Decentralized Optimization of Energy Exchanges in an Electricity Microgrid

We propose to control a local electricity grid, a "microgrid", in a decentralized fashion to reduce energy costs by coordinating the generation and consumption decisions made by the individual Distributed Energy Resources (DERs) composing the system. Our control scheme relies on the so-called Alternative Direction Method of Multipliers or ADMM, the essentials of which we summarize in the first part of the paper. Then we describe our models for loads, batteries, solar panels and wires before investigating different scenarios in which our coordination scheme may be more beneficial than independent decision-making. The first scenario demonstrates how our coordinated control strategy may reduce total expenses under variable tariffs of energy. Since this first method does not, however, ensure a gain for each user of the microgrid individually, our second scenario explores a slightly different control strategy designed to ensure an individual benefit to each user, at the cost of decreasing the total benefit.