Research On Rural Power Grid Transformation Project Based On Photovoltaic Grid Connection

Due to historical reasons,the construction of rural power grids is relatively weak.In the past,the rural power network transformation project aimed to meet the demand of load power supply,and lack of overall planning,and distributed photovoltaic grid-connected further brings new challenges to the rural power network transformation project.It is of great significance to study the transformation of rural power grid with photovoltaic and to solve the new problems of grid-connected Rural Power Grid,so as to improve the transformation benefit and ensure the safe and reliable operation of Rural Power Grid.According to the current situation of rural power grid with photovoltaic,this paper puts forward the idea of rural power grid optimizing retrofit.The idea is based on load forecasting,probabilistic power flow calculation and improved genetic algorithm.This paper mainly completes the following research work:(1)Aiming at the problem that the uneven development level of rural areas leads to the inaccurate load forecasting of rural power grids,this paper proposes a solution.Based on the regions where the rural power supply nodes are located,and forecasts are dependent on the different factors that influence each development model.The K-nearest Neighbor(KNN)algorithm is used to predict the rural development model,and then a medium and long-term rural load prediction model based on Deep Belief Network(DBN)is established according to the development model.(2)The forward and backward substitution method is improved so that it can be applied to the calculation of power flow in rural grids with photovoltaics.In order to overcome the problem of the accuracy of power flow calculation caused by the intermittency of photovoltaic output and the change of load data,this paper uses the Monte Carlo method to simulate the photovoltaic time series output characteristics of the retrofit level year and each power supply based on the probability and statistical characteristics of historical data.The time series load of the node is used as the basic data for probabilistic power flow calculation.Then,according to the calculation results of the power flow,the active power loss of the rural grid is calculated,and the line loss is used as one of the basis for judging the merits of the reconstruction plan.(3)Considering the reality of rural power grids and the operation and management requirements of power supply companies,an optimization model for rural grid transformation with photovoltaics is established,aiming at the comprehensive minimum cost of transformation of rural power grid wires and distribution transformers,as well as the cost of active power loss.An improved genetic algorithm is used to solve the model,which is mainly reflected in: the matrix coding that includes all the information of rural grid transformation;the genetic operation that reflects the adjustment of rural grid structure,the replacement of conductors and distribution transformers,the optimization of new photovoltaic grid connection points and installation capacity,etc;the further improvement of the performance of the genetic algorithm by reasonably setting the genetic crossover and mutation probability.A case study is carried out on the rural power grid reconstruction project in a certain area of ZB City,which proves that the research results in this paper can provide a certain reference value for the optimization and reconstruction of rural power grids with photovoltaics.