Mixture formulation through multivariate statistical analysis of process data in property cluster space

1Citations
Citations of this article
21Readers
Mendeley users who have this article in their library.

Abstract

Data-driven modeling approaches are suitable for representing complex processes and phenomena in cases where cause-and-effect cannot be easily described from first-principles. Chemical product formulation in industrial research and development is an area where the analysis of mixture data could be utilized more effectively. Correlation, either partial or complete, is inherent in such mixture data and requires the use of multivariate statistical tools for visualization and identification of important relationships in the data. In this paper, a systematic methodology is developed by integrating data-driven chemometric techniques and property based visualization and optimization tools to solve mixture formulation problems involving multi-block data structures. Effort has been focused on: the development of mathematical models by utilizing multivariate understanding of process and product data, visually identifying design targets a priori, and decomposition of the design problem by incorporating the concept of reverse problem formulation and property clustering techniques. A case study in industrial thermo-plastic development is presented to illustrate the methodology developed in this paper.

Cite

CITATION STYLE

APA

Hada, S., Herring, R. H., & Eden, M. R. (2017). Mixture formulation through multivariate statistical analysis of process data in property cluster space. Computers and Chemical Engineering, 107, 26–36. https://doi.org/10.1016/j.compchemeng.2017.06.017

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free