Research On The Reliability Centered Opportunistic Maintenance For Wind Turbines

Global warming requires increasing the share of renewable energy in the power sector,prompting the rapid development of wind power.As the cumulative installed capacity of wind power grows continually,a huge operation and maintenance market of wind farms has emerged around the world.Wind turbine is the core equipment of a wind farm.The harsh operating environment and inconvenient maintenance conditions make its fault rate and maintenance cost far higher than traditional power generation equipment.The choice of maintenance strategy is a significant factor that affects the maintenance cost of wind turbines.Carrying out optimization research of maintenance strategy,improving maintenance technology level,and reducing maintenance cost,are the key to enhance the enterprise competitiveness.The work in this paper is supported by the science and technology plan projects of Hebei "Research on the technology of wind power operation and maintenance and photovoltaic inverter in new energy power generation system"(No.15214307D).Taking the wind turbine as the research object,aimed at the key technical problems in optimizing the maintenance strategy for wind turbines,a research is systematically conducted by the theoretical analysis,technology improvement,simulation analysis and effect verification.The main work in this paper are as follows:(1)Studied and improved the fault criticality analysis method for subassemblies in wind turbines.The deficiencies of existing methods are analyzed,the factors affecting the fault criticality analysis are summarized,the concepts of fault vector and fault vector space are defined,and an improved criticality analysis method based on fault vector space is proposed.The difference among different criticality analysis methods is compared by a proposed difference analysis method based on inversion number.The results show that the improved method not only inherits the advantages of existing methods,but also has a high expansibility,which makes adding factors into the criticality analysis easily.In the improved method,quantitative and qualitative factors are effectively fused,the effects of factor value ranges are eliminated,the number of factors is not limited,the relative weights of each factor are considered,the obtained results have distinct physical significance,and key subassemblies can be effectively distinguished from non-key subassemblies.(2)Studied and improved the reliability modeling technique of the subassemblies in wind turbines under the condition of small fault samples.A hypothesis that the fault sample of subassemblies obeys Weibull distribution is made and tested.Taking the fault sample from other wind farms as prior information,classical reliability model,Bayesian general reliability model and Bayesian hierarchical reliability model are established for the subassemblies in the target wind farm,and the model parameters are solved.Taking the average width of confidence interval of reliability function as measurement indexes,the effect of fault sample size on modeling accuracy of each model is discussed.The results show that,compared with the classical reliability model,the shape parameter distribution given by the Bayesian general or hierarchical reliability model is more concentrated,and the former modeling accuracy is lower than the latter under the same sample size.Meanwhile,the confidence interval of the reliability function given by the Bayesian hierarchical reliability model is narrower under a larger sample size.Additionally,the Bayesian general reliability model should be adopted under a small prior sample size,conversely,the Bayesian hierarchical reliability model should be adopted.(3)Studied and improved the opportunistic maintenance strategy for subassemblies in wind turbines.Multiple state of wind turbines,maintenance opportunities and stages are considered.The optional maintenance modes and their service conditions are summarized.The reliability variation model,detailed maintenance cost model,continuous integrated cost optimization model,and implementing process are established.The comparison among the cases of implementing various strategies on subassemblies in wind turbines gives the following conclusions.Opportunistic maintenance strategy can significantly reduce the total maintenance count and cost for wind turbines.The more comprehensive the factors considered,the more significant the reduction will be.Additionally,the cost can be further reduced by setting a separate maintenance reliability threshold for each subassembly during optimization.(4)Studied and explored the reliability centered opportunistic maintenance strategy for subassemblies in wind turbines.The features of offshore wind turbines and their maintenance are summarized,the statistic on non-accessibility state is conducted,and the effect of non-accessibility state on maintenance activities is analyzed.The key and non-key subassemblies in offshore wind turbines are accurately distinguished by the improved criticality analysis method.The reliability model of non-key subassemblies under the condition of small fault sample size is established.Considering the key and non-key subassemblies synthetically,a reliability centered opportunistic maintenance strategy is finally formed.A case study show that wave height is the main key factor restricting the maintenance activities,while wind speed is secondary.Under the current strategy,the additional power loss generated by non-accessibility accounts for 17.22%of the total power loss.Compared with the current strategy,the reliability centered opportunistic maintenance strategy has more advantages in dealing with the problem of non-accessibility state of maintenance activities,which can reduce the maintenance cost of each subassembly,total maintenance count and cost significantly.