Laboratories with sensitive instruments need a low-vibration environment. It is a challenge to control the train-induced vibration impact on these instruments when a newly planned metro line is adjacent to a laboratory building. An alternative method of mitigating train-induced ground vibrations involves installing measures along the transmission path. Recent research has highlighted the potential of periodic pile barriers with specifically designed band gaps for controlling environmental vibrations. This study performed in-situ measurements of ambient vibrations inside and outside a laboratory containing various types of sensitive instruments and located adjacent to a newly designed metro line. The vibration transfer function of the laboratory was then obtained. To help design and optimize the band gaps of periodic piles, a novel band gap performance evaluation function was proposed. Finally, numerical analysis was conducted to validate the mitigation effect of the designed periodic piles. The results showed that the band gap performance evaluation function can be used to optimize the mitigation effect of periodic piles. The proposed periodic piles clearly attenuated vibrations between 52.4 and 74.3 Hz, especially those at 63 Hz. A comparison of general vibration criteria (VC) curves revealed that vibration attenuation of one level can be obtained by the designed periodic piles.

Rail corrugation occurring on tracks fitted with Egg fasteners has become a serious problem on the newly operated lines of the Beijing metro in recent years. Investigations and field measurements show that the dominant wavelengths of corrugation are 63–100 mm, inducing dominant frequencies of vibration on the rail in the frequency range of 160–310 Hz for train speeds at 60–70 km/h. Due to the influence of the dynamic behaviour of tracks on the formation of rail corrugation, various options for changing the dynamic behaviour of the track structure can be considered as possible solutions to treat corrugation. Four treatment measures have been considered: rail grinding, adding a rubber pad within the fastener, replacing the fastener with a compression-type fastener, and installing rail dampers on the rail. Field experiments have been carried out to evaluate the effectiveness of these rail corrugation treatment measures on two lines of the Beijing metro. Frequency response functions, track decay rates, and pass-by acceleration on the rail were measured, and the development of rail roughness was monitored. Experimental results show that all the four measures are effective to control or at least slow down the development of rail corrugation on tracks with Egg fastener. Rail grinding is mostly used to limit the effects of corrugation; here, it was noted that low-amplitude corrugation on some straight lines did not reappear after rail grinding. After adding a rubber pad within the fastener, the existing corrugation reduced in amplitude, while corrugation did not reappear on a newly ground rail. Replacing the fastener with a compression-type fastener and installing rail dampers on the rail were found to slow down the development of corrugation.

To deal with the issues of high computational cost and prediction uncertainty of numerical models in train-induced ground-borne vibration prediction, a prediction method based on transfer learning is proposed in this study. In this method, the vehicle–track-coupled analytical model and three-dimensional finite element model are first used to calculate the train-induced ground vibration under various condition variables, and these data were used as training samples to pre-train the deep neural network models. Numerous train-induced ground vibration experiments were then conducted along the metro lines in Beijing, and those measured vibration data were used to fine-tune the pre-trained deep neural network model with the transfer learning strategy. A random variable obeying a Gaussian distribution is assumed over the predicted vibration acceleration levels to model the randomness of train-induced vibration, and the parameters of this distribution were determined by the statistical results of vibration monitoring data in the metro tunnels. The fully trained model could complete the prediction of train-induced ground vibration in seconds. Finally, a case study was carried out, by comparing the probabilistic prediction results with the statistical results of the field measurements, and the feasibility and the improvement of the proposed method were demonstrated.