A platform for research: civil engineering, architecture and urbanism
Dehumidifier desiccant concentration soft-sensor for a distributed operating Liquid Desiccant Dehumidification System
Highlights A distributed operating LDDS is proposed. The working principle is explained and the significance of soft-sensor is emphasized. This ANFIS structured soft-sensor is simplified by a physical model and GA. This soft-sensor is validated and proved to be accurate.
Abstract Distributed operating of Liquid Desiccant Dehumidification System (LDDS) increases the flexibility in system operation, allowing one regenerator to handle multiple dehumidifier units. To meet the requirement of this operating scheme, a soft-sensor is developed for the real-time measurement of the desiccant solution concentration. A physical model of mass transfer rate in the dehumidifier is employed to reduce the number of input variables of the Adaptive Network-based Fuzzy Inference Systems (ANFIS) structure, and then the Genetic Algorithm (GA) is selected to further simplify the fuzzy interference with a constrained objective function. Owing to these simplifications, an accurate model-based soft-sensor with a concise ANFIS structure has been formulated. Concentration values obtained by this soft-sensor are validated by experimental data to prove its effectiveness and results show that the proposed method can acquire the online concentration accurately which will be beneficial in the system performance monitoring, control or optimization.
Dehumidifier desiccant concentration soft-sensor for a distributed operating Liquid Desiccant Dehumidification System
Highlights A distributed operating LDDS is proposed. The working principle is explained and the significance of soft-sensor is emphasized. This ANFIS structured soft-sensor is simplified by a physical model and GA. This soft-sensor is validated and proved to be accurate.
Abstract Distributed operating of Liquid Desiccant Dehumidification System (LDDS) increases the flexibility in system operation, allowing one regenerator to handle multiple dehumidifier units. To meet the requirement of this operating scheme, a soft-sensor is developed for the real-time measurement of the desiccant solution concentration. A physical model of mass transfer rate in the dehumidifier is employed to reduce the number of input variables of the Adaptive Network-based Fuzzy Inference Systems (ANFIS) structure, and then the Genetic Algorithm (GA) is selected to further simplify the fuzzy interference with a constrained objective function. Owing to these simplifications, an accurate model-based soft-sensor with a concise ANFIS structure has been formulated. Concentration values obtained by this soft-sensor are validated by experimental data to prove its effectiveness and results show that the proposed method can acquire the online concentration accurately which will be beneficial in the system performance monitoring, control or optimization.
Dehumidifier desiccant concentration soft-sensor for a distributed operating Liquid Desiccant Dehumidification System
Wu, Qiong (author) / Cai, WenJian (author) / Wang, Xinli (author) / Chakraborty, Anutosh (author)
Energy and Buildings ; 129 ; 215-226
2016-07-25
12 pages
Article (Journal)
Electronic Resource
English
LDDS , Dehumidifier , Concentration , Physical model , Soft-sensor , ANFIS , GA
Progress in selecting desiccant and dehumidifier for liquid desiccant cooling system
Online Contents | 2012
|