作者：张祥壮, 张帅, 李爱莲, 崔桂梅, 杨培宏

作者单位：内蒙古科技大学信息工程学院，内蒙古 包头 014010

关键词：终轧温度预测;鲸鱼优化算法;柯西变异;翻身觅食;余弦控制因子

摘要：

终轧温度是热连轧生产过程中主要控制的工艺参数，是确保带钢质量的重要前提。带钢在精轧阶段经历复杂的换热过程，现场采用的半机理模型很难提高预测精度。针对此问题，从数据驱动角度出发，建立一种基于多策略改进鲸鱼优化算法（IWOA）与极限学习机（ELM）相结合的终轧温度预测模型。融入柯西变异提升鲸鱼算法跳出局部最优的能力；借助余弦控制因子平衡鲸鱼算法全局搜索与局部开发能力；引入翻身觅食策略降低鲸鱼算法陷入局部最优的概率和提升算法的收敛速度。实验结果表明：建立的IWOA-ELM终轧温度预测模型在预报精度方面优势明显，预测终轧温度在±6℃以内的命中率为94%，具有广阔的应用前景。

Prediction of finishing rolling temperature of hot stripmill based on data driving
ZHANG Xiangzhuang, ZHANG Shuai, LI Ailian, CUI Guimei, YANG Peihong
Information Engineering Institute, Inner Mongolia University of Science and Technology, Baotou 014010, China
Abstract: The finishing rolling temperature is the main control process parameter in the hot continuous rolling process, and is an important prerequisite to ensure the quality of strip steel.The strip steel has experienced a complex heat exchange process in the finishing rolling stage, and it is difficult to improve the prediction accuracy of the semi mechanism model adopted in the field. In view of this problem, from the perspective of data-driven, a prediction model of finishing rolling temperature based on the combination of multi strategy improved whale optimization algorithm (IWOA) and extreme learning machine (ELM) is established. Incorporating cauchy variation improves the ability of whale algorithm to jump out of local optimization; Balance the global search and local development capabilities of whale algorithm with cosine control factors; The turning over for food is introduced to reduce the probability of the whale algorithm falling into the local optimum and improve the convergence speed of the algorithm.The experimental results show that the IWOA-ELM finishing rolling temperature prediction model has obvious advantages in prediction accuracy, and the hit rate of predicting the finishing rolling temperature within ±6℃ is 94%, which has broad application prospects.
Keywords: prediction of finishing rolling temperature;whale optimization algorithm;Cauchy variation;turn over for food;cosine control factor
2024, 50(3):152-159  收稿日期: 2022-09-08;收到修改稿日期: 2022-10-28
基金项目: 国家自然科学基金资助项目(61763039)；内蒙古自治区自然科学基金项目资助（2022MS06003）；内蒙古自治区重大专项（2021ZD0029）
作者简介: 张祥壮(1998-),男,山东聊城市人,硕士研究生,专业方向为复杂过程建模与优化控制研究。
参考文献
[1] 张尉. 快速抗扰动预测控制算法及其在精轧系统中的应用[D]. 武汉: 武汉科技大学, 2011.
ZHANG W. Fast disturbance resistant predictive control algorithm and its application in precision rolling systems [D]. Wuhan: Wuhan University of Science and Technology, 2011.
[2] 杨金光, 孙丽荣, 刘华强, 等. 基于PSO和DE优化算法的热轧工作辊热辊形的研究[J]. 塑性工程学报, 2018, 25(3): 289-296.
YANG J G, SUN L R, LIU H Q, et al,A study on the thermal profile of hot rolling work rolls based on PSO and DE optimization algorithms [J].Journal of Plasticity Engineering, 2018, 25 (3): 289-296.
[3] 张博, 郭强, 张飞, 等. 精轧温度模型优化算法与控制策略的研究[J]. 控制工程, 2014, 21(3): 352-356.
ZHANG B, GUO Q, ZHANG F, et al.Research on optimization algorithm and control strategy of precision rolling temperature model [J].Control Engineering of China, 2014, 21 (3): 352-356.
[4] 荆丰伟, 冯莹, 陈兆宇, 等. 数据驱动的热轧带钢头部终轧温度预计算优化[J]. 钢铁研究学报, 2022, 34(7): 655-663.
JING F W, FENG Y, CHEN Z Y, et al.Data driven optimization of pre calculation of final rolling temperature at the head of hot-rolled strip steel [J] .Journal of Iron and Steel Research, 2022, 34 (7): 655-663.
[5] HU B, ZHANG Y J, LU S H, et al. Temperature prediction for finish entry of hot strip mill based on data-driven[C]// Proceedings of the 39th Chinese Control Conference, 2020: 193-199.
[6] 陈果, 何海峰, 吴肖锋, 等. 基于非局部均值-鲸鱼优化算法的套管引线超声波检测方法研究[J]. 中国测试, 2020, 46(4): 102-108.
CHEN G, HE H F, WU X F, et al. Research on ultrasonic detection method for casing leads based on non local mean whale optimization algorithm [J].China Measurement＆ Test, 2020,46 (4): 102-108.
[7] 李爱莲, 全凌翔, 崔桂梅, 等. 融合正余弦和柯西变异的麻雀搜索算法[J]. 计算机工程与应用, 2022, 58(3): 91-99.
LI A L, QUAN L X, CUI G M, et al.Sparrow search algorithm that integrates sine and cosine and Cauchy mutation [J] .Computer Engineering and Applications, 2022, 58 (3): 91-99.
[8] 郑浩, 于俊洋, 魏上斐. 基于余弦控制因子和迭代局部搜索的蝙蝠优化算法[J]. 计算机科学, 2020, 47(s2): 68-72.
ZHENG H, YU J Y, WEI S FI .A bat optimization algorithm based on cosine control factor and iterative local search [J].Computer Science, 2020, 47 (s2): 68-72.
[9] 王正通, 程凤芹, 尤文, 等. 基于翻筋斗觅食策略的灰狼优化算法[J]. 计算机应用研究, 2021, 38(5): 1434-1437.
WANG Z T, CHENG F Q,YOU W, et al.Grey wolf optimization algorithm based on somersault foraging strategy [J].Computer Application Research, 2021, 38 (5): 1434-1437.
[10] 赵沁峰, 蔡艳平, 王新军, 等. 基于WOA-ELM的锂离子电池剩余寿命间接预测[J]. 中国测试, 2021, 47(9): 138-145.
ZHAO Q F, CAI Y P, WANG X J, et al.Indirect prediction of remaining life of lithium-ion batteries based on WOA-ELM [J].China Measurement＆ Test, 2021, 47 (9): 138-145.
[11] 田宏伟, 李志鹏, 王煜伟, 等. CEEMDAN-WOA-ELM模型风机震动趋势预测[J]. 中国测试, 2020, 46(7): 146-152.
TIAN H W, LI Z P, WANG Y W, et al. CEEMDAN-WOA-ELM model for predicting wind turbine vibration trends [J].China Measurement＆ Test, 2020, 46 (7): 146-152.
[12] ZHAO W G, ZHANG Z X, WANG L Y. Manta ray foraging optimization : an effective bio-inspired optimizer for engineering applications[J]. Engineering applications of artificial intelligence, 2020, 87: 103300.
[13] MIRJALILI S, MIRJALILI S M, LEWIS A. Grey wolf optimizer[J]. Advances in Engineering Software, 2014, 69: 46-61.
[14] HEIDARI A A, MIRJALILI S, FARIS H, et al. Harris hawks optimization: algorithm and applications[J]. Future Generation Computer Systems, 2019, 97(8): 849-872.
[15] 张帅, 王俊杰, 李爱莲, 等. 基于改进GEO-ELM的热轧带钢卷曲温度预测[J]. 电子测量技术, 2021, 44(22): 50-55.
ZHANG S, WANG J J, LI A L, et al. Prediction of curling temperature of hot-rolled strip steel based on improved GEO-ELM [J]. Electronic Measurement Technology, 2021, 44 (22): 50-55.
[16] 富永亮. 热轧带钢终轧温度控制模型的研究[D]. 沈阳: 东北大学, 2010.
FU Y L. Research on temperature control model for final rolling of hot rolled strip steel [D]. Shenyang: Northeastern University, 2010.
[17] 李秉晨, 于惠钧, 丁华轩, 等. 基于CEEMD和LSTM-ARIMA的短期风速预测[J]. 中国测试, 2022, 48(2): 163-168.
LI B C, YU H J, DING H X, et al.Short term wind speed prediction based on CEEMD and LSTM-ARIMA [J]. China Measurement＆ Test, 2022, 48 (2): 163-168.