COMPUTER NEURAL NETWORK SUPERVISORY PROCESS CONTROL SYSTEM AND METHOD

摘要

A neural network for adjusting a setpoint in process controlreplaces a human operator. The neural network operates in threemodes: training, operation, and retraining. In operation, the neu-ral network is trained using training input data along with inputdata. The input data is from the sensor(s) monitoring the process.The input data is used by the neural network to develop outputdata. The training input data are the setpoint adjustments madeby a human operator. The output data is compared with the train-ing input data to produce error data, which is used to adjust the.weights of the neural network so as to train it. After training hasbeen completed, the neural network enters the operation mode. Inthis mode, the present invention uses the input data to predict out-put data used to adjust the setpoint supplied to the regulatory con-troller. Thus, the operator is effectively replaced. The present in-vention in the retraining mode utilizes new training input data toretrain the neural network by adjusting the weight(s).