摘要
为提高充填管道磨损风险的预测精度,构建基于核主成分分析(KPCA)和自适应粒子群算法(APSO)优化的最小二乘支持向量机(LSSVM)磨损风险预测模型。首先通过KPCA对数据进行特征提取和降维处理,获取影响管道磨损的主要因素,然后应用LSSVM建立磨损风险预测模型,同时利用APSO算法对模型参数进行优化。最后,以黄陵县矿区为例,分析选取12种影响因素,建立充填管道磨损风险指标体系,借助MATLAB进行仿真训练与预测,并对预测结果进行对比分析。结果表明:KPCA-APSO-LSSVM模型与其他模型相比具有更高的预测精度及更强的泛化能力,是一种更为有效的磨损风险预测方法。
In order to improve the prediction accuracy of filling pipeline wear risk,a least square support vector machine(LSSVM)wear risk prediction model based on KPCA and APSO optimization was established.Firstly,feature extraction and dimensionality reduction were performed on the data through KPCA to obtain the main factors affecting pipeline wear.Then,LSSVM was used to establish the wear risk prediction model,and APSO algorithm was used to optimize the model parameters.Finally,taking the mining area of Huangling County as an example,12 influencing factors were analyzed and selected to establish the risk index system of filling pipeline wear.MATLAB was used for simulation training and prediction,and the prediction results were compared and analyzed.The results show that compared with other models,KPCA-APSO-LSSVM model has higher prediction accuracy and stronger generalization ability,and is a more effective method for wear risk prediction.
作者
骆正山
黄仁惠
LUO Zhengshan;HUANG Renhui(Xi′an University of Architecture and Technology,Xi′an 710055,China)
出处
《有色金属工程》
CAS
北大核心
2021年第3期96-106,共11页
Nonferrous Metals Engineering
基金
国家自然科学基金资助项目(41877527)
陕西省社科基金项目(2018S34)。
