基于改进无迹卡尔曼滤波的滑坡监测算法

Landslide Monitoring Algorithm Based on Improved Untraced Kalman filter

在露天矿开采过程中滑坡坍塌是开采过程中常见的问题,滑坡是影响露天矿边坡安全生产的重要因素,露天矿边坡稳定性研究与治理是矿山技术工作中不可缺少的组成部分。因此,提出一种基于改进灰狼算法优化的无迹卡尔曼滤波算法,解决无迹卡尔曼滤波对模型不确定性的鲁棒性差和在系统达到平稳状态时,容易丧失对突变状态的跟踪能力的问题。传统灰狼算法(GWO)易陷入局部最优、后期收敛速度慢等问题,因此提出非线性控制参数组合调整策略,形成改进的灰狼优化算法,运用改进的灰狼优化算法对无迹卡尔曼滤波进行实时优化。结果表明所提算法误差小、精度高,具有良好的预测效果。

Landslide collapse is a common problem during open-pit mining,and it is an important factor affecting the safety production of open-pit mine slopes.The study and treatment of open-pit mine slope stability is an indispensable component of mining technology work.Therefore,a new unscented Kalman filtering algorithm based on improved grey wolf algorithm optimization is proposed to solve the problems of poor robustness of unscented Kalman filtering to model uncertainty and easy loss of tracking ability to sudden changes when the system reaches a stationary state.The traditional grey wolf algorithm(GWO)is prone to problems such as local optima and slow convergence speed in the later stage.Therefore,a nonlinear control parameter combination adjustment strategy is proposed to form an improved grey wolf optimization algorithm.The improved grey wolf optimization algorithm is used for real-time optimization of the unscented Kalman filter.The results show that the proposed algorithm has small errors,high accuracy,and good predictive performance.