Energy demand forecasting for tokat gaziosmanpaşa university hospital with hybrid deep learning (CNN-LSTM) model

It is seen that electricity production and consumption will change day by day, equivalent to industrialization and growth. All investments and planning for the future will be based on accurately predicting the future. The accuracy of forecasting our future electricity consumption will also support accurate planning. Electrical energy planning regulation aims to minimize all costs from the production of electricity until it reaches the end user, to increase the efficiency of power systems, to increase the reliability of the system and to deliver electrical energy to the consumer with the highest quality. Various statistical analysis, mathematical analysis, machine learning and its subfield, deep learning methods, are used to analyze time series data and make future predictions. With the increasingly popular deep learning method, it has become more successful in time series problem solving than traditional methods and its use is increasing. In this study, a short-term demand forecasting method was developed using deep learning techniques in the Colaboratory program, with a focus on forecasting methods.This developed method was applied to Tokat Gaiosmanpaşa University hospital and the results were analyzed.