Research On Condition Monitoring And Assessment Method Of 10kV Cable Terminal Joint

In the development process of the power industry,power cables have been widely used in the distribution network system.The cable terminal joint is at the end of the cable line and plays the role of connecting the cable line and other power equipment.Because of its manufacturing process and specific structure,it is a location where insulation failures frequently occur.Power cables are responsible for the transmission and distribution of electric energy.In order to ensure the safe and reliable operation of the cable system,it is of great significance to study the condition monitoring and assessment methods of the cable terminal joints.On the one hand,in order to improve the condition monitoring level of the partial discharge and realize the effective early warning of the partial discharge of the cable terminal joint,an early warning method based on the long short-term memory(LSTM)neural network and the Mann-Kendall test method is proposed.First,the particle swarm optimization(PSO)is used to optimize the hyperparameters of the LSTM neural network,and the LSTM prediction model is constructed based on the optimized hyperparameters to predict the characteristi c parameters of partial discharge.Then the Mann-Kendall test method is used to quantitatively analyze the characteristic parameters of partial discharge to determine its trend parameters.Finally,an early warning model that comprehensively considers the amplitude and trend parameters of the partial discharge is constructed to realize active early warning.According to the early warning signal of partial discharge,the condition assessment work can be laid out in time.On the other hand,in order to evaluate the health status of cable terminal joints,a condition assessment method based on fuzzy synthetic evaluation is proposed.First,based on the existing detection and monitoring technology,from the three aspects of test data,online monitoring data and historical data,10 state characteristic quantities are selected as evaluation indicators to constructe the evaluation indicator system of the operating state of the cable terminal joint,and the deterioration is used to standardize the indicator.Then the evaluation indicator weights are obtained by combining the analytic hierarchy process and the entropy method from the subjective and objective aspects,and the fuzzy relation matrix of the evaluation indicator of each layer is determined based on the selected triangle and semi-trapezoidal membership functions.Finally,the fuzzy relation matrix of each level of the cable terminal joint and the corresponding indicator weight are compounded,and the condition assessment result of the cable terminal joint is determined based on the maximum degree of membership.This paper takes a cable terminal joint of a substation switchgear in Cangzhou City,Hebei Province as an example.Based on its characteristic parameter data,the validity of the partial discharge early warning model and condition assessment method established in this paper is verified.The discovery of potential faults p rovides a powerful method and basis for discovering potential faults and performing condition maintenance.