Search

The ultrasonic motor has nonlinear characteristics, which vary with driving conditions and possess variable dead-zone in the control input that is associated with the applied load torque. The dead-zone has a problem in accurate positioning actuator. To improve the control performance of the ultrasonic motor, the dead-zone nonlinearity should be eliminated. This article proposes a new position control scheme for the ultrasonic motors that eliminates the problem due to dead-zone by employing fuzzy neural network (FNN). To achieve the accurate position control when drive conditions vary, FNN can adjust the membership function for the antecedent part. The training of FNN is achieved using online backpropagation algorithm. The dead-zone is compensated by FNN, and PI controller performs the accurate positioning of the drive system. The validity of the proposed method is confirmed by experimental results. [ABSTRACT FROM AUTHOR]

In this article, an approach based on neural network combined with a fuzzy system is proposed for three-hours-ahead load forecasting. In this proposed prediction method, the forecasted load is obtained by adding a correction that is obtained from the neural network and a fuzzy logic to the selected similar days data. The neural network has the advantage of dealing with the nonlinear part of forecasted load curves and the fuzzy rules are constructed based on the expert knowledge. Therefore, by combining these methods, the proposed method is useful in situations where accurate forecasting models are difficult to obtain. The suitability of the proposed approach is illustrated through an application to actual load data of the Okinawa Electric Power Company in Japan. [ABSTRACT FROM AUTHOR]