Lin Hanjie, Li Li, Qiang Yue, Zhang Yi, Liang Siyu, Xu Xinlong, Li Hongjian, Hu Shengchao
Department of Civil Engineering, Chongqing Three Gorges University, Wanzhou 404100, Chongqing, China.
Heliyon. 2024 Aug 6;10(16):e35871. doi: 10.1016/j.heliyon.2024.e35871. eCollection 2024 Aug 30.
Slope instability through can cause catastrophic consequences, so slope stability analysis has been a key topic in the field of geotechnical engineering. Traditional analysis methods have shortcomings such as high operational difficulty and time-consuming, for this reason many researchers have carried out slope stability analysis based on AI. However, the current relevant studies only judged the importance of each factor and did not specifically quantify the correlation between factors and slope stability. For this purpose, this paper carried out a sensitivity analysis based on the XGBoost and SHAP. The sensitivity analysis results of SHAP were also validated using GeoStudio software. The selected influence factors included slope height ( ), slope angle ( ), unit weight ( ), cohesion ( ), angle of internal friction ( ) and pore water pressure coefficient ( ). The results showed that and were the most and least important influential parameters, respectively. GeoStudio simulation results showed a negative correlation between , , , and slope stability, while a positive correlation between , and slope stability. However, for real data, SHAP misjudged the correlation between and slope stability. Because current AI lacked common sense knowledge and, leading SHAP unable to effectively explain the real mechanism of slope instability. For this reason, this paper overcame this challenge based on the priori data-driven approach. The method provided more reliable and accurate interpretation of the results than a real sample, especially with limited or low-quality data. In addition, the results of this method showed that the critical values of , , , , and in slope destabilization are 18 Kpa, 28°, 32°, 30 m, and 0.28, respectively. These results were closer to GeoStudio simulations than real samples.
边坡失稳可能会导致灾难性后果，因此边坡稳定性分析一直是岩土工程领域的关键课题。传统分析方法存在操作难度大、耗时等缺点，为此许多研究人员开展了基于人工智能的边坡稳定性分析。然而，目前的相关研究仅判断了各因素的重要性，并未具体量化因素与边坡稳定性之间的相关性。为此，本文基于XGBoost和SHAP进行了敏感性分析。SHAP的敏感性分析结果也使用GeoStudio软件进行了验证。所选影响因素包括边坡高度（ ）、边坡角度（ ）、重度（ ）、黏聚力（ ）、内摩擦角（ ）和孔隙水压力系数（ ）。结果表明， 和 分别是最重要和最不重要的影响参数。GeoStudio模拟结果表明， 、 、 、 与边坡稳定性呈负相关，而 、 与边坡稳定性呈正相关。然而，对于实际数据，SHAP误判了 与边坡稳定性之间的相关性。由于当前人工智能缺乏常识知识，导致SHAP无法有效解释边坡失稳的真实机制。为此，本文基于先验数据驱动方法克服了这一挑战。该方法比实际样本能提供更可靠、准确的结果解释，尤其是在数据有限或质量较低的情况下。此外，该方法的结果表明，边坡失稳时 、 、 、 、 的临界值分别为18千帕、28°、32°、30米和0.28。这些结果比实际样本更接近GeoStudio模拟结果。
