Distribution network bears the important responsibility of distributing electric energy to users, and its reliability level determines whether the power system can meet the demand of users. Smart grid is different from traditional distribution network in safety, reliability, quality, economy and environmental protection. Smart grid condition monitoring is not only the monitoring management, condition maintenance and life management of power grid equipment, but also further developed into risk-based maintenance, and continuously extended to many fields such as safe operation, optimal dispatch, economic operation and service. The development of smart grid is inseparable from the understanding of existing equipment and network health, and its reliability evaluation plays a very important role in power system reliability. At present, the research on reliability evaluation of distribution network can be divided into two categories: 1) reliability evaluation of distribution equipment; 2) Reliability evaluation of distribution network.
(1)Reliability evaluation of power equipment is the basis of reliability evaluation of distribution network. Western countries first put forward the method of fault prediction and health management of power equipment, and health status assessment is one of the key technologies, that is, monitoring the safe operation of power equipment through the changes of some important characteristic parameters, relying on real data and reliable methods to judge the health status of the equipment itself, and even carrying out scientific fault diagnosis and isolation, so as to take necessary measures to alleviate the performance degradation of the equipment. Health index theory was originally put forward by EA Company in Britain, which is used to measure and characterize a quantitative value of equipment health status. Most studies on the field of health management of power and electrical equipment have this index, and the range of values is different. Based on CBRM (Condition-based Risk Maintenance), the running status of distribution equipment is evaluated . The system is based on the power operation databases of European and American countries, and has been successfully applied to many power companies in Britain, as well as countries such as the United States, Canada and Australia. Literature  a novel statistical kernel-based GRNN approach has been presented to calculate the health index of power transformers. Four conditional scores based on six important tests(TDCG, DBV, acidity, water, furan and dissipation factor) have been used to train the proposed model.Literature  presents a bayesian information fusion method for determining power probabilistic health indicators.This method synthesizes various data obtained from transformer measurement and maintenance and uses these data to establish an inference model (BBN) by using Bayesian belief network.
(2)Reliability evaluation of distribution network. China Electric Power Research Institute has carried out the special project of State Grid Corporation "Research and Demonstration Application of Modern Distribution Network Health Index Theory and Engineering Realization System". According to the characteristics of domestic distribution network, this project discusses the research of distribution network health index from four aspects: basic theory, application method, data platform and demonstration application, and puts forward the definition, classification standard, calculation method and model of distribution network health index, which is verified by actual cases. Professor Ma Zhao  summarizes the concept and research status of equipment health index, and puts forward the definition, classification, calculation method and model of distribution network health index. Secondly, taking the 110kV substation in a pilot area of Nanjing and Beijing as the network research object, the application scenario of distribution network health index is put forward. Finally, the development trend of distribution asset management in the future and the significance of deepening research on distribution network health index theory and engineering realization system for distribution asset management in the future are summarized.Literature  characterizes the main key features affecting distribution equipment and distribution network by consulting data, analyzing actual situation and calculating data. Document  introduces the key features that have an impact on the line operation, and establishes a line risk assessment model based on PCA dimension reduction for line risk assessment. In the literature , the complex distribution network was diagnosed based on analytic hierarchy process (AHP), and the 10KV power network was taken as an example for verification. Literature  studies and analyzes the health and importance of distribution network operation, and gives a specific formula for calculating distribution network health index. Literature  will summarize the evaluation process of distribution network health index, and put forward the evaluation method based on group equipment and network. Literature  discusses the key characteristics of different distribution equipment in 10kv distribution network, and establishes a health model by using Logistic regression method for verification through 30 transformers as an example.
Based on the research and application of the existing distribution network health index theory, in order to evaluate the reliability of distribution system reasonably and effectively, this paper puts forward the generation technology of distribution network health image which comprehensively considers equipment health index and operation and maintenance strategy. The main work is as follows:
(1) Analyze the distribution network health index theory and classification standard, and analyze the optimal operation and maintenance strategy of the power network.
(2) According to Q/GDW645—2011 Guidelines for Condition Evaluation of Distribution Network Equipment, the health level of distribution network is divided from [0,5] into health (4,5), sub-health (3,4), general defects (2,3), serious defects (1,2) and critical defects (0,1] .
(3) The key features of distribution network are classified from three aspects: safety, economy and reliability. Power factor, line voltage deviation and line loss rate are selected as three key feature quantities, and the key feature quantities are weighted by entropy method to calculate their respective weights.
(4) Taking the data of 7 lines in a 110kv substation in the oilfield as samples for example analysis, the calculation of health index, the classification of health degree and the generation of health degree images are completed.
(5) The optimal operation and maintenance strategy of distribution network based on health index theory is put forward, which provides a reliable basis for the planning, construction, operation and maintenance of distribution network.