摘要
由于探寻改性麦糟吸附三价砷离子As(Ⅲ)最适条件过程的复杂性,本文引入支持向量回归模型对该过程进行建模,以发现实验过程中各因素对去除率的影响,揭示该生物吸附剂对As(Ⅲ)吸附的最适条件和最佳效果。为了解决建模过程中数据采集不完整和可能存在的模型稳定性波动问题,引入了矩阵补全算法对受损数据进行补全。矩阵补全后的模型验证表明,该模型的预测结果与实际吸附率相比,具有偏差小、稳定性高、有效性和实用性,为改性麦糟吸附三价砷离子的进一步研究提供指导。
Exploring the optimum condition for modified spent grains＇ adsorption of As（III） is complex, thus this paper introduce support vector regress to model this process to find out how various factors affect the adsorption rate during the experiment so as to discover the optimum condition and efficiency. In order to solve the problem of incomplete data collection and possible fluctuation of the model stability, here adopt a matrix compete algorithm to complete the damaged data. The verification of the model after complement show that the result obtained from this model is more stable and accurate, which prove that this method is effective and practical. Therefore, study the model with machine learning could provide a guiding function for the further research of the modified spend grans＇ adsorption of As（Ill）.
出处
《计算机与应用化学》
CAS
2015年第9期1143-1148,共6页
Computers and Applied Chemistry
基金
国家自然科学基金资助项目(51164014)
江西省自然科学基金资助项目(20132BAB203020)
江西省教育厅科学技术研究资助项目(GJJ13430)
关键词
改性麦糟
生物吸附
支持向量回归机
矩阵补全
modified spent grains
adsorption
support vector regress
matrix complete
