多元高斯声束叠加模型在隐蔽工程的识别应用

吉林建筑大学 土木工程学院,长春 130118

隐蔽工程; 三维图像; 多元高斯声束叠加模型; 数据差异分析; 无损识别

Application and identification of multi-Gaussian beam model in concealed works
LI Ming,SONG Dong-zhe,LIN Ci-kai

School of civil engineering,Jilin Jianzhu university,Changchun 130118,China

concealed works; 3D image; multiple Gaussian beam superposition model; data difference analysis; nondestructive detection

备注

为准确识别地下隐蔽性建筑物的内部基本信息并实现三维图像可视化,将多元高斯声束叠加模型与相控阵超声原理结合对隐蔽性工程进行识别,为无损识别提供一种新方法.结合工程实际情况,分别对地下结构的板和外墙进行识别,绘制出识别值与设计值的数据对比直方图,并结合误差分布走向图对比分析,确保该方法的识别精度.结果表明,该方法可以清晰描述识别物的基本信息,且识别数值接近设计数值,最大误差不超过2 mm.与传统方法相比,既实现了无损识别,又为隐蔽性工程的三维信息识别提供可靠的技术手段.

In order to identify accurately the internal basic information of underground concealed buildings and achieve 3D image visualization,by combining multiple Gaussian beam superposition model with the principle of phased array ultrasound to identify the concealed engineering,which provides a new method for nondestructive identification.Based on a project with this method to discern plate and outer wall in the underground structure.The data comparison histogram of the recognized value and the designed value is drawn to compare the difference,and the error distribution trend graph is combined to ensure the recognition accuracy of the method.The result is extremely closed between identified numeral and designed numeraland the basic information is more clearly,the maximum error does not exceed 2 mm.Compared with the traditional method,both make nondestructive discerning come true and provide reliable evidence for the information identification in concealed works.