Application of an artificial neural network and genetic algorithm for determination of process orbits in the koji making process

Hanai, Taizo; Honda, Hiroyuki; Ohkusu, Eiji; Ohki, Toshihiko; Tohyama, Hisao; Muramatsu, Takahiro; Kobayashi, Takeshi

doi:10.1016/s1389-1723(99)80101-7

Cited by 16 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Presently, hybrid GA–ANN is becoming popular for fermentation parameter optimization. Hanai et al. (1999) optimized 21 variables for koji fermentation process while Hongwen et al.…”

Section: Introductionmentioning

confidence: 99%

Modelling and optimization of fermentation factors for enhancement of alkaline protease production by isolated Bacillus circulans using feed-forward neural network and genetic algorithm

et al. 2008

View full text Add to dashboard Cite

On file IMPF: 02.50 RONO: 2490 3014Aim: Modelling and optimization of fermentation factors and evaluation for enhanced alkaline protease production by Bacillus circulans. Methods and Results: A hybrid system of feed-forward neural network (FFNN) and genetic algorithm (GA) was used to optimize the fermentation conditions to enhance the alkaline protease production by B. circulans. Different microbial metabolism regulating fermentation factors (incubation temperature, medium pH, inoculum level, medium volume, carbon and nitrogen sources) were used to construct a 6-13-1 topology of the FFNN for identifying the nonlinear relationship between fermentation factors and enzyme yield. FFNN predicted values were further optimized for alkaline protease production using GA. The overall mean absolute predictive error and the mean square errors were observed to be 0what0048, 27what9, 0what001128 and 22what45 U ml-1 for training and testing, respectively. The goodness of the neural network prediction (coefficient of R2) was found to be 0what9993. Conclusions: Four different optimum fermentation conditions revealed maximum enzyme production out of 500 simulated data. Concentration-dependent carbon and nitrogen sources, showed major impact on bacterial metabolism mediated alkaline protease production. Improved enzyme yield could be achieved by this microbial strain in wide nutrient concentration range and each selected factor concentration depends on rest of the factors concentration. The usage of FFNNGA hybrid methodology has resulted in a significant improvement (>2what5-fold) in the alkaline protease yield. Significance and Impact of the Study: The present study helps to optimize enzyme production and its regulation pattern by combinatorial influence of different fermentation factors. Further, the information obtained in this study signifies its importance during scale-up studies.Peer reviewe

show abstract

“…Presently, hybrid GA–ANN is becoming popular for fermentation parameter optimization. Hanai et al. (1999) optimized 21 variables for koji fermentation process while Hongwen et al.…”

Section: Introductionmentioning

confidence: 99%

Modelling and optimization of fermentation factors for enhancement of alkaline protease production by isolated Bacillus circulans using feed-forward neural network and genetic algorithm

et al. 2008

View full text Add to dashboard Cite

show abstract

“…Once the process models were established, the optimal operation conditions (blast temperature, blast time, and rice soaking time) corresponding to a targeted performance index could be determined by using several multivariable optimization methods. The genetic algorithm was chosen in this case, because of its global optimization ability as well as its high convergence speed 4,9,14 over other non-linear programming, such as the Simplex Method. In the latter case, the initial conditions are crucial for the optimization performance.…”

Section: Genetic Algorithms (Ga)mentioning

confidence: 99%

“…Many applications on process control 7 , dynamic modeling 14 , determination of optimal substrate fed-batch strategy 9 , and optimization of culture medium 10 have been reported. Recently, reports on the combined application of both neural network techniques and genetic algorithms for optimizing the temperature trajectory of Japanese sake mashing and the Koji making process 4 and other processes 10 have appeared.…”

Section: Introductionmentioning

confidence: 99%

Optimization of Operating Conditions in Rice Heat Blast Process for Chinese Rice Wine Production by Combinational Utilization of Neural Network and Genetic Algorithms

Zhu

Zhang

Shi

et al. 2004

Journal of the Institute of Brewing

View full text Add to dashboard Cite

The rice heat blast process is a novel technique for gelatinizing raw starch. It uses heated air to replace steam for processing rice under high temperature fluidization in a short time period. This is a new technique for making rice wine with the characteristics of easy storage and zero water pollution. This study focused on three major performance indexes (starch gelatinization ratio, total fat content, and amino nitrogen content in the roasted rice), which largely affect the performance of the rice heat blast process and the rice wine quality. The relationship between these performance indexes and the corresponding operation variables were modeled by an artificial neural network (ANN) via learning sets of experimental data. Based on the ANN models obtained, genetic algorithms were used to optimize the operating conditions of the rice heat blast process. The results showed the power in determination of optimal operating conditions by the combinational utilization of artificial neural network and genetic algorithms.

show abstract

“…Many kinds of networks are developed with different properties and applications. Among these, back-propagation (BP) multilayer neural network is the most common and convenient tool (Chen 1990;Hanai et al 1999). Most applications require networks that contain at least three normal types of layers -the input, the hidden and the output.…”

Section: Introductionmentioning

confidence: 99%

“…Many kinds of networks are developed with different properties and applications. Among these, back‐propagation (BP) multilayer neural network is the most common and convenient tool (Chen 1990; Hanai et al. 1999).…”

Section: Introductionmentioning

confidence: 99%

APPLICATION OF ARTIFICIAL NEURAL NETWORK TO PREDICTION OF CANTONESE SOY SAUCE BREWING AND CHANGING PATTERN CONCERNING TOTAL NITROGEN AND α‐AMINO ACID NITROGEN

Liu

Zhao

et al. 2011

J Food Process Engineering

View full text Add to dashboard Cite

In the present study, the artificial neural network was used to predict the brewing of traditional Cantonese soy sauce and its changing pattern with regard to total nitrogen (TN) andα‐amino acidic nitrogen (AN). The effects of parameters during brewing on the contents of TN and AN were modeled by various networks. After the 7‐40‐2 network was determined statistically, it was used to extract the possible correlations between TN and AN, and the relative significance of some factors. It was found that the change of TN and AN contents with the change of time and acidic protease activity was strongly correlated with an R value of 0.9848 and 0.9916, respectively. As for the relative significances of inputs, aging time is the key factor for both outputs during the moromi aging. And for other inputs, it was temperature > pH > neutral protease activity > acid protease activity for TN, and temperature > acidic protease activity > neutral protease activity > pH for AN, respectively. PRACTICAL APPLICATIONS Because of its open‐air brewing and lengthy brewing time of Cantonese soy sauce, it was difficult to predict its quality indices, the contents of total nitrogen and α‐amino acidic nitrogen, which would change with outside factors (time and temperature) and inside factors (protease activity and pH values). So the study on brewing‐state predication and changing pattern of quality indices were performed by using the artificial neural network on an experimental scale. And results from this preliminary study would help manufacturers to foresee the change and contents of quality indices accurately in the brewing process on an industrial scale. Furthermore, it was anticipated that more effective control and improvement on the quality of traditional Cantonese would be achieved in the future study. Finally, the trained network also can be used in the research of online prediction for other time‐consuming natural fermentation.

show abstract

Application of an artificial neural network and genetic algorithm for determination of process orbits in the koji making process

Cited by 16 publications

References 7 publications

Modelling and optimization of fermentation factors for enhancement of alkaline protease production by isolated Bacillus circulans using feed-forward neural network and genetic algorithm

Modelling and optimization of fermentation factors for enhancement of alkaline protease production by isolated Bacillus circulans using feed-forward neural network and genetic algorithm

Optimization of Operating Conditions in Rice Heat Blast Process for Chinese Rice Wine Production by Combinational Utilization of Neural Network and Genetic Algorithms

APPLICATION OF ARTIFICIAL NEURAL NETWORK TO PREDICTION OF CANTONESE SOY SAUCE BREWING AND CHANGING PATTERN CONCERNING TOTAL NITROGEN AND α‐AMINO ACID NITROGEN

Contact Info

Product

Resources

About