| 引用本文: | 宋孟杰,孔德翰,余思锐,赵玉刚,陈宁立,王海东,张龙,张旋.低温表面冷凝结霜特性预测模型研究进展[J].哈尔滨工业大学学报,2025,57(5):38.DOI:10.11918/202409081 |
| SONG Mengjie,KONG Dehan,YU Sirui,ZHAO Yugang,CHEN Ningli,WANG Haidong,ZHANG Long,ZHANG Xuan.Research progress on prediction models for low-temperature surface condensation frosting characteristics[J].Journal of Harbin Institute of Technology,2025,57(5):38.DOI:10.11918/202409081 |
|
| |
|
|
| 本文已被：浏览 11次 下载 16次 |
码上扫一扫！ |
|
|
| 低温表面冷凝结霜特性预测模型研究进展 |
|
宋孟杰1,孔德翰1,余思锐1,赵玉刚2,陈宁立3,王海东4,张龙1,张旋1
|
|
(1.北京理工大学 机械与车辆学院,北京 100081;2.上海理工大学 能源与动力工程学院,上海 200093; 3.中国空气动力研究与发展中心 低速空气动力研究所,四川 绵阳 621000;4.清华大学 航天航空学院,北京 100084)
|
|
| 摘要: |
| 结霜是日常生活与工业生产中常见的低温物理现象,且往往产生负面影响。结霜模拟技术不仅有助于深入理解结霜过程,还可为防/除霜技术的发展提供理论指导,降低或避免能源、航天、交通、电力、冷藏等领域因霜导致的潜在危害。为充分理解结霜这一非均匀、变密度、移动边界、连续相变的复杂传热传质与流动耦合过程,分别基于低温表面冷凝结霜过程中液滴冷凝、凝固凸起、虚霜生长、霜层成熟四阶段及霜冻气候的既有模型研究成果进行梳理分析。结果表明:液滴冷凝阶段,既有模型对液滴粒径、成核速率等指标的模拟精度可达80%以上；凝固凸起阶段,冻结锋面高度、冻结时长等参数的模拟精度可达85.3%；虚霜生长与霜层成熟阶段,霜层厚度、霜层密度等指标的模拟精度可达82%以上；霜冻气候模拟与预测的准确率最高则可达88.4%。既有结霜模拟技术依原理差异可分为基于物理学和数学的数理模型、基于计算流体力学和数值方法的数值模拟和基于统计学和机器学习的数据分析模型3种,其中,数据分析模型类结霜模型多用于霜层生长阶段,因该阶段在结霜全周期中占时长、预测参数多且精度高而具最大发展潜力。低温表面冷凝结霜全过程中,液滴成核过程模拟因尺度小、变化快、影响因素多且处于枝晶生长前期而难度大,霜层生长中后期的枝晶周期性倒融再生过程因霜层内部微孔隙结构变化剧烈、精密测量时因物理遮挡无法观察而亦属当下挑战。本文结论为复杂场景下结/除/防/控霜等涉霜涉冰基础研究及技术开发等提供了参考与借鉴。 |
| 关键词: 冷凝结霜 数理模型 数值模拟 液滴凝固 枝晶生长 倒融再生 |
| DOI：10.11918/202409081 |
| 分类号:U492.8 |
| 文献标识码:A |
| 基金项目:国家自然科学基金(52076013)；北京市自然科学基金(3212024) |
|
| Research progress on prediction models for low-temperature surface condensation frosting characteristics |
|
SONG Mengjie1,KONG Dehan1,YU Sirui1,ZHAO Yugang2,CHEN Ningli3,WANG Haidong4,ZHANG Long1,ZHANG Xuan1
|
|
(1.School of Mechanical Engineering, Beijing Institute of Technology, Beijing 100081, China; 2.School of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China; 3.Low Speed Aerodynamics Institute, China Aerodynamics Research and Development Center, Mianyang 621000, Sichuan, China; 4.School of Aeronautics and Astronautics, Tsinghua University, Beijing 100084, China)
|
| Abstract: |
| As a common low-temperature physical phenomenon, frosting often results in negative effects in daily life and industrial production. Frosting simulation technology not only helps to deepen the understanding of the frosting process, but also provides theoretical guidance for the development of frost prevention and control technology, thereby reducing or avoiding potential hazards caused by frosting in fields such as energy, aerospace, transportation, electricity, and refrigeration. To fully understand this complex heat and mass transfer and flow coupling process, which is characterized by non-uniformity, variable density, moving boundaries, and continuous phase changes, this study systematically analyzes existing models and results of the four stages of droplet condensation, solidification tip-growth, virtual frost growth, and frost layer maturity in the low-temperature surface condensation frosting process. The results show that during the droplet condensation stage, existing models achieve a simulation accuracy of over 80% for indicators such as droplet size and nucleation rate. During the solidification tip-growth stage, the simulation accuracy of parameters such as freezing front height and freezing duration can reach 85.3%. The simulation accuracy of indicators such as frost thickness and frost density during the growth and maturity stages of frost layer can reach over 82%. Additionally, the accuracy of simulating and predicting frost climate can reach up to 88.4%. Existing frost simulation techniques can be divided into three types based on their underlying principles: mathematical models based on physical and mathematical principles, numerical simulations based on computational fluid dynamics and numerical methods, and data analysis models based on statistical and machine learning principles. Among these, the final method is mostly used in the frost growth stage, and has the greatest potential for development due to their long duration, multiple predictive parameters, and high accuracy throughout the frost formation process. During the entire condensation frosting process on low-temperature surface, the simulation of droplet nucleation in complex scenarios is difficult due to its small scale, fast changes, multiple influencing factors, and its occurrence in the early stage of dendrite growth. Similarly, the periodic reverse melting and regeneration of frost crystal in the later stage of frosting growth is also a current challenge due to the drastic changes inside the frost layer and the physical obstruction during precise measurements, which cannot be clearly observed. The conclusions of this study provide valuable references and inspiration for fundamental research and technological development related to frost and ice in complex scenarios, such as frosting, defrosting, frost prevention, and frost control, etc. |
| Key words: condensation frosting mathematical model numerical simulation droplet solidification dendritic growth reverse melting and regeneration |
|
|
|
|
