User-friendly optimization approach of fed-batch fermentation conditions for the production of iturin A using artificial neural networks and support vector machine

Abstract

Background: In the field of microbial fermentation technology, how to optimize the fermentation conditions is of great crucial for practical applications. Here, we use artificial neural networks (ANNs) and support vector machine (SVM) models to offer a series of effective optimization methods for the production of iturin A. The concentration levels of asparagine (Asn), glutamic acid (Glu) and proline (Pro) (mg/L) were set as independent variables, while the Iturin A titer (U/mL) was set as dependent variable. General regression neural network (GRNN), multilayer feed-forward neural networks (MLFN) and the SVM model were developed. Comparisons were made among different ANN models and the SVM model.

Results: The GRNN model has the lowest RMS error (457.88) and training time (1s), with a steady fluctuation after repeated experiments, whereas the MLFN models have comparatively higher RMS errors and training times, which have a significant fluctuation with the change of nodes. In terms of the SVM, it also has a relatively low RMS error (466.13), with a low training time (1s).

Conclusion: According to the modeling results, the GRNN model is considered as the most suitable ANN model for the design of the fed-batch fermentation conditions for the production of Iturin A because of its high robustness and precision. And the SVM model is also considered as an very suitable alternative model. Under the tolerance of 30%, the prediction accuracy of the GRNN and SVM models are both 100% respectively in repeated experiments.