Rate-dependent elasto-viscoplastic constitutive model for industrial powders. Part 2: Model evaluation

B. Mittal, V. M. Puri

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

The PSU-EVP model's constitutive parameters for alumina powder are presented. The PSU-EVP model was also used to back-predict the triaxial test data obtained for MZF and alumina powders using constitutive parameters such as the initial voids ratio (e0), compression index (λ), and spring-back index (κ). In the case of MZF powder, 8 out of 12 back-prediction cases had average relative difference (ARD) values below 20%. In the case of alumina powder, 7 out of 11 back-prediction cases had ARD values below 20%. Based on the back-prediction results, it was concluded that the PSU-EVP model gave fairly good results for most triaxial test data collected at 0.62 MPa/minute and 6.21 MPa/minute. However, the back-prediction results obtained at 20.7 MPa/minute had high ARD values. A sensitivity analysis was done to study the effect of changes in parameter values on the hydrostatic triaxial compression (HTC) and conventional triaxial compression (CTC) back-prediction results. From the sensitivity analysis, ± 10% (standard deviation variation from ± 0.8σ to ± 2.3σ) changes in λ and e0 mean values had marked effect on the HTC results. However, changes in the λ, κ, and e0 mean values do not produce any noticeable effect on the CTC prediction results. Overall, the PSU-EVP model can be considered to be the first step towards the development of a more robust and accurate model for prediction of stresses and strains in a dry powder compression process.

Original languageEnglish (US)
Pages (from-to)39-57
Number of pages19
JournalParticulate Science and Technology
Volume24
Issue number1
DOIs
StatePublished - Jan 2006

All Science Journal Classification (ASJC) codes

  • General Chemical Engineering

Fingerprint

Dive into the research topics of 'Rate-dependent elasto-viscoplastic constitutive model for industrial powders. Part 2: Model evaluation'. Together they form a unique fingerprint.

Cite this