Premium
NPK 2.0: Introducing tensor decompositions to the kinetic analysis of gas–solid reactions
Author(s) -
Birkelbach Felix,
Deutsch Markus,
Flegkas Stylianos,
Winter Franz,
Werner Andreas
Publication year - 2019
Publication title -
international journal of chemical kinetics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.341
H-Index - 68
eISSN - 1097-4601
pISSN - 0538-8066
DOI - 10.1002/kin.21251
Subject(s) - initialization , tensor (intrinsic definition) , mathematics , a priori and a posteriori , kinetic energy , rank (graph theory) , matlab , set (abstract data type) , mathematical optimization , nonparametric statistics , chemistry , computer science , algorithm , statistics , philosophy , physics , epistemology , quantum mechanics , combinatorics , pure mathematics , programming language , operating system
A method for deriving kinetic models of gas–solid reactions for reactor and process design is presented. It is based on the nonparametric kinetics (NPK) method and resolves many of its shortcomings by applying tensor rank‐1 approximation methods. With this method, it is possible to derive kinetic models based on the general kinetic equation from any combination of experiments without additional a priori assumptions. The most notable improvements over the original method are that it is computationally much simpler and that it is not limited to two variables. Two algorithms for computing the rank‐1 approximation as well as a tailored initialization method are presented, and their performance is assessed. Formulae for the variance estimation of the solution values are derived to improve the accuracy of the model identification and to provide a tool for diagnosing the quality of the kinetic model. The methods effectiveness and performance are assessed by applying it to a simulated data set. A Matlab implementation is available as Supporting Information.