Demand and Energy Management in Smart Grid: Techniques and Implementation

Abstract

The traditional power grid relies heavily on fuel-based power plants with one-way communication between suppliers and consumers. Smart Grid is proposed as a solution combining power grid with a robust communication network. It improves power generation, delivery and utilization, and supports renewable energy resources. In the present paper, we discuss energy management techniques used for improved power generation and utilization. We focus mainly on Demand Management System. From the consumer side, Load Shifting and Scheduling are used to avoid grid congestion and blackouts. From the power plants side, Unit Commitment and Load Forecasting create a balance between demand and supply. In order to deploy these algorithms, two architectures are discussed: centralized control and decentralized multi-agent system. Simulation results conducting by many researchers show that pricing and Demand Management System improve power generation and utilization. Also, Unit Commitment algorithms reduce the dependency on fuel-based plants. Concerning the deployment of these algorithms, multi-agent system has proven its efficiency for controlling energy demand in smart grids, alleviating concerns about cascading failures and frequent human intervention.