论文部分内容阅读
提出一种基于残差评价策略的故障检测方法,在仅有系统残差数据的基础上,不依赖于系统解析模型进行故障检测,为基于数据驱动方法进行复杂系统的故障检测提供了新的理论依据.为了提高系统故障检测的准确性,在分析故障检测漏报率和误报率的基础上,对故障检测阈值进行优化设计,并通过对车辆电源系统可能发生的励磁系统故障进行故障检测,验证了所提出方法的准确性和有效性.
A fault detection method based on residual evaluation strategy is proposed. Based on the system residual data only, the fault detection method is not dependent on the system analysis model, and provides a new theory for fault detection of complex systems based on data-driven method Based on the analysis of fault detection false negative rate and false alarm rate, the fault detection threshold is optimized and fault detection of the excitation system that may occur in the vehicle power system is carried out. The accuracy and validity of the proposed method are verified.