State-Space GMDH Neural Networks for Actuator Robust Fault Diagnosis

Advances in Electrical and Computer Engineering

关键词: fault diagnosis;    robustness;    actuators;    neural networks;    system identification;   

DOI  :  10.4316/AECE.2012.03010

来源: DOAJ

【 摘 要 】

Most fault diagnosis methods focus on the fault detection of the system or sensors and do not take into accountthe problem of the fault detection and isolation of the actuators, which are an important part of the contemporaryindustrial systems. To solve such a problem, the system outputs and inputs estimator based on a dynamic GroupMethod of Data Handling neural network in the state-space representation is proposed. In particular, the methodologyof the adaptive thresholds calculation for system inputs and outputs is presented. The approach is based on the applicationof the Unscented Kalman Filter and Unknown Input Filter is presented. This result enables performing robust fault detectionand isolation of the actuators. The final part of the paper presents an application study, which confirms the effectivenessof the proposed approach.

【 授权许可】

Unknown