Research On Uav Path Planning Algorithm For Wind Turbine Inspection

With the development of the world economy,the energy on the earth has been overexploited and utilized by human beings.On the one hand,fossil energy is increasingly exhausted,and on the other hand,a large amount of toxic gases have accumulated,leading to global warming.In order to improve the environment,China has put forward the strategic goal of carbon neutralization and carbon peak,vigorously develop clean energy and promote the construction of wind power generation facilities.With the continuous expansion of the scale of wind power plants,the traditional manual inspection method can not meet the requirements.At present,UAV is widely used in fault inspection because of its flexible and efficient characteristics.However,the manual operation of UAV has high requirements for patrol personnel.If the automatic patrol inspection of UAV is realized,the manpower and material resources can be saved and the patrol efficiency can be improved.Planning a reasonable patrol path is an important prerequisite for realizing automatic patrol of UAV.Based on the task background of wind turbine patrol,this paper studies the patrol path planning of UAV from two aspects: the overall patrol path and the local patrol path.First of all,this paper analyzes the research background,research significance and current situation at home and abroad of UAV patrol inspection,introduces the content of UAV patrol inspection and the structure of patrol inspection system of wind turbine,and analyzes the typical faults and patrol inspection methods of wind turbine blades.Secondly,the arrangement mode of wind turbine is analyzed,and the irregular arrangement mode is selected in this paper.The coordinate points are randomly selected in the two-dimensional coordinate system to simulate the distribution of wind turbines,and the planning of the overall inspection path is transformed into a TSP problem.Aiming at the shortcomings of the traditional grey Wolf algorithm,an improved grey wolf optimization algorithm is proposed,which uses Logistic chaotic sequence to initialize the population and make the population evenly distributed.Further improvement is made on the original search strategy,which improves the information exchange between individuals in the population,helps to jump out of the local optimal solution,changes the original linear convergence factor to nonlinear,and achieves a balance between the early search and the later search.The Improved Grey Wolf algorithm is applied to UAV patrol path planning,and the goal of not repeatedly traversing all wind turbines is fully achieved.Finally,the basic particle swarm optimization and sparrow search algorithm are combined to obtain an improved particle swarm optimization algorithm,which solves the shortcoming of premature convergence of the basic particle swarm optimization algorithm.Considering the actual patrol environment,two mountain models are established.Applying the improved particle swarm optimization algorithm to UAV patrol path planning can generate better paths in different three-dimensional models.Finally,through B-spline curve,the patrol path transition is smoother.Figure 58 Table 9 Reference 81...