滕旭东,滕博欣.一种声速识别空气成分波动的自监督学习方法[J].,2024,43(6):1360-1365 |
一种声速识别空气成分波动的自监督学习方法 |
Self-supervised Learning for identifying variations of air compositionBase on Speed of Sound |
投稿时间:2023-08-07 修订日期:2024-11-03 |
中文摘要: |
实际空气成分比较复杂,其声速的大小与标准空气的计算值有较大差异。文中根据维里展开式推导的声速方程,分析实际空气成分的浓度微量波动与声速变化定量关系。基于理论计算值,对比声速实测值,提出一种基于声速差值的自监督学习方法分析空气成分波动。该方法首先计算各个时刻声速差值的分布距离,接着确定分布最聚集区域,进行统计声速差发生比例,最后利用概率分布变化来发现空气成分,尤其微量成分的波动。实测结果表明:声速差发生概率曲线较好地反映空气成分浓度变化的起始、扩散和累积等阶段,通过发生概率值分布范围能够对空气成分波动状态进行识别和实时监控。 |
英文摘要: |
As the actual composition of air is quite complex, the value of its speed of sound differs significantly from the calculated value for standard air. This article discusses the calculation of the speed of sound for standard air, furthermore analyze quantitative relationship between actural air composition and changes in the speed of sound. Theoretical results show that air composition, especially trace components, have a very small contribution to variation of the speed, thus making it difficult to discern slight fluctuations in the composition. To address this, a supervised algorithm for monitoring speed of sound error distribution is proposed to analyze air composition fluctuations. Firstly, the distance matrix of the difference between the measured and calculated values of the speed of sound is calculated, then the proportion of error values is determined, and finally, the distribution probability is analyzed to determine changes in air composition. This algorithm effectively tracks the beginning, diffusion, and accumulation of concentration changes and enables prediction and monitoring of air composition fluctuations quite pricisely. |
DOI:10.11684/j.issn.1000-310X.2024.06.019 |
中文关键词: 声速 维里展开式 成分波动 概率分布 |
英文关键词: speed of sound, virial expansions, composition fluctuations probability distributions |
基金项目: |
|
摘要点击次数: 31 |
全文下载次数: 28 |
查看全文
查看/发表评论 下载PDF阅读器 |
关闭 |
|
|
|