Modelling and identification of an irrigation station using hybrid possibilistic c-means and fuzzy particle swarm optimisation
Abderrahmen Zaafouri, Mohamed Jemli, Jaouher Chrouta · International Journal of Computer Applications in Technology · 2017
Data-driven design of accurate and reliable Takagi-Sugeno (T-S) fuzzy systems has drawn the attention of several researchers in recent decades, according to an excellent ability for describing non-linear systems. In literature, several fuzzy clustering algorithms have been proposed to identify the parameters involved in the Takagi-Sugeno fuzzy model. Possibilistic C-Means (PCM) is one of the most used clustering methods because it is efficient, straightforward, easy to implement and exhibits robustness to noise. However, PCM is sensitive to initialisation and is easily trapped in local optima. Contrariwise, the PSO algorithm has strong global searching ability, and it doesn't easily get into the local minimum to overcome the drawbacks of PCM algorithm. In this paper, a hybrid fuzzy clustering method based on PCM and fuzzy PSO (FPSO) is proposed which makes use of the merits of both algorithms. Experimental results applied to an irrigation station show that the hybrid algorithm (PCM-FPSO) is efficient and can reveal encouraging results.