Multi step prediction of dense medium clean coal ash content based on time series alignment and TCNformer
-
摘要: 由于在重介分选过程中各个传感器位置不同，导致重介分选主要工艺参数与灰分存在时间滞后，影响了精煤灰分结果。基于回归模型的灰分预测方法缺乏对时间序列信息的利用，无法捕捉重介生产过程随时间变化的动态特性；基于时间序列的灰分预测方法未能充分考虑灰分和重介分选主要工艺参数之间的时间依赖关系。针对上述问题，提出了一种基于时间序列对齐和TCNformer的重介精煤灰分多步预测方法。通过滞后相关性分析来量化灰分与重介分选主要工艺参数之间的滞后步长，依此对重介分选主要工艺参数在时间维度上进行移动，使得灰分和重介分选主要工艺参数时间序列对齐，消除灰分和重介分选主要工艺参数之间的时间滞后。在Transformer模型的基础上，引入时间卷积网络（TCN）提取特征，并将单向编码器扩展为双向编码器，构建了TCNformer模型来实现精煤灰分多步预测。将时间序列对齐得到的与未来时刻灰分数据对应的过程变量序列作为解码器的输入，以提升模型预测精度。实验结果表明：该方法的平均绝对误差为0.157 9%，均方根误差为0.215 2%，平均皮尔逊相关系数为0.505 1，能有效提升精煤灰分预测精度。Abstract: Due to the different positions of various sensors during the dense medium separation process, there is a time lag between the main process parameters of dense medium separation and ash content, which affects the results of clean coal ash content. The grey prediction method based on regression models lacks the utilization of time series information and cannot capture the dynamic features of the dense medium production process over time. The time series based ash prediction method fails to fully consider the time dependence relationship between the main process parameters of ash content and dense medium separation. In order to solve the above problems, a multi step prediction method for dense medium clean coal ash content based on time series alignment and TCNformer is proposed. The method quantifies the lag step between the main process parameters of ash content and dense medium separation through lag correlation analysis. The method moves the main process parameters of dense medium separation in the time dimension accordingly, aligning the time series of the main process parameters of ash content and dense medium separation, and eliminating the time lag between the main process parameters of ash content and dense medium separation. On the basis of the Transformer model, a time convolutional network (TCN) is introduced to extract features, and the unidirectional encoder is extended to a bidirectional encoder to construct the TCNformer model for multi-step prediction of clean coal ash content. The sequence of process variables corresponding to the grey data at future moments obtained from the time series alignment is used as an input to the decoder to improve the model prediction precision. The experimental results show that the average absolute error of this method is 0.157 9%, the root mean square error is 0.215 2%, and the average Pearson correlation coefficient is 0.505 1, which can effectively improve the precision of predicting clean coal ash content.
-
图 1 某选煤厂两段两产品重介旋流器主再选工艺流程
Figure 1. The main and rewashing separation process flow of a two-stage two product dense medium cyclone in a certain coal preparation plant
图 2 灰分和重介分选主要工艺参数之间的滞后相关性
Figure 2. Time lag correlation between ash content and main process parameters of dense medium separation
表 1 重介分选主要工艺参数
Table 1. Main process parameters of dense medium separation
类型 来源 名称 过程变量 主选系统压力变送器 主选系统重介旋流器入料压力 主选系统压差式密度计 主选系统悬浮液密度 主选系统磁性物含量仪 主选系统磁性物含量 主选系统合格介质桶液位计 主选系统合格介质桶液位 再选系统压力变送器 再选系统重介旋流器入料压力 再选系统压差式密度计 再选系统悬浮液密度 再选系统磁性物含量仪 再选系统磁性物含量 指标变量 多元素煤质分析仪 灰分 铝含量 硅含量 钛含量 钾含量 硫分 水分 
下载: 导出CSV
表 2 消融实验结果
Table 2. Ablation experiments results
预测方法 时间序列对齐 输入未来时刻灰分对应过程变量数据 TCN模块 双向编码器 MAE/% RMSE/% r Transformer × × × × 0.2028 0.2715 0.2633 Transformer+TCN × × √ × 0.1808 0.2425 0.3830 TCNformer × × √ √ 0.1701 0.2304 0.4536 TCNformer+时间序列对齐 √ × √ √ 0.1753 0.2416 0.3925 本文方法 √ √ √ √ 0.1579 0.2152 0.5051
下载: 导出CSV
-
[1] 王然风,高建川,付翔. 智能化选煤厂架构及关键技术[J]. 工矿自动化,2019,45(7):28-32.WANG Ranfeng,GAO Jianchuan,FU Xiang. Framework and key technologies of intelligent coal preparation plant[J]. Industry and Mine Automation,2019,45(7):28-32. [2] 张娟莉,康文泽,张相国,等. 桃山选煤厂精煤灰分预测的探讨[J]. 洁净煤技术,2007,13(4):15-17.ZHANG Juanli,KANG Wenze,ZHANG Xiangguo,et al. Explore of clean coal ash content predicting of Taoshan Coal Washery[J]. Clean Coal Technology,2007,13(4):15-17. [3] 孔利利. 基于精煤灰分预测的重介悬浮液密度自动设定系统设计[J]. 选煤技术,2015(4):68-71.KONG Lili. The design of dense medium suspension density setting system based on clean coal ash prediction[J]. Coal Preparation Technology,2015(4):68-71. [4] 张月飞,王伟,代伟. 重介分选过程产品指标在线预测方法研究[J]. 煤炭工程,2021,53(增刊1):108-111.ZHANG Yuefei,WANG Wei,DAI Wei. On-line prediction of product indicators in dense medium coal separation[J]. Coal Engineering,2021,53(S1):108-111. [5] 李哲,孟巧荣,王然风,等. 基于EMD−RF算法的重介精煤灰分预测研究[J]. 煤炭工程,2023,55(10):174-179.LI Zhe,MENG Qiaorong,WANG Ranfeng,et al. Predictive modeling of heavy refined coal ash based on EMD-RF algorithm[J]. Coal Engineering,2023,55(10):174-179. [6] 程凯,王然风,付翔. 基于EMD−LSTM的重介分选精煤灰分时间序列预测方法研究[J]. 煤炭工程,2022,54(2):133-139.CHENG Kai,WANG Ranfeng,FU Xiang. Time series prediction method of clean coal ash content in dense medium separation based on EMD-LSTM[J]. Coal Engineering,2022,54(2):133-139. [7] YIN Xianhui,NIU Zhanwen,HE Zhen,et al. Ensemble deep learning based semi-supervised soft sensor modeling method and its application on quality prediction for coal preparation process[J]. Advanced Engineering Informatics,2020,46. DOI: 10.1016/j.aei.2020.101136. [8] ZHOU Chunxia,SUN Xiaolu,SHEN Yingsong,et al. Product quality prediction in dense medium coal preparation process based on recurrent neural network[J]. International Journal of Coal Preparation and Utilization,2024,44(3):291-308. doi: 10.1080/19392699.2023.2190098 [9] 付建新,曹师,宋卫东,等. 考虑初始缺陷的超高矿柱蠕变分析及失稳滞后时间研究[J]. 中国矿业大学学报,2017,46(2):279-284.FU Jianxin,CAO Shi,SONG Weidong,et al. Creep analysis and delay time of instability of ultrahigh pillar considering initial imperfections[J]. Journal of China University of Mining & Technology,2017,46(2):279-284. [10] YOU Yalei,MENG Huan,DONG Jun,et al. Time-lag correlation between passive microwave measurements and surface precipitation and its impact on precipitation retrieval evaluation[J]. Geophysical Research Letters,2019,46(14):8415-8423. doi: 10.1029/2019GL083426 [11] EVTUSHEVSKY O M,KRAVCHENKO V O,HOOD L L,et al. Teleconnection between the central tropical Pacific and the Antarctic stratosphere:spatial patterns and time lags[J]. Climate Dynamics,2015,44(7/8):1841-1855. [12] CHEN Jing,DING Ruilian,LIU Kangkang,et al. Collaboration between meteorology and public health:predicting the dengue epidemic in Guangzhou,China,by meteorological parameters[J]. Frontiers in Cellular and Infection Microbiology,2022,12. DOI: 10.3389/fcimb.2022.881745. [13] VASWANI A,SHAZEER N,PARMAR N,et al. Attention is all you need[J]. Advances in Neural Information Processing Systems,2017,30:5998-6008. [14] DEVLIN J,CHANG Mingwei,LEE K,et al. BERT:pre-training of deep bidirectional transformers for language understanding[EB/OL]. [2023-08-22]. https://arxiv.org/abs/1810.04805. [15] DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al. An image is worth 16×16 words:transformers for image recognition at scale[EB/OL]. [2023-08-22]. https://arxiv.org/abs/2010.11929. [16] BAI Shaojie,KOLTER J Z,KOLTUN V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling[EB/OL]. [2023-08-22]. https://arxiv.org/abs/1803.01271. [17] OORD A,DIELEMAN S,ZEN H,et al. WaveNet:a generative model for raw audio[EB/OL]. [2023-08-22]. https://arxiv.org/abs/1609.03499. [18] YU F,KOLTUN V. Multi-scale context aggregation by dilated convolutions[EB/OL]. [2023-08-22]. https://arxiv.org/abs/1511.07122. [19] WILLMOTT C J,MATSUURA K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance[J]. Climate Research,2005,30(1):79-82. [20] KINGMA D P,BA J. Adam:a method for stochastic optimization[EB/OL]. [2023-08-22]. https://arxiv.org/abs/1412.6980. [21] LOSHCHILOV I,HUTTER F. SGDR:stochastic gradient descent with warm restarts[EB/OL]. [2023-08-22]. https://arxiv.org/abs/1608.03983. -
点击查看大图
图(4) / 表(2)
计量
- 文章访问数: 71
- HTML全文浏览量: 20
- PDF下载量: 9
- 被引次数: 0
