Multi-objective Optimal Energy Consumption Scheduling in Smart Grids

Due to the increasing discrepancy of the load demand during peak hours and off peak hours, the inefficiency in the power grids is getting more prominent. In order to meet the load demand during peak hours while maintaining a reasonable security margin above the peak hours demand, additional generation capacity may be needed, which involves vast investments and deterioration of the utilization of the generation capacity. With the introduction of the Smart Grids, a two-way communication between the grid operator and the users is realized by Smart Meters (SMs). Demand Response (DR) has also been proposed to avoid installation of new power plants in the Smart Grids. DR enables real-time pricing strategy, which allows modification of users' demand profiles according to the unit cost of energy consumption. This strategy aims to effectively shift the peak hours load demand to the off peak hours, therefore to improve the current utilization of the generation capacity. In this paper, a utility company is supplying energy to a group of residential users from a community and a third-party is considered, which can be the utility manager of the community and is responsible for scheduling the consumption profiles of all users. The load scheduling problem is formulated as a constrained multi-objective optimization problem (CMOP), with two conflicting and non-commensurable objectives being first minimizing the total energy cost and second maximizing the overall utility, which is the comfort of living in the community. The two evolutionary algorithms (EAs) adopted to address the CMOP are modified version of Multi-objective Optimization Evolutionary Algorithm based on Decomposition using Differential Evolution operator (MOEA/D-DE) based on the original framework and constrained Non-Dominated Sorting Genetic Algorithm-II (NSGA II). By comparing the results obtained by these two algorithms, it is observed that the proposed MOEA/D-DE outperforms constrained NSGA II in terms of convergence and diversity preservation.