欢迎来到中国电源学会电子资源平台
会员 Online Fault Diagnosis for Single-Phase PWM Rectifier Using Data-Driven Method
  • 5
  • 0
  • 0
  • 0
  • 2022/01/01
  • 作者:
    Bin GOU  , Xiaoyun FENG  , Kun ZHANG  
  • 页数:
    9
  • 页码:
    49 - 57
  • 资源:
  • 文件大小:
    3.61M
摘要
In power electronic traction transformer, the failure of the single-phase PWM rectifier will lead to performance degradation of the system. Thus, a feasible data-driven method is proposed to realize online fault diagnosis of single-phase PWM rectifier in this paper. The principle of the data-driven method is to construct a signal predictor based on historic database utilizing the nonlinear autoregressive exogenous(NARX) model with a randomized learning algorithm named extreme learning machine (ELM). Besides, the ensemble method is employed to improve the prediction accuracy and robustness against load fluctuation. In online diagnosis, the predictor and sensor operate simultaneously and their residual is generated. Afterward, the fault detection is conducted by comparing the residual with fault threshold and the fault classification is completed based on system fault symptoms and fault residuals analysis. Several hardware-in-loop tests are implemented to verify the applicability of the proposed diagnosis method. Test results show that this data-driven method is effective to perform the online fault diagnosis with fast fault detection speed and high classification accuracy, and robust against load fluctuation.
  • 若对本资源有异议或需修改,请通过“提交意见”功能联系我们,平台将及时处理!
来源
关键词
相关推荐
可试看前3页,请 登录 后进行更多操作
试看已结束,会员免费看完整版,请 登录会员账户 或申请成为中国电源学会会员.
关闭
温馨提示
确认退出登录吗?
温馨提示
温馨提示
温馨提示
确定点赞该资源吗?
温馨提示
确定取消该资源点赞吗?
温馨提示
确定收藏该资源吗?
温馨提示
确定取消该资源收藏吗?
温馨提示
确定加入购物车吗?
温馨提示
确定加入购物车吗?
温馨提示
确定移出购物车吗?