LAYERED-INTEGRATED ANN DIAGNOSIS MODELS AND APPLICATIONS IN REMOTE DIAGNOSIS SYSTEMS

摘要

ANN is a parallel information disposal network and is widely used in remote diagnosis systems. But the single-structure ANN is too massive. So the layered-structure ANN model is applied. However, it is inconvenient to implement Internet-based remote diagnosis systems. In order to solve it effectively, layered-Integrated ANN diagnosis models are presented in the paper. The construction of the models is composed of subnetwork partition, confirming network structure, building expansion sample set, study and training. The performance of the models is analyzed from three aspects. It proves appropriate to apply the models to remote diagnosis systems through the tests and applications.