The Application of Chemometric Methods in the Production of Enzymes Through Solid State Fermentation Uses the Artificial Neural Network—a Review

Menezes, Luiz Henrique Sales de; Pimentel, Adriana Bispo; Oliveira, Polyany Cabral; Tavares, Iasnaia Maria de Carvalho; Ruíz, Héctor A.; Irfan, Murammad; Bilal, Muhammad; Chagas, Thiago Pereira das; Silva, Erik Galvão Paranhos da; Salay, Luiz Carlos; Oliveira, Julieta Rangel de; Franco, Marcelo

doi:10.1007/s12155-022-10462-w

Cited by 6 publications

(1 citation statement)

References 86 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, an artificial neural network (ANN) can be used to predict responses based on experimental results and thus produce knowledge through learning [19]. ANN models become accurate computational models of optimization techniques that have the ability to overcome research limitations such as the need to perform a large number of experiments and the planning restricted to the experimental domain [20]. In this way, the application of this methodology based on artificial intelligence results in more efficient enzyme production in solid-state fermentation compared to the classical optimization methodology [21,22].…”

Section: Introductionmentioning

confidence: 99%

Saccharification of Agricultural Wastes and Clarification of Orange Juice by Penicillium rolfsii CCMB 714 Pectinase

Macedo,

Azevedo,

da Silva

et al. 2023

Fermentation

View full text Add to dashboard Cite

Pectinases are enzymes used in several industrial processes. Seven agroindustrial wastes—jackfruit seed meal (Artocarpus heterophyllus), cocoa seed peel (Theobroma cacao), cocoa husks (Theobroma cacao), passion fruit husks (Passiflora edulis), mangosteen husks (Garcinia mangostana), malt residue (Hordeum vulgare) and the peach palm waste (Bactris gasipaes Kunth.)—were evaluated to produce a crude extract containing pectinase activity by Penicillium rolfsii CCMB 714. The jackfruit seed meal was chosen as the best substrate for solid-state fermentation, which was optimized with 4 mL of water as a wetting agent for 2 days at 35 °C and with a 0.5% nitrogen source, whereby the pectinase production increased by 44% (362.09 U/g). The obtained crude extract was characterized and applied to wastes saccharification and orange juice clarification. The pectinase showed better activity at a pH of 3.0 to 5.0 and 55 °C, it stably maintained over 80% of activity at 30–50 °C for up to 60 min and 1 mM CuSO4 increased the pectinase activity by 17%. The saccharification of agroindustrial wastes (cocoa husks, mangosteen husks and passion fruit husks) resulted in 126.55 µmol/mL of reducing sugars from passion fruit husks, which represents an increase of 126% after optimization (45 °C for 22 h). For the clarification of orange juice, it was possible to reduce the absorbance of the juice by 55%. These results elucidate the potential of the low-cost pectinase solution from P. rolfsii CCMB 714 cultivated in jackfruit seed meal for both the enzymatic pretreatment of plant biomass and the application in beverage industries.

show abstract

Section: Introductionmentioning

confidence: 99%

Saccharification of Agricultural Wastes and Clarification of Orange Juice by Penicillium rolfsii CCMB 714 Pectinase

Macedo,

Azevedo,

da Silva

et al. 2023

Fermentation

View full text Add to dashboard Cite

show abstract

Temperature control of beer fermentation based on variable domain fuzzy PID and neural network technology and its application analysis

2024

Adv Control Appl

View full text Add to dashboard Cite

Temperature control in the beer manufacturing process is crucial for product quality. Given the gap between China's automation in beer production and the international level, improving the technology in this area has gradually become a core issue in optimizing domestic beer production. This study combines a proportional integral derivative controller with a fuzzy modeling strategy and incorporates a variable‐domain structure to propose a variable‐domain fuzzy proportional integral derivative controller control method. To cope with the challenges of production interaction, the study also introduces neural network technology. The experimental data indicated that the variable‐domain fuzzy proportional integral derivative controller outperforms the conventional proportional integral derivative controller and the fuzzy proportional integral differential controller in terms of overshooting, with a maximum overshoot of only 1.0, compared with 0.50 and 0.70, respectively. The variable‐domain fuzzy proportional integral differential controller exhibited a minimal overshoot of only 0.01 when the model parameter is increased by 20%. In comparison, the other methods reach overshoot values of 0.92 and 1.0. The proposed method maintained superior stability even under the influence of impulse disturbance, step disturbance, and modeling variations. These results demonstrated that the research method is significantly more stable than both the proportion integration differentiation (PID) controller and fuzzy PID controller in complex dynamic parameter environments. The proposed method involved 60 rounds of neural network control, which was successfully implemented. The temperature readings T1 and T2 remained stable within the range of 1.0%–1.02% throughout the experiments. The study demonstrates that the proposed methods have higher accuracy and less fluctuation in actual application, making them more available. Taken together, the above results show that the combination of variable‐domain fuzzy PID controller and neural network technology in beer production has achieved excellent control results. This study not only provides a strong technical support for the progress of beer production technology in China, but also has important industrial application value and wide promotion prospects.

show abstract

Solid-State Fermentation as a Green Technology for Biomass Valorization: Optimization Techniques for Bioprocess—An Overview

de Menezes,

Oliveira,

do Espírito Santo

et al. 2023

Bioenerg. Res.

View full text Add to dashboard Cite

The Application of Chemometric Methods in the Production of Enzymes Through Solid State Fermentation Uses the Artificial Neural Network—a Review

Cited by 6 publications

References 86 publications

Saccharification of Agricultural Wastes and Clarification of Orange Juice by Penicillium rolfsii CCMB 714 Pectinase

Saccharification of Agricultural Wastes and Clarification of Orange Juice by Penicillium rolfsii CCMB 714 Pectinase

Temperature control of beer fermentation based on variable domain fuzzy PID and neural network technology and its application analysis

Solid-State Fermentation as a Green Technology for Biomass Valorization: Optimization Techniques for Bioprocess—An Overview

Contact Info

Product

Resources

About