Performance investigation of micro- and nano-sized particle erosion in a 90° elbow using an ANFIS model

Shahaboddin Shamshirband, Amir Malvandi, Arash Karimipour, Marjan Goodarzi, Masoud Afrand, Dalibor Petković, Mahidzal Dahari, Naghmeh Mahmoodian

Research output: Contribution to journalArticlepeer-review

Abstract

The accuracy of soft computing technique was employed to predict the performance of micro- and nano-sized particle erosion in a 3-D 90° elbow. The process, capable of simulating the total and maximum erosion rate with adaptive neuro-fuzzy inference system (ANFIS), was constructed. The developed ANFIS network was with three neurons in the input layer, and one neuron in the output layer. The inputs included particle velocity, particle diameter, and volume fraction of the copper particles. The size of these particles was selected in the range of 10nm to 100μm. Numerical simulations have been performed with velocities ranging from 5 to 20m/s and for volume fractions of up to 4%. The governing differential equations have been discretized by the finite volume method for ANFIS training data extraction. The ANFIS results were compared with the CFD results using root-mean-square error (RMSE) and coefficient of determination (R2). The CFD results show that an improvement in predictive accuracy and capability of generalization can be achieved by the ANFIS approach. The following characteristics were obtained: ANFIS model can be used to forecast the maximum and total erosion rate with high reliability and therefore can be applied for practical purposes.

Original languageEnglish (US)
Pages (from-to)336-343
Number of pages8
JournalPowder Technology
Volume284
DOIs
StatePublished - Nov 1 2015
Externally publishedYes

Keywords

  • ANFIS
  • CFD
  • Erosion corrosion
  • Estimation

ASJC Scopus subject areas

  • General Chemical Engineering

Fingerprint

Dive into the research topics of 'Performance investigation of micro- and nano-sized particle erosion in a 90° elbow using an ANFIS model'. Together they form a unique fingerprint.

Cite this