Power quality prediction designed as binary classification in AC coupling Off-Grid system

One of the most critical problems of autonomous energy systems (so called Off-Grid systems) is the keeping of the power quality parameters (PQP) in the requested limits. This paper focuses on development of a simple binary classification as a tool for the forecasting of a PQP. This tool will help as an advisor for a shifting of the load in the Off-Grid system, which keeps the PQP in the requested limits. Some of the most frequently applied machine learning (ML) algorithms like artificial neural network, support vector machine and decision trees with bagging and boosting technique are used in this paper to demonstrate this way of solving the PQP forecasting.