锻压技术

基于混合核支持向量机的带钢凸度预测

英文标题：Strip convexity prediction based on hybrid kernel support vector machine

单位：1.首钢京唐钢铁联合有限责任公司 2.华北理工大学电气工程学院 3.北京首钢股份有限公司

分类号：TP335.5

出版年,卷(期):页码：2025,50(7):132-142

摘要：

为了解决热轧带钢凸度预测精度低、泛化能力差的问题，提出了高斯核和多项式核混合的支持向量机（SVM）预测模型。针对混合核支持向量机参数难以确定的问题，提出了使用佳点集、不完全伽玛函数自适应权重和可选择反向学习策略改进的河马算法（IHO）对混合核参数进行寻优。仿真实验结果表明，改进的河马算法的寻优速度快、收敛精度高。在凸度预测实验中，与随机森林、核极限学习机、单一高斯核支持向量机、多项式核支持向量机预测模型相比，混合核支持向量机预测模型的精度分别提高了18.49%、15.75%、28.76%和10.27%，对于实现轧制参数精准优化、有效改善板形边浪、楔形等缺陷具有重要意义。

In order to solve the problems of low prediction accuracy and poor generalization ability of hot-rolled strip convexity, a support vector machine (SVM) prediction model with a mixture of Gaussian kernel and polynomial kernel was proposed. For the problem that the parameters of the hybrid kernel support vector machine were difficult to determine, an improved hippopotamus optimization algorithm (IHO) was proposed to optimize the hybrid kernel parameters by using good point sets, incomplete gamma function adaptive weights and optional reverse learning strategy. Simulation experiment results show that the IHO algorithm has a fast optimization speed and high convergence accuracy. In the convexity prediction experiment, compared with the random forest, kernel extreme learning machine, single Gaussian kernel support vector machine and polynomial kernel support vector machine prediction models, the accuracy of the hybrid kernel support vector machine prediction model is improved by 18.49%, 15.75%, 28.76% and 10.27%, respectively, which is of great significance for achieving the accurate optimization of rolling parameters and effectively improving the defects such as plate edge waves and wedges.

基金项目：

河北省自然科学基金资助项目(F2018209201)；唐山市科技局科技计划资助项目（22130213G)

作者简介：刘文广（1978-），男，硕士，高级工程师 E-mail：464710757@qq.com 通信作者：李子轩（2000-），男，硕士研究生 E-mail：2941512970@qq.com

参考文献：

[1]宋君,董广,丁相阳,等.UCMW机组冷连轧过程板形形成机理及预报模型[J].锻压技术,2024,49(7):243-250.

Song J，Dong G，Ding X Y，et al.Formation mechanism and prediction model of strip flatness in cold tandem rolling process of UCMW mill[J]. Forging & Stamping Technology，2024，49(7)：243-250.

[2]张振,李欣,刘颂,等.基于多类别生产状态的烧结矿转鼓指数预测模型[J].中国冶金,2022,32(1):27-35.

Zhang Z, Li X, Liu S, et al. Predictive model of sinter drum index based on multicategory production status[J].China Metallurgy, 2022, 32(1): 27-35.

[3]汪龙军,丁成砚,范宇超,等.基于深度森林的热轧带钢凸度预测模型[J].轧钢,2023,40(1):90-96.

Wang L J, Ding C Y, Fan Y C, et al. Hot rolled strip crown prediction model based on deep forest[J]. Steel Rolling, 2023,40(1): 90-96.

[4]曹建国,江军,赵秋芳,等.基于数据挖掘的宽厚板板凸度控制[J].中南大学学报,2019,50(11):2743-2752.

Cao J G, Jiang J, Zhao Q F, et al. Wide and heavy plate crown control based on data mining[J]. Journal of Central South University,2019,50(11):2743-2752.

[5]马红,熊雯,张雪荣.基于极限学习机的热轧板凸度预测方法[J].装备维修技术,2023(2):74-77.

Ma H,Xiong W, Zhang X R. Prediction method of hot rolled plate crown based on limit learning machine[J].Equipment Technology, 2023(2):74-77.

[6]郝学斌.深度信念网络算法下的热连轧板凸度预测模型[J].材料与冶金学报,2022,21(5):364-370.

Hao X B. Prediction model of crown of hot rolling based on deep belief network algorithm[J]. Journal of Materials and Metallurgy, 2022,21(5):364-370.

[7]陈楠,李旭,栾峰,等.基于机理与数据驱动的热连轧板凸度组合预测[J].哈尔滨工业大学学报, 2023,55(10):74-81.

Chen N, Li X, Luan F, et al. Prediction of hot strip crowns of hot tandem rolling based on mechanism and data driving[J]. Journal of Harbin Institute of Technology,2023,55(10): 74-81.

[8]王优龙,李维刚,王永强.基于集成特征选择和SVR的热连轧板凸度预测[JL].钢铁,2024,59(1):99-107.

Wang Y L, Li W G, Wang Y Q. Crown prediction of hot strip steel based on integrated feature selection and SVR[J]. Iron & Steel,2024,59(1):99-107.

[9]刘元铭,王振华,王涛,等.热轧带钢出口凸度数据驱动建模及智能化预测分析[J].中国机械工程,2020,31(22):2728-2733.

Liu Y M, Wang Z H, Wang T, et al. Datadriven modeling and intelligent prediction analysis for hot strip outlet crowns[J]. China Mechanical Engineering,2020,31(22): 2728-2733.

[10]姬亚锋,宋乐宝,原浩,等. 基于KPLS与SVM的热连轧板凸度预测[J]. 中国冶金,2021,31(1): 20-24,30.

Ji Y F, Song L B, Yuan H, et al. Strip crown prediction of hotrolled strip based on KPLS integrate SVM[J]. China Metallurgy,2021,31(1): 20-24,30.

[11]郭云川,张长胜,段青娜,等.融合多策略的改进秃鹰搜索算法[J].控制与决策,2024,39(1):69-77.

Guo Y C,Zhang C S,Duan Q N, et al. Improved bald eagle search algorithm fused with multiple rategies[J]. Control and Decision,2024,39(1):69-77.

[12]于军琪,薛志璐,赵安军,等.基于多策略改进麻雀搜索算法的并联冷机系统节能优化[J].控制与决策,2024,39(6):1810-1818.

Yu J Q,Xue Z L,Zhao A J, et al. Optimization of parallel chillers system based on multistrategy improved sparrow search algorithm for energy saving[J]. Control and Decision, 2024,39(6):1810-1818.

[13]李硕,于洋,宋浩源,等.基于凸度转移理念的冷轧支撑辊辊型设计[J].中国冶金,2023,33(9):112-117.

Li S, Yu Y, Song H Y, et al. Roll contour design of cold rolling backup roll based on crown transfer concept[J].China Metallurgy, 2023, 33(9): 112-117.

[14]江德文,王振阳,戴建华,等.基于支持向量机的高炉煤气利用率预测建模[J].中国冶金,2021,31(4):55-63,67.

Jiang D W, Wang Z Y, Dai J H, et al.Forecast modeling of gas utilization rate of blast furnace based on support vector machine[J].China Metallurgy, 2021, 31(4): 55-63,67.

[15]梁盛楠,刘文博,李雅芝.基于改进局部极化准则的多核SVM模型[J].东北师大学报(自然科学版),2023,55(4):32-38.

Liang S N,Liu W B,Li Y Z. Multiple kernel SVM model based on improved local polarization[J]. Journal of Northeast Normal University(Natural Science Edition),2023,55(4):32-38.

[16]崔兆亿,耿秀丽.基于随机森林和量子粒子群优化的SVM算法[J].计算机集成制造系统,2023,29(9):2929-2936.

Cui Z Y, Geng X L. Support vector machine algorithm based on random forest and quantum particle swarm optimization[J]. Computer Integrated Manufacturing Systems,2023,29(9): 2929-2936.

[17]Amiri M H, Mehrabi Hash N, Montazeri M, et al. Hippopotamus optimization algorithm: A novel natureinspired optimization algorithm [J]. Scientific Reports, 2024, 14(1): 5032.

[18]包金山,杨定坤,张靖,等.基于特征提取与INGOSVM的变压器故障诊断方法[J].电力系统保护与控制,2024,52(7):24-32.

Bao J S,Yang D K,Zhang J,et al. Transformer fault diagnosis method based on feature extraction and INGOSVM[J].Power System Protection and Control,2024,52(7):24-32.

[19]黄鹤,高永博,茹锋,等.基于自适应黏菌算法优化的无人机三维路径规划[J].上海交通大学学报,2023,57(10):1282-1291.

Huang H, Gao Y B, Ru F, et al. 3D Path planning of UAV based on adaptive slime mould algorithm optimization[J]. Journal of Shanghai Jiao Tong University, 2023, 57(10): 1282-1291.

[20]蒋宇飞,许贤泽,徐逢秋,等.多策略融合改进的自适应蜉蝣算法[J].北京航空航天大学学报,2024,50(4):1416-1426.

Jiang Y F，Xu X Z，Xu F Q，et al. Multistrategy fusion improved adaptive mayfly algorithm[J]. Journal of Beijing University of Aeronautics and Astronautics，2024，50（4）：1416-1426.

[21]李得恺,张长胜,杨雪松.融合多策略改进的黏菌优化算法[J].模式识别与人工智能,2023,36(7):647-660.

Li D K, Zhang C S, Yang X S. Improved slime mould algorithm fused with multistrategy[J]. Pattern Recognition and Artificial Intelligence, 2023, 36(7): 647-660.

[22]张秀玲,代景欢,康学楠,等.基于多RBF神经网络的板形数据建模[J].矿冶工程,2019,39(6):124-128.

Zhang X L, Dai J H, Kang X N, et al. Plate data modeling based on multiple RBF neural networks[J]. Mining and Metallurgical Engineering, 2019,39(6):124-128.

服务与反馈：

【文章下载】【加入收藏】