Intelligent HVAC Control Systems Based on Machine Learning and Deep Learning Models Ganesh B. Murade, Ankit Kumar Sharma, Bhanu Pratap Soni, Arvind S. Pande, Ganesh K. Shirsat Driving Affordable and Clean Energy Through AI and Intelligent Systems, 2026 Buildings are responsible for nearly 35% of global energy consumption, with heating and cooling accounting for a significant share. As urbanization and population growth accelerate, energy demand in cities continues to rise, leading to increased CO2 emissions, higher electricity costs, and greater stress on power grids. To address these challenges, real-time control systems and building automation have shown strong potential in improving energy efficiency in the built environment. In particular, optimized Heating, Ventilation, and Air Conditioning (HVAC) control systems can significantly reduce energy consumption, lower operational costs, and mitigate environmental impacts. This study investigates the application of machine learning models—including Linear Regression, Random Forest, Support Vector Machine (SVM), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU)—to predict HVAC performance. The models are evaluated using Mean Squared Error (MSE) and R2 score, with the goal of identifying the most effective approach for enabling data-driven energy optimization in buildings.
A TWO-STAGE FORECASTING APPROACH FOR HVAC SYSTEMS: COMPARATIVE ANALYSIS OF ML AND DL MODELS WITH FORECASTED TEMPERATURE INPUTS Ganesh Murade International Journal of Applied Mathematics, 2025 Heating, ventilation, and air-conditioning (HVAC) systems are among the most energy-intensive components in commercial buildings. Accurate forecasting of HVAC energy consumption is essential for implementing energy-efficient control strategies. This study provides a detailed comparative analysis of classical machine learning (ML) and deep learning (DL) models for forecasting HVAC energy consumption using both predicted and actual indoor temperature data. Leveraging an open-source, high-resolution dataset from the Oak Ridge National Laboratory's FRP-2 multizone building, we implement a two-stage framework. The first stage involves temperature prediction using relative humidity and airflow data, while the second stage forecasts HVAC energy consumption using either predicted or actual temperature inputs. We evaluate Linear Regression, Random Forest, Support Vector Machine, GRU, and LSTM models. The results reveal that while models with actual temperature inputs yield superior accuracy (R² up to 0.884), models using predicted temperatures still achieve competitive performance. Our findings suggest that ML and DL models, particularly LSTM, offer promising capabilities for real-time energy forecasting in smart building environments.
HVAC Hybrid Control methods for HEE in Buildings: Overview Ganesh B. Murade, Bhanupratap Soni, Aniruddha Mukherjee 2021 7th IEEE International Conference on Advances in Computing Communication and Control Icac3 2021, 2021 Inhabitants of business building will in general have limited methods for influencing HVAC activity frameworks addresses the biggest segment of energy use in business structures. Central air likewise addresses the best chance for investment funds, as structures inefficiently over-condition spaces. Heating, Ventilation and air conditioning (HVAC) units are answerable for keeping up the temperature and stickiness setting in a structure. Investigations of HVAC represent approx. half energy utilization in a structure and 10% of worldwide power use. Centrally air advancement accordingly the possibility contributes to all towards maintainability objectives diminishing energy utilization and CO2 emissions.
