为了解决大残量、小残量或零残量问题和普通迭代方法收敛速度慢的问题,引入NL2SOL优化算法,并与GaussNewton法相结合,提出了一种多项式加速的正则化迭代电容层析成像算法.在介绍12电极电容层析成像系统基本原理的基础上,给出了该算法的数学模型,并利用谱分析对该算法的收敛性进行了证明.仿真实验验证了该算法满足收敛条件且重建图像误差小,在电容层析成像应用中具有可行性.%To solve the larger residual,small residual or zero residual problem and ihe slow convergence speed problem in ordinary iter-ation method, introducing NL2SOL optimization algorithm integrate the GaussNewton method, onv polynomial acceleration regularization it-erative capacitance tomography algorithm is proposed. On the basic of introducing the ultimate principle of 12-electrode electrical capacitance tomography system, the mathematical model of this algorithm is presented, and the convergence of this algorithm is proved by using the spectrum analysis. The proposed algorithm satisfy convergence conditions and small reconstruction error and having feasibility in the electri-cal capacitance tomography application is demonstrated through experiments.
展开▼
