A NEURAL NETWORK IN AN EXPERT DIAGNOSTIC SYSTEM

An Expert System (ES) has been used to improve the fault diagnosis in a particle physics detector. A part of this expert diagnostic system is composed of a neural network, performing a classification of failures through a suitable visual pattern recognition. The classification is performed in order to recognize failures in the streamer tube system, which is a particle position detector in the experimental apparatus. An histogram of the particle position during several events is sent to a neural network. The neural network is able to give a response with regard to some typical malfunctions in the streamer tube system, like an operator who analyses the histogram at a workstation. The neural network outputs are used by the ES and integrated with other information to perform an efficient diagnosis.