An application of genetic algorithms to integrated system expansion optimization

This paper presents the application of a Genetic Algorithm (GA) based method to integrated system expansion optimization. Given an existing system model, the projected load growth in a target year, and various system expansion options, this method finds the optimal mix of system expansion options to minimize a generalized cost function subject to various system constraints. The system expansion options considered include whether to build new transmission lines/transformers and how much capacity to build, if existing lines/transformers should be upgraded and how much to upgrade, and if distributed generations should be installed and where and how much to install. The GA based method is implemented and tested on a real US system. The optimization results are compared with the successive elimination method (SEL) to demonstrate the performance improvement. A unique offspring selection procedure is used in the GA implementation to maintain genetic diversity in the solution population and to prevent premature convergence.