No abstract

No abstract

Abstract-There is a need to improve the delivery of energy services, and utilizing distributed energy resources offers significant potential. We propose an energy service modeling technique that would capture temporal variations of its demand and value, and differentiate it from the electric energy consumed by the end-use equipment. We then use this technique with a novel energy service simulation platform that aims to maximize the net benefit derived from energy services. The simulation platform creates a strategy for how available distributed resources should be operated in order to provide the desired energy services while minimizing the cost of consumption. The corresponding optimization problem is solved using particle swarm optimization. The simulation platform proved capable of creating an operation schedule that maximizes net benefit under a range of challenging conditions.

Abstract-This paper describes a methodology for making robust day-ahead operational schedules for controllable residential distributed energy resources (DER) using a novel energy service decision support tool. The tool is based on the consumers deriving benefit from energy services and not on electric energy. It maximizes consumer net benefit by scheduling the operation of DER. The robust schedule is derived using a stochastic programming approach formulated for the DER scheduler: the objective function describing the consumer net benefit is maximized over a set of scenarios that model the range of uncertainty. The optimal scenario set is derived using heuristic scenario reduction techniques. Robust operational schedules are formulated for a 'smart' home case study with four controllable DER under stochastic energy service demand, availability of storage DER, and status of dynamic peak pricing. The robust schedule results in a lower expected cost but at the expense of long computation times. The computation period however is not much of a disadvantage because schedules are computed off-line. The consumer can prepare several DER schedules and simply choose the one to implement according to their perception of the coming day. The robust schedules are formulated using an improved version of co-evolutionary particle swarm optimization.

No abstract