An expert system for load allocation in distribution expansion planning

The design of a rule-based expert system for load allocation in the expansion planning of a distribution system is reported. The heuristic rules followed by distribution planners are incorporated into the knowledge base using the artificial language PROLOG. Algorithms aimed at minimizing power loss, minimizing investment cost, and computing service reliability, respectively, are developed to support the inference engine to reach proper lead allocation plans. The expert system using the developed algorithms and knowledge base is built on a personal computer and is applied to reallocate some loads in a distribution system in Taiwan, which consists of three substations and twenty-eight feeders, to a substation that is expected to be operating in the near future. It is demonstrated that the expert system can yield several suitable load allocation plans and can assist the planners in reaching an optimal load allocation plan by providing valuable information. The expert system is not designed to replace the system planner. Rather, it suggests several good plans, and the final decision is left to the system planners