Determination of Improved Efficiency in Shell and Tube Heat Exchangers by Nature Inspired Optimization Techniques

Abstract

The heat exchanger is a multipurpose energy transport system that needs periodic newlineimprovements to enhance its performance, and profitability of enterprises. The objective newlineof this research is to delineate a tube heat exchanger with different methodologies and soft computing techniques to improve the efficiency of existing heat exchangers. The Organic Rankine Cycle (ORC) is an advanced innovation for producing power from low and moderate temperature heat sources. The cycle parameters for STHX are propelled with ORC technology. The nature-based-optimisation (NBO) methods are employed in different studies to tackle heat transfer and energy issues for thermal systems. In this research, prospective NBO techniques are proposed for the model to foresee the execution of their set-up in various working conditions. The design setup is segmented by potential NBO methods such as GA, PSO, GWO, IGWO, and KHO algorithms. A comprehensive investigation is conducted to study significant outputs like exergetic plant efficiency, energetic cycle efficiency, electrical energy, cost and fouling rate. The output parameters are assessed for optimal input configuration using tube setup as well as different liquids and temperature. The exergetic plant efficiency, energetic cycle efficiency, and electrical energy achieved the accuracy of 98.11%, 97.40% and 96.35 % respectively with back propagation network. The model also forecasts considerable decrease in heat exchanger cost and fouling rate. The accuracy levels achieved are 99.11%, for exergetic plant efficiency, 97.40% for energetic cycle efficiency and 98.35% for electrical energy with KHO strategy. The KHO strategy predicts better results and higher accuracy makes it an efficient optimisation tool for heat exchanger investigation. The optimisation methods are implemented using MATLAB codes and finally executes with high performance INTEL processor with 4GB RAM.