Simulation-Optimization Approaches for Demand Side Management in Smart Grids

In smart and connected communities (S&CCs), energy, transportation, water, public safety, and all other services need to be managed efficiently to support smooth operation while providing a clean, economical, and safe environment for citizens. The power system infrastructure is the most important part of S&CCs that affects the functionality of all components. Over the past two decades, power systems have witnessed significant changes in the use of renewable and distributed energy resources, energy control technologies, and technical advances in communication and computation, which has led to the development of smart grids. Smart grids provide sustainable power grids with the capabilities of self-healing and automatic execution in an isolated mode, but the operation and control planning of smart grids are challenging procedures. The main challenges arise from energy load scheduling of customers using interruption load management (ILM) and load shifting strategies; communication between customers and utility companies or third-party aggregators to increase customer satisfaction and decrease costs; and the need to provide reliable and high-quality energy to customers. In this doctoral study, novel simulation and optimization approaches for demand side management (DSM) in smart grids are introduced to address the main challenges in the operation and control planning of smart grids. To this end, this study investigates efficient DSM programs for smart grids and addresses the primary challenges in two broad frameworks: (1) a deterministic optimization framework for load shifting in smart grids, which can obtain optimal scheduling for time-shiftable energy loads using an advanced ϵ-constraint optimization method, and (2) a stochastic and dynamic simulation optimization framework for DSM programs in smart grids that finds near-optimal solutions for ILM with uncertain loads. These two frameworks provide a simulation and optimization tool for utility companies or third-party aggregators to provide day-ahead energy load scheduling based on desirable DSM strategies. The proposed frameworks are applied in two synthetic smart grid case studies. The results of the case studies show that the proposed frameworks are able to meet the desired energy load curves while resulting in better objective functions. This doctoral research reveals that both deterministic and stochastic DSM programs are promising tools to optimize and boost the implementation of DSM programs and attain several benefits