Support vector machine classification applied to the parametric design of centrifugal pumps. Issue 8 (3rd August 2018)

• Record Type: Journal Article
• Authors: Riccietti, E.; Bellucci, J.; Checcucci, M.; Marconcini, M.; Arnone, A.
• Is Part Of: Engineering optimization. Volume 50:Issue 8(2018)
• Page Start: 1304
• Page End: 1324
• Publication Date: 2018-08-03
• Subjects: Support vector machines -- parametric design -- binary classification -- centrifugal pumps -- unbalanced datasete; Engineering design -- Periodicals; Mathematical optimization -- Periodicals; 620.0042
• DOI: 10.1080/0305215X.2017.1391801
• Languages: English
• ISSNs: 0305-215X

Abstract

ABSTRACT: In this article the parametric design of centrifugal pumps is addressed. To deal with this problem, an approach based on coupling expensive Computational Fluid Dynamics (CFD) computations with artificial neural networks as a regression meta-model was proposed in 2015 by Checcucci, Schneider, Marconcini, Rubechini, Arnone, De Franco, and Coneri, 'A novel approach to parametric design of centrifugal pumps for a wide range of specific speeds'—Proceedings of the 12th international symposium on experimental and computational aerothermodynamics of internal flows, Lerici (SP), Italy. Paper No. 121. Here, the previously proposed approach is improved by also including the use of support vector machines as a classification tool. The classification process is aimed at identifying parameter combinations corresponding to manufacturable machines among the much larger number of unfeasible ones. A binary classification problem on an unbalanced dataset has to be faced. Numerical tests show that the addition of this classification tool helps to reduce considerably the number of CFD computations required for the design, providing large savings in computational time.