Welfare maximization under a three-choice minority game model for energy demand allocation

Microgrids have been increasingly researched as a promising solution for providing energy to off-grid rural communities. Among the challenges of deploying microgrids to these communities is the problem of developing simple demand management schemes so that peaks and troughs in the consumers’ load demand can be minimized. A usual demand management scheme involves awarding consumers who utilize energy at times of low demand and penalizing those who utilize energy at times of high demand. Such schemes, however, may not always maximize user welfare. In this work, welfare analysis for a three time-window scheme for a simple microgrid is conducted. The study investigates whether ideal agent distributions that maximize social welfare across these time windows of energy usage can be computed assuming homogeneous agent utility and linear costs. A three-choice minority game was then developed as an allocation scheme for the grid. The game was simulated to determine whether these ideal agent distributions can be achieved within the game. Simulation results show that agent attendances over time approach the ideal agent distributions for some penalization schemes. The developed model therefore has a promising potential to be applied as a demand management scheme.