【摘要】 常规位场异常向梯度张量转换的过程稳定性差。基于Tikhonov正则化原理及不同方向导数换算的内在关系,提出位场梯度张量转换的频率域正则化迭代法。根据L曲线基本原理,给出确定最佳迭代次数的曲率函数法。模型实验结果表明,位场梯度张量转换的频率域正则化迭代法具有较高的计算精度与稳定性,曲率函数法可以准确界定最佳迭代次数。实际资料处理结果验证了方法的有效性。该研究可为位场梯度张量反演提供高质量数据。更多还原

【Abstract】 The paper proposes a regularization iterative method for potential field tensor conversion in frequency domain, based on Tikhonov regularization and the relationship of different direction derivatives so as to improve the stability of conversion and presents curvature function method for determining the optimal number of iterations based on the theory of the L-curve. The model tests show that this method features a higher computational accuracy and stability thanks to the curvature function method capable of an accurate definition of the optimal number of iterations. The actual data processing results prove that the method has a good application effect. The study could provide high-quality data for potential field tensor inversion.更多还原