面向多工序锂电池制造过程的多源异构数据缺失值填充方法

doi:10.3969/j.issn.0255-8297.2025.05.006

摘要/Abstract

摘要： 在多工序锂电池制造过程中，数据分析面临多源异构和缺失挑战。针对这一问题，本文提出一种融合自回归积分滑动平均（auto regressive integrated moving average,ARIMA）模型与插值技术的填充方法。该方法通过ARIMA模型提取时间序列数据的趋势与周期性特征，结合插值技术修复因设备故障或数据采集不完整导致的缺失值，增强了对复杂数据变化规律的捕捉能力。通过多组实验表明，该方法在填充精度和数据完整性上均优于均值填充、K近邻填充和单独的插值填充方法。本文提出的ARIMA-插值混合填充模型能够有效提高锂电池制造过程的多源异构的缺失值填充质量，为后续特征提取和分析提供可靠数据基础。

关键词: 缺失值填充, 数据预处理, 多工序生产, 自回归积分滑动平均

Abstract: In multi-process lithium battery manufacturing, data analysis faces challenges of multi-source heterogeneity and missing data. To address these issues, this article proposes a filling method that combines autoregressive integrated moving average model (ARIMA) with interpolation techniques. The proposed method extracts trends and periodic features of time series data through ARIMA models, and integrates interpolation techniques to repair missing values caused by equipment failures or incomplete data collection, enhancing the ability to capture complex data change patterns. Multiple experiments have shown that the proposed ARIMA-interpolation method outperforms traditional techniques such as mean filling, K-nearest neighbor filling, and standalone interpolation in terms of filling accuracy and data integrity. The results confirm that the proposed method effectively improves the quality of data preprocessing in lithium battery manufacturing, providing a reliable data foundation for subsequent feature extraction and analysis.

Key words: missing value filling, data preprocessing, multi-process production, autoregressive integrated moving average (ARIMA)

中图分类号:

TP181

林佳岸, 唐小勇. 面向多工序锂电池制造过程的多源异构数据缺失值填充方法[J]. 应用科学学报, 2025, 43(5): 785-798.

LIN Jiaan, TANG Xiaoyong. Multi-source Heterogeneous Data Missing Value Filling Method for Multi-process Lithium Battery Manufacturing Process[J]. Journal of Applied Sciences, 2025, 43(5): 785-798.

参考文献

[1] 严思思. 智能制造背景下转底炉生产线智能化改造策略[J]. 冶金与材料, 2024, 16(11): 177-179. Yan S S. Intelligent transformation strategy of rotary hearth furnace production line under the background of intelligent manufacturing [J]. Metallurgical and Materials, 2024, 16(11): 177-179. (in Chinese)
[2] 肖睿. 智能制造与工业4.0时代的产业资本配置优化[J]. 商讯, 2024(12): 175-178. Xiao R. Optimization of industrial capital allocation in the era of intelligent manufacturing and industry 4.0[J]. Business News, 2024(12): 175-178. (in Chinese)
[3] 李一凡, 黄景涛, 关海平. 基于aFCM-KNN的风电功率缺失值填补[J]. 计算机仿真, 2024, 41(8): 52-57. Li Y F, Huang J T, Guan H P. Wind power data imputation based on aFCM-KNN [J]. Computer Simulation, 2024, 41(8): 52-57. (in Chinese)
[4] 曹旭. 基于数据挖掘的人力资源数据缺失值填补方法[J]. 自动化技术与应用, 2024, 43(6): 133-136, 155. Cao X. Method of filling missing values in human resource data based on data mining [J]. Techniques of Automation and Applications, 2024, 43(6): 133-136, 155. (in Chinese)
[5] 杜学平, 赵清华, 秘琳, 等. 基于BP-ARIMA的中国月度LNG出厂价格预测模型[J]. 油气储运, 2024, 43(10): 1173-1179, 1188. Du X P, Zhao Q H, Mi L, et al. Prediction model for China’s monthly LNG ex-factory prices based on BP-ARIMA [J]. Oil & Gas Storage and Transportation, 2024, 43(10): 1173-1179, 1188. (in Chinese)
[6] 袁淑娟. 基于时间序列模型ARIMA的校园供水管网暗漏检测研究[J]. 科学技术创新, 2024(17): 94-97. Yuan S J. Research on hidden leakage detection of campus water supply pipe network based on time series model ARIMA [J]. Scientific and Technological Innovation, 2024(17): 94-97. (in Chinese)
[7] Chen X Y, Li Q R, Zeng X H, et al. A hybrid ARIMA-GABP model for predicting sea surface temperature [J]. Electronics, 2022, 11(15): 2359.
[8] Holt C C. Forecasting seasonals and trends by exponentially weighted moving averages [J]. International Journal of Forecasting, 2004, 20(1): 5-10.
[9] Fekade B, Maksymyuk T, Kyryk M, et al. Probabilistic recovery of incomplete sensed data in IoT [J]. IEEE Internet of Things Journal, 2018, 5(4): 2282-2292.
[10] Liu Z Y, Liu X, Meng H, et al. Numerical analysis of the thermal performance of a liquid cooling battery module based on the gradient ratio flow velocity and gradient increment tube diameter [J]. International Journal of Heat and Mass Transfer, 2021, 175: 121338.
[11] 陈新岗, 赵龙, 马志鹏, 等. 基于ISSA-CNN-BiGRU-Attention的锂电池健康状态评估[J]. 电子测量技术, 2024, 47(8): 45-52. Chen X G, Zhao L, Ma Z P, et al. State of health assessment of lithium batteries based on ISSA-CNN-BiGRU-Attention [J]. Electronic Measurement Technology, 2024, 47(8): 45-52. (in Chinese)
[12] 丁杰雄, 李菲, 吕强, 等. 五轴多工序加工检测夹具下的误差成本优化分析[J]. 计算机集成制造系统, 2018, 24(12): 2941-2952. Ding J X, Li F, Lyu Q, et al. Cost analysis and optimization relate to error transmission for multistage manufacturing process with detecting fixture on five-axis machine tool [J]. Computer Integrated Manufacturing Systems, 2018, 24(12): 2941-2952. (in Chinese)
[13] Sarvestani S E, Hatam N, Seif M, et al. Forecasting blood demand for different blood groups in Shiraz using auto regressive integrated moving average (ARIMA) and artificial neural network (ANN) and a hybrid approaches [J]. Scientific Reports, 2022, 12: 22031.
[14] Phung Duy Q, Nguyen Thi O, Le Thi P H, et al. Estimating and forecasting Bitcoin daily prices using ARIMA-GARCH models [J]. Business Analyst Journal, 2024, 45(1): 11-23.
[15] Wang J, Ji T Y, Li M S. A combined short-term forecast model of wind power based on empirical mode decomposition and augmented dickey-fuller test [J]. Journal of Physics: Conference Series, 2021, 2022(1): 012017.
[16] Choudhary A, Kumar S, Sharma M, et al. A framework for data prediction and forecasting in WSN with auto ARIMA [J]. Wireless Personal Communications, 2022, 123(3): 2245-2259.
[17] 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 K X, Qin X, Wang J H, et al. A quality soft sensing method designed for complex multi-process manufacturing procedures [J]. Advanced Engineering Sciences, 2024, 56(6): 3-14. (in Chinese)
[18] 孙天宇, 黄魁东, 杨富强, 等. 多工序制造中的误差流模型及其应用研究进展[J]. 计算机集成制造系统, 2024, 30(7): 2251-2269. Sun T Y, Huang K D, Yang F Q, et al. Regulation and application of stream of variation modeling for multistage manufacturing [J]. Computer Integrated Manufacturing Systems, 2024, 30(7): 2251-2269. (in Chinese)
[19] Sun J. Missing value filling research based on ensemble learning [J]. International Journal of Mathematics and Systems Science, 2024, 7(3): 2578-1839.
[20] Serefoglu Cabuk K, Cengiz S K, Guler M G, et al. Chasing the objective upper eyelid symmetry formula; R2, RMSE, POC, MAE, and MSE [J]. International Ophthalmology, 2024, 44(1): 303.