Dynamic optimization of chilled water pump operation to reduce HVAC energy consumption

Energy consumption in building has been increasing significantly due to higher demand on cooling system. It consumes a large share of total building energy usage. A proper operation control strategy for this system promises the large energy saving. This paper proposes the optimization of chilled water pump operation to improve the cooling system efficiency in a building. The method is developed by combining the artificial neural network and genetic algorithm. This study aims to develop a control method that can determine the optimal chilled water flow rate to absorb the heat on cooling system during the operation. The building energy performances are modeled using EnergyPlus software, while the control strategy is developed in MATLAB. Input output data sharing between MATLAB and EnergyPlus is performed by BCVTB (Building Controls Virtual Test). The dynamic optimization of chilled water flow rate is carried out continuously during real time operation. The developed control performance has been tested under various operating condition including variable outdoor temperature, sudden change of load condition, and different indoor temperature set point. The results show that the optimization of chilled water flow rate on large cooling system could reduce the energy consumption of chilled water pump as 51.11%.