Energy management of renewable based power grids using artificial intelligence: Digital twin of renewables

Abstract

This paper proposes fog computing as a method of minimizing latency and maximizing performance by storing the information in the cloud to give a wide range of services to users using cloud computing. Also, consumer requests have been handled by several cloud Data Centers (DCs). Due to the fog's ability to attract more users and provide more services, load balancing has become increasingly crucial. It is therefore necessary to enhance the method that balances the fog's load. The study proposes a three-layer model comprising a cloud layer, a fog layer, and a user layer for optimal energy management in renewable power grids. The renewable energy sources are all modeled within the digital twin environment to make sure that very accurate monitoring is made. It is proposed for balancing the fog load using an artificial intelligence based optimization algorithm called the Whale optimization algorithm (WOA). Services are provided by fog servers in response to requests from users. In case of failure of the fog layer, all consumers' records are saved and services are provided to them through the cloud. Moreover, the service broker policies (SBP) have been applied to select the DCs efficiently. This paper compares the suggested algorithm to three previously developed algorithms including the particle swarm optimization (PSO), differential devolution (DE), and teaching–learning-based optimization (TLBO). It uses the three SBPs of closest data center, optimize response time (RT), reconfigure dynamically with load, and an improved SBP. It also minimizes RT and process time to improve the efficiency. The results show that there is an improvement in WOA's performance by approximately 5% compared to that of other algorithms.