A relaxed constrained decentralised demand side management system of a community-based residential microgrid with realistic appliance models

Abstract Reducing the environmental impacts caused by conventional power sources in smart grids, achieving socio-economic sustainability, and effectively addressing the rapidly increasing energy demand are some of the critical characteristics of demand-side management systems. In this paper, a multi-agent-based decentralised relaxed-constrained energy management strategy for a community-based residential microgrid system using demand-side management is presented. The proposed demand-side management system controls the creative decision-making process of the residential customer agents interconnected within the proposed residential microgrid system. The main objectives of the proposed demand-side management controllers are to make decisions that reduce the peak demand of the load to each agent and to reshape the profile of the power load based on their energy consumption pattern. In addition to this, the novel realistic appliance models with discrete operational levels and on–off capabilities proposed in this research makes the optimisation process a non-convex mixed-integer problem. The proposed decentralised optimisation scheme addressed this issue, by initially relaxing the constraints on the appliances and then using the gradient descent algorithm to decompose and solve the realistic schedules for the devices in the scheduling period. Results indicated that the proposed decentralised relaxed constrain approach is more feasible, effective, economical and efficient in addressing the energy management problem of a residential community microgrid.