Application of Statistical and Machine Learning Techniques for Laboratory-Scale Pressure Filtration

Modeling and Analysis of Cake Moisture

Gireesh S.S. Raman, Mark Stephen Klima

Research output: Contribution to journalArticle

Abstract

This study deals with the modeling and analysis of the pressure filtration process using statistical and machine learning techniques. The effects of externally controllable process-influencing factors such as pressure, pH, temperature, solids concentration, filtration time, air-blow time, and cake thickness on filtration performance, measured in terms of cake moisture, were modeled. A 9-factor regression model based on an exhaustive search algorithm and a 7-6-1 artificial neural network (ANN) model based on a resilient backpropagation algorithm were developed and gave R 2 values of 0.84 and 0.94, respectively. Relative importance of input variables was analyzed using novel methods such as added-variable plots based on the regression model and Olden’s method based on the ANN model. Results from both methods established a negative correlation for pressure, solids concentration, filtration time, temperature, and air-blow time and a positive correlation for cake thickness and pH. Analysis from regression and ANN models indicated pH to be the most significant process-influencing factor. Even though both models served as good interpretable models, the ANN model outperformed the regression model in terms of predictive capability, with an R 2 value of 0.965 compared with the regression model’s 0.750 for the test dataset.

Original languageEnglish (US)
Pages (from-to)148-155
Number of pages8
JournalMineral Processing and Extractive Metallurgy Review
Volume40
Issue number2
DOIs
StatePublished - Mar 4 2019

Fingerprint

Learning systems
Moisture
moisture
modeling
artificial neural network
Neural networks
laboratory
analysis
machine learning
Backpropagation algorithms
air
Air
temperature
Temperature

All Science Journal Classification (ASJC) codes

  • Chemistry(all)
  • Geotechnical Engineering and Engineering Geology
  • Mechanical Engineering
  • Economic Geology

Cite this

@article{131b24cef1e741f39e78325841c69dd5,
title = "Application of Statistical and Machine Learning Techniques for Laboratory-Scale Pressure Filtration: Modeling and Analysis of Cake Moisture",
abstract = "This study deals with the modeling and analysis of the pressure filtration process using statistical and machine learning techniques. The effects of externally controllable process-influencing factors such as pressure, pH, temperature, solids concentration, filtration time, air-blow time, and cake thickness on filtration performance, measured in terms of cake moisture, were modeled. A 9-factor regression model based on an exhaustive search algorithm and a 7-6-1 artificial neural network (ANN) model based on a resilient backpropagation algorithm were developed and gave R 2 values of 0.84 and 0.94, respectively. Relative importance of input variables was analyzed using novel methods such as added-variable plots based on the regression model and Olden’s method based on the ANN model. Results from both methods established a negative correlation for pressure, solids concentration, filtration time, temperature, and air-blow time and a positive correlation for cake thickness and pH. Analysis from regression and ANN models indicated pH to be the most significant process-influencing factor. Even though both models served as good interpretable models, the ANN model outperformed the regression model in terms of predictive capability, with an R 2 value of 0.965 compared with the regression model’s 0.750 for the test dataset.",
author = "Raman, {Gireesh S.S.} and Klima, {Mark Stephen}",
year = "2019",
month = "3",
day = "4",
doi = "10.1080/08827508.2018.1497628",
language = "English (US)",
volume = "40",
pages = "148--155",
journal = "Mineral Processing and Extractive Metallurgy Review",
issn = "0882-7508",
publisher = "Taylor and Francis Ltd.",
number = "2",

}

TY - JOUR

T1 - Application of Statistical and Machine Learning Techniques for Laboratory-Scale Pressure Filtration

T2 - Modeling and Analysis of Cake Moisture

AU - Raman, Gireesh S.S.

AU - Klima, Mark Stephen

PY - 2019/3/4

Y1 - 2019/3/4

N2 - This study deals with the modeling and analysis of the pressure filtration process using statistical and machine learning techniques. The effects of externally controllable process-influencing factors such as pressure, pH, temperature, solids concentration, filtration time, air-blow time, and cake thickness on filtration performance, measured in terms of cake moisture, were modeled. A 9-factor regression model based on an exhaustive search algorithm and a 7-6-1 artificial neural network (ANN) model based on a resilient backpropagation algorithm were developed and gave R 2 values of 0.84 and 0.94, respectively. Relative importance of input variables was analyzed using novel methods such as added-variable plots based on the regression model and Olden’s method based on the ANN model. Results from both methods established a negative correlation for pressure, solids concentration, filtration time, temperature, and air-blow time and a positive correlation for cake thickness and pH. Analysis from regression and ANN models indicated pH to be the most significant process-influencing factor. Even though both models served as good interpretable models, the ANN model outperformed the regression model in terms of predictive capability, with an R 2 value of 0.965 compared with the regression model’s 0.750 for the test dataset.

AB - This study deals with the modeling and analysis of the pressure filtration process using statistical and machine learning techniques. The effects of externally controllable process-influencing factors such as pressure, pH, temperature, solids concentration, filtration time, air-blow time, and cake thickness on filtration performance, measured in terms of cake moisture, were modeled. A 9-factor regression model based on an exhaustive search algorithm and a 7-6-1 artificial neural network (ANN) model based on a resilient backpropagation algorithm were developed and gave R 2 values of 0.84 and 0.94, respectively. Relative importance of input variables was analyzed using novel methods such as added-variable plots based on the regression model and Olden’s method based on the ANN model. Results from both methods established a negative correlation for pressure, solids concentration, filtration time, temperature, and air-blow time and a positive correlation for cake thickness and pH. Analysis from regression and ANN models indicated pH to be the most significant process-influencing factor. Even though both models served as good interpretable models, the ANN model outperformed the regression model in terms of predictive capability, with an R 2 value of 0.965 compared with the regression model’s 0.750 for the test dataset.

UR - http://www.scopus.com/inward/record.url?scp=85052086221&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052086221&partnerID=8YFLogxK

U2 - 10.1080/08827508.2018.1497628

DO - 10.1080/08827508.2018.1497628

M3 - Article

VL - 40

SP - 148

EP - 155

JO - Mineral Processing and Extractive Metallurgy Review

JF - Mineral Processing and Extractive Metallurgy Review

SN - 0882-7508

IS - 2

ER -