Analysis Of Optimal Site Selection For Distributed Photovoltaic Access To Rural Distribution Network

The rural distribution system has serious network losses and voltage reduction issues.Especially,the large number of distributed photovoltaic applications in the rural distribution system will have a certain impact on system node voltage,power transmission and safety.The reasonable distribution and constant capacity of distributed photovoltaic can effectively reduce the network loss and improve the voltage quality.Solving this problem is a multi-objective optimization problem,and there are difficulties and challenges in solving a reasonable optimal configuration scheme.To solve this problem,under the condition that the total amount of distributed photovoltaic installation and its installation locations and corresponding installation scale are not determined,this paper constructs a mathematical optimization model considering the environmental cost of low-carbon environmental protection cost,investment cost,total power purchase cost,network loss cost and other factors,and solves the weight coefficient of each sub factor through the improved AHP,the traditional hierarchical division The analysis method has the disadvantage of subjective influence when writing the contrast matrix,Therefore,the Delphi method and fuzzy trigonometric function are introduced to improve it.The choice of weight is more objective and reasonable.The particle swarm optimization algorithm with non-linearly changing weights was used to verify that the nonlinearly changing inertial weights helped to improve the convergence performance of the particle swarm optimization algorithm.and can better coordinate the global optimization and local optimization.In order to retain the independence of each objective function and analyze its internal relations to the maximum extent,a multi-objective function of optimal configuration is established,which takes into account the total cost of construction and operation of distributed photovoltaic,network loss index and voltage deviation index.An adaptive multi-objective particle swarm optimization algorithm is proposed,which is suitable for solving multi-objective mathematical model.The algorithm introduces the fast sorting non dominated method,crowding degree and differential variation,crossover,inertia weight non-linear change operation to improve,increase the difference of each particle in the population,prevent particle aggregation and other problems,improve the ability of the algorithm to get rid of falling into local optimum.At the same time,it also The convergence speed of the algorithm is improved.Then,using the proposed intelligent algorithm to find the solution set,because there are many effective solutions,i.e.manyplanning schemes,facing the situation of final decision-making difficulties,an optimal decision-making based on the shortest Euclidean distance method of normalized space is proposed to make up for the shortcomings of the algorithm,and the optimal solution selection is realized through the decision-making measures,so as to obtain the overall optimal application implementation scheme.Taking a rural distribution system in Chengde,Hebei Province as an example,the effectiveness of the proposed planning model and algorithm improvement is tested.The results show that the feasibility of the mathematical planning model of location and capacity determination adopted in this paper.The reasonable location and capacity of distributed photovoltaic can significantly reduce the investment in distribution system lines and network loss costs,improve the voltage quality,and bring huge social benefits and economic benefits Economic benefits.At the same time,it shows that the improved algorithm has good convergence and accuracy.