论文部分内容阅读
在边坡的稳定性评价中保证边坡分类的准确度十分重要.人工非洲野狗群体智能算法,能够通过模拟野狗的觅食行为来对目标函数进行寻优.结合投影寻踪算法以及阻滞增长曲线函数,建立人工非洲野狗优化投影寻踪模型,利用非洲野狗算法AAWDA优化投影指标函数及阻滞增长曲线函数参数,提高了模型求解的准确性,通过对求解后的结果建立回归模型,然后根据分级阈值对边坡等级进行分类,测试结果显示较好的精度.将模型应用于案例边坡的稳定性分析,并同PSO-PP模型,ANN模型所得结果进行比较分析,得出运用AAWDA-PP回归模型预测结果与经验值之间的误差最小,说明模型在研究边坡稳定性评价分级中更加准确有效.
It is very important to assure the accuracy of slope classification in slope stability evaluation.The Artificial African Bobcats intelligent algorithm can optimize the objective function by simulating the foraging behavior of wild dogs.By combining the projection pursuit algorithm and the resistance The model of Artificial African Dusky Optimized Projection Pursuit is established. The projection index function and retardation growth curve function parameters are optimized by African Dachshund algorithm AAWDA, which improves the accuracy of the model. Through the regression of the result of the solution, Model, and then classified the slope grade according to the classification threshold, the test results show good accuracy.The model is applied to the stability analysis of the case slope and compared with the results obtained by the PSO-PP model and the ANN model Using AAWDA-PP regression model, the error between prediction result and experience value is the smallest, which shows that the model is more accurate and effective in studying slope stability evaluation classification.