Research On Power Supply Reliability Of Distribution Network Considering Various Factors

With the optimization of main network structure and the improvement of power supply construction in China,the weak links in distribution network have become the bottleneck of power system development in China,and the stable supply of power has become an increasingly concerned issue.Distribution system is at the end of power system,which is an important link between the whole power system and users.Research on the reliability of distribution system is an important measure to ensure the power supply quality of power system and improve the modernization level of power industry.It plays an important guiding role in improving economic and social benefits as well as urban power network construction and transformation.Firstly,the paper expounds the basic concept of power system reliability,analyzes the characteristics of power system reliability research,introduces the evaluation indicators that affect the reliability of distribution network power from two aspects of load and system,and introduces the current reliability.The main methods of concentration assessment,including simulation and analytical methods,provide the basis for the subsequent reliability assessment analysis and calculation of the distribution system.Then,starts from the grid structure,technical equipment level,equipment quality,causes of failure,and operation and maintenance level of distribution network,the paper obtains the important impact indicators of power supply reliability based on the sample data of Southern Power Grid Power Supply Bureau,which lays the foundation for the research of mathematical analysis and evaluation model of this paper.The mathematical statistics method based on correlation analysis processes the data to measure the influence of different indicators on the reliability of power supply,and extracts the main factors affecting the reliability of power supply.Based on the grey correlation analysis,the influencing factors related to the reliability evaluation index of power supply are found.Finally,considering various factors affecting power supply reliability,this paper establishes two kinds of power supply reliability evaluation models for multi-input variables: BP neural network model and DE-SVR model.The sample data of the Southern Power Grid Power Supply Bureau is used to analyze the effectiveness of the two evaluation models.The research results of the paper provide a scientific and effective reference for the power supply enterprise to develop a reliability improvement strategy.