| 结合人工神经网络的翅片管式冷凝器快速仿真模型 |
| |
| 引用本文： | 丁国良,张春路,刘浩. 结合人工神经网络的翅片管式冷凝器快速仿真模型[J]. 工程热物理学报, 2002, 23(1): 75-78 |
| |
| 作者姓名： | 丁国良 张春路 刘浩 |
| |
| 作者单位： | 上海交通大学制冷与低温工程研究所,上海,200030 |
| |
| 基金项目： | 国家重点基础研究发展规划项目(No．G2000026309），国家教委留学回国人员基金资助项目(教外司留[1997］832号） |
| |
| 摘 要： | 为了提高翅片管式冷凝器仿真精度和计算速度，提出了含基本数学模型与两个神经网络的复合冷凝器模型．基本数学模型采用分区集中参数模型，简化用神经网络用以反映分区集中参数模型与分布参数模型间的非线性关系，精度改进用神经网络通过分析比较模型与实验值的差别，提高复合模型的精度．实际应用表明，复合模型的计算速度比分布参数
|
| 关 键 词： | 冷凝器 翅片管 模型 神经网络 |
| 文章编号： | 0253-231X(2002)01-0075-04 |
| 修稿时间： | 2001-01-06 |
FAST SIMULATION MODEL COMBINING WITH ARTIFICIAL NEURAL NETWORKS FOR TUBE-IN-FIN CONDENSER |
| |
| DING Guo-Liang ZHANG Chun-Lu LIU Hao. FAST SIMULATION MODEL COMBINING WITH ARTIFICIAL NEURAL NETWORKS FOR TUBE-IN-FIN CONDENSER[J]. Journal of Engineering Thermophysics, 2002, 23(1): 75-78 |
| |
| Authors: | DING Guo-Liang ZHANG Chun-Lu LIU Hao |
| |
| Abstract: | In order to improve predicting precision and increase computation speed of simulation for tube-in-fin condenser, a novel method integrating the basic mathematical model with artificial neural network is presented. A three-zone model is used as the basic mathematical model. An artificial neural network maps the nonlinear relation between the distributed-parameter model and the three-zone one. Another artificial neural network is involved in the present model to reflect the bias between the model and experimental results, and to improve the simulation accuracy. Practical utilization of the present model shows that the computation speed of the present model is two orders of magnitude faster than the distributed-parameter model, while the precision is also improved. |
| |
| Keywords: | condenser tube-in-fin model artificial neural network |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
