摘要
为了优化脱硫波纹板除雾器叶片的结构设计,按照正交实验方法设计的工况,使用Fluent模拟不同结构参数和运行工况下除雾器叶片内部流场.以数值模拟结果为样本,建立了基于最小二乘支持向量机除雾效率和除雾器压降特性模型,模型回归值与数值模拟计算值最大相对误差在2%以内.模型预测结果分析表明,叶片间距、转折角度、烟气流速和烟气含液量对除雾效率和压降有显著影响,与实验和理论分析结论一致.采用遗传算法对除雾器参数优化模型进行求解,结果表明在优化结构参数组合下除雾器性能有明显提高.提出的预数值计算与人工智能算法结合的方法为获取除雾器叶片最佳结构参数组合设计提供了新思路.
According to the orthogonal experimental design method, Fluent was used to numerically simulate the two-phase flow of gas and liquid in wave-plate mist eliminator with different structural parameters and operation conditions in wet flue gas desulfurization system in order to optimize the design of mist eliminators. A prediction model for removal efficiency and pressure drop was estab- lished applying least square support vector machine from the results of numerical computation. The highest relative error between the predicted output and measured value is smaller than 2%. The pre- diction results show that not only the vane spacing and vane turning angles, but also flue gas velocity and water entrainment load play an important role in influencing the removal efficiency and pressure drop. It is consistent with some experimental and simulation conclusions. Based on the prediction model, a mist eliminator parameters optimization model was created employing genetic algorithm (GA) and the results show that eliminator performance can be improved obviously. The optimal so- lution and data analysis show that the model can direct the optimum design of mist eliminators.
出处
《东南大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2013年第1期76-82,共7页
Journal of Southeast University:Natural Science Edition
关键词
除雾器
数值模拟
除雾效率
压降
最小二乘支持向量机
遗传算法
mist eliminator
numerical simulation
removal efficiency
pressure drop
least square support vector machine
genetic algorithm