School of Chemical and Petroleum Engineering, Shiraz University, Molasadra Ave., Shiraz, Iran.
Bioresour Technol. 2013 Nov;148:47-52. doi: 10.1016/j.biortech.2013.08.060. Epub 2013 Aug 14.
In this study, pretreatment of sugarcane bagasse and subsequent enzymatic hydrolysis is investigated using two categories of pretreatment methods: dilute acid (DA) pretreatment and combined DA-ozonolysis (DAO) method. Both methods are accomplished at different solid ratios, sulfuric acid concentrations, autoclave residence times, bagasse moisture content, and ozonolysis time. The results show that the DAO pretreatment can significantly increase the production of glucose compared to DA method. Applying k-fold cross validation method, two optimal artificial neural networks (ANNs) are trained for estimations of glucose concentrations for DA and DAO pretreatment methods. Comparing the modeling results with experimental data indicates that the proposed ANNs have good estimation abilities.
在这项研究中，使用两种预处理方法（稀酸（DA）预处理和组合 DA-臭氧化（DAO）方法）对甘蔗渣进行预处理，并随后进行酶水解。这两种方法都是在不同的固液比、硫酸浓度、高压釜停留时间、蔗渣含水量和臭氧化时间下进行的。结果表明，与 DA 方法相比，DAO 预处理可以显著提高葡萄糖的产量。应用 k 折交叉验证方法，为 DA 和 DAO 预处理方法的葡萄糖浓度估计训练了两个最优的人工神经网络（ANNs）。将模型结果与实验数据进行比较表明，所提出的 ANN 具有良好的估计能力。
