Automatic orbit analysis with the use of a neuron network mechanism.

摘要

A system for diagnosing equipment malfunction uses a neural network to perform orbit analysis. An orbit plot is input via a graphic input unit or as deflection data from X-Y probes which detect deflection of a rotating shaft in a piece of equipment, such as a motor or turbine. The neural network is trained using orbit plots produced by known malfunctions or defects in the equipment. The trained neural network is able to identify similar patterns in the input orbit plots to diagnose equipment malfunctions.