PSO优化算法在尾矿库空间预测中的应用  

Application of PSO optimization algorithm in tailings pond spatial prediction

作　　者：吕世玮 黄德镛[1] 高聪 贾子月 LYU Shiwei;HUANG Deyong;GAO Cong;JIA Ziyue(Faculty of Land Resources Engineering,Kunming University of Science and Technology,Kunming 650093,China)

机构地区：[1]昆明理工大学国土资源工程学院,昆明650093

出　　处：《有色金属（矿山部分）》2024年第1期126-132,共7页NONFERROUS METALS(Mining Section）

基　　金：“十三五”国家重点研发计划项目(2017YFC0804601)。

摘　　要：由于尾矿库的空间变异特殊性,传统的插值法预测精度不高,误差很大。为解决这一问题,考虑到粒子群算法在求解非线性问题时具有更强的全局搜索能力、并行性和多样性、适应性和自适应性,提出了一种基于粒子群算法优化Kriging模型来拟合尾矿库内部参数数据的克里格空间预测算法,旨在提高尾矿库内部参数信息的拟合精度,减少误差。为验证改进算法的有效性,采用多种预测评价指标进行分析,并与传统Kriging进行交叉验证分析,确定优化算法的可靠性。结果表明:相比于传统的加权最小二乘法拟合法,采用PSO算法拟合Kriging模型参数时,内摩擦角平均误差率降低了23.36%,黏聚力平均误差率降低了8.45%,证明算法可行和有效。Due to the spatial variability of tailings reservoir,traditional interpolation methods have low prediction accuracy and large errors.To address these issues,considering the stronger global search capability,parallelism,diversity,adaptability,and self-adaptability of the particle swarm optimization(PSO)algorithm in solving nonlinear problems,this study proposes a PSO-optimized Kriging spatial prediction algorithm for fitting internal parameter data of tailings dams.The aim is to improve the fitting accuracy of internal parameter information and reduce errors.To verify the effectiveness of the proposed algorithm,multiple prediction evaluation indicators are used for analysis,and cross-validation analysis is performed against traditional Kriging to determine the reliability of the optimization algorithm.The experimental results show that compared to the traditional weighted least squares fitting method when using the PSO algorithm to fit Kriging model parameters,the average error rate of the internal friction angle is reduced by 23.36%and the average error rate of the cohesion is reduced by 8.45%.This demonstrates the feasibility and effectiveness of the proposed algorithm.

关 键 词：尾矿库 空间预测:Kriging 粒子群优化 误差分析 

分 类 号：TD926.4[矿业工程—选矿]