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

To optimize the synthetic process of microorganism inspired multistage porous hydrogel, the mathematical analysis strategies was applied to select the optimal experimental parameters. One-factor-at-a-time (OFAT) design was used to optimize the preparation process of yeast fermentation multistage porous hydrogels. The analysis of variance (ANOVA) was applied to determine the significant influencing factors, and the results revealed that the mass ratio of yeast to glucose (Ryeast/glucose), gelation temperature of yeast fermentation (Tgelation) and reaction time (treaction) had a significant influence on responses. Box-Behnken design (BBD) based response surface methodology (RSM) was used to design experiments and build the relationship between the input parameters and output responses. Ideal point method was used to transform a multi-objective optimization problem into a single-objective optimization problem. Artificial neural network (ANN) coupled genetic algorithm (GA) were employed to further optimize and predict the optimal preparation conditions of yeast fermentation multistage porous hydrogels. The results showed ANN coupled GA was a more effective tool in the modelling and optimization of the preparation of yeast fermentation multistage porous hydrogels. The optimized preparation conditions are Ryeast/glucose 1.84, Tgelation 25.00 °C, treaction 239.97 min. These values are expected to give us the minimum density, the maximum swelling degree and compressive strength. The research content of this paper provides theoretical support and factual basis for process optimization with complex influencing factors.