Neural observer for the hot isostatic pressing nonlinear system

The equations describing the hot isostatic pressing (HIP´ing) system dynamics are nonlinear, complicating any controller design for the system. Further complications arise because one of the system´s states, grain size, cannot be measured directly as a system output. The goal is to reconstruct or estimate the grain size. It is first necessary to show that it is possible to reconstruct grain size from direct measurements of the system´s other parameters. Then, a working methodology must be found to accomplish the reconstruction task. Recent results show that the system is observable, and therefore grain size may be estimated during the HIP´ing process without directly measuring it as an output of the system. To accomplish the reconstruction task, the use of a backpropagation trained neural observer was investigated. Results of simulations indicate that use of a backpropagation trained neural observer yields an accurate estimate of grain size for use in the feedback control loop