硅酸盐学报

基于LVC-TimeMixer的浮法玻璃多类别缺陷预测方法

PDF下载

郑立召 ¹ , 许世清 ¹ , 张志勇 ² , 刘世民 ¹^,² , 王辉 ¹ , 窦丽姣 ¹ , 郝晓辰 ¹

硅酸盐学报 | 第二十八届中国科协年会智能建造关键无机非金属材料–先进玻璃材料智能设计与产业前沿专题-研究论文 2026,54(10): 3578-3592

收起

硅酸盐学报 |第二十八届中国科协年会智能建造关键无机非金属材料–先进玻璃材料智能设计与产业前沿专题-研究论文 2026 , 54 (10) : 3578 -3592

基于LVC-TimeMixer的浮法玻璃多类别缺陷预测方法

全屏

郑立召¹, 许世清¹, 张志勇², 刘世民¹^,², 王辉¹, 窦丽姣¹, 郝晓辰¹

作者信息

^1.燕山大学，河北　秦皇岛　066004

^2.河北省浮法玻璃制造数智化技术创新中心，河北　邢台　054100

通讯作者:

许世清（1985-），男，博士，副教授。

作者简介:

郑立召（1993-），男，博士研究生。

A Multi-Class Defect Prediction Method for Float Glass Based on LVC-TimeMixer

Lizhao ZHENG¹, Shiqing XU¹, Zhiyong ZHANG², Shimin LIU¹^,², Hui WANG¹, Lijiao DOU¹, Xiaochen HAO¹

Affiliations

^1.Yanshan University, Qinhuangdao 066004, Hebei, China

^2.Hebei Technology Innovation Center for Digital and Intelligent Float Glass Manufacturing, Xingtai 054100, Hebei, China

Corresponding Author:

XU Shiqing (1985-), male, Ph.D., Associate Professor. E-mail: ysuxsq@ysu.edu.cn

ZHENG Lizhao (1993-), male, Doctoral candidate. E-mail: zlzbxk@stumail.ysu.edu.cn

出版时间: 2026-09-03 doi: 10.14062/j.issn.0454-5648.20260332

文章导航

摘要

收起

浮法玻璃熔化成形过程具有高温连续运行、多变量强耦合和缺陷响应长时滞等特点，现有在线缺陷检测多属于事后识别，难以及时支撑工艺前馈调节。针对多变量耦合及时滞不确定带来的多类别缺陷预测难题，提出一种时滞感知变量耦合TimeMixer模型（Lag-aware Variable-Coupled TimeMixer，LVC-TimeMixer）。该模型以TimeMixer的多尺度分解与趋势波动混合结构为主干，引入时滞感知编码器和变量分组耦合混合器：前者利用5、10、30、60、120 min和240 min候选滞后尺度显式表征工艺扰动到缺陷响应的延迟关系，后者按照温度、压力、流量、气氛和历史缺陷等工艺语义构建变量组内表征与组间耦合，从而建立熔化成形工艺参数、历史缺陷序列与未来缺陷变化趋势之间的动态映射。基于某浮法玻璃生产线历史数据的验证结果表明，LVC-TimeMixer在平均绝对误差（Mean absolute error，MAE）、均方根误差（Root mean square error，RMSE）、决定系数（Coefficient of determination，R²）和30 min累计缺陷平均绝对误差（30-min cumulative MAE）方面表现最优，对称平均绝对百分比误差（Symmetric mean absolute percentage error，sMAPE）和严重缺陷F1值（F1-score）虽非全部最优但接近优势水平；消融实验进一步表明，时滞感知编码与变量分组耦合具有协同增益。该方法可为浮法玻璃缺陷事前预警、工艺参数优化和智能质量控制提供数据驱动支撑。

关键词

浮法玻璃 / 缺陷预测 / 时滞感知编码 / 变量分组耦合 / 时间序列预测

Abstract

收起

Introduction

Float glass is a fundamental material for architecture, automotive applications, electronic displays and photovoltaics, and its quality stability directly affects downstream processing efficiency and end-use performance. In continuous float glass production, defects such as bubbles, knots, stones and point defects reduce product grade, increase cutting losses and rework costs, and disturb stable operation. Existing online defect inspection systems usually identify defects after they have already formed, so they cannot provide sufficient support for feed-forward adjustment of furnace thermal conditions, combustion state, glass level or tin-bath atmosphere. The melting and forming process is a highly nonlinear, strongly coupled, and long-lag industrial thermal process. Process disturbances are transmitted through the temperature field, pressure field, atmosphere field and glass flow field, and their influence may appear at the inspection end only after tens of minutes or longer. Therefore, this study focuses on short-term prediction of four types of defects with 10 levels in float glass and proposes a lag-aware variable-coupled TimeMixer model, named LVC-TimeMixer, to establish a dynamic mapping from historical process states and defect sequences to future defect trends.

Methods

Historical records from a float glass production line were used to construct the prediction dataset. The input data consisted of 31 process variables covering furnace thermal state, pressure and liquid level, combustion and atmosphere, tin-bath protective gas, feeding and control settings, together with 40 historical defect labels corresponding to Bubble, Knot, Stone and PDefect at ten levels. The raw 10 s records were sorted by timestamp and resampled to a 1 min interval. Window averaging was applied to process variables to suppress high-frequency noise, while defect labels were aggregated by counts in each window; short missing values were treated using forward filling and linear interpolation. Supervised samples were generated by a sliding-window strategy, with a 360 min historical input window and a 30 min prediction horizon. On the basis of the original TimeMixer framework, LVC-TimeMixer introduces a lag-aware encoder and a variable-coupled mixer before the multiscale mixing backbone. The lag-aware encoder explicitly extracts delayed representations under candidate lag scales of 5, 10, 30, 60, 120 min and 240 min and uses a gating mechanism to adaptively select effective lag information. The variable-coupled mixer groups variables according to process semantics, including temperature, pressure, flow, atmosphere and historical defects, and then learns intra-group representations and inter-group coupling relationships. The encoded representation is subsequently fed into the TimeMixer backbone, where multiscale decomposition, Past-Decomposable-Mixing and Future-Multipredictor-Mixing are used to predict the multi-step defect sequence.

Results and discussion

The proposed model was compared with DLinear, PatchTST, iTransformer, the original TimeMixer and WPMixer under the same data split, input window, prediction horizon and evaluation metrics. The results show that LVC-TimeMixer achieved the best overall performance in MAE, RMSE, R² and 30 min cumulative defect MAE, with corresponding values of 10.525, 30.368, 0.616 and 5 194.10, respectively. Although its sMAPE and severe-defect F1 were not the best among all models, they remained close to the leading level, indicating that the proposed lag-aware and variable-coupled structures improve trend fitting and cumulative risk estimation while maintaining competitive severe-defect warning capability. At prediction horizons of 5, 10, 15, 20 min and 30 min, LVC-TimeMixer consistently maintained the lowest MAE and a relatively high R², demonstrating stable short- and medium-term prediction capabilities.

For different defect categories, the optimal model varied because Bubble, Knot, Stone and PDefect have different temporal patterns and process response mechanisms. LVC-TimeMixer did not achieve the best MAE for every single defect type, but its errors for all four defect categories were close to those of the corresponding optimal models, and its overall defect prediction and 30 min cumulative risk estimation were superior. Rolling prediction curves and predicted-vs.-true scatter distributions further indicate that LVC-TimeMixer can better track the rise, fall and local peaks of total defects on chronological test data. Ablation experiments confirm that the lag-aware encoder improves prediction by aligning process disturbances with delayed defect responses, whereas the variable-coupled mixer used alone may be limited when lag alignment is absent. When both modules were introduced simultaneously, the MAE and 30 min cumulative defect MAE were further reduced by 21.22% and 16.60%, respectively, compared with TimeMixer-Lag, proving the synergistic effect between lag-aware encoding and variable-group coupling. Engineering evaluation showed that the average and maximum inference times for a single prediction window were 11.6 ms and 21.3 ms, respectively, on a computing platform equipped with custom 2 × 24 GB VRAM NVIDIA GeForce RTX 4090 GPUs, which is far below the 1 min rolling warning threshold in industrial applications.

Conclusions

In this study, an LVC-TimeMixer model was proposed for short-term multiclass defect prediction in the float glass melting-forming process. By embedding a lag-aware encoder and a variable-coupled mixer into the TimeMixer multiscale trend-fluctuation modeling framework, the model explicitly describes both the delayed relationship between process disturbances and defect responses and the coupling relationship among temperature, pressure, flow, atmosphere and historical defect variables. The comparative experiments, prediction-curve analysis and ablation study demonstrate that the proposed model improves overall prediction accuracy, cumulative risk estimation and prediction stability. The model can provide data-driven support for rolling defect early warning, furnace operating-condition optimization and intelligent quality control in float glass production. However, further research is still required to improve generalization under extreme operating conditions, to verify long-term real-time deployment, and to integrate synchronized stripe-image diagnosis or other multi-source visual information.

Key words

float glass / defect prediction / lag-aware encoding / variable group coupling / time series forecasting

引用本文

郑立召, 许世清, 张志勇, 刘世民, 王辉, 窦丽姣, 郝晓辰. 基于LVC-TimeMixer的浮法玻璃多类别缺陷预测方法. 硅酸盐学报, 2026 , 54 (10) : 3578 -3592 . DOI: 10.14062/j.issn.0454-5648.20260332

Lizhao ZHENG, Shiqing XU, Zhiyong ZHANG, Shimin LIU, Hui WANG, Lijiao DOU, Xiaochen HAO. A Multi-Class Defect Prediction Method for Float Glass Based on LVC-TimeMixer[J]. Journal of the Chinese Ceramic Society, 2026 , 54 (10) : 3578 -3592 . DOI: 10.14062/j.issn.0454-5648.20260332

基金

收起

国家自然科学基金面上项目(52272029)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

[1]

LIU Z Y. Research status on the structure, performance, and defects of float glass[J]. Adv Engineering&PrecisionManufacturing, 2026, 2(1): 1-7.

[2]

QI F L, TAN X, ZHANG Z Z, et al. Glass makes blurs: Learning the visual blurriness for glass surface detection[J]. IEEE Trans Ind Inf, 2024, 20(4): 6631-6641.

[3]

金明芳, 何峰, 谢峻林, 等. 700 t/d浮法玻璃熔窑中玻璃液流动特性与澄清过程研究[J]. 过程工程学报, 2024, 24(4): 425-434.

JIN Mingfang, HE Feng, XIE Junlin, et al. Chin J Process Eng, 2024, 24(4): 425-434.

[4]

彭开香, 秦昕, 王佳浩, 等. 一种面向多工序复杂制造过程的质量软测量方法[J]. 工程科学与技术, 2024, 56(6): 3-14.

PENG Kaixiang, QIN Xin, WANG Jiahao, et al. Adv Eng Sci, 2024, 56(6): 3-14.

[5]

许世清, 陈天麟, 刘林林, 等. 人工智能在平板玻璃生产领域的研究进展与展望: 技术融合与产业变革[J]. 硅酸盐学报, 2025, 53(10): 2912-2928.

XU Shiqing, CHEN Tianlin, LIU Linlin, et al. J Chin Ceram Soc, 2025, 53(10): 2912-2928.

[6]

CHEN T Q, GUESTRIN C. XGBoost: A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco California USA. ACM, 2016: 785-794.

[7]

BECK M, PÖPPEL K, SPANRING M, et al. xLSTM: Extended long short-term memory[C]//Advances in Neural Information Processing Systems 37. Vancouver, BC, Canada. Neural Information Processing Systems Foundation, Inc. (NeurIPS), 2024: 107547-107603.

[8]

JHIN S Y, KIM S, PARK N. Addressing prediction delays in time series forecasting: A continuous GRU approach with derivative regularization[C]//Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. ACM, 2024: 1234-1245.

[9]

LI Z X, PENG K X, CHEN J H, et al. CNN-LSTM-CVA based quality monitoring of waviness for float glass manufacturing process[C]//2023 35th Chinese Control and Decision Conference (CCDC). Yichang, China. IEEE, 2023: 76-81.

[10]

VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems. 2017, 30: 5998-6008.

[11]

ZHOU H Y, ZHANG S H, PENG J Q, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting[J]. Proc AAAI Conf Artif Intell, 2021, 35(12): 11106-11115.

[12]

LIN L, SHANG X N. Time series forecasting of train axle fatigue crack acoustic emission signals by integrating multi-head attention mechanism into DLinear model[J]. Appl Acoust, 2025, 240: 110922.

[13]

QU Z J, MENG Y, HOU X X, et al. Integrated energy short-term multivariate load forecasting based on PatchTST secondary decoupling reconstruction for progressive layered extraction multi-task learning network[J]. Expert Syst Appl, 2025, 269: 126446.

[14]

LIU Y, HU T, ZHANG H, et al. iTransformer: Inverted transformers are effective for time series forecasting[C]//Int Conf Learn Represent. 2024: 11116-11140.

[15]

TANG W, CHEN J, CHEN D, et al. Predicting EV battery state of health using long short term degradation feature extraction and FEA TimeMixer[J]. Sci Rep, 2025, 15: 2200.

[16]

WANG Y X, WU H X, DONG J X, et al. TimeXer: Empowering transformers for time series forecasting with exogenous variables[C]//Advances in Neural Information Processing Systems 37. Vancouver, BC, Canada. Neural Information Processing Systems Foundation, Inc. (NeurIPS), 2024: 469-498.

[17]

MURAD M M N, AKTUKMAK M, YILMAZ Y. WPMixer: Efficient multi-resolution mixing for long-term time series forecasting[J]. Proc AAAI Conf Artif Intell, 2025, 39(18): 19581-19588.

[18]

陈淑勇, 李从云, 王照猛, 等. 浮法工艺玻璃带成形过程的模拟分析[J]. 硅酸盐学报, 2021, 49(2): 340-346.

CHEN Shuyong, LI Congyun, WANG Zhaomeng, et al. J Chin Ceram Soc, 2021, 49(2): 340-346.

2026年第54卷第10期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.14062/j.issn.0454-5648.20260332

接收时间：2026-05-15
首发时间：2026-10-22
出版时间：2026-09-03

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2026-05-15
修回日期：2026-06-21

基金

国家自然科学基金面上项目(52272029)

作者信息

^1.燕山大学，河北　秦皇岛　066004

^2.河北省浮法玻璃制造数智化技术创新中心，河北　邢台　054100

通讯作者:

许世清（1985-），男，博士，副教授。

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/gsyxb/CN/10.14062/j.issn.0454-5648.20260332

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT