Author : Shikha Sharma, Praveen Kumar and Ruchika Manchanda

Fact Gathering means generating rule base from available numerical data. The intelligence of a fuzzy system lies in its rule base. Number of rule base generation methods are used for fuzzy system such as neural networks, as Hybrid Learning, genetic algorithms, Wang & Mendel Approach, biogeography based optimization approach and particle swarm optimization found in the literature. Designing fuzzy systems is an optimization problem. This paper presents a recent nature-inspired algorithm named Ant Colony optimization (ACO) approach for automatic generation of optimized fuzzy rule base from available numerical data i.e. Fact Gathering. The ACO is inspired by real ant colony observations. It is a multi-agent approach. In the ACO, artificial ant colonies cooperate in finding good solutions for difficult discrete optimization problems. Here, Ant paths help to determine the consequent parameters of generated rules. Extracted rules are printed and then system performance is evaluated using MSE value. This approach is applied on well-known fuzzy control problem of battery charger. The proposed approach provides fuzzy models described with reduced number of rules as compared to initial fuzzy models. Also, this approach is much more efficient in terms of computational time and MSE as compare to other approaches.

Keywords: Ant colony optimization, designing fuzzy systems, Fact Gathering, pheromone, rule base, system performance.

Article published in International Journal of Current Engineering and Technology, Vol.4, No.6 (Dec-2014)