Research On Prognostics And Health Management Technology Of Wind Turbine Based On SCADA

With increasing emphasis on environmental protection, domestic wind power industry has shown a trend of rapid development. Domestic wind power industry started late. Although the industrial development is rapid, it is very extensive and many aspects of power equipment manufacturing and wind farm operators are not mature. So that causes the accidents of wind power equipment and electric field occur frequently. Now that has become an important reason for restricting the healthy and rapid development of China’s wind power industry. Therefore, the study of PHM of wind power equipment has important practical significance and application value.In this paper, a domestic wind farm for the object, a PHM technology research framework is proposed for the domestic wind turbine health management based on the full investigation of typical wind farm health management. First of all, according to the FMECA, the FMECA analysis of wind turbine systems is completed and the main components of the wind turbine, the failure mode, the main reason for failure and severe degree are given. Secondly, a preprocessing method of SCADA data of the wind farm is proposed. After extracting the state characteristics of important parts, a weighted D_S evidence theory fusion technology is used in the paper. Then, a variable weight fuzzy comprehensive evaluation model is established based on the fuzzy theory. After the actual data validation, the results are close to the actual situation of the wind turbine and that verify the validity of the model. According to the different device characteristics of wind turbine, this paper uses appropriate diagnostic methods to study fault diagnosis. That provides support for fault prediction. Finally, wind turbine fault prediction is carried out. The same dimension gray dynamic forecasting model is established based on gray theory. Simulation and practical application show that the model improves the prediction accuracy of wind turbine failure prediction. The state evaluation, fault diagnosis and state prediction of wind turbine are verified by the sample of real-time monitoring data that were collected in a domestic wind farm. The fault prediction and health management prototype system of wind turbine is designed and developed by J2 EE architecture. This system integrates state evaluation, fault diagnosis and state prediction, and it realizes the management of wind turbine.Experimental results show that the data fusion method, the variable weight fuzzy comprehensive evaluation model and the dynamic dimension gray forecast model that are improved in this paper are effective and feasible. PHM prototype system achieves the state evaluation, fault diagnosis and state prediction of wind turbine very well. The results of this paper improve the reliability of wind turbine operation, reduce the failure rate and improve the operational efficiency of wind farm. These results have very good application value.