Distributed intelligent load management and control system

Abstract

Demand response (DR) is becoming a key component of future smart grid that can reduce peak load and adapt elastic demand to fluctuating generations. While reducing energy bills for the participant, DR usually decreases its utility, which is different for distributed occupants inside a participating entity. A two-level distributed intelligent load management and control system is proposed in this paper to minimize the cost of the participant, where the cost is a measurement of disutility, considering differences across plug loads, with the load reduction constraint of a DR event. The control system contains a smart DR controller and distributed intelligent gateways. In the system, the cost function of a load is modeled to reflect the dissatisfaction of the occupant for switching off or dimming the load, and a two-level optimization method is deployed to minimize the participant's aggregated cost. Each intelligent gateway collects the cost functions of loads in the neighborhood of an occupant, generates its optimal cost function and sends to the smart DR controller. The smart DR controller utilizes those cost functions to allocate the load reduction among the gateways, which can then optimize the load reduction among loads for the distributed optimum. While the cost function of the loads can be modeled as either continuous or discrete functions based on the type of the load, Lagrange multipliers and particle swarm optimization (PSO) are utilized for optimization, respectively. This innovative method is implemented in a DR system of a building, and tests results show that the proposed distributed DR method is practical and promising.