【作者】 何坚；

【导师】 诸凯；

【作者基本信息】 天津大学 ， 工程热物理， 2006， 硕士

【摘要】 生物传热学主要是研究生物体的传热传质规律,是生物学、热物理学与临床医学的交叉。本文运用跨学科的研究方法,进行了舌体导热系数实验与计算的研究。这是生物传热学领域众所关心的课题之一,同时也是从热科学的角度研究中医舌诊机理必须解决的一个重要问题。本论文研究的内容是,基于组织光学和人工神经网络理论,设计一种新的导热系数测量方法,不以导热微分方程和导热系数的表达式为依据,只需借助与导热系数相关的某些参量,应用数值模拟的方法获得导热系数。本论文应用FT-IR光谱仪和双积分球测量系统设计实验,获得不同条件下生物体的组织光学特性参数。将组织光学特性参数的实验数据和热物性参数的经典数据,共同作为构建人工神经网络的数据基础。即将吸收系数、散射系数、各向异性因子定义为输入变量,将相应的导热系数定义为目标变量,按照选取的网络学习方法,对神经网络进行学习训练,利用其仿真特性得到生物体导热系数。经研究发现,基于神经网络算法,可以构建组织光学参数与生物热物性参数之间的非线性映射。由实验及计算可知,猪舌温度20℃,其导热系数的平均值为0.68 W /( m·K),在10℃至25℃的温度范围内,舌体导热系数变化不大,约从0.67 W /( m·K)变化至0.68 W /( m·K),温度对猪舌导热系数的影响不明显;猪舌纵向截面的导热系数略高于横向截面的导热系数,说明由舌中向舌两侧的导热导热性能优于由舌根到舌尖的导热性能,这种微观的差异用传统的实验方法是无法了解的,而且从实验数据上证实生物组织属于各项异性物质。本文的创新点在于,利用组织光学的方法间接获得生物体热物性参数,使组织光学领域和工程热物理领域实现成功交叉;应用人工神经网络的黑箱思想,建立了一种全新的生物体导热系数测量计算方法。本论文课题是国家自然科学基金(No:54076067)的资助项目。更多还原

【Abstract】 Bio-heat transfer science, which mainly studies the heat and mass transfer law, is an interdisciplinary science of biology, thermophysics and clinical medicine. The research on experiment and numerical calculation of the thermal conductivity of tongue in this paper is done with the method of interdisciplinary science, which is well known in the field of Bio-heat transfer science. Furthermore, as a topic of common concern, it is a key problem that has to be solved during the tongue inspection in Traditional Chinese Medicine mechanism by adopting theory and method of thermal science. Based on theories of tissue optics and artificial neural network, this paper designs a new method of measuring thermal conductivity, with which numerical calculation is applied to acquire thermal conductivity just in virtue of the related parameters instead of differential equation and expression of heat transfer.The FT-IR optical spectrometer and the measuring system of double integrating spheres are used in the experiment in order to acquire the optical parameters of organism in different experiment conditions. Based on the experiment data of optical parameters and the classic data of thermal conductivities, the artificial neural network is established. Absorption coefficient, dissipation coefficient, and aeolotropism coefficient are defined as input variable, and related thermal conductivity is defined as target variable. Consequently the neural network is trained, according to selected learning methods. Finally the thermal conductivity of organism is acquired by emulation course.From the research, the author comes to realize that nonlinear mapping on optical parameters and thermal conductivity could be established by artificial neural network. The result shows, on the condition that the temperature of pig tongue is 20℃, the average of the thermal conductivity is 0.68 W /( m·K). Moreover, in the range of 10℃to 25℃, the thermal conductivity of pig tongue changes from 0.67 W /( m·K)to 0.68 W /( m·K). So it is not evident that the temperature has an influence on the thermal conductivity of pig tongue. The conductivity of pig tongue at vertical section is higher than that at horizontal更多还原