Optimal Real-time Dispatch for Integrated Energy Systems
This report describes the development and application of a dispatch optimization algorithm for integrated energy systems (IES) comprised of on-site cogeneration of heat and electricity, energy storage devices, and demand response opportunities. This work is intended to aid commercial and industrial sites in making use of modern computing power and optimization algorithms to make informed, nearoptimal decisions under significant uncertainty and complex objective functions. The optimization algorithm uses a finite set of randomly generated future scenarios to approximate the true, stochastic future; constraints are included that prevent solutions to this approximate problem from deviating from solutions to the actual problem. The algorithm is then expressed as a mixed integer linear program, to which a powerful commercial solver is applied. A case study of United States Postal Service Processing and Distribution Centers (P&DC) in four cities and under three different electricity tariff structures is conducted to 1) determine the added value of optimal control to a cogeneration system over current, heuristic control strategies; 2) determine the value of limited electric load curtailment opportunities, with and without cogeneration; and 3) determine the trade-off between least-cost and least-carbon operations of a cogeneration system. Key results for the P&DC sites studied include 1) in locations where the average electricity and natural gas prices suggest a marginally profitable cogeneration system, optimal control can add up to 67% to the value of the cogeneration system; optimal control adds less value in locations where cogeneration is more clearly profitable; 2) optimal control under real-time pricing is a) more complicated than under typical time-of-use tariffs and b) at times necessary to make cogeneration economic at all; 3) limited electric load curtailment opportunities can be more valuable as a compliment to the cogeneration system than alone; and 4) most of the trade-off between least-cost and least-carbon IES is determined during the system design stage; for the IES system considered, there is little difference between leastcost control and least-carbon control.