An intelligent framework for drive system tuning

It is described how artificial neural networks, when combined with a rule-based mechanism, can provide a practical and effective method for servo drive tuning. An artificial neural network is used as a modeling tool in system identification, and an integrated rule-base provides a control parameter adjustment mechanism to tune conventional feedback controllers. The proposed tuning strategy greatly simplifies the processes involved in establishing suitable parameters for servo drive systems and negates the need for detailed knowledge of the system being controlled. A DC motor drive and a commercial digital motion controller are used to evaluate the proposed strategy. The results confirm that the strategy can be easily implemented, and can enable an effective solution with minimum effort and time