An on-line training neural network for pro-cess control system and method trains by retrieving training sets from the stream of process data. The neural network detects the availability of new train-ing data, and constructs a training set by retrieving the corresponding input data. The neural network is trained using the training set. Over time, many training sets are presented to the neural network.
When multiple presentations are needed to effec-tively train, a buffer of training sets is filled and up-date as new training data becomes available. The size of the buffer is selected in accordance with the training needs of the neural network. Once the buf-fer is full, a new training set bumps the oldest train-ing set off the top of the buffer stack. The training sets in the buffer stack can be presented one or more times each time a new training set is con-structed. A historical database of timestamped data can be used to construct training sets when training input data has a time delay from sample time to availability for the neural network. The timestamps of the training input data are used to select the ap-propriate timestamp at which input data is re-trieved for use in the training set. Using the histori-cal database, the neural network can be trained ret-rospectively by searching the historical database and constructing training sets based on past data.