Artificial Neural Network (FFBP-ANN) Based Grey Relational Analysis for Modeling Dyestuff Solubility in Supercritical CO2 with Ethanol as the Co-Solvent

Abstract

The research on dye solubility modeling in supercritical carbon dioxide is gaining prominence over the past few decades. A simple and ubiquitous model that is capable of accurately predicting the solubility in supercritical carbon dioxide would be invaluable for industrial and research applications. In this study, we present such a model for predicting dye solubility in supercritical carbon dioxide with ethanol as the co-solvent for a qualitatively diverse sample of eight dyes. A feed forward back propagation - artificial neural network model based on Levenberg-Marquardt algorithm was constructed with seven input parameters for solubility prediction, the network architecture was optimized to be [7-7-1] with mean absolute error, mean square error, root mean square error and Nash-Sutcliffe coefficient to be 0.026, 0.0016, 0.04 and 0.9588 respectively. Further, Pearson-product moment correlation analysis was performed to assess the relative importance of the parameters considered in the ANN model. A total of twelve prevalent semiempirical equations were also studied to analyze their efficiency in correlating to the solubility of the prepared sample. Mendez-Teja model was found to be relatively efficient with root mean square error and mean absolute error to be 0.094 and 0.0088 respectively. Furthermore, Grey relational analysis was performed and the optimum regime of temperature and pressure were identified with dye solubility as the higher the better performance characteristic. Finally, the dye specific crossover ranges were identified by analysis of isotherms and a strategy for class specific selective dye extraction using supercritical CO2 extraction process is proposed.

Seepage Time Soft Sensor Model of Nonwoven Fabric Based on the Extreme Learning Machine Integrating Monte Carlo

“…Due to the disadvantages of conventional methods and the rapid development of artificial intelligence, many neural network methods are used in dyeing cloth and related industries. Srinidhi et al put forward a Levenberg-Marquardt-based BP method to predict the dye solubility in supercritical carbon dioxide accurately, which overcame the problem in the dyeing industry [10]. Laura et al proposed a system that can generate color formula when combined with the BP algorithm, which could control small changes in color and reduce the subjectivity of visual evaluation of the color of textile products [11].…”

Response Surface Methodology and Artificial Neural Network Methods Comparative Assessment for Fuel Rich and Fuel Lean Catalytic Combustion

“…The number of input neurons signifies and is equal to the independent variables of the system and the number of output neurons denotes and is equal to the response of the system. All hidden units are attached to each input unit and to the output layer and there is no communication between neurons in the same layer but information only moves in the forward direction, from the input towards the output layer [32] [62] [67].…”

“…The output neurons add their own biases to the weighted sum they receive and return the network response for the input data provided to the network. The network parameters to be determined during the training process are the weights which characterize the connection between the neurons and the bias of each neuron [32] [42] [62].…”

“…Ayodele et al [31] demonstrated the robustness of back propagation artificial neural network for predictive modeling of photodegradation of organic pollutants beside the determination of the level of importance of the process parameters. Also, Srinidhi et al [32] stated that ANN is showing promise for tackling multivariate and complex modeling problems.…”