摘要
变形是影响混凝土坝安全运行的重要因素之一,对其中径向位移的预测能够保证混凝土坝在运行期间的安全。然而,大坝的变形数据并不是线性变化且容易陷入局部最优,为了解决这一问题,利用非线性灰狼优化算法(TGWO)对轻量梯度提升机(LightGBM)进行优化,构建了一种以统计模型为基础的TGWO-LightGBM混凝土坝变形预测模型。仿真结果表明,TGWO-LightGBM模型相较于GWO-LightGBM模型、LightGBM模型,能够较好地搜索并优化轻量梯度提升机神经网络的参数、更好地平衡全局和局部性能,从而使得LightGBM预测模型具有更高的预测精度。
Deformation is one of the important factors affecting the safe operation of concrete dams,and the prediction of radial displacement can ensure the safety of concrete dams during operation.However,the deformation data of dam are not linear and easily fall into local optimization.In order to solve this problem,the nonlinear Gray Wolf optimization algorithm(TGWO)is used in this paper to optimize the lightweight gradient elevator(LightGBM),and a TGWO-LighTGBM concrete dam deformation prediction model based on statistical model is constructed.The simulation results show that the TGWO-LightGBM model can better search and optimize the parameters of the light gradient elevator neural network,balance the global and local performance,and make the LightGBM prediction model have higher prediction accuracy than the GWO-LightGBM model and LightGBM model.
作者
黄姿慧
顾冲时
王岩博
顾昊
HUANG Zihui;GU Chongshi;WANG Yanbo;GU Hao(College of Water Resources and Hydropower,Hohai University,Nanjing 210024,Jiangsu,China;State Key Laboratory of Hydrology,Water Resources and Hydraulic Engineering Science,Hohai University,Nanjing 210024,Jiangsu,China)
出处
《水力发电》
CAS
2024年第8期89-93,102,共6页
Water Power
基金
国家自然科学基金资助项目(U2243223,52379122)
河海大学中央高校基本科研业务费项目(B230201011)
新疆维吾尔自治区水利科技专项(XSKJ-2023-23)
江苏省科协青年科技人才托举工程(TJ-2022-076)
安徽省自然科学基金联合基金(2208085US17)。
