Integrating AI Applications in an Energy Management System

DESPITE THE OVERWHELMING satisfaction utilities report in implementing pioneer expert systems in energy-management systems, the number of actual implementations of such systems in control centers remains quite small compared to the number of systems successfully demonstrated in prototype form.1–3 This discrepancy suggests that utilities have unanswered concerns about the cost and difficulty of integrating expert systems into an EMS. Applications developers must address these concerns before this technology finds a wider acceptance in power-system operation. As part of that effort, we first needed to determine why integrating an expert system or other type of AI application in an EMS is more complex and costly than adding a conventional advanced application program. Based on this analysis, we then defined the interfaces needed to achieve an effective integration. This groundwork led us to believe that AI applications should not be integrated directly into the EMS. Rather, they should plug into an AI environment that supports the necessary interfaces with the EMS and provides common services to the applications. We have designed and implemented such an environment, which currently supports three distinct AI applications—intelligent alarm processing, fault diagnosis, and power-system restoration. To illustrate the benefits of this approach, this article presents case studies based on implementations of this environment in three different EMS architectures.