A frequently used learning algorithm for self-organising fuzzy logic control has been analysed and found to be inadequate for some applications. The key problem is how to gain the credit value for control rule modification. A new type of learning algorithm is proposed. Some supervision rules are given to secure rule modifications. A new approach is derived to calculate the credit value for the rule generation and modification. Simulation studies for controlling different process models and different model parameters are demonstrated.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (1992) studied this question.