[1]刘 艳,黄亚博,曾建航.doi: 10.3969/j.issn.1001-3849.2026.03.008多尺度特征融合下锌镍合金电镀层厚度估计[J].电镀与精饰,2026,(03):62-69.
　LIU Yan,HUANG Yabo,ZENG Jianhang.Thickness estimation of Zinc nickel alloy coating under multi scale feature fusion[J].Plating & Finishing,2026,(03):62-69.
点击复制

doi: 10.3969/j.issn.1001-3849.2026.03.008多尺度特征融合下锌镍合金电镀层厚度估计()

《电镀与精饰》[ISSN:1001-3849/CN:12-1096/TG]

卷:
期数:
2026年03
页码:
62-69
栏目:
出版日期:
2026-03-31

文章信息/Info

Title:
Thickness estimation of Zinc nickel alloy coating under multi scale feature fusion
作者:
刘 艳1; 黄亚博2; 曾建航1
(1. 郑州工程技术学院 信息工程学院,河南 郑州 450044 ;2. 河南大学 计算机与信息工程学院,河南 开封 475004)
Author(s):
LIU Yan1;  HUANG Yabo2;  ZENG Jianhang1
(1. College of Information Engineering, Zhengzhou University of Technology, Zhengzhou 450044, China; 2. College of Computer and Information Engineering, Henan University, Kaifeng 475004, China)
关键词:
多尺度特征; 锌镍合金; 电镀层; 厚度估计; 小波变换
Keywords:
multi-scale features;  zinc nickel alloy;  electroplating layer;  thickness estimation;  wavelet transform
分类号:
TG335 TQ153.2
文献标志码:
A
摘要:
锌镍合金电镀过程存在时序滞后效应，仅基于时域或频域单一尺度分析，难以融合镀层厚度的低频趋势与高频波动特征，导致厚度估计精度受限。针对这一问题，提出了一种多尺度特征融合的锌镍合金电镀层厚度估计方法。基于锌镍合金电镀过程，选取气刀压力、气刀与带钢距离以及生产线速度作为影响镀层厚度的因素。引入离散小波变换提取这些影响因素数据的多尺度特征，获取低频和高频分量特征，将二者分别作为深度信念网络(Deep Belief Network，DBN)和长短期记忆网络(Long Short-Term Memory，LSTM)的输入。利用DBN捕捉低频分量长期厚度累积效应，采用LSTM处理高频分量解析工艺瞬态特征，分别输出厚度估计结果。通过加权融合两个模型的输出，得到最终厚度估计结果，有效解决传统单一分析方法难以同时处理时序滞后效应和周期性波动问题。结果表明，该方法能准确反映工艺参数的影响规律：气刀压力固定为0.5 MPa时，镀层厚度随气刀距离(10～19 mm)增加而上升；气刀距离固定为16 mm时，厚度随气刀压力(0.3～0.6 MPa)增大而下降。在不同生产线速度下，该方法估计结果与实际工艺规律相符，且相比其它方法具有更高的估计精度。
Abstract:
Due to temporal lag effect in the zinc nickel alloy electroplating process, it is difficult to integrate the low-frequency trend and high-frequency fluctuation characteristics of the coating thickness based solely on a single scale analysis in the time or frequency domain, resulting in limited accuracy in thickness estimation. A multi-scale feature fusion method for estimating the thickness of zinc nickel alloy electroplating layer was proposed. Based on the zinc nickel alloy electroplating process, gas knife pressure, distance between gas knife and strip steel, and production line speed were selected as factors affecting the thickness of the coating. Discrete wavelets transform to extract multi-scale features of these influencing factor data were introduced. Low-frequency and high-frequency component features were obtained. Inputs for deep belief network (DBN) and long short term memory (LSTM) were used, respectively. The long-term thickness accumulation effects from low-frequency components were captured by DBN, while the transient process characteristics were analyzed through the processing of high- frequency components by LSTM, with independent thickness estimation outputs being generated by both models. By weighting and fusing the outputs of the two models, the final thickness estimation result was obtained, effectively addressing the difficulty faced by traditional single analysis methods in simultaneously handling temporal lag effects and periodic fluctuations. The experimental results demonstrate that the influence of process parameters can be accurately reflected by this method . When the air knife pressure is fixed at 0.5?MPa, the coating thickness is observed to increase with the air knife distance in the range of 10 to 19?mm. When the air knife distance is fixed at 16?mm, the thickness is seen to decrease as the air knife pressure rises from 0.3 to 0.6?MPa. At different production line speeds, the estimation results of this method are consistent with actual process rules. Compared with other methods, higher estimation accuracy is achieved, providing reliable technical support for coating thickness control

参考文献/References:

[1].文思倩, 闫焉服, 周慧, 等. 电镀参数对电沉积Sn-Ag-Cu合金镀层的影响[J]. 稀有金属与硬质合金, 2024, 52(1): 63-68, 80.
[2].王文建, 常皓, 拜颖乾, 等. 电火花加工金刚石-镍复合电极制备参数优化研究[J]. 模具技术, 2024(3): 57-63.
[3].刘军松, 潘秀, 田驰. 次磷酸钠对锌–镍合金电沉积和镀层耐蚀性的影响[J]. 电镀与涂饰, 2022, 41(2): 77-84.
[4].刘嘉楠, 黄嘉乐, 阳颖飞, 等. CTAB对锌-镍合金基复合电镀液中PTFE分散性的影响[J]. 电镀与涂饰, 2022, 41(17): 1208-1213.
[5].田庚凡, 黎敏, 张颖, 等. 应用超声显微镜技术检测饰品镀金层厚度[J]. 贵金属, 2022, 43(增1): 170-175.
[6].张小刚, 俞东宝, 汤慧, 等. 便携式X射线荧光光谱仪检测贫铀棒材表面镍锌镀层厚度[J]. 冶金分析, 2022, 42(8): 55-61.
[7].张颖, 李雪, 刘彦伯, 等. 贵金属超薄镀层厚度的超声显微测量[J]. 贵金属, 2022, 43(4): 56-63.
[8].张晓东, 史靖文, 白广芝, 等. 基于双通道特征融合的热连轧厚度预测补偿[J]. 锻压技术, 2024, 49(2): 161-171.
[9].刘北元. 垂直电镀挂具对TO封装陶瓷外壳表面镀镍层厚度均匀性的影响[J]. 微纳电子技术, 2024, 61(8): 88-93.
[10].丁浩, 张慧, 孙锡科, 等. 电刷镀-电接触强化复合工艺制备厚镀层研究[J]. 热加工工艺, 2022, 51(24): 92-96.
[11].王爱红, 杨鸿建, 闫建升, 等. 可溶性阳极高速镀锡机组锡层均匀性控制策略研究[J]. 电镀与精饰, 2025, 47(2): 17-22.
[12].文思倩, 闫焉服, 周慧, 等. 电镀参数对电沉积Sn-Ag-Cu合金镀层的影响[J]. 稀有金属与硬质合金, 2024, 52(1): 63-68, 80.
[13].郑乐怡, 王荣, 姚佳雯, 等. ZHL-02置换镀和电镀银层的耐蚀性能研究[J]. 材料保护, 2024, 57(1): 58-64.
[14].林珊玲, 彭雪玲, 王栋, 等. 多尺度增强特征融合的钢表面缺陷目标检测[J]. 光学精密工程, 2024, 32(7): 1075-1086.
[15].李想, 杨宁, 刘伟锋, 等. 基于多尺度特征融合的功耗建模分析方法[J]. 四川大学学报(自然科学版), 2024, 61(3): 188-200.
[16].李立柱, 孟明, 高云园, 等. 基于小波变换的EEG-fNIRS多模态数据融合方法[J]. 传感技术学报, 2023, 36(7): 1064-1072.
[17].刘朕, 朱炳宇, 张景祥. EEG信号识别中可调优化Q因子小波变换的多特征融合算法[J]. 控制理论与应用, 2022, 39(12): 2302-2312.
[18].贺婷, 周宁, 吴啸宇. 基于深度全连接神经网络的储层有效砂体厚度预测[J]. 吉林大学学报(地球科学版), 2023, 53(4): 1262-1274.
[19].朱林青, 张涛, 吕灼恒, 等. 基于LSTM预测模型的应用性能异常检测[J]. 计算机仿真, 2024, 41(5): 536-542.

相似文献/References:

[1]刘明昊,孙玉凤,任春宇,等.碱性锌镍电镀液主要成分的快速在线分析[J].电镀与精饰,2019,(1):27.[doi:10.3969/j.issn.1001-3849.2019.01.006]
　LIU Minghao,SUN Yufeng,REN Chunyu,et al.Rapid Online Analysis of Main Components of Alkaline Zinc?Nickel Alloy Electroplating Solution[J].Plating & Finishing,2019,(03):27.[doi:10.3969/j.issn.1001-3849.2019.01.006]
[2]赵明*,汪长,方代号,等. 滚镀碱性锌镍合金低电区沉积速率改善方案的探讨 [J].电镀与精饰,2023,(9):78.[doi:10.3969/j.issn.1001-3849.2023.09.013]
　Zhao Ming*,Wang Chang,Fang Daihao,et al.Investigation on the improvement scheme of deposit speed in LCD area of barrel alkaline Zn-Ni process[J].Plating & Finishing,2023,(03):78.[doi:10.3969/j.issn.1001-3849.2023.09.013]
[3]李雨昕,丁运虎*,秦 锐,等.doi: 10.3969/j.issn.1001-3849.2025.07.001工艺参数对碱性锌镍合金镀层性能的影响[J].电镀与精饰,2025,(07):1.
　Wang Yipin,He Guangyao.Study on the effect of process parameters on Zn-Ni alloy coating properties Li Yuxin1, Ding Yunhu1,2*, Qin Rui3Hu Xialin1, Huang Chaozhi1,[J].Plating & Finishing,2025,(03):1.

更新日期/Last Update: 2026-03-11