基于IWOA-BP神经网络的滑坡区塔线体系应力预测模型

Stress prediction model of tower-line system in landslide area based on IWOA-BP neural network

边坡变形会导致滑坡区输电铁塔基础发生变形,从而引起塔线体系内力增大,若此时再遇强风天气,就极易发生断线、倒塔事故,建立边坡变形和风荷载作用下塔线体系的应力预测模型,可有效预防线路事故的发生。首先利用改进的Circle混沌映射、自适应惯性权重以及Levy飞行策略对鲸鱼算法(WOA)进行优化,进而利用改进的鲸鱼优化算法(IWOA)对BP神经网络的权值和阈值进行参数寻优,得到滑坡区塔线体系应力预测模型,将风向角、风速和塔腿支座位移作为模型输入,塔线体系杆件最大应力作为输出。预测结果表明,提出的IWOA-BP模型具有较高的收敛速度与预测精度,与WOA-BP模型相比,平均绝对误差下降了77.4%,均方根误差下降了82.6%,平均相对误差下降了79.1%。

The deformation of the slope will lead to the deformation of the foundation of the transmission tower in the landslide area,thus causing an increase in the internal force of the tower line system,and if there is a strong wind at this time,it will be very easy to break the line and collapse the tower.Establishing the stress prediction model of the tower line system under the action of slope deformation and wind load can effectively prevent the occurrence of line accidents.Firstly,the improved Circle chaotic mapping,adaptive inertia weights and Levy flight strategy are used to optimise the whale algorithm,and then the improved whale optimisation algorithm is used to optimise the weights and thresholds of the BP neural network to obtain a stress prediction model for the tower line system in the landslide area,with the wind angle,wind speed and tower leg support displacement as model inputs and the maximum stress of the tower line system pole as output.The results of the prediction demonstrate that the proposed IWOA-BP model in this study exhibits remarkable convergence speed and accuracy.Compared with WOA-BP model,the average absolute error decreases by 77.4%,the root mean square error decreases by 82.6%and the average relative error decreases by 79.1%.