东华理工大学核资源与环境国家重点实验室;铀资源探采与核遥感全国重点实验室;东华理工大学;
地浸采铀作为铀矿的绿色开采技术，在生产运行中产生海量数据，利用这些海量数据进行大数据分析和趋势预测，能够提升技术人员制定生产计划的可靠性。目前采用的基于编码器-解码器结构的时序预测模型，由于存在注意力机制，导致计算复杂、内存消耗大。本研究提出深度可分离卷积混合模型，通过动态序列分割模块降低固定分割带来的语义破坏，通过深度可分离卷积混合模块降低模型运行时间并捕获局部和全局特征。结果表明，深度可分离卷积混合网络模型的均方误差(Mean Square Error, MSE)与平均绝对误差(Mean Absolute Error, MAE)相较于时间序列分块自注意力模型(Patch Time Series Transformer, PatchTST)分别降低了1.04%和4.13%,提出的动态序列分割模块的MSE与MAE相较于原有模型分别降低了7.32%和5.03%;在性能对比分析上，深度可分离卷积混合模型的训练速度相较于趋势季节分解线性模型(Decomposition Linear, DLinear)提高了59.91%。建立的模型能够准确预测采区生产运行中硫酸注液量的变化趋势，改善了现有预测模型针对地浸铀矿数据集存在的运行时间长、运行内存大、数据拟合差的问题，可为地浸铀矿生产决策提供理论和实践参考。
| 266 | 0 | 311 |
| 下载次数 | 被引频次 | 阅读次数 |
[1] 阳奕汉，苏学斌，陈梅芳，等.“数字建井”在砂岩型铀矿地浸开发中的探索与应用[J].中国矿业，2022,31(12):64-69.YANG Yihan,SU Xuebin,CHEN Meifang,et al.Exploration and application of "digital well construction" in in-situ leaching of sandstone uranium deposit[J].China Mining Magazine,2022,31(12):64-69(in Chinese).
[2] 张明林，刘洋，吴建勇，等.中国铀矿地质勘查信息化建设现状及“十四五”发展思路[J].世界核地质科学，2021,38(3):287-294.ZHANG Minglin,LIU Yang,WU Jianyong,et al.Construction situation of uranium exploration informatization and development idea for the 14th Five-Year Plan in China[J].World Nuclear Geoscience,2021,38(3):287-294(in Chinese).
[3] 毛鑫磊.信息化技术在内蒙古某地浸采铀矿山中的应用[J].化工管理，2024,694(6):82-84.MAO Xinlei.Application of information technology in a in-situ leaching of uranium mine in Inner Mongolia[J].Chemical Enginering Management,2024,694(6):82-84(in Chinese).
[4] 袁志华，孙占学，周义朋，等.绿色铀矿山研究进展[J].有色金属(冶炼部分),2021(11):85-95.YUAN Zhihua,SUN Zhanxue,ZHOU Yipeng,et al.Research progress of green uranium mines[J].Nonferrous Metals(Extractive Metallurgy),2021(11):85-95(in Chinese).
[5] 王亮，王海峰，霍建党，等.纳岭沟矿床地浸采铀现场试验浸出液铀浓度变化研究[C]//中国核学会.中国核科学技术进展报告(第三卷)——中国核学会2013年学术年会论文集：第2册.北京：核工业北京化工冶金研究院，2013:7.
[6] KUCHIN Y,MUKHAMEDIEV R,YUNICHEVA N,et al.Application of machine learning methods to assess filtration properties of host rocks of uranium deposits in Kazakhstan[J].Applied Sciences,2023,13(19):10958.
[7] 贾皓，廖文胜，杜志明，等.基于人工智能算法的地浸采区产量预测[J].铀矿冶，2023,42(3):10-16.JIA Hao,LIAO Wensheng,DU Zhiming,et al.Production prediction of in-situ leaching mining area based on artificial intelligence algorithm[J].Uranium Mining and Metallurgy,2023,42(3):10-16(in Chinese).
[8] 余东原，罗跃，梁大业，等.基于机器学习方法预测地浸过程中铀浸出金属量的变化[J].有色金属(冶炼部分),2024(2):92-98.YU Dongyuan,LUO Yue,LIANG Daye,et al.Pred-icting the variation of uranium leaching metal content in ground-leaching process based on machine learning methods[J].Nonferrous Metals (Extractive Metallurgy),2024(2):92-98 (in Chinese).
[9] 贾明滔，谭笑，苏学斌，等.基于卷积和长短期记忆网络的地浸开采铀浓度预测研究[J].铀矿地质，2024,40(3):578-586.JIA Mingtao,TAN Xiao,SU Xuebin,et al.Uranium concentration prediction for in-situ leaching based on convolution and long short-term memory networks[J].Uranium Geology,2024,40(3):578-586(in Chinese).
[10] 张宇，马亮，李召坤，等.基于多目标优化算法的“水平井-直井”系统流量决策方法[J/OL].铀矿冶，2024:1-7[2024-11-03].https://doi.org/10.13426/j.cnki.yky.2024.06.05.ZHANG Yu,MA Liang,LI Zhaokun,et al.Flow decision-making method for the "horizontal well-vertical well" system based on multi-objective optimization algorithm[J/OL].Uranium Mining and Metallurgy,2024:1-7[2024-11-03](in Chinese).
[11] 周渊凯，刘祜.基于深度学习方法的测井岩性识别研究[J].铀矿地质，2024,40(2):336-345.ZHOU Yuankai,LIU Hu.Deep learning based lithology recognition of well logging data[J].Uranium Geology,2024,40(2):336-345(in Chinese).
[12] ULYANOV D,VEDALDI A,LEMPITSKY V.Instance normalization:The missing ingredient for fast stylization[EB/OL].(2016-07-27).https://doi.org/10.48550/arXiv.1607.08022.
[13] KIM T,KIM J,TAE Y,et al.Reversible instance normalization for accurate time-series forecasting against distribution shift[C].Kigali:International Conference on Learning Representations (ICLR),2022.
[14] NIE Y Q,NGUYEN N H,SINTHONG P,et al.A time series is worth 64 words:Long-term forecasting with transformers[C].Kigali:International Conference on Learning Representations (ICLR),2023.
[15] DAI J F,QI H Z,XIONG Y W,et al.Deformable convolutional networks[C].Venice:2017 Institute of Electrical and Electronics Engineers (IEEE) International Conferenceon Computer Vision,2017:764-773.
[16] CHEN Z Y,ZHU Y S,ZHAO C Y,et al.DPT:Deformable patch-based transformer for visual recognition[C].Beijing:Association for Computing Machinery,2021:2899-2907.
[17] WANG W X,CHEN W,QIU Q B,et al.Crossformer++:A versatile vision transformer hinging on cross-scale attention[J].Institute of Electrical and Electronics Engineers (IEEE) Transactions on Pattern Analysis and Machine Intelligence,2024,46(5):3123-3136.
[18] ZHOU H Y,ZHANG S H,PENG J Q,et al.Informer:Beyond efficient transformer for long sequence time-series forecasting[J].Association for the Advancement of Artificial Intelligence (AAAI)-21 Technical Tracks,2021,35(12):11106-11115.
[19] DAS A,KONG W,LEACH A,et al.Long-term forecasting with TiDE:Time-series dense encoder[EB/OL].arXiv preprint arXiv:2304.08424,2023.
[20] ZENG A L,CHEN M X,ZHANG L,et al.Are Transformers effective for time series forecasting[J].Proceedings of the 37-th Association for the Advancement of Artificial Intelligence(AAAI) Conference on Artificial Intelligence,2023,37(9):11121-11128.
基本信息:
DOI：10.13426/j.cnki.yky.2024.10.12
中图分类号:TD868;TP183
引用信息:
[1]刘志锋,唐俊贤,林芝宁等.基于深度可分离卷积混合网络模型的地浸采铀注液量预测研究[J].铀矿冶,2025,44(01):9-17.DOI:10.13426/j.cnki.yky.2024.10.12.
基金信息:
中国铀业有限公司-东华理工大学核资源与环境国家重点实验室联合创新基金(2022NRE-LH-14); 国家国防科技工业局核能开发项目“铀裂变瞬发n-γ融合测井及航空监测关键技术研究”; 江西省自然科学基金(20242BAB25084)