Linguistic Membership Functions Research Articles

Design of PSO‐based fuzzy logic controller for single axis magnetic levitation system

AbstractIn this paper, a particle swarm optimization (PSO) method is proposed to design an optimal robust fuzzy logic controller (FLC). The objective of this paper is to design a nonlinear optimal robust controller for the single axis magnetic levitation system with high accuracy. PSO algorithm is applied to search globally optimal parameters of FLCs. Three different FLCs are designed. First, proportional derivative (PD)‐like FLC. Second, the FLC is based on the PSO algorithm to find the optimal range of the eight linguistic membership functions (FLC1 with PSO algorithm). Finally, the FLC is based on the PSO algorithm to find the optimal range and shape of the four linguistic membership functions (FLC2 with PSO algorithm). The performances of three different FLCs are compared. Simulation results show that PSO‐based optimal FLCs find the optimal range and shape of the four linguistic membership functions and achieved better performance than the other proposed controllers, minimizing 48 fuzzy rules. © 2011 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Short-term load forecasting by a neuro-fuzzy based approach

An approach based on an artificial neural network (ANN) combined with a fuzzy system is proposed for short-term load forecasting. This approach was developed in order to reach the desired short-term load forecasting in an efficient manner. Over the past few years, ANNs have attained the ability to manage a great deal of system complexity and are now being proposed as powerful computational tools. In order to select the appropriate load as the input for the desired forecasting, the Pearson analysis method is first applied to choose two historical record load patterns that are similar to the forecasted load pattern. These two load patterns and the required weather parameters are then fuzzified and input into a neural network for training or testing the network. The back-propagation (BP) neural network is applied to determine the preliminary forecasted load. In addition, the rule base for the fuzzy inference machine contains important linguistic membership function terms with knowledge in the form of fuzzy IF–THEN rules. This produces the load correction inference from the historical information and past forecasted load errors to obtain an inferred load error. Adding the inferred load error to the preliminary forecasted load, we can obtain the finial forecasted load. The effectiveness of the proposed approach to the short-term load-forecasting problem is demonstrated using practical data from the Taiwan Power Company (TPC).

Autotuning Method of Membership Function in a Fuzzy Learning Controller

A learning algorithm for fuzzy inference rules that can tune both the control input and the linguistic membership function of a fuzzy controller is presented. The algorithm introduces a reference model to generate a desired output and minimizes a performance index based on the error between the reference and plant output. To minimize the performance index, the control input and the membership functions are simultaneously updated by a gradient method. Using the algorithm, we can easily achieve good control rules with a minimal amount of prior information about the environment. In particular, when this algorithm is applied to a tracking problem with a nonlinear plant we obtain efficient control rules with fewer number of rules and higher learning speed than those of the fixed membership function.