Operation Condition Evaluation And Maintenance Decision Of Offshore Wind Turbine Group

Offshore wind energy resources are abundant and have many advantages over onshore wind power,offshore wind power development has become the focus of global renewable energy development.The operating environment and accessibility of offshore wind turbines make the operation and maintenance cost of offshore wind turbines higher,and the operation and maintenance of offshore wind turbines has become the urgent problem to be solved in the large-scale construction of offshore wind power.Domestic and foreign scholars and research institutions have done a lot of work in the offshore wind turbine monitoring and diagnosis,data statistics and analysis,operation&maintenance management optimization.So the systematically study of the operation condition evaluation and maintenance decision method of offshore wind turbine group has theoretical significance and practical value.There are four key problems for theoretical research and analysis:turbine condition evaluation and fault prognosis,unit maintenance mode selection and scheduled preventive maintenance interval decision,optimization of operation and maintenance resource allocation for offshore wind farm,maintenance operation scheduling optimization.The main work of this dissertation is summarized as follows.In research of wind turbine condition evaluation and fault prognosis,proposed a real-time multi-level fuzzy synthetic condition evaluation model for offshore wind turbines.Based on the online operation data monitored by Supervisory Control and Data Acquisition(SCADA)system of wind turbine,the model integrates the analytic hierarchy process and correlation coefficient method to calculate the combination weights of indicators at all levels,and adjusts the weights according to the deterioration grade,which improved the accuracy of turbine status evaluation.Structure and evaluation index system of typical offshore double-fed asynchronous wind turbines are analyzed,the application examples indicated the evaluation results can reflect the actual operation condition of turbines.A wind turbine pitch system on-line degradation evaluation model was established based on the analysis of typical variable-speed pitch-to-feather control principle and monitoring parameters for pitch system,which was taken as a nonlinear Multi Input Multi Output(MIMO)system.The difference between the on-line measurement and the calculated value from the MIMO regression model applying Least Square Support Vector Machines(LSSVM)method was defined as the Observed Vector of the system.The Gaussian Mixture Model(GMM)was applied to fitting the distribution of the multi dimension Observed Vectors.The target detection method in image processing named"background subtraction" was applied to detect the degradation status of the system.The Degradation Index was calculated using the SC AD A data of a wind turbine damaged its pitch bearing retainer and rolling body,which illustrated the feasibility of the provided method.In research of unit maintenance mode selection and scheduled preventive maintenance interval decision,the logic decision method of Reliability Centered Maintenance method was applied to select the appropriate maintenance method ofdifferent wind turbine and its subsystem.Based on the analysis of the disadvantage of traditional time based maintenance applied to offshore wind turbine,an estimation method of maintenance interval was also proposed.In research of optimization of operation and maintenance resource allocation for offshore wind farm,the offshore wind turbine group operation and maintenance process is divided into several independent sub modules,models for each sub module are established to reflect the internal operation mechanism of each sub module.The sub modules are integrated to form a complete operation and maintenance process by using the time series method.The simulation analysis using random sampling algorithm was conducted to evaluate the total maintenance cost of turbine group.The optimization proposal of operation and maintenance resource(such as ship model,speed,number,staff population)allocation for built wind farm was achieved through sensitivity analysis.Finally,the application in the virtual offshore wind farm project verified the feasibility and accuracy of the model.In the aspect of maintenance operation scheduling optimization,on purpose of enhancing the maintenance efficiency and reducing the generation cost of offshore wind turbine group,an on-line multi-objective decision model for the daily maintenance scheduling of offshore wind turbine group was proposed considering generation prediction and site weather forecast,which minimizes the total maintenance cost and the workload balance with the constraints of budgets and human resources.The Kruskal spatial analysis algorithm of Geographic Information System(GIS)was applied to find the minimum spanning tree of return path of each maintenance operations combination.The Binary Particle Swarm Optimization method was employed to solve the model,and obtains all non-dominated solution set of maintenance turbines combination and their exact maintenance time.The simulation result shows the feasibility of the proposed model,and the effectiveness of the algorithm is verified.In summary,this dissertation takes full advantage of the data from the SCADA system installed by the offshore wind turbines,focused on the unique operation and maintenance features of offshore wind turbine,four critical issues are further researched,which enhances the on-site operation and maintenance management ability to reduce the total cost of operation and maintenance of wind turbines.