Multiple distributed energy storage scheduling using constructive evolutionary programming

Deregulation of the electricity supply industry is promoting the increased use of electrical energy storage. However, to achieve the system-wide benefits of competition, techniques for optimal scheduling of distributed storage resources are required. In this paper, we use constructive evolutionary programming to minimise the cost of operating a power system with multiple distributed energy storage resources. The evolutionary technique combines the advantages of both dynamic and evolutionary programming by evolving piecewise linear convex cost-to-go functions (i.e. the storage content value curves). The multi-stage scheduling problem is thus decomposed into many smaller one-stage sub-problems with evolved cost-to-go functions. Evolutionary programming is shown to be suitable for both decentralised computing and for market applications. Case studies demonstrate that the technique is robust and efficient for this type of scheduling problem.