基于APSO-SVR-GRU模型的白水河滑坡周期项位移预测

Displacement prediction of periodic term of Baishuihe landslide based on APSO-SVR-GRU model

滑坡周期项位移的预测,是研究地质灾害中滑坡变形至关重要的一步。由于单一模型易受偶然因素影响,且无法充分利用有效信息,导致其预测精度不高,适用性不强。基于此,文中提出了一种结合自适应粒子群算法(APSO)、支持向量机回归算法(SVR)、门控神经网络算法(GRU)的组合模型。该模型通过自适应粒子群优化算法对支持向量机回归算法进行参数寻优,确定最优参数组合,然后利用最小二乘法对APSO-SVR模型与GRU模型赋权建立最优权重比组合模型。以三峡白水河滑坡作为研究对象,选取降雨量、库水位及位移量作为周期项位移的影响因子,对模型进行训练验证,结果表明:在白水河滑坡周期项位移预测中,文中所提出的APSO-SVR-GRU组合模型与单一模型相比,具有更高的预测精度和稳定性。

The prediction of landslide periodic term displacement is a crucial step in the study of landslide deformation in geological disasters.Since single prediction model is susceptible to accidental factors and cannot make full use of effective information,its prediction accuracy is not high and its applicability is not strong.In this paper,a combined prediction model combining adaptive particle swarm optimization(APSO),support vector machine regression(SVR)and gated neural network(GRU)algorithm is proposed.The model uses the adaptive particle swarm optimization algorithm to optimize the parameters of the support vector machine regression algorithm,determines the optimal parameter combination,and then uses the least square method to weight the APSO-SVR model and the GRU model to establish the optimal weight ratio combination model.Taking the Baishuihe landslide of the Three Gorges as the research object,selecting precipitation,reservoir water level and displacement as the influence factors of the periodic term displacement,the model is trained and verified.The results show that:in the Baishuihe landslide periodic term displacement prediction,the APSO-SVR-GRU compared with a single model has higher prediction accuracy and stability.