Noise reduction and deflection of pipe tunnel data via intelligent optimization and modal decomposition

Urban underground engineering construction widely uses pipe jacking tunnels due to their trenchless characteristics. However, environmental noise easily affects the monitoring data, making it difficult to extract deflection data. Therefore, this study proposes a model for monitoring data noise reduction and deflection extraction by combining the improved Whale Optimization Algorithm with Variational Mode Decomposition. The improved Whale Optimization Algorithm addresses Variational Mode Decomposition’s parameter selection challenge through enhanced global search capability, while Variational Mode Decomposition provides superior non-stationary signal decomposition compared to traditional methods. Experiments show that under a strong noise environment with a Signal-to-Noise Ratio of -5 dB, the Root Mean Square Error of the reconstructed signal is only 0.143, which demonstrate that the proposed method achieves superior signal reconstruction accuracy even under severe noise conditions, outperforming traditional empirical mode decomposition approaches. Application to actual pipe jacking tunnel projects demonstrates high accuracy in deflection extraction with excellent structural anomaly identification capability and effective early warning performance. The proposed model significantly improves the Signal-to-Noise Ratio and deflection extraction accuracy of pipe jacking tunnel monitoring data. It enhances the reliability of structural health assessment and provides support for the safe operation and maintenance of pipe jacking tunnels.