Research On Reliability Analysis And Maintenance Decision Of Wind Power Equipment Based On Bayesian Random Sampling

China's wind power industry has developed rapidly,and its cumulative installed capacity and installed capacity have become the world's first.After nearly 20 years of development,the wind turbines that were initially built and operated have entered the end of life,the failure rate and failure rate have increased year by year,and the pressure for wind power operation and maintenance has also increased.At the same time,after years of accumulation,the wind turbine fault database has begun to take shape.In addition to the steady development of onshore wind power,China is still developing the offshore wind power platform.How to use the accumulated experience of land wind power operation and maintenance for offshore wind power operation and maintenance Conducting guidance has become the mainstream topic of current wind power operation and maintenance.Based on the above background,this paper uses the wind turbine operating fault data of a wind farm in Zhangjiakou area of Hebei Province,and uses the Weibull distribution model to model the reliability of the main components of the wind turbine.According to the fault distribution parameters and related reliability of the wind turbine components.Quantitative indicators to judge the failure mode of wind turbine components and optimize their maintenance decisions.At the same time,combined with Bayesian reliability method,it focuses on solving the wind power operation and maintenance decision-making problem of the newly commissioned wind farm under the condition that the main system component fault data is insufficient or no fault data.The main research work of the thesis is as follows:(1)To evaluate and master the reliability of wind turbine operation,based on the failure data of wind turbine operation,establish a two-parameter Weibull distribution model to realize operational reliability analysis,and introduce the parameter solving method and related reliability indicators of Weibull distribution.In this paper,the least squares method and the maximum likelihood estimation method are used to solve the unknown parameters of the Weibull distribution.When the number of fault data samples is small,the accuracy of the estimation results of the parameters is poor.Under these conditions,the Bayesian method is used to solve the Weibull distribution parameters of small sample fault data.The goal is to achieve accurate estimation of Weibull distribution parameters.(2)After verifying the validity of the Bayesian method under the condition of small sample fault data,a Weibull distribution parameter estimation method under the condition of fault-free sample data is proposed,and the Weibull distribution is established by using the fault data of other wind turbines under similar working conditions.The parameter is a priori estimate,and the Monte Carlo method is used to solve the a priori estimate.(3)After obtaining the Weibull distribution parameters and related reliability quantitative indicators of the main components system of the wind turbine,the Monte Carlo simulation method is used to determine the maintenance parameters of the wind turbine,and the reasonable maintenance is carried out under the premise of obtaining the Weibull distribution parameters.Decision-making to achieve safe and stable operation of wind farms and to improve economic efficiency.