A temperature prediction model for coal spontaneous combustion based on PSO-SRU deep artificial neural networks
-
摘要: 针对传统煤自燃温度预测模型泛化能力不强、鲁棒性较差的问题，提出了一种基于改进粒子群(PSO)优化简单循环单元(SRU)的煤自燃温度预测模型(PSO−SRU模型)。首先，对煤自燃程序升温实验中采集的气体浓度数据进行预处理，选取与煤温相关性较强的O2，CO，CO2，CH4，C2H4作为煤温预测指标，并将预测指标划分为训练集和测试集；其次，构建SRU预测模型拟合训练集中煤自燃温度与气体指标间非线性规律，将平均绝对误差(MAE)作为适应度函数，利用改进的PSO算法优化SRU预测模型参数；最后，将测试集数据输入参数最优的SRU预测模型，利用SRU计算得到煤自燃温度预测值。实验结果表明：通过指标择优和参数寻优后，PSO−SRU模型在测试集上的MAE相较于基于支持向量回归(SVR)、随机森林（RF）和反向传播（BP）的煤自燃温度预测模型分别降低了12.58，7.65，5.91 ℃，表明PSO−SRU模型在一定程度上提高了预测精度；均方根误差（RMSE）分别降低了22.65，17.45，8.94 ℃，PSO−SRU模型在训练集和测试集上的决定系数(R2)仅相差0.03，表明PSO−SRU模型具有良好的泛化性和鲁棒性。Abstract: Traditional temperature prediction models for coal spontaneous combustion typically have low generality and robustness. This paper improves them by proposing a coal spontaneous combustion temperature prediction model based on particle swarm optimization and simple recurrent unit(PSO-SRU). It firstly pre-processes the gas concentration data collected from temperature programmed oxidation tests, selects the concentration data of O2, CO, CO2, CH4, C2H4 that highly relate to the coal temperature as the prediction indicators, and further separates the indicators into training and testing data sets. Then, a SRU based prediction model over the training data set is trained to learn the nonlinear relationship between the coal spontaneous combustion temperature and the indicators. Mean absolute error(MAE) forms the fitness function and PSO algorithms are involved to optimize the SRU prediction model's parameters. Finally, the PSO-SRU model with optimized parameters are applied over the testing data set to predict the coal spontaneous combustion temperature. Experiments show the PSO-SRU model can improve the prediction accuracy, as the model's MAE and root mean square error(RMSE), comparing with those generated by support vector regression(SVR), random forest(RF), and back propagation(BP), decreases by 12.58, 7.65, 5.91 ℃, and 22.65, 17.45, 8.94 ℃ respectively. The PSO-SRU model also demonstrates a good generality and robustness, as the difference of determination coefficient (R2) of the model over the training and testing data sets is only 0.03.
-
图 1 SRU内部结构和网络结构
Figure 1. Simple recurrent units(SRU) interior structure and network structure
图 2 PSO−SRU模型架构
Figure 2. Temperature prediction model framework for coal spontaneous combustion based on particle swarm optimization and simple recurrent unit (PSO-SRU)
图 5 不同隐藏层数SRU预测模型MAE与时间对比
Figure 5. Comparison of mean absolute errors(MAE) and running time under various hidden layer of simple recurrent units(SRU) perdiction model
图 6 PSO算法改进前后适应度变化曲线
Figure 6. Fitness value change curves before and after improving particle swarm optimization(PSO) algorithm
图 7 不同模型测试样本真实煤温与预测煤温对比
Figure 7. Comparison of real and predicted temperatures of testing samples using different models
表 1 温度与气体指标间的相关性
Table 1. Correlation between temperature and gas indexes
名称 温度 O2 CO CO2 CH4 C2H4 温度 1.00 −0.74 0.80 0.82 0.81 0.72 O2 −0.74 1.00 −0.69 −0.85 −0.72 −0.65 CO 0.80 −0.69 1.00 0.83 0.68 0.61 CO2 0.82 −0.85 0.83 1.00 0.67 0.60 CH4 0.81 −0.72 0.68 0.67 1.00 0.92 C2H4 0.72 −0.65 0.61 0.60 0.92 1.00 
下载: 导出CSV
表 2 不同预测模型的性能预测结果对比
Table 2. Comparison of predictions using various models
模型 MAE/℃ RMSE/℃ R2 训练集 测试集 训练集 测试集 训练集 测试集 SVR 17.35 18.01 27.21 31.34 0.86 0.83 RF 10.81 13.08 14.86 26.14 0.91 0.86 BP 7.49 11.34 12.51 17.63 0.97 0.89 POS−SRU 3.15 5.43 6.27 8.69 0.99 0.96
下载: 导出CSV
-
[1] 谢和平,吴立新,郑德志. 2025年中国能源消费及煤炭需求预测[J]. 煤炭学报,2019,44(7):1949-1960.XIE Heping,WU Lixin,ZHENG Dezhi. Prediction on the energy consumption and coal demand of China in 2025[J]. Journal of China Coal Society,2019,44(7):1949-1960. [2] 林柏泉,李庆钊,周延. 煤矿采空区瓦斯与煤自燃复合热动力灾害多场演化研究进展[J]. 煤炭学报,2021,46(6):1715-1726.LIN Baiquan,LI Qingzhao,ZHOU Yan. Research advances about multi-field evolution of coupled thermodynamic disaster in coal mine goaf[J]. Journal of China Coal Society,2021,46(6):1715-1726. [3] 邓军,白祖锦,肖旸,等. 煤自燃灾害防治技术现状与挑战[J]. 煤矿安全,2020,51(10):118-125.DENG Jun,BAI Zujin,XIAO Yang,et al. Present situation and challenge of coal spontaneous combustion disasters prevention and control technology[J]. Safety in Coal Mines,2020,51(10):118-125. [4] 程卫民,张孝强,王刚,等. 综放采空区瓦斯与遗煤自燃耦合灾害危险区域重建技术[J]. 煤炭学报,2016,41(3):662-671.CHENG Weimin,ZHANG Xiaoqiang,WANG Gang,et al. Reconstruction technology of gas and coal spontaneous combustion coupled hazard in fully mechanized caving goaf[J]. Journal of China Coal Society,2016,41(3):662-671. [5] 张玉涛,李亚清,邓军,等. 煤炭自燃灾变过程突变特性研究[J]. 中国安全科学学报,2015,25(1):78-84.ZHANG Yutao,LI Yaqing,DENG Jun,et al. Study on catastrophe characteristics of coal spontaneous combustion[J]. China Safety Science Journal,2015,25(1):78-84. [6] 秦波涛,仲晓星,王德明,等. 煤自燃过程特性及防治技术研究进展[J]. 煤炭科学技术,2021,49(1):66-99.QIN Botao,ZHONG Xiaoxing,WANG Deming,et al. Research progress of coal spontaneous combustion process characteristics and prevention technology[J]. Journal of Coal Science and Technology,2021,49(1):66-99. [7] 邓军,徐精彩,陈晓坤. 煤自燃机理及预测理论研究进展[J]. 辽宁工程技术大学学报,2003,22(4):455-459.DENG Jun,XU Jingcai,CHEN Xiaokun. Perspectives on spontaneous combustion mechanism and prediction theory of coal[J]. Journal of Liaoning Technical University,2003,22(4):455-459. [8] 朱红青,王海燕,王斐然,等. 煤堆测温技术研究进展[J]. 煤炭科学技术,2014,42(1):50-54.ZHU Hongqing,WANG Haiyan,WANG Feiran,et al. Research progress on coal stockpile temperature measuring technology[J]. Coal Science and Technology,2014,42(1):50-54. [9] 陈欢,杨永亮. 煤自燃预测技术研究现状[J]. 煤矿安全,2013,44(9):194-197.CHEN Huan,YANG Yongliang. Research status of predicting coal spontaneous combustion[J]. Safety in Coal Mines,2013,44(9):194-197. [10] 李林,陈军朝,姜德义,等. 煤自燃全过程高温区域及指标气体时空变化实验研究[J]. 煤炭学报,2016,41(2):444-450.LI Lin,CHEN Junchao,JIANG Deyi,et al. Experimental study on temporal variation of high temperature region and index gas of coal spontaneous combustion[J]. Journal of China Coal Society,2016,41(2):444-450. [11] 郝宇,叶正亮. 不同甲烷气氛下煤自燃指标气体及活化能研究[J]. 工矿自动化,2019,45(11):65-69.HAO Yu,YE Zhengliang. Research on index gas and activation energy of coal spontaneous combustion under different methane atmosphere[J]. Industry and Mine Automation,2019,45(11):65-69. [12] 周冬,刘贞堂,钱继发,等. 采空区内煤自燃气体特征及产生规律分析[J]. 工矿自动化,2019,45(3):18-22.ZHOU Dong,LIU Zhengtang,QIAN Jifa,et al. Analysis of gas characteristics and generation rules of coal spontaneous combustion in goaf[J]. Industry and Mine Automation,2019,45(3):18-22. [13] 金永飞,郭军,文虎,等. 煤自燃高温贫氧氧化燃烧特性参数的实验研究[J]. 煤炭学报,2015,40(3):596-602.JIN Yongfei,GUO Jun,WEN Hu,et al. Experimental study on the high temperature lean oxygen oxidation combustion characteristic parameters of coal spontaneous combustion[J]. Journal of China Coal Society,2015,40(3):596-602. [14] 高峰,王文才,李建伟,等. 浅埋煤层群开采复合采空区煤自燃预测[J]. 煤炭学报,2020,45(增刊1):336-345.GAO Feng,WANG Wencai,LI Jianwei,et al. Prediction of coal spontaneous combustion in compound gob of shallow seam group mining[J]. Journal of China Coal Society,2020,45(S1):336-345. [15] LEI Changkui,DENG Jun,CAO Kai,et al. A random forest approach for predicting coal spontaneous combustion[J]. Fuel,2018,223(1):63-73. [16] 邓军,雷昌奎,曹凯,等. 煤自燃预测的支持向量回归方法[J]. 西安科技大学学报,2018,38(2):175-180.DENG Jun,LEI Changkui,CAO Kai,et al. Support vector regression approach for predicting coal spontaneous combustion[J]. Journal of Xi'an University of Science and Technology,2018,38(2):175-180. [17] 刘宝,穆坤,叶飞,等. 基于相关向量机的煤自燃预测方法[J]. 工矿自动化,2020,46(9):104-108.LIU Bao,MU Kun,YE Fei,et al. Prediction method of coal spontaneous combustion based on relevance vector machine[J]. Industry and Mine Automation,2020,46(9):104-108. [18] 昝军才,魏成才,蒋可娟,等. 基于BP神经网络的煤自燃温度预测研究[J]. 煤炭工程,2019,51(10):113-117.ZAN Juncai,WEI Chengcai,JIANG Kejuan,et al. Prediction of coal spontaneous combustion temperature based on BP neural network[J]. Coal Engineering,2019,51(10):113-117. [19] 郑学召,李梦涵,张嬿妮,等. 基于随机森林算法的煤自燃温度预测模型研究[J]. 工矿自动化,2021,47(5):58-64.ZHENG Xuezhao,LI Menghan,ZHANG Yanni,et al. Research on the prediction model of coal spontaneous combustion temperature based on random forest algorithm[J]. Industry and Mine Automation,2021,47(5):58-64. [20] TAO Lei, YU Zhang, WANG Sidai, et al. Simple recurrent units for highly parallelizable recurrence[C]// Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, 2018: 4470-4481. [21] KENNEDY J, EBERHART R C. Particle swarm optimization[C]//Proceedings of the IEEE International Conference on Neural Networks, Perth, 1995: 1942-1948. [22] 闫群民,马瑞卿,马永翔,等. 一种自适应模拟退火粒子群优化算法[J]. 西安电子科技大学学报,2021,48(4):120-127.YAN Qunmin,MA Ruiqing,MA Yongxiang,et al. Adaptive simulated annealing particle swarm optimization algorithm[J]. Journal of Xidian University,2021,48(4):120-127. [23] 敖永才,师奕兵,张伟,等. 自适应惯性权重的改进粒子群算法[J]. 电子科技大学学报,2014,43(6):874-880.AO Yongcai,SHI Yibing,ZHANG Wei,et al. Improve particle swarm optimization with adaptive inertia weight[J]. Journal of University of Electronic Science and Technology of China,2014,43(6):874-880. [24] 汪雅文, 钱谦, 冯勇, 等. 融合吸引排斥和双向学习的改进粒子群算法[J/OL]. 计算机工程与应用: 1-9. [2021-07-20]. http://kns.cnki.net/kcms/detail/11.2127.TP.20210618.0839.002.html.WANG Yawen, QIAN Qian, FENG Yong, et al. An improved particle swarm optimization algorithm based on attraction-repulsion and bidirectional learning strategies[J/OL]. Computer Science and Application: 1-9. [2021-07-20]. http://kns.cnki.net/kcms/detail/11.2127.TP.20210618.0839.002.html. [25] 姚成玉,赵哲谕,陈东宁,等. 有向动态拓扑混合作用力微粒群优化算法及可靠性应用[J]. 机械工程学报,2017,53(10):166-179. doi: 10.3901/JME.2017.10.166YAO Chengyu,ZHAO Zheyu,CHEN Dongning,et al. Unidirectional dynamic topology hybrid force PSO algorithm and its applications in reliability optimization[J]. Journal of Mechanical Engineering,2017,53(10):166-179. doi: 10.3901/JME.2017.10.166 -
点击查看大图
图(7) / 表(2)
计量
- 文章访问数: 209
- HTML全文浏览量: 97
- PDF下载量: 32
- 被引次数: 0
