Research On Planning Of Cleaning Robot System For Track Connected Photovoltaic Power Station

The surface area ash of power generation modules in photovoltaic power station will reduce the efficiency of photovoltaic power generation.Regular cleaning is very important for the operation and maintenance of photovoltaic power plant.At present,there is little research on autonomous space and time planning of photovoltaic power station operation and maintenance system.Realizing the intelligent operation and maintenance of the number,path and cleaning time of photovoltaic panel cleaning robots can reduce the operation and maintenance cost and improve the work efficiency.This paper plans and studies a set of track connected photovoltaic power station cleaning robot system.The main research contents include:1 The cleaning route planning of track connected photovoltaic cleaning robot based on NSGA-Ⅱ algorithm is established.Based on the basic pattern of photovoltaic power station,a track connected photovoltaic cleaning system is designed.Taking the total cost and total time of multiple robots as the planning goal,and considering the number of robots and cleaning path,the Pareto optimal solution set is solved by NSGA-Ⅱ algorithm.The feasibility of this method is proved by simulation examples.2 The cleaning time planning of track connected photovoltaic cleaning robot based on meteorological factors is established.The sensor collects and processes the environmental meteorological data in different periods and the dust accumulation data in the same period,establishes the GWO-SVR prediction model to predict the dust accumulation,uses the GWO to optimize the parameters of the model,compares the SVR prediction model,and finally evaluates the prediction results of the model.So as to set the standard value of dust accumulation,specify that if it exceeds the standard value,carry out cleaning work and realize the cleaning time planning.3 The planning and management software system of photovoltaic power plant cleaning robot based on Visual Studio is established.Complete the analysis,design and construction of the software platform.The software system is based on the cleaning mechanism of track connected photovoltaic power station to realize the functions of real-time weather station data display,cleaning time planning,cleaning path planning,historical data query and so on.Experiments show that the planning method designed in this paper can provide a variety of optimization schemes for the number of robots and path decision-making,so as to meet the optimization requirements of different objective functions.The fitting degree of the prediction model of photovoltaic panel dust deposition is 95%.