近地表空间梯度瞬变电磁粒子群优化反演研究

空间梯度瞬变电磁测量方法利用以发射线圈为对称平面的两个同规格接收线圈信号相减,消除了一次场互感影响,使得早时段瞬变电磁响应得到精确测量,为近地表电性结构层探测提供技术支持。瞬变电磁法一维正演理论研究表明:相比于重叠回线装置,空间梯度瞬变电磁观测模式在早时段(1 ms内)记录具有优势,对近地表30 m深度范围内的介质地电参数变化反映更为灵敏。为实现近地表空间梯度瞬变电磁高精度反演,提出一种基于反向学习策略的自适应混沌粒子群算法(OBL-ACPSO)。利用该算法对理论模型的无噪和含噪瞬变电磁数据进行反演,结果表明,改进的算法显著提高反演精度,抗噪性能突出,为近地表空间梯度瞬变电磁数据的快速反演解释提供了新算法。

Research on PSO Inversion of Spatial Gradient TEM on the Near-surface

In the spatial gradient transient electromagnetic measurement(TEM)method,the signals of two receiving coils of the same specification with the transmitting coil as the symmetry plane are subtracted,so that the mutual inductance of the primary field is eliminated,and TEM response can be accurately measured in the early period,which provides technical support for the detection of near-surface electrical structural layers.1D forward theory of TEM shows that compared with overlapping loop devices,the spatial gradient observation mode has advantages in early recording(within 1ms),and is more sensitive to the change of medium geoelectric parameters within the depth range of 30 meters under the surface.To realize high-precision inversion of near-surface spatial gradient TEM,an opposition-based learning adaptive chaotic particle swarm optimization algorithm(OBL-ACPSO)is proposed.The algorithm is used to invert the noiseless and noisy transient electromagnetic data of the theoretical model,and the results show that the algorithm significantly improves the inversion accuracy and has outstanding anti-noise performance,which provides a new algorithm for fast inversion and interpretation of near-surface spatial gradient TEM data.