论文部分内容阅读
针对漏磁检测中的缺陷反演重构问题,引入了一种新型启发式优化算法—布谷鸟搜索算法,提出了以径向基函数神经网络为前向模型,布谷鸟搜索算法用作迭代算法的漏磁反演方法.为验证该反演方法的有效性,分别使用了不含噪声和含噪声的漏磁仿真信号以及实测漏磁信号.实验结果表明,与粒子群优化算法和差分进化算法相比,布谷鸟搜索算法的处理误差最小,而且对含噪声仿真漏磁信号和实测漏磁信号的重构结果依然能够较好地逼近真实缺陷.因此,基于布谷鸟搜索算法的反演方法对噪声具有一定的鲁棒性,是一种有效可行的漏磁反演方法.
Aiming at the problem of defect inversion and reconstruction in MFL testing, a new heuristic optimization algorithm called cuckoo search algorithm is proposed. The radial basis function neural network is used as the forward model. Cuckoo search algorithm is used as iterative algorithm To verify the validity of this inversion method, the leakage magnetic leakage (EMF) signal without noise and noises and the measured MFL signal are respectively used.The experimental results show that the proposed method is compatible with the PSO and the differential evolution algorithm Compared with the cuckoo search algorithm, the processing error is the smallest, and the reconstructed result of the noise-induced leakage signal and the measured leakage signal still can approach the real defect well.Therefore, the method based on cuckoo search algorithm The noise has a certain robustness and is an effective magnetic flux leakage inversion method.