Empowering Smart Cities: SARIMA Forecasting of Power Consumption in Tetouan's Urban Grid

Abstract

Predicting power consumption is crucial for optimizing energy management in smart grids, where intricate patterns and seasonality characterize consumption dynamics. This study focuses on forecasting power consumption within Tetouan city, leveraging the Seasonal Autoregressive Integrated Moving Average (SARIMA) model. SARIMA's efficacy in handling time series data with inherent seasonality makes it a fitting choice for capturing the nuanced patterns in energy usage. The dataset, collected at 10-minute intervals from Tetouan, provides a comprehensive representation of consumption patterns. The SARIMA model is configured using the training dataset to adapt to Tetouan's unique energy dynamics. Forecasting is conducted on the validation dataset, and model accuracy is assessed using metrics such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and R-squared. This study contributes insights into power consumption forecasting, with implications for smart grid operations, resource allocation, and overall grid efficiency in the urban context of Tetouan.