03-09-2009, 04:51 AM
MULTIPLE ENERGY RESOURCE AGENT
COORDINATION BASED ON ELECTRICITY PRICE
ABSTRACT
This paper describes an algorithm we have developed for coordinating large num-
bers distributed energy resource agents - customer loads and generators - to pro-
vide a demand-response service for electricity retailers, network operators, and
other businesses. Electricity customer price is used as an encouragement for cus-
tomers to make use of an agent-based system. In this paper we report on a new
genetic algorithm optimization of a variable supply cap on the power drawn from
the grid. The cap is used to optimize responses of a large number of cool rooms
which are representative of customer loads. Such responses can minimize the total
cost and also be used to reduce the fluctuation of wholesale electricity prices and
assist constrained networks during demand peaks. The management of large num-
bers distributed energy resources can be used to improve the efficiency of the elec-
tricity network. In our system, the coordination of energy resources is decentral-
ized. Such a decentralized architecture improves grid reliability, allowing consum-
ers to play a more active role in their energy usage. The information exchange is
through indirect communications between resource agents and a broker agent. The
coordination mechanism is asynchronous and adapts to change in an unsupervised
manner, making it intrinsically scalable and robust.