Modelling and optimization for a novel liquid desiccant dehumidification system

• Main Author: Zhou, Siqi
• Other Authors: Cai Wenjian
• Format: Theses and Dissertations
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/72571
• Institution: Nanyang Technological University

With the improvement of people’s life in modern society, requirements on Indoor Air Quality (IAQ), especially on the indoor air temperature and humidity are increasing. Meanwhile, conventional energy inefficient air conditioning systems consume lots of energy during the process of air humidity control. Liquid desiccant dehumidification system (LDDS) can independently control air temperature and humidity, save operating cost and utilize low grade energy in process of desiccant solution regeneration. With these advantages, LDDS has been considered as a novel and comprising air conditioning system which gains much attention from the investigators. This dissertation concerns the design and fabrication of a novel and energy efficient LDDS. Studies on system modelling and real-time optimization technologies are carried out. The developed real-time optimization strategies are successfully applied on a real LDDS to significantly improve system performance and energy efficient and fully explore its potential in energy savings. This research offers new research ideas and interests on the area of liquid desiccant air conditioning. The main innovations and contributions in this project are listed as follows: (a). Design and develop a novel energy efficient LDDS. To overcome the shortcomings of the existing LDDS, the ideas of heat pipe energy recovery, energy storage and hybrid system operating mode are designed into LDDS. Compared with the existing system, the newly designed system can improve the performance of energy efficiency, dehumidification efficiency and applications in LDDS. (b). Starting from the energy and mass conservation and the fundamental of heat and mass transfer, analysis of the heat and mass transfer process in both the dehumidifier and the regenerator is carried out to develop the LDDS models by hybrid methods. Experiment results show that the prediction relative errors of proposed models are within 15%. The proposed models have the advantages of simplicity, lower computing burden, no iterative calculation and high accuracy, which can be utilized in system performance prediction and real-time optimization for LDDS. (c). Develop the real-time optimization strategy of dehumidifier and apply the strategy into a real LDDS. The hybrid energy models are built for the chiller, the dehumidification fan and pump. The energy consumption of the dehumidifier is considered as the objective function. Desiccant solution flow rate and temperature are selected as the optimization variables. Experimental results verify that the realtime optimization strategy for the dehumidifier can achieve the system energy savings by 12%. Therefore, the energy efficiency of LDDS is significantly improved and its potential of energy savings is fully explored. (d). Develop the multi-objective optimization strategy for regenerator and employ the multi-objective optimization method to analyse optimization of regenerator. By analysing the feature of the adiabatic regenerator and components’ characteristics, the desiccant regeneration rate and energy consumption are considered as the two objective functions to be optimized. Desiccant solution flow rate and temperature, regenerating airflow rate are the three optimization variables for the formulated multi-objective problem of the regenerator. Experimental results show that when ambient air temperature is higher, the presented strategy can achieve energy savings up to 17.7%. After analysing different time segments, it is observed that the higher the temperature of outdoor air, the lower the relative humidity. In this Project, firstly I designed 3D and 2D models for LDDS using Solid-works and AutoCAD according to the results of the heat and mass transfer of both dehumidifiers and regenerators. The detailed parameters will be obtained from experimental data during the actual running process. Then in order to optimize LDDS, I created the hybrid models for chillers, pumps and fans in dehumidifier and regenerator, respectively. And within the help of my supervisor, the prediction relative errors of proposed models are within 15% which means that the optimization strategy can be applied in LDDS. Finally, I utilized the created models to predict the energy consumption during different time periods in running process. The results show that the new LDDS is more energy-saving than the original one. Keywords: Liquid desiccant; model and real-time optimization; heat and mass transfer model; energy-saving optimization