Lessen Pressure Drop and Forecasting Thermal Performance in U-Tube Heat Exchanger Using Chimp Optimization and Deep Belief Neural Network

Abstract

In the chemical, pharmaceutical, and petroleum industries, Shell and U-Tube Heat Exchangers (STHX) were extensively utilized. Baffles must be positioned at the right distance and angle to increase the heat exchangers' capacity to convey heat and, as a result, lower pressure in the shell. The rate of heat transfer in an STHX has been improved, and pressure drop has been reduced using a variety of models. But those methods are not provided satisfactory pressure drop reduction. In the proposed model, an optimal Unilateral Ladder-Type Helical Baffles (ULHB) design and intelligent performance prediction system based U-tube heat exchanger was designed to reduce the pressure drop as well as predict the heat exchanger performance. The shell and tubes were made up of steel and copper material, respectively. A baffle was placed above tubes to barrier the flow of cold water. The design of the baffle was accomplished by using Chimp Optimization Algorithm (ChOA) and is motivated by the hunting behaviour of chimpanzees. After designing the exchanger, its fluid analysis was verified, and the parameter values of the heat exchanger were collected to create a dataset. Based on that data, the intelligent performance prediction-system was designed. The controlling system analysed the given data to predict the performance of the heat exchanger. The suggested model has a pressure drop of 55 Pa, a heat transfer coefficient of 411 U, and 86% accuracy for the thermal performance prediction process. The proposed model provides better performance by improving heat transfer efficiency and significantly reduces pressure drop.