节点文献
氧化沟液固两相湍流反应动力学仿真系统
Simulation System of Fluid and Solid Two-Phase Turbulent Flow and Biological Reaction in Oxidation Ditch
【作者】 陈园；
【导师】 张代钧；
【作者基本信息】 重庆大学 ， 环境科学， 2008， 硕士
【摘要】 废水生物处理工程涉及复杂水力学和生物化学过程,其设计优化和运行控制是当前国际水科学技术研究的前沿领域。在变水力条件下污泥沉降的研究至今尚未形成公认的试验方法,现有的水环境模拟软件仍无法对复杂边界条件下的多相湍流生物反应过程进行有效的模拟与运算。论文基于图像分析方法,开发了一套新的污泥沉降速率测量方法与装置,应用于不同水力条件下(静/动水)各类污泥(离散颗粒和絮凝颗粒)沉降速率的测试。采用两相湍流混合物模型描述活性污泥-废水两相流的水力特性,将两相流模型与活性污泥2号模型(ASM2)相耦合建立液固两相湍流生物反应动力学模型;以matlab7.0为开发平台编辑相关的模拟程序,开发相应的模拟仿真系统,实现对反应器内部液固两相三维流场及水质浓度场的模拟。应用所开发的模拟仿真系统对一个工业规模的卡鲁塞尔氧化沟污水处理厂进行模拟计算,采用灵敏度分析方法与基于拉丁超立方抽样的最大似然法相结合对模型参数进行校核。结果表明,所采用的模型校核方法能有效地减少模拟值与实测值的误差。
【Abstract】 The optimal design and operation control, involving complicated hydraulics and biochemical process, of wastewater biologic treatment project is the frontier research field of current international water science and technology, but there is no generally accepted experiment method about sludge various sedimentation under different hydraulic conditions. Furthermore, the existing water environment modelling software can not simulate the multiphase flow and biological reaction under complex boundary conditions.Based on the image analysis technology, this paper developed a new experiment method and device for the measurement of sludge settling velocity, which is adapted to settling velocity measurement of scattered sludge or flocculated sludge under hydrostatic or turbulent hydraulic conditions.The 2-phase turbulence mixture model was used to discribe the hydraulic characteristics of 2-phase flow, sludge and wastewater. This hydraulic model was coupled with activated sludge model NO.2 (ASM2) to define a model of liquid and solid 2-phase turbulenc and biological recation. A simulation system was developed to model the 3-demention fileds of solid-liquid 2-phase flow and materials concentrations distribution in the reactor, where, Matlab 7.0 was used to to compile the modeling programs.A full-scale carrousel oxidation ditch wastewater treatment plant was investigated with this coupling model. A generalized likelihood uncertainty estimation (GLUE), combined with sensitivity analysis and Latin hypercube sampling (LHS), was used to calibrate the model parameters. The results shown that this calibration method can efficiently reduce the error between modeling values and measured values.