An Intelligent Approach to Develop, Assess and Optimize Energy Consumption Models for Air-Cooled Chillers using Machine Learning Algorithms

The building sector accounts for more than 70% of thetotal electricity use. Chillers consume 50% electrical energyduring seasonal periods With the growth sectorand climate change, it's essential to develop energy-efficient HVAC systemsthat optimize ever-increasing energy demand. This study aims anenergy consumption prediction model air-cooled chillers using machinelearning algorithms. is done by developing different static and dynamicdata-driven regressive neural network models comparing accuracy oftheir identify most accurate modeling algorithm 3 maininputs chilled water return temperature, outside drybulb andcooling load. proposed structure was then optimized in terms thenumber neurons, epochs, time delays aswell as number input variables a genetic algorithm. Training andtesting were real data obtained from fully instrumented 4-tonair-cooled chiller. Results show that artificialneural can predict with high level ofaccuracy compared conventional techniques. development ofhighly self-tuning be powerful tool use otherapplications such fault detection diagnosis, assessment, systemoptimization. Further studies are necessary evaluate effectiveness ofusing deep learning algorithms hidden layers cross-validationtechniques.