The main goal of controlling an electrical energy system is generating electricity and creating a reliable interconnected system where it maintains the voltage and frequency in an allowable range. Also, increasing the use of renewable distributed energy sources, the integration of various measurement, monitoring, control, and communication infrastructures in the network of modern power systems, provides an occasion to build an efficient and resilient network. On the other hand, due to the increasing size, the dimension of data, complexity, and uncertainty of the modern power system, traditional methods sometimes do not overcome the control issues of a power system. Also, timely evaluation of the power system situation requires adequate response time and reasonable performance to allow controllers and network operators to take preventive and/or corrective actions. Therefore, the need to select efficient methods for faster discovering and identification of power system control problems have always been a priority. In this regard, machine learning and deep learning methods are used to predict events, the amount of variables, and states, their classifications, and features selection when confronted with power system control problems in prevention, normal, emergency, and restoration modes. Machine learning methods are used to extract patterns and order in data to analyze, process, predict, and categorize large data related to the assessment of complex issues of power system dynamical security in order to maintain reliability.
