Cepload: A load-allocation program for army central energy plants. Final report

Abstract : Significant energy savings may be achieved through improved coordination of boiler operation in Army central energy plants. Historically, plant operators have tended to run their facilities conservatively to cover the uncertainty of imminent loads while ensuring the reliability of the plant. Because a properly adjusted boiler's operating efficiency depends primarily on its current load, and because most boilers produce their peak efficiencies in the range of 80 percent to 100 percent of their rated capacity, a preferred operating method would maintain each boiler's load as close as possible to the point of maximum efficiency. Most Army heating and cooling loads are related to the weather. An accurate forecast of loads into the near future should make it possible to adjust the boilers to handle those loads more efficiently. Given a reliable forecast model for future loads and an evaluation of boiler operating parameters, an optimum boiler load allocation strategy may be developed. Such a strategy could help the Army improve energy efficiency and reduce the operating costs. The overall objective of this research is to develop a computer-based expert system to help central energy plant personnel optimize boiler operations based on accurate load forecasts. This report documents the development of an accurate load-forecasting model and a prototype expert system called CEPLOAD, which can use the model to help energy plant personnel optimize boiler load allocation.