针对电容层析成像技术中的"软场"效应和病态问题,提出了一种改进信赖域的新电容层析成像算法。在分析电容层析成像基本原理的基础上,推导出了求解ECT反问题的信赖域算法的计算步骤,同时利用BFGS公式对迭代过程中产生的Hesse矩阵进行校正。在此基础上对信赖域算法的收敛性进行了分析和证明,算法满足收敛条件且重建图像误差小。仿真和实验结果表明,和LBP、Landweber和共轭梯度算法相比,对于简单流型该算法兼备成像质量高、边界均匀稳定等优点,为ECT图像重建算法的研究提供了一个新的方法。
Aiming at the "soft field" effect and ill-posed problem in electrical capacitance tomography, an improved trust region based new algorithm is proposed for electrical capacitance tomography. Based on the analysis of the basic principle of electrical capacitance tomography, the calculation steps of the trust region algorithm are deduced for solving ECT inverse problem, and the Hesse matrix produced in the iterative process is corrected using BFGS formula. The convergence characteristic of the trust region algorithm is analyzed; result shows that the proposed algorithm meets the convergence condition and the error of image reconstruction is small. Experiment and simulation results indicate that for simple flow pattern, compared with LBP, Landweber and conjugate gradient algorithms, this new algorithm provides high quality images and favorable stability, which presents a feasible and effective way to the research on image reconstruction for electrical capacitance tomography system.