Algorithm for searching optimal set values of absorption chiller system using Bayesian optimization

HVAC systems account for at least 40% of the energy consumption general office buildings. Therefore, reducing use is indispensable. used in public buildings mostly adopt a central air conditioning system. To save system, high-efficiency heat source machines have been adopted, and inverter control pumps has generally introduced. As further measure to energy, an optimal proposed. Model-based mainly studied. However, target system must be appropriately equipped with sensors model-based previous operation data. solve this problem, we proposed model-free method using Bayesian optimization. We verified its effectiveness one-on-one (one cooling tower one chiller) systems. result, energy-saving ratio when compared rated specification 10.38% our 11.34% approach, which shows equivalent performance. In addition, training results indicate that set values can automatically determined 4 weeks as proceeds under