Application of machine learning techniques in modern hybrid power systems-a case study

Due to the rapid adoption of intelligent power electronic devices and digital technologies, traditional vertically designed power systems are being phased out and replaced by modern hybrid power systems. Since its inception, the power system has undergone numerous changes that have increased system efficiency, increased the share of renewable energy, and made it easier to control. However, such a rapid revolution in electrical power systems during the current Industrial Revolution has increased its complexity. The primary concerns are cybersecurity, forecasting supply and demand, optimal power allocation, power quality maintenance, and a skilled workforce shortage. Digital tools aid in load management and the optimization of various power resources. Modern hybrid power systems, artificial intelligence techniques such as machine learning and optimization algorithms, are emerging in the power sector for better control. Nonetheless, little research has been conducted on machine learning applications in industries with integrated power resources. Machine learning techniques will be used in the industries to forecast supply and demand, make the best use of energy resources, etc. This chapter aims to discuss the use of machine learning techniques in modern hybrid power systems. A well-known large industry with multiple energy resources has been considered for this purpose. All components of the power system network are modeled, and simulations are run to determine the best way to use them under various generation and load scenarios, weather conditions, and financial conditions. For the case study considered, the simulated results are validated using field data and ETAP software, and the results are encouraging.