Performance improvement of building heating, cooling and ventilation systems

Heating, Ventilation, and Air Conditioning (HVAC) systems are responsible for a substantial share of the energy consumed in commercial buildings. Energy used by HVAC systems has increased over the years due to its broader application in response to the growing demand for better thermal comfort within the built environment. While existing case studies demonstrate the energy saving potential of efficient HVAC operation, there is a lack of studies quantifying energy savings from optimal operation of HVAC systems when considering indoor environmental conditions. This research aims to improve the performance of HVAC systems by optimizing its energy consumption without compromising indoor environmental conditions. The concept of maintaining indoor environmental conditions poses new challenges to the optimal operation of HVAC systems. While the primary objective of ensuring optimal operation is to minimize energy consumption, controlling the indoor environmental parameters, e.g., temperature, humidity, the level of carbon dioxide (CO2), and volatile organic compounds (VOCs) to remain within the acceptable range imposes excess energy use. These two conflicting objectives constitute a multi-variable constrained optimization problem that has been solved using a particle swarm algorithm (PSO). A real-time predictive model has been developed for individual indoor environmental parameters and HVAC energy consumption using Nonlinear Autoregressive Exogenous (NARX) neural network (NN). During model development, efforts have been paid to optimize the performance of the model in terms of complexity, prediction results, and ease of application to a real system. The proposed predictive models are then optimized to provide an optimal control setting for HVAC systems taking into account seasonal variations. An extensive case study analysis has been performed in a real commercial building to demonstrate the effectiveness of developing predictive models and evaluating the relevance of integrating indoor air quality (IAQ) within the optimization problem. Results show that it is possible to minimize 7.8% energy consumption from HVAC systems without compromising indoor environmental conditions. This study demonstrates that the proposed optimal control settings maintain the indoor environment within the acceptable limit of thermal comfort conditions (indoor air temperature between 19.60 to 28.20C and indoor air humidity between 30 to 65 %RH as per ASHRAE Standard 55-2017) and air quality (CO2 ≤ 800 ppm and VOC ≤ 1000ppm as per Australian Standard AS 1668.2 2016). The outcomes of this research will act as a guideline for energy management practices, not only for energy efficient building design and retrofitting but also for building energy performance analysis. This research provides insight into the aspects that affect the performance of predictive models for indoor temperature. The proposed feature selection approach establishes its efficacy to determine salient and independent input parameters without compromising prediction performance. The application of this approach will minimize the measurement and data storing cost of variables. Further, using fewer numbers of input parameters in the model will reduce the computational cost and time. Thus, the proposed model establishes its applicability in a real system for a more extended period of advanced prediction. In addition, the need to better account for building-occupant interactions as an important step to maintain a healthy indoor environment has been recognized through evaluating a real-life demand control (DCV) system. Lastly, the proposed optimization approach, where four defined environmental parameters are considered simultaneously presents a new outlook within the HVAC control system by eliminating the unseen interface between thermal comfort and IAQ. Overall, this unexploited potential to simultaneously improve the performance of HVAC systems and indoor environmental conditions drives the discussion on reconsidering the set-point configuration standards of HVAC in commercial buildings, either as part of individual building retrofit planning or as part of building regulatory applications.