节点文献
基于遗传编程的吸附床传热状态预测
Heat Transfer Prediction of Absorption Tower Based on Genetic Programming
【摘要】 脱硝吸附塔是烟气处理系统中的重要环节。塔内的热交换受到多方因素的非线性影响,给准确快速的评估吸附剂换热状态带来巨大挑战。为解决此问题,使用数值模拟与均一介质方法,对一种插管式吸附塔换热进行了模拟与多元遗传编程回归,获得了白盒表达式并验证模型准确性。分析了因变量、遗传编程参数等设定对回归的影响。最终结果表明使用遗传编程的回归效果优异,在工业生产与探究变量物理关系中,具有强大的应用潜力。
【Abstract】 The adsorption tower of denitration is an important step in the flue treatment system.The heat exchange in the tower is affected by nonlinear factors, which brings great challenges to accurately and rapidly evaluating the heat transfer of adsorbents. To solve this change, the numerical simulation and homogeneous medium were adopted to simulate the heat transfer of a cannulated adsorption tower. Based on the data, regression of multi-genetic programming was carried out. The white-box expression was obtained and the accuracy of the model was verified. The influence of the dependent variable and the parameter of regression was analyzed. The final results show that the regression of genetic programming is excellent, and has strong application potential in industrial production and exploring the physical relationship between variables.
【Key words】 heat transfer in absorption tower; genetic programming; error analysis; machine learning; symbolic regression;
- 【文献出处】 工程热物理学报 ,Journal of Engineering Thermophysics , 编辑部邮箱 ,2023年06期
- 【分类号】TK124
- 【下载频次】22