Research On Real-time Reliability State Evaluation And Prediction Of Wind Turbine

In recent years, the wind turbine industry has a fast development, it is the indispensable strength to solve the world's energy problems,and it is on a large capacity and high quality direction. Along with the increase of the unit capacity, failures on the wind turbine's gear box, spindle, yaw system and generator system have appeared, even tower collapsed accident. These failures of the unit bring a large impact on the normal operation, power generation and quality,and will also bring a huge economic losses.Previous regular maintenance system has failed to meet its request of safe,reliable and economic operation,so the state maintenance is the future trend of development. Reliability evaluation technology is the important foundation for state maintenance, only to grasp the equipments real-time detailed state, can we make the fault warning, fault diagnosis and life prediction and so on according to the function of equipment.The traditional reliability analysis technology of the unit cannot analysis the real-time reliability quantitatively, and cannot satisfy the requirement of state maintenance. This topic research on reliability evaluation and real-time prediction method of the wind unit equipments,then set up the real-time reliability analysis model, to make a warning for the equipments whose real-time reliability is downso as to prevent the happening of the fault.The second chapter basically measures the real-time running deviation degree with the real time monitoring parameters, establishes the analysis model, and use the principal component analysis method to eliminate the relationship between the parameters of the same equipment, which will make the result more accurate. The reliability trend prediction model is also established.Through the analysis and comparison of various methods, the second exponential smoothing method of predicting is more suitable to predict the short-term change of operation state, then support the accuracy of the real-time reliability evaluation. And then, according to the characteristics of the wind turbine, establishe the wind generator model, and use the hierarchy analysis on the wind generator's system division, and at the same time use the grey hierarchy process to analyse and identify the importance of each of the equipment.Combined with the method proposed in the second chapter, analyse and predicte the real-time reliability of the wind generator step by step to make a more accurate evaluation of the wind generator's real-time reliability state, and predicte its short-term reliability trend,thus the timely discovery of the state degeneration, can take measures to reduce or avoid major accident in time, and make a big contribution to the safe, efficient, reliable operation. Finally the research contents of the paper is summarized, and puts forward the direction of future research proposals.