AI Models for Predicting Urban Electricity Consumption: A Step Towards Smarter Cities

The rapid growth of urbanization and industrialization has led to an unprecedented demand for electricity, making accurate forecasting of energy consumption a challenge for modern cities. Effective energy management is essential not only for meeting consumer needs but also for ensuring sustainability, reducing costs, and optimizing resource allocation. With the increasing availability of data and advances in artificial intelligence (AI), predictive models have emerged as powerful tools for addressing this challenge. This study explores the potential of AI models in predicting electricity consumption across different states regions, focusing on India as a case study. Leveraging multivariate data, including variables such as average precipitation, annual income, temperature, population, area, and historical electricity consumption, we aim to develop a reliable predictive framework. Such models can provide actionable insights, empowering city planners and policymakers to make informed decisions regarding energy distribution, infrastructure development, and environmental sustainability. In this research, a regression approach was employed to model electricity consumption due to its ability to handle multivariate datasets with varying conditions effectively. The model was trained on data from ۲۶ states, with an additional ۶ states used for testing its performance. Comprehensive analysis of data distribution, correlation adjustments, and optimization techniques were applied to enhance the model's accuracy and reliability. By integrating AI into energy forecasting, this study aims to demonstrate the practical applicability of machine learning techniques in real-world. The results not only highlight the strengths and limitations of the proposed approach but also pave the way for further advancements in creating smarter cities.

کلیدواژه ها:

AI ، Electricity consumption ، Smart cities ، Machine Learning Techniques