Application of Virtual Sample Generation and Screening in Process Parameter Optimization of Botanical Medicinal Materials.

BACKGROUND: The small sample problem widely exists in the fields of the chemical industry, chemistry, biology, medicine, and food industry. It has been a problem in process modeling and system optimization. The aim of this study is to focus on the problems of small sample size in modeling, the process parameters in the ultrasonic extraction of botanical medicinal materials can be obtained by optimizing the extraction rate model. However, difficulty in data acquisition results in problem of small sample size in modeling, which eventually reduces the accuracy of modeling prediction. METHODS: A virtual sample generation method based on full factorial design (FFD) is proposed to solve the problem ofa small sample size. The experiments are first conducted according to the Box- Behnken Design (BBD) to obtain small-size samples, and the response surface function is established accordingly. Then, virtual sample inputs are obtained by the FFD, and the corresponding virtual sample outputs are calculated by the response surface function. Furthermore, a screening method of virtual samples is proposed based on an extreme learning machine (ELM). The connection weights of ELM are used for further optimization and screening of the generated virtual samples. RESULT: The results show that virtual sample data can effectively expand the sample size. The precision of the model trained on semi-synthetic samples (small-size experimental simples and virtual samples) is higher than the model trained merely on small-size experimental samples. CONCLUSION: The virtual sample generation and screening methods proposed in this paper can effectively solve the modeling problem of small samples. The reliable process parameters can be obtained by optimizing the model trained by the semi-synthetic samples.

Optimization of process parameters for anaerobic fermentation of corn stalk based on least squares support vector machine.

In order to improve the yield and efficiency of biogas produced from anaerobic fermentation of corn stalk, least squares support vector machine (LS-SVM) was used to optimize the pretreatment process parameters. Weight of corn stalk, ultrasonic duration time, alkali pretreatment (2% NaOH) time, and single/dual-frequency ultrasound were selected as the experimental factors of orthogonal experimental design (OED). A new modeling method combining LS-SVM and OED was proposed to establish the predictive model between cumulative biogas production (CBP) and pretreatment process parameters. The effect of experimental factors on CBP was analyzed by two-dimensional (2D) and three-dimensional (3D) contour maps of the predictive model. The optimum parameters for process pretreatment were as follows: weight of corn stalk 53 g, dual-frequency ultrasound, ultrasonic duration time 33 min, alkali pretreatment time 56 h. The CBP of the optimal conditions obtained was 22.69 L and was 14.13% higher than that of optimal conditions for OED.

AO-BBO: A Novel Optimization Algorithm and Its Application in Plant Drug Extraction.

INTRODUCTION: It is well-known that the biogeography-based optimization (BBO) algorithm lacks searching power in some circumstances. MATERIAL & METHODS: In order to address this issue, an adaptive opposition-based biogeography-based optimization algorithm (AO-BBO) is proposed. Based on the BBO algorithm and opposite learning strategy, this algorithm chooses different opposite learning probabilities for each individual according to the habitat suitability index (HSI), so as to avoid elite individuals from returning to local optimal solution. Meanwhile, the proposed method is tested in 9 benchmark functions respectively. RESULT: The results show that the improved AO-BBO algorithm can improve the population diversity better and enhance the search ability of the global optimal solution. The global exploration capability, convergence rate and convergence accuracy have been significantly improved. Eventually, the algorithm is applied to the parameter optimization of soft-sensing model in plant medicine extraction rate. CONCLUSION: The simulation results show that the model obtained by this method has higher prediction accuracy and generalization ability.