Abstract:Paddy straw (PS)–based lignocellulosic biomaterials (LBM) generate biofuel, which minimizes climate change, global warming, CO 2 , and nursery gas emissions. The current work is primarily focused on generation of bioethanol from PS to reduce energy costs and supplement non-renewable energy sources with alternative energy sources from specific agricultural biowaste sources. Due to its simplicity and effectiveness, the bioethanol PS was produced by employing the process of dilute acid hydrolysis fermentation (PDAHF). In this research, at constant temperature and hydrolysis time, the impacts of acid concentration and fermentation time on bioethanol yield were examined. Experiments revealed that the greatest production of ethanol under the best operating circumstances was 38.64%. Furthermore, to determine the impact of fermentation time, temperature, and pH on the bioethanol yield from PS biomass, the process variables were optimized using RSM, and its interaction effect on maximizing ethanol yield was investigated. From ANOVA analysis, a highly significant adjusted and R -squared correlation factor of 0.9963 and 0.9790, respectively, with low p -value of less than 0.0001 and F -value of 299.1 regression quadratic model equation was obtained. The maximum predicted and experimental bioethanol yields are 38.497 and 38.517%, respectively. Additionally, two types of artificial neural networks (ANN) were developed in this study: a feedforward artificial neural network (FANN) and a nonlinear autoregressive network with exogenous inputs (NARX). The Levenberg-Marquardt (LM) training algorithm for the FANN model provided best prediction of bioethanol yield with regression (R) of 0.966 and 0.515, 0.441, 0.394, and 0.011 for root mean squared error (RMSE), mean absolute error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE), respectively, while gradient descent backpropagation (GD) for the NARX model generated R of 0.869 and RMSE, MSE, MAE, and MAPE of 0.515, 0.441, 0.394, and 0.011, respectively. In comparison to NARX, FANN has higher R-value and lower RMSE, MSE, MAE, and MAPE, making it a better choice for predicting the output of bioethanol.

Neural networks and genetic algorithm as robust optimization tools for modeling the microbial production of poly-β-hydroxybutyrate (PHB) from Brewers' spent grain

Optimization of Phenyllactic Acid Biosynthesis and Separation by Machine Learning with Neural Network and Overlay Sampling Uniform Design

Response Surface Methodology and Artificial Neural Network Optimization and Modeling of the Saccharification and Fermentation Conditions of the Polyhydroxybutyrate from Corn Stover

Bioprocess Modeling Using A Robust Genetic Programming

Application of Artificial Neural Network Coupling Particle Swarm Optimization Algorithm to Biocatalytic Production of GABA.

Modeling and optimization of tannase production with Triphala in packed bed reactor by response surface methodology, genetic algorithm, and artificial neural network

Optimization of fermentative hydrogen production process using genetic algorithm based on neural network and response surface methodology

A Novel Integrated Approach to the Enhanced Production of Polyhydrobutyrate with Mixed Culture in Activated Sludge

Optimization of medium composition for two-step fermentation of vitamin C based on artificial neural network–genetic algorithm techniques

Optimization of Culture Medium for Maximal Production of Spinosad Using an Artificial Neural Network - Genetic Algorithm Modeling

Metabolic modeling of Halomonas campaniensis improves polyhydroxybutyrate production under nitrogen limitation

Enhanced production of mycelium biomass and exopolysaccharides of Pleurotus ostreatus by integrating response surface methodology and artificial neural network

Microorganism inspired hydrogels: Optimization by response surface methodology and genetic algorithm based on artificial neural network

Applying machine learning and genetic algorithms accelerated for optimizing ethanol production

Optimization of fungal secondary metabolites production via response surface methodology coupled with multi-parameter optimized artificial neural network model

Optimized phenol degradation and lipid production by Rhodosporidium toruloides using response surface methodology and genetic algorithm-optimized artificial neural network

Optimization of biomass production from sugar bagasse in anaerobic digestion using genetic algorithm

Paddy straw as a biomass feedstock for the manufacturing of bioethanol using acid hydrolysis and parametric optimization through response surface methodology and an artificial neural network

Optimizing Polyhydroxyalkanoate production using a novel Bacillus paranthracis isolate: A response surface methodology approach

Modeling and Optimization of Triticale Wort Production Using an Artificial Neural Network and a Genetic Algorithm

Preparation of potential organic fertilizer rich in γ-polyglutamic acid via microbial fermentation using brewer's spent grain as basic substrate