AI-Driven Energy Consumption Forecasting and Optimization System

The AI-Driven Energy Consumption Forecasting and Optimization System successfully demonstrates an end-to-end solution for predicting and optimizing electricity usage in buildings. The system integrates data processing, multiple forecasting models, optimization techniques, and a user-friendly dashboard to provide actionable insights.   The model evaluation, the Prophet model outperformed  LSTM  and  Transformer, achieving lower error metrics and better reliability, indicating strong seasonal patterns in the dataset. The explainability analysis revealed that occupancy and temperature are the most influential factors affecting energy consumption, aligning with real-world behavior. The optimization module effectively reduced energy costs while maintaining operational and comfort constraints, achieving significant savings (up to ~77%). This highlights the system’s capability to make intelligent, cost- efficient decisions.   Overall, the project not only predicts energy demand but also recommends optimal usage strategies, making it a practical decision-support system for smart buildings. Future enhancements can include real-time IoT integration, improved deep learning models, and deployment in large- scale smart grid environments.

[1]             Afram and F. Janabi-Sharifi, “Theory and applications of HVAC control systems A review of model predictive control (MPC),” Building and Environment, vol. 72, pp. 343– 355, 2014. [2]             C. Wei and Y. Li, “Reinforcement Learning for Building Energy Management: Review and Case Studies,” IEEE Transactions on Industrial Informatics, vol. 17, no. 9, pp. 6083–6093, Sep. 2021 [3]             Y. Zhang, N. Good, and P. Mancarella, “Data-Driven Control for Smart Building Energy Management: A Reinforcement Learning Approach,” Applied Energy, vol. 221, pp. 372–386, 2018   [4]             B. Yildiz, J. Bilbao, and A. Sproul, “A review and analysis of regression and machine learning models on commercial building electricity load forecasting,” Renewable and Sustainable Energy Reviews, vol. 73, pp. 1104– 1122, 2017 [5]             [2005/32/EC] N.N.: Directive 2005/32/EC of the European Parliament and of the Council, July 6, 2005, available from: http:// ec.europa.eu/enterprise/ eco_design/dir2005- 32.htm, 2005. [6]             K. Amasyali and N. M. El-Gohary, “A review of data-driven building energy consumption prediction studies,” Renew. Sustain. Energy Rev., vol. 81, pp. 1192–1205, Jan. 2018. [7]             L. Benali, G. Notton, A. Fouilloy, C. Voyant, and R. Dizene, “Solar radiation forecasting using artificial neural network and random forest methods: Application to normal beam, horizontal diffuse and global components,” Renew. Energy, vol. 132, pp. 871–884, Mar. 2019. [8]             N. Amjady, “Short-term hourly load forecasting using time-series modeling with peak load estimation capability,” IEEE Trans. Power Syst., vol. 16, no., pp. 498–505, Aug. 2001. [9]             S. Bouktif, A. Fiaz, A. Ouni, and M. A. Serhani, “Optimal deep learning LSTM model for electric load forecasting using feature selection and genetic algorithm: Comparison with machine learning approaches,” Energies vol. 11, no. 7, p. 1636, Jun. 2018. [10]         J. R. Driscoll, N. Sarnak, D. D. Sleator, and R. E. Tarjan, “Making data structures persistent,” J. Comput. Syst. Sci., vol. 38, no. 1, pp. 86–124, 1989.