TY - JOUR AU - Luo, Lin AU - Li, Pengfei AU - Xing, Zhenhua AU - Zhou, Qian AU - Yang, Hu PY - 2026 TI - Dynamic Inversion for the Mechanical Parameters of Tunnel Surrounding Rock Based on UBPSO Algorithm JO - Chinese Journal of Underground Space and Engineering SN - 1673-0836 SP - 437 EP - 447 VL - 22 IS - 2 AB - With the excavation of tunnels, the surrounding rock is disturbed and damaged, resulting in changes in its geotechnical parameters. To obtain dynamic variation laws of geotechnical parameters accurately, an inversion method for mechanical parameters of tunnel surrounding rock based on updated boundaries particle swarm optimization (UBPSO) algorithm is proposed. Due to the large fluctuation of optimization results with the particle swarm algorithm, an updated boundaries particle swarm algorithm is proposed. By updating the upper and lower limits of search boundaries dynamically, performing reversal and mutation operations on individual historical optimal values, and updating inertia weights adaptively, the updated boundaries particle swarm algorithm achieves high-precision and fast optimization. The stability test of the Ackley function indicates that compared to the particle swarm optimization algorithm, the updated boundaries particle swarm algorithm has the following advantages, fast optimization speed, high accuracy of results, small fluctuation of outcomes, and less sensitivity to getting stuck in local optima. Based on the updated boundaries particle swarm optimization algorithm, a reverse analysis model for mechanical parameters of tunnel surrounding rock using MATLAB-PYTHON-ABAQUS and on-site monitoring data is established. Taking the YK76+470 to YK76+502.5 section of the Tongluoshan tunnel as an example, the model was used to analyze the on-site monitoring data of the tunnel arch settlement. It was found that the elastic modulus of the colluvial soil layer decreased from 0.200 GPa to 0.106 GPa, the internal friction angle dropped from 14° to 12.072°, and the cohesion value fell from 22 kPa to 19.373 kPa. UR - https://doi.org/10.20174/j.JUSE.2026.02.06 DO - 10.20174/j.JUSE.2026.02.06