Research on Microseismic Periodic Noise Suppression Method Based on Long Short-Term Memory Network

The signal-to-noise ratio of ground-truth microseismic data is relatively low. Most of the current noise suppression methods are effective in dealing with random noise but neglect the periodic noise present in the microseismic data, leading to poor denoising effects. To address this issue, this paper proposes a new noise suppression method that combines short-time stationarity tests with Long Short-Term Memory (LSTM) algorithms to suppress periodic noise in microseismic data. By processing both simulated and field data and comparing the results with the traditional Variational Mode Decomposition (VMD) algorithm, the experimental results demonstrate that the method proposed in this paper can more effectively suppress periodic background noise in microseismic data, thereby enhancing the signal-to-noise ratio of the data.