A method for predicting equipment failure rate by combining Weibull distribution with ARMA is disclosed

Abstract

The invention belongs to the technical field of equipment failure rate prediction of electric power system, in particular to a method for predicting equipment failure rate by combining Weibull distribution and ARMA, wherein the obtained equipment failure rate data are arranged according to time series order to obtain time series data of equipment failure rate. According to the development trend ofequipment failure rate, the failure period of the equipment is judged. Based on the judgement results of failure period, the Weibull distribution function of different stages is selected to model thefailure rate data, and the trend item of failure rate time series data is determined. The time series data of failure rate stochastic term are obtained by the difference between the time series dataof failure rate and its trend term, and the time series ARMA model is established for the stochastic term based on the time series analysis method. The trend item prediction result and the random itemprediction result are added together to obtain the prediction result of equipment failure rate. This method combines the Weibull distribution function with the time series analysis model, and realizes the accurate prediction of equipment failure rate.

Description

technical field [0001] The invention belongs to the technical field of equipment failure rate prediction in electric power systems, and in particular relates to an equipment failure rate prediction method combining Weibull distribution and ARMA. Background technique [0002] The failure rate of electrical equipment is a set of random sequences formed over time, which is generally obtained by recording the actual operating conditions of the equipment for a long time and performing statistical analysis on the obtained data. On the one hand, due to the complexity of the geographical distribution of power equipment and the random diversity of operating states, the failure rate of power equipment has the characteristics of time-varying and random, and on the other hand, it has its own changing rules. Equipment failure rate is the basis of equipment reliability evaluation, so research on equipment failure rate prediction methods has an important impact on improving the reliability...

AI Technical Summary

Problems solved by technology

[0004] (1) There are many factors affecting the equipment failure rate and the influencing principles are complex, so it is difficult to establish a prediction model for the equipment failure rate and all influencing factors. The impact of equipment failure rate, which makes prediction more difficult

[0005] (2) The collection of equipment failure rate data is not strictly at equal time intervals, so it is used as data at equal time intervals for modeling analysis, ignoring the impact of sampling intervals on equipment failure rates, resulting in poor prediction accuracy

[0006] (3) Due to the limitation of statistical data on the reliability of power equipment, the number of available statistical samples is small, and it is difficult to find a suitable fitting function for modeling and fitting, resulting in poor fitting accuracy of equipment failure rate data

Images