论文部分内容阅读
真空玻璃保温性能取决于真空度的影响,由传热系数(U值)来表征,却难以测量。真空度也伴随着时间而不断衰减,也会影响到真空玻璃的保温性能。真空玻璃热传导过程具有高度复杂性和非线性。为了实现真空玻璃保温性能的快速评定,采用RBF神经网络对真空玻璃热传导过程进行建模,基于此模型对真空玻璃非热源一侧中心温度进行较为准确的智能软测量,该方法将复杂的传热过程模型化,且具备较快的学习速度,该方法对具有不同传热系数的真空玻璃具有良好的自适应性,将为以后研究真空玻璃真空度以及保温性能检测的智能软测量提供一定的理论基础。
Vacuum glass insulation performance depends on the impact of vacuum, the heat transfer coefficient (U value) to characterize, but difficult to measure. Vacuum is also accompanied by time and constantly decay, will also affect the insulation properties of vacuum glass. Vacuum glass heat transfer process is highly complex and non-linear. In order to quickly evaluate the thermal insulating properties of vacuum glass, the RBF neural network is used to model the thermal conductivity of vacuum glass. Based on this model, the soft center of the vacuum glass on the non-heat source side is measured accurately. The method combines complex heat transfer Process modeling and fast learning speed. The method has good adaptability to vacuum glass with different heat transfer coefficient, and will provide certain theory for the future research on vacuum softness of vacuum glass and thermal insulation performance testing basis.