基于Fluent与GA-BP的窑炉氮气填充过程优化  

Optimisation of Nitrogen Filling Process in Kilns Based on Fluent and GA-BP Neural Networks

作　　者：刘新宇 包建东 霍李均 翟国平 LIU Xin-yu;BAO Jian-dong;HUO Li-jun;ZHAI Guo-ping(College of Automation,Nanjing University of Science and Technology,Nanjing Jiangsu 210000,China;Jiangsu Botao Intelligent Thermal Engineering Co.,Ltd,Suzhou Jiangsu 215000,China)

机构地区：[1]南京理工大学自动化学院,江苏南京210000 [2]江苏博涛智能热工股份有限公司,江苏苏州215000

出　　处：《计算机仿真》2025年第2期330-337,共8页Computer Simulation

摘　　要：针对高温固相法烧结锂电池材料前期的窑炉气氛快速填充过程,提出了一种最高效并且最大限度节约氮气原料的优化设计方法。通过Fluent软件对不同进气口数量与进气速度下的填充过程进行仿真并提取数据集,选取了五种不同回归算法进行填充过程数据集的拟合对比实验。实验表明GA-BP神经网络的效果最优,并以此建立窑炉氮气填充过程的数学模型。以氮气消耗量为目标函数,将填充过程转化为非线性规划问题进行求解。结果表明本文提出的最优化设计方法能够建立有效的数学模型,并可以求解得到最佳的窑炉设计参数,可以提高生产效率,并降低生产成本。An optimised design method is proposed for the rapid flling process of the kiln atmosphere during the pre-sintering of Li-ion battery materials by the high-temperature solid phase method,which is the most efficient way and the most economical way to save nitrogen.The filling process is simulated by Fluent software for different numbers of inlets and inlet velocities and the data set is extracted,and five different regression algorithms are selected to fit the filling process data set for comparison experiments.The experiments show that the GA-BP neural network is the most effective,and this is used to establish a mathematical model of the nitrogen filling process in the kiln.Using nitrogen consumption as the objective function,the flling process is transformed into a non-linear programming problem to be solved.The results show that the optimal design method proposed in this paper can establish an effective mathematical model and can be solved to obtain the best kiln design parameters,which can improve production efficiency and reduce production cost.

关 键 词：窑炉 回归算法 神经网络 非线性规划 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]