Optimization Using Horizon-Scan Technique: A Practical Case of Solving an Industrial Problem

Abstract

This chapter introduces a new Computational Intelligence algorithm called Horizon Scan. Horizon Scan is a heuristic based technique designed to search for optimal solution in non-linear space. It is a variant of the Hill-Climbing technique and works in contrary to the temperature-cooling scheme used in Simulated-Annealing. Initial experiments on the application of Horizon Scan to standard test cases of linear and non-linear problems have indicated promising results (Chand & Sugianto, 2003a; Chand & Sugianto, 2003b; Chand & Sugianto, 2004). In this chapter, the technique is described in detail and its application in finding the optimal solution for the Scheduling-Pricing-Dispatch problem in the Australian deregulated electricity market context is demonstrated. It is hoped that the proposed approach will enrich the existing literature on Computational Intelligence, in particular to solve optimization problems, such as those that exist in the deregulated electricity industry around the globe.