A switching decentralized and distributed extended Kalman filter for pressure swing adsorption processes. (28th December 2016)

• Record Type: Journal Article
• Authors: Fakhroleslam, Mohammad; Fatemi, Shohreh; Bozorgmehry Boozarjomehry, Ramin
• Is Part Of: International journal of hydrogen energy. Volume 41:Number 48(2016)
• Page Start: 23042
• Page End: 23056
• Publication Date: 2016-12-28
• Subjects: Pressure swing adsorption -- Observer -- Extended Kalman filter -- Distributed parameter systems -- Switching systems -- Hydrogen purification; Hydrogen as fuel -- Periodicals; Hydrogène (Combustible) -- Périodiques; Hydrogen as fuel; Periodicals; 665.81
• DOI: 10.1016/j.ijhydene.2016.08.120
• Languages: English
• ISSNs: 0360-3199

Abstract

Abstract: A continuous-discrete Distributed and Decentralized Switching Kalman Filter (DDSKF) is designed for estimation of spatial profiles in Pressure Swing Adsorption (PSA) processes. The introduced observer is an integral part of the control strategy of hybrid systems in general and PSA systems in particular. A reduced order model is developed based on the mechanistic model of the process. The sensors are optimally located and observability of the process is studied. The proposed observer is used to estimate the spatial profiles of various states of a two-bed, six-step PSA system used for production of pure H2 from a H2 –CH4 gas mixture. The spatial profiles of the system have been estimated using the proposed observer quite accurately and rapidly based on noise corrupted measured temperature and pressure of a few points in the process. Highlights: A mechanistic model was developed for a hydrogen purification PSA process. A continuous-discrete distributed and decentralized switching observer was designed. Cascade sub-observers reduced the complexity and computational cost of the observer. The sensors were optimally located and observability of the process was studied. All spatial profiles were estimated based on noisy temperature and pressure measurements.