Small-World Particle Swarm Optimizer for Real-World Optimization Problems

Many real-world problems from different domains, viz. engineering, data mining, biology, can be formulated as the optimization of a continuous function. These problems require the estimation of a set of model parameters or state variables that provide the best possible solution to a predefined cost or objective function, or a set of optimal trade-off values in the case of two or more conflicting objectives. Locating global optimal solutions becomes challenging especially in the presence of high dimensionality, nonlinear parameter interaction, insensitivity, and multi-modality of the objective function. These conditions make it very difficult for any search algorithm to find high-quality solutions quickly without getting stuck in local optima. Unfortunately, these difficulties are frequently encountered in real-world optimization problems when traversing the search space en route to the global optimum. Small-world PSO has been proven to be effective in solving global function optimization problems. After all, every optimization algorithm has to be applied to some real-world problems. This paper evaluates the performance of small-world PSO algorithm on two real-world function optimization problems. Comparative study with state of the art demonstrates the effectiveness of small-world PSO.