The vibration produced by blasting excavation in urban underground engineering has a significant influence on the surrounding environment, and the strength of vibration intensity involves many influencing factors. In order to predict the space-time effects of blasting vibration more accurately, an automatic intelligent monitoring system is constructed based on the rough set fuzzy neural network blasting vibration characteristic parameter prediction model and the network blasting vibrator (TC-6850). By setting up the regional monitoring network of monitoring points, the obtained monitoring data are analyzed. An artificial intelligence model is used to predict the influence of stratum condition, excavation hole, and high-rise building on blasting vibration velocity and frequency propagation. The results show that the artificial intelligence prediction model based on a rough set fuzzy neural network can accurately reflect the formation attenuation effect, hollow effect, and building amplification effect of blasting vibration by effectively fuzzing and standardizing the influencing factors. The propagation of blasting vibration in a soil–rock composite stratum is closely related to the surrounding rock conditions with a noticeable elastic modulus effect. The hollow effect is regional, which has a significant influence on the surrounding ground and buildings. Besides, the blasting vibration of the excavated area is stronger than that of the unexcavated area. The propagation of blasting vibration on high-rise buildings was complicated, of which the peak vibration velocity is maximum at the lower level of the building and decreased with the rise of the floor gradually. The whip sheath effect appears at the top floor, which is related to the blasting vibration frequency and the building’s natural vibration frequency.
Read full abstract