Research On Algorithm For Path Planning Of Substation Inspection Robot

As the industrial scale develops and new voltage levels continue to increase,the development,making unattended or less-attlarge-scale interconnection of power grids and power systems has become a trend of ended substations develop in a large-scale,modern,and complex direction.The use of traditional manual inspection methods can no longer meet the needs of modern substation inspections.In order to achieve automation,intelligence and efficiency of substation inspections,the introduction of inspection robots in substations to assist or replace manual inspections can effectively compensate for traditional manual inspections.Defects in inspection methods.Substation global path planning is the prerequisite for the inspection robot to complete inspection tasks.Therefore,it is of great significance to study the global path planning of substation inspection robots.In this paper,the following researches are done on the path planning of the substation inspection robot and its key technologies:First,the principle and advantages and disadvantages of six typical environmental modeling methods including viewable method,tangent chart method,Voronoi chart method,free space method,topological method,and grid method are analyzed.With characteristics,it is determined that the grid method is adopted for the environment modeling in this paper.Secondly,it analyzes the four common positioning methods of substation inspection robots for magnetic navigation,GPS navigation,laser navigation,and visual navigation,and selects magnetic navigation plus RFID smart tags as the positioning and navigation methods for inspection robots in this paper.Finally,the characteristics of the actual environment of the substation were extracted for simplified processing,and a two-dimensional model of the substation environment was created by combining the gridmethod and magnetic navigation positioning and navigation methods.Second,the six intelligent algorithms commonly used in global path planning,namely ant colony algorithm,genetic algorithm,Dijkstra algorithm,BFS algorithm,A~* algorithm,and B-A~* algorithm are analyzed for algorithm principles and defects,and the advantages and disadvantages of the six algorithms are compared.According to different environmental characteristics,A~* algorithm and B-A~* algorithm were selected as the basic algorithms for global path planning of the substation inspection robot in this paper.Third,the A~* algorithm and the B-A~* algorithm are improved respectively to improve the efficiency of the algorithm in planning the path and the smoothness and rationality of the generated path.The A~* algorithm has been improved in three aspects,namely,the weight ratio of the algorithm evaluation function,the algorithm path generation strategy,and the obstacle expansion strategy.For the improvement of the B-A~* algorithm,the EB-A~* algorithm is further proposed.The EB-A~* algorithm uses the improved A~*algorithm of this paper,and introduces intelligent sampling functions,elite search strategies and path smoothing strategies.Through simulation,the effectiveness of the improved A~* and EB-A~* algorithms is verified.Fourth,according to the actual work of the inspection robot in the substation,the global path planning is divided into three types: single-task point path planning,multi-task point real-time planning,and multi-task point path planning in advance.Secondly,according to the created two-dimensional model of the substation environment,multiple substation environment models are constructed in the simulation software.Finally,the modular annealing algorithm is combined with the improved A~* algorithm and the EB-A~*algorithm,and the single-task point path planning simulation,the multi-task point path real-time planning simulation,and the multi-task point path advance are performed in the constructed substation environment model.The planning simulation verified the true effectiveness of the algorithm.