Abstract
This paper considers the on-line implementation of the modulating function method, for parameter and state estimation, for the model of an air-handling unit, the central element of HVAC systems. After recalling the few elements of the method, more attention is paid on issues related to its on-line implementation, issues for which we use two different techniques. Experimental results are obtained after implementation of the algorithms on a heat flow experiment, and they are compared with conventional techniques (conventional tools from Matlab for parameter estimation, and a simple Luenberger observer for state estimation) for their validation.
Highlights
The challenges of the clear evidence of climate change (Edenhofer and Seyboth, 2013) bring in the front-line new regulations which aim to reduce the green-house gas emissions in all sectors
Different models have been proposed in the literature for modeling an air handling unit (AHU): physical models described by partial differential equations used by most of the simulation tools available (Crawley et al, 2008), transfer functions (Afroz et al, 2017), data-driven models (Afram and Janabi-Sharifi, 2014), and hybrid models or reduced-order models (ROM) (Afram and Janabi-Sharifi, 2015)
We investigate the use of the so-called modulating function method to perform both state and parameter estimation on a continuous-time ROM- based on a simple thermal-electrical analogy (Fraisse et al, 2002)
Summary
The challenges of the clear evidence of climate change (Edenhofer and Seyboth, 2013) bring in the front-line new regulations which aim to reduce the green-house gas emissions in all sectors. Different models have been proposed in the literature for modeling an AHU: physical models described by partial differential equations used by most of the simulation tools available (Crawley et al, 2008), transfer functions (Afroz et al, 2017), data-driven models (Afram and Janabi-Sharifi, 2014), and hybrid models or reduced-order models (ROM) (Afram and Janabi-Sharifi, 2015) The latter lead to many possibilities as they link to estimation and control scenarios where realtime capabilities are important. The modulating function technique, originally introduced by Shinbrot (Shinbrot, 1957), allows to circumvent this issue elegantly, with the use of fixed-length time integrals of the measured signals, akin to continuous-time finite impulse response (FIR) filtering This offers the additional advantages of giving estimates after a fixed and predetermined amount of time, contrary to usual Kalman-based or observer-based methods.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
