研究报告
谢彬,马邕文,万金泉,王艳,渠艳飞.基于混合智能算法的多目标优化在厌氧氨氧化与反硝化协同脱氮除碳中的应用[J].环境科学学报,2018,38(4):1467-1473
基于混合智能算法的多目标优化在厌氧氨氧化与反硝化协同脱氮除碳中的应用
- Application research of multi-objective optimization in carbon and nitrogen removal by anaerobic ammonia oxidation and denitrification based on multi-intelligence-algorithm
- 基金项目：国家自然科学基金（No.31570568，31670585）；制浆造纸工程国家重点实验室项目（No.201535）；广东市科技计划项目（No.201607010079，201607020007）；广东省科技计划项目（No.2016A020221005）
- 作者
- 单位
- 谢彬
- 华南理工大学环境与能源学院, 广州 510006
- 马邕文
- 1. 华南理工大学环境与能源学院, 广州 510006;2. 华南理工大学教育部工业聚集区域污染控制与修复重点实验室, 广州 510006;3. 华南理工大学制浆造纸国家重点实验室, 广州 510006
- 万金泉
- 1. 华南理工大学环境与能源学院, 广州 510006;2. 华南理工大学教育部工业聚集区域污染控制与修复重点实验室, 广州 510006;3. 华南理工大学制浆造纸国家重点实验室, 广州 510006
- 王艳
- 1. 华南理工大学环境与能源学院, 广州 510006;2. 华南理工大学教育部工业聚集区域污染控制与修复重点实验室, 广州 510006;3. 华南理工大学制浆造纸国家重点实验室, 广州 510006
- 渠艳飞
- 华南理工大学环境与能源学院, 广州 510006
- 摘要：针对厌氧氨氧化与反硝化协同实现脱氮除碳优化问题，采用UASB反应器处理不同进水条件下的氨氮废水，基于BP神经网络分别建立NH4+-N去除模型和COD去除模型，同时为了提高模型的鲁棒性和运算速度，使用PCA算法降低输入变量维数.仿真结果表明，基于PCA-BP的预测模型具有较好的预测能力，检验样本中模型预测值与实际真实值的相关系数分别为0.9164和0.9987，且两模型的平均预测误差都保持在在10%以内.进一步结合NSGA-Ⅱ算法建立以去除NH4+-N和COD最大化的优化模型，以优化结果为条件建立的出水效果接近实际真实值，表明该模型给出的优化解决方案有效可行，可为实现厌氧氨氧化与反硝化协同脱氮除碳工艺的设计和操作提供参考和指导.
- Abstract：In this paper, an up-flow anaerobic sludge bed reactor (UASB) was operated to investigate the optimization of carbon and nitrogen removal via anaerobic ammonia oxidation (ANAMMOX) and denitrification under different flow conditions, a soft-predicting model was employed to simultaneously predict the effluent ammonia nitrogen removal and COD removal based on back propagation (BP) neural network. In order to improve the robustness and speed of operation models, principal component analysis (PCA) was used to reduce the dimensions of input variables. The results revealed that the proposed PCA-BP models were capable of dynamically predicting the effluent ammonia nitrogen removal and COD removal with correlation coefficient were 0.9164 and 0.9987, respectively. The average prediction error of the two models was kept within 10%. Furthermore, the optimization model of maximizing in ammonia nitrogen removal and COD removal was developed by integrating the non-dominated sorting genetic algorithms-Ⅱ (NSGA-Ⅱ).The experimental results showed that the effluent established on the basis of the optimization results was close to the actual real value. As a result, the proposed model is effective and feasible, which can provide reference and guidance for the design and operation of the process of ANAMMOX and denitrification for carbon and nitrogen removal.
摘要点击次数： 909 全文下载次数： 1659
