Performance assessment of heat exchanger using intelligent decision making tools

Process and manufacturing industries today are under pressure to deliver high quality outputs at lowest cost. The need for industry is therefore to implement cost savings measures immediately, in order to remain competitive. Organizations are making strenuous efforts to conserve energy and explore alternatives. This paper explores the development of an intelligent system to identify the degradation of heat exchanger system and to improve the energy performance through online monitoring system. The various stages adopted to achieve energy performance assessment are through experimentation, design of experiments and online monitoring system. Experiments are conducted as per full factorial design of experiments and the results are used to develop artificial neural network models. The predictive models are used to predict the overall heat transfer coefficient of clean/design heat exchanger. Fouled/real system value is computed with online measured data. Overall heat transfer coefficient of clean/design system is compared with the fouled/real system and reported. It is found that neural net work model trained with particle swarm optimization technique performs better comparable to other developed neural network models. The developed model is used to assess the performance of heat exchanger with the real/fouled system. The performance degradation is expressed using fouling factor, which is derived from the overall heat transfer coefficient of design system and real system. It supports the system to improve the performance by asset utilization, energy efficient and cost reduction in terms of production loss. This proposed online energy performance system is implemented into the real system and the adoptability is validated.