Session: K20-01: APPLICATIONS OF MACHINE LEARNING/AI FOR HEAT TRANSFER

Paper Number: 121705

121705 - Machine Learning Algorithm for Predicting Heat Transfer Coefficient and Pressure Drop in Dimpled Ducts 

Abstract: 

Conducting experiments and/or simulations to collect large numbers of data points over extensive ranges of design parameters and operating conditions requires substantial experimental efforts and extremely high computational resources even by using the current advanced computational facility. Besides, it is highly possible that the corresponding bank of data to represent the performance data points of a cooling system contains only limited numbers of data. Moreover, such limited numbers of data may be scattered as such they have been acquired under very different ranges of operating and design conditions by different researchers. Since there are no relations among the scattered data points, developing correlations through them is challenging by using interpolation techniques. To overcome the challenge of expensive experiments and simulations for collecting data points, machine learning (ML) has been identified as a powerful technique. In this work, the machine learning-based prediction of hydrothermal performances of water-cooled dimpled ducts is presented. An artificial neural network (ANN) is used as the ML algorithm. The significance of the present study is to develop the ANN-based model using limited number of performance data without any existing relations/correlations between input variables and outputs. The input dataset for training the ANN model was prepared through a computational fluid dynamics (CFD) approach. The present study provides a practical insight to predict hydrothermal performance of any thermal management solution subject to limited available performance data and with a complex physics behind the operation of the system without any need to identify the key parameters affecting the physics of the problem.

Presenting Author: Daksh Adhikari Advanced Cooling Technologies, Inc.

Presenting Author Biography: Research and Development Engineer

Authors:

Mohammad Reza Shaeri Advanced Cooling Technologies, Inc.
Andoniaina M. Randriambololona University of Maryland
Daksh Adhikari Advanced Cooling Technologies, Inc.