Renewable Energy Global Innovations features: Optimization of A Three-Bed Adsorption Chiller by Genetic Algorithms and Neural Networks

Significance Statement

Recovery of waste heat is one of the main techniques of improving maximum energy efficiency utilization of a variety of processes with low parameters of heat generated as by-product. The use of waste heat driven pumps is slowly overtaking the predominantly applied mechanical coolers. Increased attention is currently being paid to adsorption chillers since they can be powered with low-temperature heat sources and yet allow to be integrated into cogenerative systems. Adsorption cycles of the applied systems have a distinctive advantage over other systems in that they can use low grade waste heat of near ambient temperature. A Tri-bed twin-evaporator adsorption chiller comprise of a ground-breaking design in cooling production which allows more efficient conversion and management of low grade sources of thermal energy due to more effective way of utilization adsorptive abilities of the beds during a single work. Although it is the most effective way in chilled water production, the intricacy of the Tri-bed twin evaporator adsorption chiller operation is still not sufficiently recognized and the enhancement in cooling capacity of the cooler is still a puzzling task.

A team of researchers led by professor Jaroslaw Krzywanski at the Jan Dlugosz University in Poland optimized a three-bed adsorption chiller by applying genetic algorithms and neural networks. They introduced an artificial intelligence approach for the optimization study of the Tri-bed twin evaporator adsorption chiller using low-temperature heat from cogeneration. Their research work is now published in the peer-reviewed journal, Energy Conversion and Management.

The research team began by developing genetic algorithms and artificial neural networks. The developed algorithm and network were then tested and validated before they were used to develop the model. The researchers then used the developed model to estimate the behavior of the adsorption heat pump by assessing the effects of: the time cycle, cooling and heating inlet water temperatures and that of temperatures in low-high pressure evaporators. The team also examined the cooling capacity as one of the major energy efficiency factors in cooling production during the study for various scenarios.

The authors also observed that the highest value which could be obtained for the considered range of input operational parameters was equal to 93 kW. It was also noted that such a value was only attainable where specific: cycle time, cooling water temperature, heating water temperature, high pressure inlet temperature and low pressure inlet temperatures as specified in this paper were used.

Results of their study showed the cooling capacity evaluated using the model, is in good agreement with the experimental data. The maximum relative error between the measured and calculated results is lower than ±10%. Therefore, the developed model in Krzywanski  and colleagues is an easy to use and powerful optimization tool which allows to estimate the cooling capacity of the Tri-bed twin evaporator adsorption chiller, integrated into multi-generative systems.

The structure of the [5-2-2-1] type of neural network for optimization of a Tri-bed twin evaporator adsorption chiller

About The Author

Jaroslaw Krzywanski is an Associate Professor the head the Division of Advanced Computational Methods at the Faculty of Mathematics and Natural Science of Jan Dlugosz University in Czestochowa, Poland.

He received the M.Sc. degree from Czestochowa University of Technology, Department of Mechanical Engineering and Computer Sciences, Institute of Thermal Machinery, Poland and Ph.D. degree from Silesian University of Technology, Faculty of Energy and Environmental Engineering, Poland.

He has published more than 120 refereed works, including papers, a monograph, conference proceedings and serves as an editorial board member of several international journals. He has participated in the scientific committee of several conferences and serves as a reviewer in a wide range of international journals.

He is interested in modeling of energy devices and processes, including solid fuels combustion, waste driven adsorption chillers as well as gas emissions and hydrogen production from biomass combustion and gasification, respectively.

Reference

Krzywanski, K. Grabowska, F. Herman, P. Pyrka, M. Sosnowski, T. Prauzner, W. Nowak. Optimization of a three-bed adsorption chiller by genetic algorithms and neural networks. Energy Conversion and Management, volume 153 (2017) pages 313–322.

Go To Energy Conversion and Management Read more research excellence studies on: Renewable Energy Global Innovations (http://ift.tt/21cCPA4)