Comparison of Cost Forecasting Algorithms for Industrial Loads

Providing consumers with information on electrical power usage combined with forecasts for power consumption, energy prices, and predicted electricity costs makes the data more useful. It will help the locals change their ways of utilizing energy, which will be beneficial for the corporations that produce electricity as well as for the distribution system and the electricity grid. The cost of electrical energy usage is forecasted in this paper using machine learning methods. This study analyses the performance of two machine learning algorithms: Auto Regressive Integrated Moving Average (ARIMA) and Seasonal Auto Regressive Integrated Moving Average (SARIMA). To determine which machine learning method is more accurate and efficient, the actual cost and predicted cost values of each algorithm are compared. The study also analyses the difference in error parameters of the individual algorithms. It is found that ARIMA outperforms SARIMA in terms of effectiveness.