Research On Preventive Maintenance Strategy Of Wind Turbines Based On Condition Monitoring

Wind is a broad prospect and great potential for renewable energy. Nowadays, due to the global ecological environment deterioration and oil crisis, wind power as one of the most importance alternative energy source, is developing rapidly. Wind power generation is the most important forms of wind power applications, making wind power develop the real alternative energy and playing the scale benefit. Wind power plants are generally established in harsh outdoor environments, and remoteness, so the wind turbines' maintenance is very difficult, resulting in expensive maintenance costs. Therefore, how to effectively carry out maintenance on wind turbines is an importance problem of wind power applications.The traditional preventive maintenance is timed based, and inspection interval do not change any more once identified. It does not reflect the true state of the wind turbines, and often lead to over-maintenance or under repair. Condition based preventive maintenance is to determine whether the wind turbines needs to be repaired according to its status. It can reflect the true state of the wind turbines and there is no over-maintenance or under repair. But because this activity is random, it will make wind turbines outage during maintenance, making wind farms disrupting production. And it will cause interference for the safe operation of wind farms. So it needs to control the checking activities of preventive maintenance, so that normal production activities will not be disrupted due to the preventive maintenance.The dissertation analyzes the operation principle of wind turbines, wind turbine preventive maintenance life cycle model is established. Based on this, the state transition diagram is drawn and state transition equation is written. Reliablity, stability and asymptotic stability of system, parameters' impact on system's steady-state value are analyzed.Two commonly used wind turbines preventive maintenance strategies are analyzed deeply. That is time based maintance(TBM) and condition based maintance(CBM). For TBM, based on the shortest gross downtime time and the largest gross profit per unit time, the optimal detecting times per unit time are desigend. CBM is based on the monitored state of the wind turbines to determine inspection intervals.It real time good but strong randomness, and may lead to undesirable interruption. Therefore, it is necessary to control such maintenance activities to ensure normal production activities from interference. Based on CBM, combined linear control theory, control law of checking rate is designed, making it possible to follow the desired trajectory, to ensure stability and security of the production process. In addition, based on the frequency and duration method, the gross expected cost of TBM model, CBM model and redundant part maintenance model are analyzed.Taking the first wind power plants of Jiuquan Qiaoxi for example, using the random numbers which subjects to uniformly distributed to simulate CBM preventive maintenance activities and compared with the proposed control strategy in the paper. The results show that the control strategy proposed in this paper is more ideal, and have the shortest payback period.