Sampling Frequency and Digital Noise Filtering for Transient Kinetic Study in CSTR via Temperature Scanning and Composition Modulation Technique

Main Article Content

Amaraporn Kaewchada
Kwanhatai Jiradechkhajorn
Attasak Jaree

Abstract

Temperature Scanning and Composition Modulation technique (TS-CM) is a technique for estimating reaction parameters under transient conditions in a CSTR reactor. Studies of reaction kinetics performed under transient conditions can lead to savings of time and resources as compared to the conventional steady-state method. This is because a collection of data necessary for extracting the reaction kinetics can be obtained within a single experiment. This research focused on the refinement of TS-CM technique as an efficient tool for reaction kinetics studies. The alkaline hydrolysis of ethyl acetate was used as a model reaction. Different degrees of noise intensity were added to the output signal of NaOH concentration in order to study the effect of noise (0.5, 1.0, 3.0 and 5.0%) and sampling period (1, 30 and 60 seconds) on the accuracy of reaction kinetics. Results revealed that the sampling period of 1 second yielded only slight errors of kinetic parameters compared to that of without noise.The second part dealt with noise filtering. In order to smooth the NaOH concentration data, TS-CM technique was applied with three digital filters: a) the first-order filter or first-order filter, b) the second-order filter or second-order filter and c) the Savitzky-Golay filter. Results showed that the best filter was the Savitzky-Golay, which filtered with 81 points of window size and 2nd order of polynomial degree.

Article Details

How to Cite
Kaewchada, A., Jiradechkhajorn, K., & Jaree, A. (2015). Sampling Frequency and Digital Noise Filtering for Transient Kinetic Study in CSTR via Temperature Scanning and Composition Modulation Technique. Applied Science and Engineering Progress, 8(4), 283–292. Retrieved from https://ph02.tci-thaijo.org/index.php/ijast/article/view/67487
Section
Research Articles