摘要
针对电磁调速系统的实际特点,提出了一种具有建模不确定性的非线性系统在线故障检测方法.假定该系统仅是输入、输出可测量的,并把故障建模为状态变量和输入变量的函数.文中用一种基于RBF神经网络的在线非线性估计器来跟踪调速系统中出现的故障,从理论上证明了该方法对有建模不确定性的非线性系统的故障检测具备良好的鲁棒性.仿真实例说明了该故障检测方法的有效性.
In this paper, a kind of on-line fault detection scheme is presented for nonlinear systems with modeling uncertainty. It is applied to speed control system of induction motor. The faults are assumed to be a function of the state and inputs. Only the inputs and outputs of the system can be measured. An on-line approximator based on RBF neural network is used to track the faults of the system. The approximator is proved theoretically to work with good robustness for the fault detection of the nonlinear system with modeling uncertainty. A simulated example on the speed control system demonstrates the efficiency of the proposed fault detection scheme.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2002年第3期317-321,共5页
Pattern Recognition and Artificial Intelligence
基金
江苏省应用科学基金(B199021)
