Comparative Study between Neural Network Model and Mathematical Models for Prediction of Glucose Concentration during Enzymatic Hydrolysis

Abstract

Enzymatic hydrolysis of cellulose is a complex process because of a number of inhibition and enzyme inactivation reactions which happen during hydrolysis. Artificial Neural Networks (ANNs) are very effective in developing predictive models for processes involving complex reaction kinetics that would otherwise be difficult to be modeled by more traditional deterministic approaches. The present investigation was carried out to study the application of Artificial Neural Network as a tool for predicting glucose production by enzymatic hydrolysis of pure cellulose and comparison with mathematical models and experimental results. A feed forward neural network with one hidden layer was trained and used to predict the glucose production. Comparing the R 2 (coefficient of determination), MSE (mean square error) and ARD (average relative deviation) values of the neural network model with the mathematical model, it was concluded that the neural network is more accurate than the mathematical models. The obtained results show that the ANN can be a useful method for the design of the enzymatic hydrolysis.

Artificial neural networks to model the production of blood protein hydrolysates for plant fertilisation

“…This algorithm computes numerically the Jacobian matrix of the system, which permits to update weight and bias iteratively until minimising the error function (in our case MSE). Similarly, the Levenberg–Marquardt training algorithm was chosen by Nikzad et al to model the enzymatic hydrolysis of cellulose with an ANN comprising seven neurons in the hidden layer. This model was found to be more accurate than two phenomenological models chosen from literature.…”

Bioreactor design for enzymatic hydrolysis of biomass under the biorefinery concept

“…However, enzymatic hydrolysis is considered the most promising strategy because it presents several advantages over chemical saccharification including high efficiency, low energy consumption due to the requirement of mild operation conditions, not environmental damage because enzymes are biodegradable, not toxic compounds formation and, low deterioration and corrosion problems for the equipment. Additionally, enzymatic hydrolysis produces fewer fermentation inhibitors [12,[32][33][34][35][36].…”

“…In another investigation, Choi et al [164] adapted the overall mass transfer coefficient in an air-lift bioreactor to the equation (31) for a temperature of 20°C as reference. Moreover, equations (32) and (33) represent mathematical expressions to determine k a L considering the diameter in bubble column reactors (D ) c . Shah et al [161] developed correlation shown in equation (34) to calculate the mass transfer coefficient using Froud and Galileo dimensionless numbers.…”

Lignocellulose-Based Bioproducts

“…β-glucosidase catalyzes cleavage of cellobiose, which plays a significant role in the hydrolysis process, since cellobiose is an end product inhibitor of many cellulases including both exo-and endoglucanases Lee 1997;Rabinovich et al 2002;Sun and Cheng 2002). β-Glucosidase, in turn, is inhibited by glucose and, therefore, enzymatic hydrolysis is sensitive to the substrate concentration (Nikzad et al 2012). In addition to substrate concentration, pretreatment of cellulosic materials and hydrolyzing conditions such as temperature and pH are among factors influencing the efficiency of enzymatic hydrolysis.…”