NEURAL NETWORKS FOR PNEUMATIC ACTUATOR FAULT DETECTION

The suitability of artiicial neural networks (ANNs) for detecting fault conditions in pneumatic control valve actuators is investigated. Speciically, the ability of a neural network to act as a predictor of correct valve behaviour is examined. Experimental results indicate that standard network architectures are unsuitable for temporal prediction of non-linear system behaviour. An original recurrent network architecture, designed speciically as a predictor and based on autoregres-sive models and functional approximation is therefore proposed. The performance of this network is evaluated using both measured data and data from simulations based on a mathematical model of the valve. Laboratory implementation of the fault detection system produced encouraging qualitative results, including high success rates for the detection of faults corresponding to valve Coulomb friction changes and input pressure oosets. 1. INTRODUCTION Pneumatically-actuated control valves occur frequently as a basic component of control systems in many processing and manufacturing plants. The wear and tear to which industrial control valve actuators are subjected leads to degeneration of performance and eventually to failure. In modern automated plants, an unrevealed actuator fault may have serious consequences. Although the detection of sudden failures is usually easily accomplished , this is seldom the case when deterioration is gradual. In fact, the use of feedback control to maintain desirable process operation may compensate for, and thus obscure, a developing fault.