人工智能优化紫外-过碳酸钠法降解间甲酚废水过程
刘一楠 , 张婧 , 陈晓飞 , 陈平 , 孙啸林 , 慕朝 , 马磊
现代化工 ›› 2025, Vol. 45 ›› Issue (7) : 119 -125.
人工智能优化紫外-过碳酸钠法降解间甲酚废水过程
Optimization of UV-sodium percarbonate route M-cresol-containing wastewater degradation process by artificial intelligence
本研究采用紫外-过碳酸钠(UV-SPC)氧化反应体系处理间甲酚废水,并借助人工智能方法进行工艺优化,利用响应面法(RSM)进行实验设计,考察溶液初始pH、反应时间、间甲酚初始浓度、SPC投加量、催化剂用量和反应温度等因素对TOC去除率的影响。基于RSM实验结果,分别使用RSM模型和人工神经网络(ANN)模型进行优化,并进行了对比分析,考察2个模型差异,结果显示,ANN模型准确度比RSM模型高50%以上。在ANN模型模拟优化所得最佳反应条件下,实验中TOC去除率为91.48%,明显高于以RSM模型模拟优化所得的最优结果,验证了ANN模型法优异的学习能力和泛化能力。
In this study,the UV-sodium percarbonate (UV-SPC) oxidation reaction system is used to treat m-cresol-containing wastewater,and the treatment process is optimized by means of artificial intelligence method.The response surface method (RSM) is mainly used to perform experimental design.The influences of initial pH of solution,reaction time,initial concentration of m-cresol,SPC dosage,catalyst dosage and reaction temperature on the removal rate of total organic carbon (TOC) are deeply investigated.Based on RSM experiment results,RSM model and AI model are respectively used for the optimization,and a comparative analysis is carried out to evaluate the differences between two models.The results demonstrate that the accuracy of ANN model is more than 50% higher than that of RSM model.Under the optimal conditions from the optimization by ANN model,the removal rate of TOC reaches 91.48% in the experiment,which is significantly higher than the optimal result obtained from RSM model optimization.It is also verified that AI model method has an excellent learning ability and a generalization ability.
人工智能 / 响应面 / 过碳酸钠 / 紫外线 / 人工神经网络
artificial intelligence / response surface / sodium percarbonate / ultraviolet / artificial neural network
| [1] |
韩培威, 刘伟军, 赵颖, |
| [2] |
孙文静, 王亚旻, 卫皇曌, |
| [3] |
|
| [4] |
|
| [5] |
王特, 杨鹏辉, 屈撑囤, |
| [6] |
王平. 高盐含酚废水生物处理及微生物群落结构研究[D]. 大连: 大连理工大学, 2009. |
| [7] |
|
| [8] |
刘聪. 高浓度含酚废水构成分析及芬顿氧化技术应用[J]. 净水技术, 2020, 39(2):91-97. |
| [9] |
冯思慧. Al2O3负载CuO催化臭氧氧化含酚污水[D]. 哈尔滨: 哈尔滨工程大学, 2016. |
| [10] |
|
| [11] |
郑怀礼, 相欣奕. 光助Fenton氧化反应降解染料罗丹明B[J]. 光谱学与光谱分析, 2004,(6):726-729. |
| [12] |
李纪华, 薛韫坤, 何坚. 过碳酸钠降解水相中罗丹明B的实验研究[J]. 环境工程, 2012, 30(S2):176-178. |
| [13] |
|
| [14] |
|
| [15] |
赵颖, 王亚旻, 卫皇曌, |
| [16] |
|
| [17] |
|
| [18] |
|
| [19] |
|
| [20] |
|
| [21] |
|
| [22] |
|
| [23] |
|
| [24] |
|
| [25] |
|
| [26] |
|
| [27] |
|
| [28] |
|
| [29] |
张婧, 张橙, 卫皇瞾, |
| [30] |
|
| [31] |
|
| [32] |
张婧. 人工智能在废水高级氧化处理技术中的应用[D]. 北京: 北京石油化学院, 2024. |
| [33] |
|
| [34] |
|
| [35] |
|
| [36] |
|
| [37] |
|
| [38] |
|
国家自然科学基金(52100072)
石家庄高层次科技创新创业人才项目(08202303)
/
| 〈 |
|
〉 |