Control Optimization of HVAC Systems in Smart City Commercial Buildings for Sustainable Biotechnological Operations

Abstract

Efficient control and management of HVAC systems in commercial buildings are essential for reducing energy consumption, lowering operational costs, and supporting sustainable biotechnological operations in smart cities. This study proposes an optimized control method for HVAC systems in smart city commercial buildings, designed to meet the stringent environmental requirements of biotechnology facilities, where precise temperature and humidity control are critical for research, production, and commercialization. The research first examines existing control algorithms for HVAC systems in smart buildings, identifying key limitations in responsiveness and energy efficiency. To address these challenges, a deep reinforcement learning-based control method is introduced. The approach incorporates a Smith predictor to compensate for time lag in the control loop, enhancing system responsiveness. A Proportional-Integral (PI) controller is applied to improve the alignment between the reference and actual models within the Smith predictor, ensuring robust time-lag compensation. Additionally, a fuzzy nonlinear PI online optimization method is employed to dynamically adjust the PI controller parameters, further improving control precision and efficiency. Experimental results demonstrate that the proposed method achieves significant energy savings, with energy consumption reductions of 53.6% and 76.4% compared to traditional methods when controlling the pump regulating valve. The method also effectively shortens regulation time, minimizes control process oscillations, and reduces overshoot. These improvements make the proposed HVAC control strategy highly suitable for biotechnological commercial buildings in smart cities, where energy-efficient, stable, and precise environmental control systems are essential for supporting the reliable commercialization and operation of biotechnological innovations.