A Decision-Making Model for Self-adaptation of Cyber-Physical Systems: Application to Smart Grids

Cyber-physical systems (CPSs) are systems that combine physical components with computational elements, working together in real-time to achieve specific goals. These systems are widely utilized in various areas, such as industrial automation, autonomous vehicles, and smart infrastructure. In this study, we focus on the field of smart grids, which are advanced electrical power networks featuring intelligent capabilities. We recognize the significance of self-adaptation in CPS, particularly in the context of smart grids. The proposed method involves developing a decision making model for self-adaptation in CPS specifically tailored to the smart grid domain. We employ machine learning techniques, such as neural networks, along with feedback mechanisms and reconfiguration strategies, targeting four key aspects of the smart grid: voltage, current, temperature, and consumption/production. The aim is to enable the system to, automatically, adjust to fluctuations in electricity demand and supply, as well as changes in network conditions. This involves granting the network the ability to independently monitor, analyze, and act upon real-time data, leading to the optimization of electrical energy production, distribution, and consumption. We evaluate the performance of our proposal through various scenarios. Thus, the simulation results demonstrate the effectiveness of our method.