A Statistical Data-Driven Method for Remaining Life Prediction of Operating Accessories of Universal Circuit Breakers

Abstract

The invention relates to a method for predicting the remaining service life of a universal circuit breaker operating accessory based on statistical data driving. The prediction method comprises the steps that distribution test is carried out on the relative increment of the operating time of the operating accessory from both qualitative and quantitative aspects to verify whether the relative increment conforms to normal distribution; on the basis, the maximum likelihood estimation method is used to estimate parameters in a degradation model; based on the concept of first arrival time, a remaining service life predicting model is established; and an analytical formula of the probability density function of the remaining service life is derived to predict the remaining service life of the circuit breaker operating accessory. According to the invention, the method considers the influence of uncertainties in the operation process of a device on the service life and can acquire the remaining service life distribution required for health management; the method does not use single point estimation; the method is convenient for quantifying the uncertainty of the prediction result of the remaining service life; and the efficiency of the operation and maintenance of a universal circuit breaker can be effectively improved.

Description

Technical field [0001] The technical scheme of the present invention relates to the technical field of prediction of remaining life of operating accessories of circuit breakers, and specifically is a method for predicting remaining life of operating accessories of universal circuit breakers based on statistical data. Background technique [0002] The low-voltage distribution network is the terminal power grid of the transmission network, directly connected to the majority of users, and is a very important link between the transmission grid and the users. It is the key for the power system to distribute and regulate electric energy to the majority of power users. The operation and maintenance level and health of low-voltage distribution networks often directly determine the power quality level of power users. As one of the most important electrical equipment in the low-voltage power distribution network, the universal circuit breaker is directly related to the stable and safe oper...

AI Technical Summary

Problems solved by technology

In addition, most of the existing remaining life prediction methods are artificial intelligence methods, which usually require a large number of training samples, and the prediction results are difficult to obtain the probability distribution function that reflects the random uncertainty characteristics of the remaining life, such as Shen Zhongjie et al. (Shen Zhongjie, Chen Xuefeng ,He Zhengjia,Sun Chuang,Zhang Xiaoli,Liu Zhiwen.Residual life prediction of rolling bearings based on relative features and multivariate support vector machines[J].Chinese Journal of Mechanical Engineering,2013,49(2):183-189.) Using unaffected bearings The relative root mean square value affected by individual differences evaluates the law of bearing performance decline, uses correlation analysis to select sensitive features as input, and constructs an MSVM model that takes into account the dual advantages of multivariate regression and small sample prediction to predict the remaining life of bearings

At present, most of the remaining prediction methods only realize the single-point estimation of the remaining life or the prediction of the equipment performance degradation curve, and the uncertainty in the equipment degradation process is not reflected in the prediction results, which cannot provide an effective premise basis for determining health management measures

Images