PSYCHROMETRIC BIN ANALYSIS FOR DATA CENTER COOLING MODES USING ARTIFICIAL NEURAL NETWORKS

Abstract

With an increase in the need for energy efficient data centers, a lot of research is being done to increase the use of Air Side Economizers (ASEs), Direct Evaporative Cooling (DEC), Indirect Evaporative Cooling (IEC) and multistage I/DEC cooling. The cooling strategies used to control these systems is based on typical meteorological year (TMY) weather data and thermodynamic principles. But the main drawback of these control strategies is that they do not account for the nonlinearities developed by the conditions inside the data center. So, the primary objective of this study is to use Artificial Neural Networks (ANN) for predicting the CA humidity and temperatures for different modes of cooling. These results can then be studied and then utilized to come up with new bins for each cooling mode. These results will account for the nonlinearities in the data center which are difficult to model using traditional methods.