Comparison of Response Surface Methodology (Rsm) and Artificial Neural Networks (Ann) Towards Efficient Optimization of Flexural Properties of Gypsum-Bonded Fiberboards

Nazerian, Morteza; Kamyabb, Meysam; Shamsian, Mohammad; Dahmardehb, Mohammad; Kooshaa, Mojtaba

doi:10.1590/01047760201824012484

Cited by 30 publications

(11 citation statements)

References 22 publications

(19 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It is noticed that RSM values were lower for RMSE, MPE, MAE and SEP. This indicates that RSM model is more precise in predicting specific energy requirement for bio-fiber comminution process; although this finding is contrary to those reported elsewhere [29], it may be attributed to the finite experimental data used in network training.…”

Section: Comparison Of Response Surface Methods and Artificial Neural contrasting

confidence: 96%

Predictive Energy Requirement Models in Bio-fiber Comminution Process

2020

View full text Add to dashboard Cite

Comminution of bio-fibers is necessary for their effective application in development of reinforced composite and feedstock; however, comminution is an energy intensive process and minimization of specific energy required in the size reduction of bio-fibers is an optimization task in design of various particulate based materials. The objective of this study was to model specific energy requirement in bio-fiber comminution process. The banana and coir bio-fibers used in the study were respectively extracted from post-harvest banana stem and outer shell of a coconut. Characterization of treated bio-fibers was done using Fourier Transform Infrared Spectroscopy and comminution process was carried out in a laboratory-scale knife mill. Electrical energy was recorded using a plug-in power energy meter while Kick's, Rittinger's and Bond's energy laws provide insight into the relationship between particle size reduction and energy consumption during comminution process. The Rosin Rammler-Bennett (RRB) and Gaudin-Schuhmann (GS) models were fitted with MATLAB ® R2014a. The Artificial Neural Networks (ANN) and Response Surface Method (RSM) was utilised in predicting specific energy requirements as controlled by fiber treatment conditions, milling speed, initial fiber length and drying temperature. The predictive accuracy of the RSM and ANN models were adjudged adequate in terms of the process independent variables. A global optimal condition was found at 2% NaOH concentration, 29 mm fiber length, 1981 rpm mill speed and 41 °C drying temperature. At these conditions, specific energy was 1.430 kJ/kg for banana fiber and 1.109 kJ/ kg for coir fiber having desirability of unity.

show abstract

Section: Comparison Of Response Surface Methods and Artificial Neural contrasting

confidence: 96%

Predictive Energy Requirement Models in Bio-fiber Comminution Process

2020

View full text Add to dashboard Cite

show abstract

“…Table 8 shows the error comparison obtained from both and ANN predictions. The comparative error analysis was used to verify the prediction accuracy and generalization capacity of both models in optimizing the bioprocess 53 , 54 . Overall, the ANN model showed lower error values than the RSM, indicating lower computational deviations and an advanced generalization capability 11 , 54 .…”

Section: Resultsmentioning

confidence: 99%

A modeling method for the development of a bioprocess to optimally produce umqombothi (a South African traditional beer)

Hlangwani

Doorsamy

Adebiyi

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Bioprocess development for umqombothi (a South African traditional beer) as with other traditional beer products can be complex. As a result, beverage bioprocess development is shifting towards new systematic protocols of experimentation. Traditional optimization methods such as response surface methodology (RSM) require further comparison with a relevant machine learning system. Artificial neural network (ANN) is an effective non-linear multivariate tool in bioprocessing, with enormous generalization, prediction, and validation capabilities. ANN bioprocess development and optimization of umqombothi were done using RSM and ANN. The optimum condition values were 1.1 h, 29.3 °C, and 25.9 h for cooking time, fermentation temperature, and fermentation time, respectively. RSM was an effective tool for the optimization of umqombothi’s bioprocessing parameters shown by the coefficient of determination (R2) closer to 1. RSM significant parameters: alcohol content, total soluble solids (TSS), and pH had R2 values of 0.94, 0.93, and 0.99 respectively while the constructed ANN significant parameters: alcohol content, TSS, and viscosity had R2 values of 0.96, 0.96, and 0.92 respectively. The correlation between experimental and predicted values suggested that both RSM and ANN were suitable bioprocess development and optimization tools.

show abstract

“…The Levenberg-Marquardt algorithm (LM), known to be the fastest back-propagation algorithm, was employed. In particular, the LM shows outstanding performance in nonlinear regression problems and is well-suited for mean-squared error training neural networks [34,35]. The activation function consists of a nonlinear neural network, the sigmoid function is applied to the hidden layer, and the purlin function is applied to the output layer as a linear activation function.…”

Section: Ann Analysis Resultsmentioning

confidence: 99%

Comparison between Multiple Regression Analysis, Polynomial Regression Analysis, and an Artificial Neural Network for Tensile Strength Prediction of BFRP and GFRP

Kim

2021

Materials

View full text Add to dashboard Cite

In this study, multiple regression analysis (MRA) and polynomial regression analysis (PRA), which are traditional statistical methods, were applied to analyze factors affecting the tensile strength of basalt and glass fiber-reinforced polymers (FRPs) exposed to alkaline environments and predict the tensile strength degradation. The MRA and PRA are methods of estimating functions using statistical techniques, but there are disadvantages in the scalability of the model because they are limited by experimental results. Therefore, recently, highly scalable artificial neural networks (ANN) have been studied to analyze complex relationships. In this study, the prediction performance was evaluated in comparison to the MRA, PRA, and ANN. Tensile strength tests were conducted after exposure for 50, 100, and 200 days in alkaline environments at 20, 40, and 60 °C. The tensile strength was set as the dependent variable, with the temperature (TP), the exposure day (ED), and the diameter (D) as independent variables. The MRA and PRA results showed that the TP was the most influential factor in the tensile strength degradation of FRPs, followed by the exposure time (ED) and diameter (D). The ANN method provided the best correlation between predictions and experimental values, with the lowest error and error rate. The PRA method applied to the response surface method outperformed the MRA method, which is most commonly used. These results demonstrate that ANN can be the most efficient model for predicting the durability of FRPs.

show abstract

Comparison of Response Surface Methodology (Rsm) and Artificial Neural Networks (Ann) Towards Efficient Optimization of Flexural Properties of Gypsum-Bonded Fiberboards

Cited by 30 publications

References 22 publications

Predictive Energy Requirement Models in Bio-fiber Comminution Process

Predictive Energy Requirement Models in Bio-fiber Comminution Process

A modeling method for the development of a bioprocess to optimally produce umqombothi (a South African traditional beer)

Comparison between Multiple Regression Analysis, Polynomial Regression Analysis, and an Artificial Neural Network for Tensile Strength Prediction of BFRP and GFRP

Contact Info

Product

Resources

About