CN-121979242-A - Flight control method and system of inspection unmanned aerial vehicle in insulator degradation state

CN121979242ACN 121979242 ACN121979242 ACN 121979242ACN-121979242-A

Abstract

The invention relates to the technical field of unmanned aerial vehicle inspection, and discloses a flight control method and a flight control system for an unmanned aerial vehicle for inspecting an insulator in a deteriorated state. The method at least comprises the steps of constructing an objective function based on the square sum of first derivatives of flight parameters of the minimized unmanned aerial vehicle, solving the objective function according to constraint conditions formed by spatial distribution characteristics, spatial electric field intensity and temperature field distribution to obtain initial optimal unmanned aerial vehicle flight parameters, dynamically correcting the initial optimal unmanned aerial vehicle flight parameters according to an adjustment strategy generated by reinforcement learning to obtain real-time optimal unmanned aerial vehicle flight parameters, calculating an initial optimal unmanned aerial vehicle inspection path after optimizing a bottleneck path and a redundant path based on the real-time optimal unmanned aerial vehicle flight parameters, and dynamically adjusting the initial optimal unmanned aerial vehicle inspection path according to real-time detection data of insulator degradation states. The invention realizes the precision, high efficiency and safety of insulator degradation state detection under a complex inspection scene.

Inventors

WU WEIHUA
DING JIAN
SHI YUCHAO
SU LIANGZHI
GAO YUFENG
LI JIASI
BIAN XUEJING
HUANG JIABIN
WAN YANZHEN

Assignees

国网浙江省电力有限公司杭州供电公司

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (10)

1. The flight control method of the unmanned aerial vehicle for inspecting the degradation state of the insulator is characterized by comprising the following steps of: Acquiring multidimensional monitoring data of a target inspection area before inspection of the unmanned aerial vehicle, and obtaining spatial distribution characteristics, spatial electric field intensity and temperature field distribution related to an insulator; Constructing an objective function based on the square sum of first derivatives of the minimized unmanned aerial vehicle flight parameters, and solving the objective function according to constraint conditions formed by the spatial distribution characteristics, the spatial electric field intensity and the temperature field distribution to obtain initial optimal unmanned aerial vehicle flight parameters; The unmanned aerial vehicle is controlled to take off according to the initial optimal unmanned aerial vehicle flight parameters and enters the target inspection area to perform inspection, and real-time state information of the unmanned aerial vehicle, real-time electric field measurement data and real-time temperature measurement data related to an insulator are obtained; dynamically correcting the initial optimal unmanned aerial vehicle flight parameters according to an adjustment strategy generated by reinforcement learning based on the unmanned aerial vehicle real-time state information, the electric field measurement data and the real-time temperature measurement data to obtain real-time optimal unmanned aerial vehicle flight parameters; And calculating an initial optimal unmanned aerial vehicle inspection path after optimizing a bottleneck path and a redundant path by combining the current inspection position and the target inspection position of the unmanned aerial vehicle based on the real-time optimal unmanned aerial vehicle flight parameters, and dynamically adjusting the initial optimal unmanned aerial vehicle inspection path according to real-time detection data of the insulator degradation state.
2. The method for controlling the flight of an unmanned aerial vehicle for inspecting an insulator degradation state according to claim 1, wherein the acquiring multi-dimensional monitoring data of a target inspection area before inspection of the unmanned aerial vehicle to obtain spatial distribution characteristics, spatial electric field intensity and temperature field distribution related to the insulator comprises: Acquiring images and point cloud data of a target inspection area before inspection of the unmanned aerial vehicle, space three-dimensional electric field data, insulator surface and surrounding environment temperature data; extracting spatial distribution characteristics related to the insulator from the image and the point cloud data; Calibrating the space three-dimensional electric field data by combining the simulation reference data of the power system to obtain the space electric field intensity related to the insulator; and calibrating the temperature data of the surface and the surrounding environment of the insulator by using a thermodynamic simulation model to obtain the temperature field distribution related to the insulator.
3. The method for controlling the flight of the unmanned aerial vehicle for inspecting the degradation state of the insulator according to claim 1, wherein the constructing the objective function based on the sum of squares of first derivatives of the minimized unmanned aerial vehicle flight parameters, solving the objective function according to constraint conditions composed of the spatial distribution characteristics, the spatial electric field intensity and the temperature field distribution, and obtaining initial optimal unmanned aerial vehicle flight parameters comprises: taking unmanned aerial vehicle flight parameters including flight track, flight altitude and flight speed as optimization targets, and constructing an objective function based on the square sum of first derivatives of the minimized unmanned aerial vehicle flight parameters; converting the spatial distribution characteristics, the spatial electric field intensity and the temperature field distribution into constraint conditions for determining the safety boundary of the unmanned aerial vehicle flight parameter after prediction and evaluation; And solving the objective function according to the constraint condition based on the multidimensional monitoring data to obtain initial optimal unmanned aerial vehicle flight parameters.
4. The method for controlling the flight of the unmanned aerial vehicle for inspecting the degradation state of the insulator according to claim 1, wherein the dynamically correcting the initial optimal unmanned aerial vehicle flight parameter according to the adjustment strategy generated by reinforcement learning based on the real-time state information of the unmanned aerial vehicle, the electric field measurement data and the real-time temperature measurement data to obtain the real-time optimal unmanned aerial vehicle flight parameter comprises: calculating unmanned aerial vehicle state deviation, electric field deviation and temperature deviation based on the unmanned aerial vehicle real-time state information, the electric field measurement data and the real-time temperature measurement data; Inputting the unmanned plane state deviation, the electric field deviation and the temperature deviation into a pre-constructed flight parameter adjustment strategy generation model to perform strategy reasoning so as to obtain a real-time unmanned plane flight parameter adjustment strategy; and dynamically correcting the initial optimal unmanned aerial vehicle flight parameters according to the unmanned aerial vehicle flight parameter real-time adjustment strategy to obtain the real-time optimal unmanned aerial vehicle flight parameters.
5. The method for controlling the flight of the unmanned aerial vehicle for inspecting the degradation state of an insulator according to claim 4, wherein the construction process of the flight parameter adjustment strategy generation model comprises the following steps: The unmanned aerial vehicle state deviation, the electric field deviation and the temperature deviation are taken as inputs, and an adjustment strategy aiming at unmanned aerial vehicle flight parameters is taken as output, so that an SAC reinforcement learning strategy network is constructed; designing a cumulative reward function which comprises an electric field tracking precision weight coefficient, an unmanned plane state stability weight coefficient and a flight smoothness weight coefficient and introduces an entropy regularization term; the SAC reinforcement learning strategy network is interacted with a multi-physical field coupling environment in the target inspection area to obtain experience data containing flight parameter adjustment strategies under different insulator inspection scenes; and performing iterative training on the SAC reinforcement learning strategy network based on the experience data with the aim of maximizing the cumulative reward function value to obtain a flight parameter adjustment strategy generation model.
6. The method for controlling the flight of the unmanned aerial vehicle for inspecting the degradation state of the insulator according to claim 1, wherein the calculating the initial optimal unmanned aerial vehicle inspection path after optimizing the bottleneck path and the redundant path by combining the current inspection position and the target inspection position of the unmanned aerial vehicle based on the real-time optimal unmanned aerial vehicle flight parameter, and dynamically adjusting the initial optimal unmanned aerial vehicle inspection path according to the real-time detection data of the degradation state of the insulator comprises: identifying a bottleneck path which obstructs traffic in the inspection process and a redundant path which repeatedly covers or deviates from the target inspection area in the inspection process based on the historical inspection data of the target inspection area; Based on the real-time optimal unmanned aerial vehicle flight parameters, combining the current inspection position and the target inspection position of the unmanned aerial vehicle, calculating an initial optimal unmanned aerial vehicle inspection path after eliminating the bottleneck path and avoiding the redundant path; and if the real-time detection data of the insulator degradation state shows that the suspected insulator degradation region exists, dynamically adjusting the initial optimal unmanned aerial vehicle inspection path.
7. The method for controlling the flight of the unmanned aerial vehicle for inspecting the degradation state of an insulator according to claim 6, wherein the identifying a bottleneck path blocking traffic during inspection and a redundant path repeatedly covering or deviating from the target inspection area during inspection based on the historical inspection data of the target inspection area comprises: based on the historical inspection data of the target inspection area, a bottleneck path which obstructs traffic in the inspection process is identified by adopting a sliding window method, and a redundant path which repeatedly covers or deviates from the target inspection area in the inspection process is identified by dynamic time warping.
8. The method for controlling the flight of the unmanned aerial vehicle for inspecting the degradation state of the insulator according to claim 6, wherein the calculating the initial optimal unmanned aerial vehicle inspection path after eliminating the bottleneck path and avoiding the redundant path based on the real-time optimal unmanned aerial vehicle flight parameter and combining the current inspection position and the target inspection position of the unmanned aerial vehicle comprises: and calculating and eliminating the bottleneck path and an initial optimal unmanned aerial vehicle routing inspection path after the redundant path is avoided by adopting a path search algorithm based on the real-time optimal unmanned aerial vehicle flight parameters and combining the current routing inspection position and the target routing inspection position of the unmanned aerial vehicle.
9. The method of claim 6, wherein the dynamically adjusting includes increasing a patrol path node density of the suspected insulator degradation region.
10. The utility model provides an unmanned aerial vehicle's flight control system is patrolled and examined to insulator degradation state which characterized in that includes: the data acquisition module is used for acquiring multidimensional monitoring data of a target inspection area before inspection of the unmanned aerial vehicle to obtain spatial distribution characteristics, spatial electric field intensity and temperature field distribution related to the insulator; The initial parameter determining module is used for constructing an objective function based on the square sum of first derivatives of the minimized unmanned aerial vehicle flight parameters, and solving the objective function according to constraint conditions formed by the spatial distribution characteristics, the spatial electric field strength and the temperature field distribution to obtain initial optimal unmanned aerial vehicle flight parameters; The inspection control module is used for controlling the unmanned aerial vehicle to take off according to the initial optimal unmanned aerial vehicle flight parameters and enter the target inspection area to perform inspection, so that real-time state information of the unmanned aerial vehicle, real-time electric field measurement data and real-time temperature measurement data related to an insulator are obtained; The real-time parameter determining module is used for dynamically correcting the initial optimal unmanned aerial vehicle flight parameter according to an adjustment strategy generated by reinforcement learning based on the unmanned aerial vehicle real-time state information, the electric field measurement data and the real-time temperature measurement data to obtain the real-time optimal unmanned aerial vehicle flight parameter; And the path generation and adjustment module is used for calculating an initial optimal unmanned aerial vehicle inspection path after optimizing the bottleneck path and the redundant path by combining the current inspection position and the target inspection position of the unmanned aerial vehicle based on the real-time optimal unmanned aerial vehicle flight parameters, and dynamically adjusting the initial optimal unmanned aerial vehicle inspection path according to real-time detection data of the insulator degradation state.

Description

Flight control method and system of inspection unmanned aerial vehicle in insulator degradation state Technical Field The invention relates to the technical field of unmanned aerial vehicle inspection, in particular to a flight control method and a flight control system for an unmanned aerial vehicle inspection in an insulator degradation state. Background The traditional insulator inspection mode mainly comprises manual tower climbing detection and ground telescope observation, and has the problems of high labor intensity, low inspection efficiency, high risk of overhead operation, obvious limitation of topography and climate conditions and the like. More importantly, manual inspection is difficult to accurately capture early degradation characteristics of the insulator, and a power transmission line with remote or complex layout cannot be covered comprehensively, so that a blind area exists in inspection, and the intelligent operation and maintenance requirements of a power grid are difficult to meet. In recent years, unmanned aerial vehicle inspection technology gradually becomes a mainstream solution of insulator inspection by virtue of the advantages of strong flexibility, wide operation range, lower cost and the like. Through carrying on equipment such as vision sensor, electric field sensor, unmanned aerial vehicle can reach the region of patrolling and examining fast, gathers the outward appearance image and the surrounding environment data of insulator, has promoted the efficiency of patrolling and examining and has enlarged the coverage to a certain extent. However, the insulator inspection scene faces a complex multi-physical field environment, and the existing unmanned aerial vehicle inspection technology still has a plurality of technical challenges to be solved. On one hand, the prior art only focuses on the image feature recognition of the insulator, but fails to fully consider the coupling relation between key environment parameters such as electric field intensity, temperature field distribution and the like and unmanned aerial vehicle flight parameters, on the other hand, the conventional path planning scheme is mostly executed based on a preset route, and fails to effectively recognize a bottleneck path and a redundant path in the inspection process, so that the inspection path is long and low in efficiency, and meanwhile, a dynamic path adjustment mechanism aiming at real-time environment change is lacked, so that the comprehensiveness and the high efficiency of inspection are difficult to be considered. Disclosure of Invention When the existing unmanned aerial vehicle inspection technology is used for dealing with complex multi-physical field environments of insulator inspection, the problems of poor adaptability of flight parameters and environments, insufficient optimization of path planning and the like exist, and therefore the precision and efficiency of insulator degradation state inspection still have a large improvement space. In order to solve the technical problems, the invention provides a flight control method and a flight control system for an unmanned aerial vehicle for inspecting an insulator in a deteriorated state. In a first aspect, an embodiment of the present invention provides a flight control method for an unmanned aerial vehicle for inspecting an insulator in a degraded state, including: Acquiring multidimensional monitoring data of a target inspection area before inspection of the unmanned aerial vehicle, and obtaining spatial distribution characteristics, spatial electric field intensity and temperature field distribution related to an insulator; Constructing an objective function based on the square sum of first derivatives of the minimized unmanned aerial vehicle flight parameters, and solving the objective function according to constraint conditions formed by the spatial distribution characteristics, the spatial electric field intensity and the temperature field distribution to obtain initial optimal unmanned aerial vehicle flight parameters; The unmanned aerial vehicle is controlled to take off according to the initial optimal unmanned aerial vehicle flight parameters and enters the target inspection area to perform inspection, and real-time state information of the unmanned aerial vehicle, real-time electric field measurement data and real-time temperature measurement data related to an insulator are obtained; dynamically correcting the initial optimal unmanned aerial vehicle flight parameters according to an adjustment strategy generated by reinforcement learning based on the unmanned aerial vehicle real-time state information, the electric field measurement data and the real-time temperature measurement data to obtain real-time optimal unmanned aerial vehicle flight parameters; And calculating an initial optimal unmanned aerial vehicle inspection path after optimizing a bottleneck path and a redundant path by combining the current inspection position and t