Chapter 15 - Applications of artificial intelligence techniques in heat exchanger systems

Abstract

The performance evaluation of a counter-flow heat exchanger is complex because of the unsteady and nonlinear nature of its operating conditions. Numerous artificial intelligence techniques are being introduced day by day. However, the search for simple and efficient algorithms still persists. In this chapter, an efficient neural network (NN)-based tool was developed for estimating the outlet temperatures of working fluid from such heat exchangers. For this purpose, a series of repeated experiments were conducted and compared with generalized regression neural network (GRNN) modeling. GRNN architecture proposed in this study consists of four inputs (cold fluid flow rate, cold fluid inlet temperature, hot fluid flow rate, and hot fluid inlet temperature) and two outputs (cold fluid outlet temperature, and hot fluid outlet temperature). Such GRNN architecture was trained, tested, and validated with real-time experimental data sets. The results of the NN model are in good agreement with experimental results. The overall accuracy of the proposed GRNN model in predicting the outputs are 98.50% and 96.46%.