58th International Universities Power Engineering Conference, UPEC 2023, Dublin, Ireland, 30 August - 01 September 2023, (Full Text)
This paper solves the Non-Intrusive Load Monitoring problem by using two machine learning based models: Xgboost and Recurrent Neural Network. We utilize and develop models using a publicly available dataset. To improve the performance we have implemented hyperparameter optimization using grid search. The numerical simulation results show that proposed Xgboost model outperforms the RNN based model. With the implementation of hyperparameter optimization an improved numerical accuracy is obtained.