基于人工神经网络和遗传算法的黑木耳糙米醋发酵条件优化

Optimization of fermentation conditions for Auricularia auricular and brown rice vinegar based on artificial neural networks and genetic algorithm

以黑木耳糙米为原料,研究醋酸发酵的工艺条件,对接种量、摇床转速和装料量参数进行了单因素和正交试验,摸索出参数优化范围,并利用人工神经网络和遗传算法联合对发酵条件进行优化,结果表明,当接种量为8.1%,摇床转速为169r/min,装料量为250mL三角瓶装料61mL时,醋酸含量可达3.855g/100mL,该方法与正交试验设计得出的结果相比,发酵酸度上升了8.0%,为发酵优化控制提供了一种更加准确的方法。

Using Auricularia auricular and brown rice as raw material, acetic fermentation conditions were studied. Ranges of different parameters, such as inoculum, rotating speed and filling volume were determined by single-factor and orthogonal experimental tests. Then, optimal fermentation conditions were optimized by artificial neural networks and genetic algorithm. The results indicated that the highest acidity could reach 3.855g/100ml when fermentation conditions were as follows： inculum 8.1%, rotating speed 169r/min, and filling liquid medium volume 61 ml. Compared with the results of orthogonal test, the acidity increased by 8.0% and a higher accurate method was proposed for fermentation conditions optimization.