution map; And the tenth processing module is used for integrating all diagnostic information and outputting a comprehensive diagnostic report containing fault statistics, positioning information and maintenance suggestions.

Description

Unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system Technical Field The application relates to the technical field of operation and maintenance of photovoltaic power stations, in particular to a photovoltaic power station hot spot fault diagnosis system based on an unmanned aerial vehicle. Background In recent years, photovoltaic power generation has been an important component of clean energy, and its installed capacity has been continuously increasing. The photovoltaic module is a core unit for power generation of a power station, and the health condition of the photovoltaic module directly influences the power generation efficiency and the income. The hot spot fault is a typical defect of a photovoltaic module, and can cause local overheating, power attenuation and even fire disaster, so that accurate and efficient diagnosis of the hot spot is a key link of operation and maintenance of a photovoltaic power station. However, the existing hot spot detection scheme lacks a real-time analysis and rechecking mechanism for suspected faults in the detection stage of manual inspection, is easy to misjudge and miss the faults caused by environmental interference, and fails to link with a diagnosis result in the path planning stage, so that targeted approaching observation for fault points cannot be realized. Disclosure of Invention The application aims to solve the technical problems, provides a photovoltaic power station hot spot fault diagnosis system based on an unmanned aerial vehicle, and aims to primarily identify suspected hot spots and generate a recheck instruction by acquiring multi-mode data and carrying out airborne real-time analysis; the method comprises the steps of dynamically adjusting a flight path based on a rechecking instruction, realizing targeted secondary data acquisition of suspected fault areas, accurately identifying hot spots on secondary data, classifying faults and geographically locating the faults, and finally generating a visual hot spot distribution map and a comprehensive diagnosis report containing fault statistics, locating information and maintenance suggestions. The application realizes the process automation from preliminary screening, accurate rechecking to intelligent diagnosis and decision support, remarkably improves the accuracy, efficiency and automation level of the hot spot fault diagnosis of the photovoltaic power station, and provides reliable technical support for the preventive maintenance and asset management of the power station. In some embodiments of the present application, there is provided an unmanned aerial vehicle-based photovoltaic power station hot spot fault diagnosis system, including: the unmanned aerial vehicle acquisition unit is used for acquiring multi-mode data of the photovoltaic power station; The airborne computing unit is used for analyzing the multi-mode data in real time, identifying suspected hot spot faults and generating corresponding rechecking instructions; The path planning unit is used for dynamically adjusting a preset flight path of the unmanned aerial vehicle according to the rechecking instruction to obtain a targeted inspection path of the suspected hot spot fault, and collecting rechecking data of the suspected hot spot fault area to generate rechecking data; The diagnosis analysis unit is used for receiving the re-detection data and carrying out accurate judgment and geographic positioning on the hot spot fault; And the decision and visualization unit is used for generating a diagnosis report according to the accurate decision and the geographic positioning of the hot spot fault. In some embodiments of the present application, the unmanned aerial vehicle acquisition unit is configured to acquire multi-mode data of a photovoltaic power station, and includes: the first processing module is used for acquiring the space pose data of the unmanned aerial vehicle platform in real time under the navigation coordinate system of the photovoltaic power station; And the second processing module is used for synchronously acquiring an infrared thermal imaging sequence and a visible light image sequence of the photovoltaic module. In some embodiments of the present application, the on-board computing unit is configured to perform real-time analysis on the multi-mode data, and when a suspected hot spot fault is identified and a corresponding recheck instruction is generated, the on-board computing unit includes: the third processing module is used for preprocessing the infrared thermal imaging sequence, and carrying out preliminary screening on the hot spots to generate a preliminary screening result; and the fourth processing module is used for generating a rechecking instruction aiming at the suspected fault area according to the preliminary screening result and the space pose data. In some embodiments of the present application, the path planning unit is configured to dynamically adjust a preset flig