Development of an Enthalpy and Carbon Dioxide Based Demand Control Ventilation for Indoor Air Quality and Energy Saving with Neural Network Control

Abstract

An enthalpy and carbon dioxide level based demand control ventilation (EDCV) algorithm has been developed. This takes into account both the indoor occupancy level and the energy content of the fresh air and return air while controlling the fresh air supply. It has been applied under various operating conditions to ensure that the most effective control strategy was used. A back propagation (BP) neural network was used to tune the proportional, integral and differential (PID) parameters in order to obtain a good control performance. Experiments were conducted in a mediumsized lecture theatre to verify the performance of the developed EDCV algorithm in a real application. The results showed that acceptable indoor air quality could be obtained with less energy consumption. Under the optimum experimental conditions, about 15.4% of the total cooling energy was saved. The control performance was found to be good with the PID parameters tuned via the neural network