摘要
随着现代矿山对安全性和生产效率要求的不断提高,露天矿边坡变形监测与预测已成为保障矿山安全生产的关键技术课题。传统预测方法在复杂地质条件下的预测精度和实时性往往难以满足工程需求。为此本文提出了一种基于河马优化算法优化BP神经网络的预测模型。为验证该模型的有效性,以西南某露天矿边坡为研究对象,进行为期一年的变形监测,对采集的变形数据进行系统处理与分析后,基于此监测数据,对该模型进行训练测试与试验。为验证所提算法的有效性,在试验时将HO-BP模型与BP神经网络模型、GA-BP模型以及SSA-BP模型进行对比,通过对比预测精度指标对模型综合性能进行评估。实验结果表明,HO-BP模型通过结合层次优化策略和反向传播算法,在各项评价指标上均表现出显著优势。HO-BP模型RMSE为1.049,MAE为0.889,MAPE为0.82%,R^(2)为0.990,相较于传统的BP神经网络、GA-BP预测模型和SSA-BP预测模型,HO-BP预测模型在所有评价指标上均有显著提升。这表明,HO-BP预测模型在处理边坡变形预测问题时,具有更高的准确性和可靠性,为矿山安全管理提供新的思路及技术支持。
With the continuous improvement of safety and productivity requirements in modern mines,the monitoring and prediction of slope deformation in open pit mines has become a key technical subject to ensure the safe production of mines.The prediction accuracy and real-time performance of traditional prediction methods under complex geological conditions often difficult to meet the engineering needs.For this reason,a prediction model based on Hippo optimization algorithm optimized BP neural network was proposed.To verify the effectiveness of the model,an open-pit mine slope in Southwest China was taken as the research object,and deformation monitoring was carried out for one year,and the deformation data collected were systematically processed and analyzed,and then based on this monitoring data,training tests and experiments were carried out on the model.To verify the effectiveness of the proposed algorithm,the HO-BP model was compared with the BP neural network model,GA-BP model and SSA-BP model,and the comprehensive performance of the model was evaluated by comparing the prediction accuracy index.The experimental results show that the HO-BP model,by combining the hierarchical optimization strategy and the back-propagation algorithm,exhibits significant advantages in all evaluation indices.The RMSE of the HO-BP model is 1.049,the MAE is 0.889,the MAPE is 0.82%,and the R^(2) is 0.990,which is higher than that of the traditional BP neural network model,the GA-BP prediction model,and the SSA-BP prediction model.The HO-BP prediction model shows significant improvement in all evaluation indexes.This indicates that the HO-BP prediction model has higher accuracy and reliability in dealing with the slope deformation prediction problem,which provides new ideas and technical support for mine safety management.
作者
张焕雄
张成良
王良成
邓涛
ZHANG Huanxiong;ZHANG Chengliang;WANG Liangcheng;DENG Tao(Faculty of Land Resources Engineering,Kunming University of Science and Technology,Kunming 650093,China)
出处
《有色金属(中英文)》
北大核心
2025年第8期1408-1420,共13页
Nonferrous Metals
关键词
露天矿边坡
变形预测
河马优化算法
BP神经网络
矿山安全
open-pit mine slope
deformation prediction
hippopotamus optimization algorithm
BP neural network
mine safety
