中文摘要：

混凝土超声层析成像检测中,成像结果的稳定性和精确性仍很难满足实际要求。针对这个问题,提出一种基于模拟退火遗传算法的层析成像技术,根据超声波在混凝土中的传播特点,首先在反演算法中引入物理意义明确的自然权函数,形成二次增广目标函数方程;而后加入模拟退火搜索,使层析过程加速,并改善算法收敛性和稳定性;最后通过多位变异自适应遗传算法计算最优反演层析成像结果。数值仿真实验表明,由于利用了正确的先验信息,加上遗传模拟退火算法在寻找函数极值方面具有优良的特性,计算结果准确有效,数值稳定,成像结果能真实有效地反映对象内部缺陷,分辨力得到了明显改善。

英文摘要：

In ultrasonic computerized tomography（CT） imaging test of concrete structures,the stability and accuracy of the imaging results were still difficult to meet practical demands.A CT imaging technology based on simulated annealing genetic algorithm（SAGA） is presented here to solve the problems.Firstly,a natural weight matrix with clear physical meaning was introduced in the inverse algorithm to form the quadric broadening objective function according to the propagating character of ultrasound in concrete.And then the simulated annealing searching is added to speed up inverse process and improve the convergence and stability of the algorithm.And finally,the optimal inverse imaging results could be achieved by variable ectopic adaptive genetic algorithm.It is shown from numerical simulation experiment that because of contribution with the usage of the correct priori information and the excellent characteristic of the SAGA in searching the global minimum value of function,the calculated results are accurate and effective and the numerical value is stable,and the imaging results reflects the inner defections of the tested object reliably and exactly and image resolution is improved obviously.