摘要
针对直驱式风力发电机变频器的易故障性和故障难以识别定位的问题提出一种智能的变频器故障诊断方法。在分析直驱式风力发电机及全功率因数变换器数学模型的基础上讨论了变频器的故障机理和常见的故障分类,结合PSIM软件建立能模拟各种全额变频器故障的直驱式风力发电系统仿真模型并获取仿真结果 ,在MATLAB下设计一个基于神经网络的故障诊断系统并运用遗传算法优化神经网络初始权值、阀值以此提高变频器故障诊断系统的性能。
The direct-driven wind turbine generator frequency converter is easy to fault and the faults are difficult to be identified. For solving this problem, an intelligent fault diagnosis method is proposed herein. The fault mechanism and common fault classification are discussed by analyzing the mathematical model of direct-driven wind turbine generator and the full power factor converter. A direct-driven wind power system simulation model which can simulate various inverter faults is established by using PSIM software and the simulating results are achieved. A neural network based fault diagnosis system is also designed under the circumstance of MATLAB, in which, the original weights and thresholds of neural network are optimized by genetic algorithm to improve the performances of fault diagnosis system.
出处
《水力发电》
北大核心
2011年第12期78-80,96,共4页
Water Power
基金
国家自然科学基金资助项目(50767003
50867004)
