构建优化模型对羰基还原酶的产酶培养条件进行优化,获得较优的发酵产酶条件。通过Plackett-Burman设计筛选出羰基还原酶产酶培养条件的4个显著影响因子;设计正交试验进一步探究显著影响因子对羰基还原酶产酶的影响;以正交试验结果为样本数据,采用支持向量机(SVM)模型对正交试验结果进行拟合回归;构建SVM-PSO优化模型对拟合函数进行全局寻优,寻优结果即为羰基还原酶较优产酶条件。实验结果表明在预测的最优产酶条件下培养菌体,羰基还原酶酶活达到299.19 U/L,比优化前提高了16.81%,证明构建的SVMPSO模型对于产酶条件优化问题有较好处理能力。

To illustrate the optimal cultural condition for carbonyl reductase production,a novel optimal method based on SVM-PSO theory was proposed. Firstly,four significant impact factors of enzyme production were picked out with the Plackett-Burman experimental design. Then orthogonal design was carried out in order to investigate the effects of the significant impact factors on carbonyl reductase activity. On the basis of orthogonal experiment,SVM model was used for the fitted regression,and SVM-PSO optimal model was constructed for the global optimization. The results of above global optimization were regared as the theoretically optimal fermentation conditions. Under the optimized conditions,carbonyl reductase activity reached 299. 19 U / L,with a 16. 81% increase compared to that before optimization. The result of this experiment showed that the SVM-PSO model was feasible for condition optimization of enzyme production.