半航空瞬变电磁法(SATEM)是一种高效的地球物理勘技术，在矿产资源勘探、地下水及地热资源调查等方面得到了广泛的应用。然而，所采集的数据常常受到噪声干扰，对后续的数据处理和解释精度产生了显著影响。为了解决噪声残留及有效信号丢失的问题，提升去噪效果，并减少主观因素的影响，将U-Net运用到处理半航空瞬变电磁数据降噪领域，提出一种基于U-Net深度学习架构的半航空瞬变电磁数据降噪方法。该方法使用U型编解码网络结构，通过端到端的训练方式，自动学习并提取数据中的噪声特征。编码结构学习并提取数据中的噪声信息特征，解码结构重组数据特征还原去噪后数据尺寸。通过在编码和解码结构的对称层上引入跳跃连接，有效融合了包含丰富空间信息的低级特征与包含语义信息的高级特征，从而实现对噪声的准确去除。

实际算例表明，经U-Net去噪后的数据信噪比提升约10 dB，与传统去噪方法相比，U-Net在瞬变电磁数据的噪声去除效果上具有明显优势。在广西贺州至巴马高速公路(来宾至都安段)凤凰2号隧道的实测数据降噪工作中，降噪后的多测道图和视电阻率成像结果的可解释性显著增强。证明本方法在半航空瞬变电磁数据降噪中的重要实际意义，为未来的地球物理勘探提供了有效的技术支持。

The semi-airborne transient electromagnetic (SATEM) method, an efficient geophysical exploration technique, has been extensively applied to mineral resource exploration, groundwater surveys, and geothermal resource surveys. However, the collected data are frequently susceptible to noise interference, significantly affecting the accuracy of subsequent data processing and interpretation. To address issues such as residual noise and the loss of effective signals, enhance denoising effects, and reduce the influence of subjective factors, this study proposed a denoising method for SATEM data based on the U-Net deep learning architecture (also referred to as the U-Net-based method) by applying U-Net to SATEM data denoising. In this method, a U-shaped encoder-decoder architecture is employed to automatically learn and extract noise features from the data through an end-to-end training approach. The encoder learns and extracts noise features from data, while the decoder reconstructs the noise features and restores denoised data. By introducing skip connections to the symmetric layers in the encoder and the decoder, the U-Net-based method effectively integrates the low-level features bearing rich spatial information with the high-level features containing semantic information, thus achieving accurate denoising.

Practical calculation cases indicate that the U-Net-based method can improve the signal-to-noise ratio (SNR) of data by approximately 10 dB after denoising, proving significant advantages of denoising SATEM data compared to traditional denoising methods. This method has been employed to denoise the measured data of the No.2 Fenghuang tunnel in the Laibin-Du'an section of the Hezhou-Bama expressway in Guangxi, significantly enhancing the interpretability of the multi-channel diagrams and apparent resistivity images after data denoising. Therefore, the U-Net-based method holds great practical significance for SATEM data denoising, thus providing effective technical support for future geophysical exploration.