储罐液位的红外定量辨识  

Quantitative identification for liquid level based on infrared technology

作　　者：宋春霞[1] 陈林 罗兵 SONG Chun-xia;CHEN Lin;LUO Bing(Weaponry Department,Naval Petty Officer Academy,Bengbu 233000,China)

机构地区：[1]海军士官学校兵器系,安徽蚌埠233000

出　　处：《激光与红外》2021年第2期196-201,共6页Laser & Infrared

摘　　要：为了提高储罐液位检测的准确性,提出了结合共轭梯度反演算法利用红外检测技术来定量识别储罐液位的方法。针对研究问题,建立了二维导热下的储罐液位识别模型,并利用有限体积法实现了对正问题模型的求解,在此基础上,利用共轭梯度算法实现了对储罐液位的定量识别,利用数字算例分析了初始假设、测温误差和换热系数对识别结果的影响,数字算例的识别结果表明识别方法有很好的准确性和适应性,最后采用实验验证了识别方法的可行性。In order to improve the accuracy of liquid level detection,the method was proposed by using the infrared technology with the conjugate gradient recognition algorithm.In view of the research problem,the identification model of liquid level detection was established in two-dimensional heat conduction.And on this basis,the direct problem was solved by using finite volume method.The quantitative detection for liquid level was proposed by using conjugate gradient algorithm.Influences of initial guess,the error of temperature measurement and heat transfer coefficient were analyzed by adopting numerical examples.The accuracy and adaptability of identification method were indicated through the identification results of numerical examples.At last the feasibility of identification method was proved through the experiment method.

关 键 词：储罐液位 红外检测 定量检测 共轭梯度 

分 类 号：TN219[电子电信—物理电子学]