基于自适应人工蜂群算法优化的最小二乘支持向量机在变形预测中的应用

Application in deformation prediction based on LSSVR model optimized by AABC

针对人工蜂群算法(ABC)机制中步长更新及概率选择方面的不足,提出了自适应人工蜂群算法(AABC).在AABC中设计了新的自适应步长更新公式及自适应概率选择公式来平衡算法的勘探和开发能力.采用AABC算法对最小二乘支持向量回归机(LSSVR)进行参数优化,进而构建出基于AABC优化的LSSVR变形预测模型并应用于滑坡变形预测.实验结果表明,AABC有效的解决了ABC过早收敛、收敛精度不高等缺点,且较对比模型来讲,AABC_LSSVR模型预测精度更高,预测趋势更符合实际.

To overcome the shortages of artificial bee colony algorithm（ABC） in step and probability selection,we designed a new adaptive step update formula and adaptive probability selection formula in adaptive artificial bee colony algorithm（AABC） to balance the exploration and development ability of the algorithm.Optimizing the parameters of least squares support vector regression（LSSVR） by AABC, then a deformation prediction model is built and applied to landslide deformation prediction. The experimental results show that AABC effectively solves the shortcomings of premature convergence and low convergence accuracy of ABC.Compared with the former model, the deformation prediction model which optimize by AABC is in higher accuracy and the prediction trend is more grounded in actual situation, and AABC had effectively solved the shortcomings of premature convergence and low convergence accuracy of ABC.