An integrated optimization and agent-based framework for the U.S. power system

Today, many of the engineered systems are comprised of a large number of components that interact with each other and have the ability to exhibit emergent behavior thus enabling a system to adapt to changing environments. Using the example of the US power grid as a complex adaptive system, we demonstrate how components in a multi-layered power grid structure dynamically interact, evolve and adapt over time. In our model, electricity regulators strive to balance workload by dynamically adjusting service attributes in response to demand fluctuations. Additionally, they seek to change long-term consumption patterns by providing incentives and social education. Moreover, consumer agents focus on maximizing quantitative and qualitative utilities. By embedding a non-convex optimization model with the agent-based framework we study cooperativeness or competition in the consumers game environment. Our framework allows us to study the behavior of consumers under different control and incentive strategies. We expand model dynamics to include intrinsic environment and control factors. This study also examines circumstances in which agent-based and equilibrium models present similar outcomes or are unable to converge to same results. This method is used to study the robustness of the results, present equilibriums of interoperability equations, and study dynamics of traits.