Experimental Study and Modeling Approach of Response Surface Methodology Coupled with Crow Search Algorithm for Optimizi
- PDF / 992,053 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 19 Downloads / 201 Views
RESEARCH ARTICLE-CHEMICAL ENGINEERING
Experimental Study and Modeling Approach of Response Surface Methodology Coupled with Crow Search Algorithm for Optimizing the Extraction Conditions of Papaya Seed Waste Oil S. M. Z. Hossain1 · S. Taher1 · A. Khan1 · N. Sultana2 · M. F. Irfan1 · B. Haq3 · S. A. Razzak4 Received: 23 November 2019 / Accepted: 16 April 2020 © King Fahd University of Petroleum & Minerals 2020
Abstract Papaya seed waste can be a reliable feedstock for producing valuable bioproducts (biodiesel, biolubricants, beauty products, etc.) due to its high oil content. This article focuses to explore the effects of Soxhlet extraction process conditions (extraction time and seed particle size) on the percent oil yield obtained from papaya seeds. Initially, two mathematical models were developed using response surface methodology (RSM) via central composite design and regression analysis (generalized linear model, GLM) to predict the oil yield. The prediction performance of RSM model was found to be superior than GLM. The extracted oil was characterized by gas chromatography–mass spectrometry (GC–MS) analysis. The analysis of variance results indicated that both factors were strongly significant. Later, crow search algorithm (nature-motivated metaheuristic algorithm) articulated with RSM was utilized for global optimal solution. The maximum yield of 29.96% was obtained at extraction time of 6.5 h and seed particle size of 0.85 mm. The similar results were obtained by desirability function-based optimization approach. The predicted optimal set was also validated further by experimental yield of 31.1% with the variation of < 5%. Keywords Papaya seed waste oil · Solvent extraction · Optimization · Response surface methodology · Crow search algorithm
List of symbols OFAT RSM CCD BBD
One-factor-at-a-time Response surface methodology Central composite design Box–Behnken design
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s13369-020-04551-1) contains supplementary material, which is available to authorized users.
B
S. M. Z. Hossain [email protected]
1
Department of Chemical Engineering, College of Engineering, University of Bahrain, Zallaq, Kingdom of Bahrain
2
Department of Computer Science, College of Computer Science and Information Technology, Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia
3
Department of Petroleum Engineering, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia
4
Department of Chemical Engineering, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia
GLM A B ANOVA y CSA AP fl N GC–MS RE MAE RMSE DF
Generalized linear model Extraction time (T ) Seed particle size (S) Analysis of variance Predicted response Crow search algorithm Awareness probability Flight length Flock size Gas chromatography–mass spectrometry Relative error Mean absolute error Root mean squared error Desirability function
1 Introduction The leading contribution to the greenhouse effect is due to the immense usage
Data Loading...
