基于PSO−Elman神经网络的井底风温预测模型

程磊; 李正健; 史浩镕; 王鑫

doi:10.13272/j.issn.1671-251x.2023090062

基于PSO−Elman神经网络的井底风温预测模型

doi: 10.13272/j.issn.1671-251x.2023090062

程磊^{1, 2,},
李正健¹,
史浩镕¹,
王鑫¹

1.
河南理工大学安全科学与工程学院，河南焦作　454003
2.
河南理工大学煤炭安全生产与清洁高效利用省部共建协同创新中心，河南焦作　454003

基金项目: 国家自然科学基金资助项目（U1904210）。

详细信息

作者简介:
程磊（1970—），男，安徽砀山人，教授，博士，主要从事矿山通风与安全方面的研究工作，E-mail：cheng@hpu.edu.cn

中图分类号: TD727.2
计量
- 文章访问数: 119
- HTML全文浏览量: 30
- PDF下载量: 18
- 被引次数: 0
出版历程
- 收稿日期: 2023-09-20
- 修回日期: 2024-01-21
- 网络出版日期: 2024-01-31

A bottom air temperature prediction model based on PSO-Elman neural network

1.
School of Safety Science and Engineering, Henan Polytechnic University, Jiaozuo 454003, China
2.
Henan Collaborative Innovation Center of Coal Work Safety and Clean-efficiency Utilization, Henan Polytechnic University, Jiaozuo 454003, China

摘要

摘要: 目前井下风温预测大多采用BP神经网络，但其预测精度受学习样本数量的影响，且容易陷入局部最优，Elman神经网络具备局部记忆能力，提高了网络的稳定性和动态适应能力，但仍然存在收敛速度过慢、易陷入局部最优的问题。针对上述问题，采用粒子群优化（PSO）算法对Elman神经网络的权值和阈值进行优化，建立了基于PSO−Elman神经网络的井底风温预测模型。分析得出入风相对湿度、入风温度、地面大气压力和井筒深度是井底风温的主要影响因素，因此将其作为模型的输入数据，模型的输出数据为井底风温。在相同样本数据集下的实验结果表明：Elman模型迭代90次后收敛，PSO−Elman模型迭代41次后收敛，说明PSO−Elman模型收敛速度更快；与BP神经网络模型、支持向量回归模型和Elman模型相比，PSO−Elman模型的预测误差较低，平均绝对误差、均方误差（MSE）、平均绝对百分比误差分别为0.376 0 ℃，0.278 3，1.95%，决定系数$ {{R}}^{\text{2}} $为0.992 4，非常接近1，表明预测模型具有良好的预测效果。实例验证结果表明，PSO−Elman模型的相对误差范围为−4.69%～1.27%，绝对误差范围为−1.06～0.29 ℃，MSE为0.26，整体预测精度可满足井下实际需要。
- 井下热害防治 /
- 井底风温预测 /
- 粒子群优化算法 /
- Elman神经网络 /
- PSO−Elman
Abstract: Currently, most underground wind temperature predictions use BP neural networks. But their prediction precision is affected by the number of learning samples and they are prone to falling into local optima. Elman neural networks have local memory capability, which improves the stability and dynamic adaptability of the network. However, there are still problems such as slow convergence speed and easy falling into local optima. In order to solve the above problems, the particle swarm optimization (PSO) algorithm is used to optimize the weights and thresholds of the Elman neural network. A bottom air temperature prediction model based on the PSO Elman neural network is established. The analysis shows that the relative humidity of the inlet and outlet wind, the surface inlet wind temperature, the surface atmospheric pressure, and the depth of the shaft are the main influencing factors of the bottom air temperature. Therefore, they are used as input data for the model, and the output data of the model is the bottom air temperature. The experimental results on the same sample dataset show that the Elman model converges at 90 iterations and the PSO Elman model converges at 41 iterations. It indicates that the PSO-Elman model converges faster. Compared with the BP neural network model, support vector regression (SVR) model, and Elman model, the prediction error of the PSO-Elman model is significantly reduced. The mean absolute error, mean square error (MSE), and mean absolute percentage error are 0.376 0 ℃, 0.278 3, and 1.95%, respectively. The determination coefficient R² is 0.992 4, which is very close to 1, indicating that the prediction model has good predictive performance. The verification results of the example show that the relative error range of the PSO-Elman model is −4.69%-1.27%, the absolute error range is −1.06-0.29 ℃, and the MSE is 0.26. The overall prediction precision can meet the actual needs of the underground.
- underground heat hazard prevention and control /
- bottom air temperature prediction /
- particle swarm optimization algorithm /
- Elman neural network /
- PSO-Elman

HTML全文

图 1 Elman神经网络结构

Figure 1. Elman neural network structure

下载: 全尺寸图片幻灯片

图 2 PSO算法流程

Figure 2. Flow of particle swarm optimization algorithm

下载: 全尺寸图片幻灯片

图 3 PSO−Elman模型流程

Figure 3. Flow of PSO−Elman model

下载: 全尺寸图片幻灯片

图 4 预测模型的进化曲线

Figure 4. Evolution curves of prediction models

下载: 全尺寸图片幻灯片

图 5 4种预测模型在测试集上的预测结果

Figure 5. Prediction results of four prediction models on test set

下载: 全尺寸图片幻灯片

图 6 4种预测模型在测试集上的预测误差

Figure 6. Prediction errors of four prediction models on test set

下载: 全尺寸图片幻灯片

图 7 井底风温预测值与真实值对比

Figure 7. Comparison between predicted and actual values of bottom air temperature

下载: 全尺寸图片幻灯片

表 1 样本数据

Table 1. Sample data

序号	地面大气压力/Pa	入风温度/℃	入风相对湿度/%	井筒深度/m	井底风温/℃
1	99862.8	26.8	80.23	558.90	25.6
2	99936.2	27.7	76.78	558.90	27.8
3	99868.9	26.5	80.05	552.50	26.5
4	99982.3	27.6	74.26	552.50	27.6
5	99965.1	26.8	78.88	673.20	25.3
$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $
63	92240.0	14.0	51.70	417.54	16.2
64	91860.0	13.8	58.40	417.54	16.0
65	91420.0	12.2	67.60	417.54	15.2

下载: 导出CSV

表 2 4种模型的井底风温预测结果及误差

Table 2. Prediction results and errors of bottom air temperature of four models

样本编号	真实值/℃	BP神经网络模型			SVR模型			Elman模型			PSO−Elman模型
样本编号	真实值/℃	预测值/℃	绝对误差/℃	相对误差/%	预测值/℃	绝对误差/℃	相对误差/%	预测值/℃	绝对误差/℃	相对误差/%	预测值/℃	绝对误差/℃	相对误差/%
1	27.9	27.2131	−0.6869	−2.46	27.8983	−0.0017	−0.01	27.7709	−0.1291	−0.46	27.8026	−0.0974	−0.35
2	27.6	26.9595	−0.6405	−2.32	27.7180	0.1180	0.43	27.0653	0.0053	0.02	27.6048	0.0048	−0.02
3	28.9	27.2982	−1.6018	−5.54	27.9765	−0.9235	−3.20	27.8461	−1.0539	−3.65	27.8981	−1.0019	−3.47
4	27.4	27.1367	−0.2633	−0.96	27.6248	0.2248	0.82	27.5246	0.1246	0.45	27.6149	0.2149	0.78
5	28.0	27.4548	−0.5452	−1.95	28.3935	0.3935	1.41	28.1950	0.1950	0.70	28.1494	0.1494	0.53
6	16.2	16.4894	0.2894	1.79	15.2606	−0.9394	−5.80	15.1536	−1.0464	−6.46	16.3820	0.1820	1.12
7	15.8	14.8908	−0.9092	−5.75	14.6212	−1.1788	−7.46	14.6948	−1.1052	−6.99	14.7649	−1.0351	−6.55
8	16.2	15.2269	−0.9731	−6.01	15.0037	−1.1963	−7.38	14.9791	−1.2209	−7.54	15.4598	−0.7402	−4.57
9	16.0	16.4385	0.4385	2.74	14.8663	−1.1337	−7.09	14.8427	−1.1573	−7.23	15.7786	−0.2214	−1.38
10	15.2	17.7870	2.5870	17.02	13.7536	−1.4464	−9.52	13.9971	−1.2029	−7.91	15.3130	0.1130	−0.74

下载: 导出CSV

表 3 4种预测模型的评估指标

Table 3. Evaluation indicators of four prediction models

模型	MAE/℃	MSE	MAPE/%	$ {R}^{2} $
BP神经网络	0.8935	1.2557	4.65	0.9658
SVR	0.7556	0.8153	4.31	0.9985
Elman	0.7241	0.7774	4.14	0.9788
PSO−Elman	0.3760	0.2783	1.95	0.9924

下载: 导出CSV

表 4 井下实测数据

Table 4. Underground measured data

测点位置	入风温度/℃	入风相对湿度/%	地面大气压力/Pa	井筒深度/m	井底风温/℃
副井	20.8	71.40	104610	521.5	24.3
副井	21.2	82.00	104660	521.5	23.0
回风联络巷	21.6	88.00	102668	521.5	22.6
一车场	21.8	86.90	104633	521.5	22.8
运输大巷	22.4	78.80	106604	521.5	24.6

下载: 导出CSV

表 5 井底风温预测数据评估结果

Table 5. Evaluation results of prediction data of bottom air temperature

真实值/℃	预测值/℃	绝对误差/℃	相对误差%	MSE
24.3	24.46	0.14	0.66	0.26
23.0	22.82	−0.18	−0.78
22.6	21.54	−1.06	−4.69
22.8	23.09	0.29	1.27
24.6	24.38	−0.22	−0.89

下载: 导出CSV

参考文献(21)

[1]	王超,李岳峰,张成良. 基于不同指标无量纲化方法的岩爆预测模型优选[J]. 中国安全生产科学技术,2020,16(2):24-29. WANG Chao,LI Yuefeng,ZHANG Chengliang. Optimization of rockburst prediction model based on different index dimensionless methods[J]. Journal of Safety Science and Technology,2020,16(2):24-29.
[2]	杨小彬,裴艳宇,程虹铭,等. 基于SOFM神经网络模型的岩爆烈度等级预测方法[J]. 岩石力学与工程学报,2021,40(增刊1):2708-2715. YANG Xiaobin,PEI Yanyu,CHENG Hongming,et al. Prediction method of rockburst intensity grade based on SOFM neural network model[J]. Chinese Journal of Rock Mechanics and Engineering,2021,40(S1):2708-2715.
[3]	付华,付昱,赵俊程,等. 基于KPCA−ARIMA算法的瓦斯涌出量预测[J]. 辽宁工程技术大学学报(自然科学版),2022,41(5):406-412. FU Hua,FU Yu,ZHAO Juncheng,et al. Prediction of gas emission based on KPCA-ARIMA algorithm[J]. Journal of Liaoning Technical University(Natural Science),2022,41(5):406-412.
[4]	马恒,任美学,高科. 基于随机搜索优化XGBoost的瓦斯涌出量预测模型[J]. 中国安全生产科学技术,2022,18(5):129-134. MA Heng,REN Meixue,GAO Ke. Prediction model of gas emission amount based on XGBoost optimized with random search algorithm[J]. Journal of Safety Science and Technology,2022,18(5):129-134.
[5]	汤国水,张宏伟,韩军,等. 基于MABC−SVM的含瓦斯煤体渗透率预测模型[J]. 中国安全生产科学技术,2015,11(2):11-16. TANG Guoshui,ZHANG Hongwei,HAN Jun,et al. Prediction model on permeability of gas-bearing coal based on MABC-SVM[J]. Journal of Safety Science and Technology,2015,11(2):11-16.
[6]	谢丽蓉,路朋,范文慧,等. 基于LVQ−CPSO−BP算法的煤体瓦斯渗透率预测方法研究[J]. 采矿与安全工程学报,2017,34(2):398-404. XIE Lirong,LU Peng,FAN Wenhui,et al. LVQ-CPSO-BP-based prediction technique of coal gas permeability rate[J]. Journal of Mining & Safety Engineering,2017,34(2):398-404.
[7]	高佳南,吴奉亮,马砺,等. 矿井淋水井筒风温PSO-SVR预测方法[J]. 西安科技大学学报,2022,42(3):476-483. GAO Jianan,WU Fengliang,MA Li,et al. PSO−SVR prediction method of airflow temperature of shaft with water dropping in mine[J]. Journal of Xi'an University of Science and Technology,2022,42(3):476-483.
[8]	纪俊红,马铭阳,崔铁军,等. GSK−XGBoost模型在井底风温预测中的应用[J]. 中国安全生产科学技术,2022,18(3):131-136. JI Junhong,MA Mingyang,CUI Tiejun,et al. Application of GSK-XGBoost model in prediction of wind temperature at well bottom[J]. Journal of Safety Science and Technology,2022,18(3):131-136.
[9]	吕品,左金宝,倪小军. 基于BP神经网络的矿井淋水井筒风温预测[J]. 煤矿安全,2008,39(12):11-13. LYU Pin,ZUO Jinbao,NI Xiaojun. BP neural network for predicting air temperature of watering well in mine[J]. Safety in Coal Mines,2008,39(12):11-13.
[10]	张翔,王佰顺,徐硕,等. 基于PSO−BP的矿井淋水井筒风温预测[J]. 煤矿安全,2012,43(11):178-181. ZHANG Xiang,WANG Baishun,XU Shuo,et al. Prediction of airflow temperature of shafts with water dropping based on PSO−BP neural network[J]. Safety in Coal Mines,2012,43(11):178-181.
[11]	马恒,刘亮亮. 基于T−S模糊神经网络的淋水井筒温度预测分析[J]. 世界科技研究与发展,2015,37(3):226-229. MA Heng,LIU Liangliang. Prediction analysis of temperature of wellbore with spay water based on T-S fuzzy neural network[J]. World Sci-tech R & D,2015,37(3):226-229.
[12]	段艳艳. 基于支持向量机的矿井风温预测[D]. 西安:西安科技大学,2013. DUAN Yanyan. Prediction of the mine air temperature based on support vector machine[D]. Xi'an:Xi'an University of Science and Technology,2013.
[13]	张群. 潘三矿热害调查及风温预测研究[D]. 淮南:安徽理工大学,2015. ZHANG Qun. The thermal damage investigation and wind temperature prediction research of Panji No. 3 Colliery[D]. Huainan:Anhui University of Science and Technology,2015.
[14]	何启林,任克斌. 深井建井期入风井筒风温的预测[J]. 煤炭工程,2002,34(8):47-48. doi: 10.3969/j.issn.1671-0959.2002.08.020 HE Qilin,REN Kebin. Prediction of air temperature in the inlet well during deep well construction period[J]. Coal Engineering,2002,34(8):47-48. doi: 10.3969/j.issn.1671-0959.2002.08.020
[15]	顾伟红,赵雪. 基于GRA−SSA−Elman的隧道施工瓦斯安全性预测评价[J]. 自然灾害学报,2023,32(6):230-239. GU Weihong,ZHAO Xue. Gas safety prediction and evaluation of tunnel construction based on GRA-SSA-Elman[J]. Journal of Natural Disasters,2023,32(6):230-239.
[16]	乔寅威,贾新春,关燕鹏,等. 基于DE−WOA的Elman神经网络的空气质量预测方法及应用 [J/OL]. 控制工程,1-8[2023-08-17]. https://doi.org/10.14107/j.cnki.kzgc.20220929. QIAO Yinwei,JIA Xinchun,GUAN Yanpeng,et al. Air quality prediction method and its application of Elman neural network based on DE-WOA[J/OL]. Control Engineering of China,1-8[2023-08-17]. https://doi.org/10.14107/j.cnki.kzgc.20220929.
[17]	吴定会,朱勇,范俊岩,等. 基于双重注意力IJAYA−Elman的高炉煤气柜位预测[J/OL]. 控制工程,1-9[2023-08-17]. https://doi.org/10.14107/j.cnki.kzgc.20220463. WU Dinghui,ZHU Yong,FAN Junyan,et al. Prediction of blast furnace gas holder lever based on double attention IJAYA-Elman[J/OL]. Control Engineering of China,1-9[2023-08-17]. https://doi.org/10.14107/j.cnki.kzgc.20220463.
[18]	李练兵,朱乐,李思佳,等. 基于差分电压和ICS−Elman神经网络的锂离子电池剩余使用寿命预测方法[J]. 太阳能学报,2023,44(12):433-443. LI Lianbing,ZHU Le,LI Sijia,et al. Remaining useful life prediction of lithium-ion batteries based on differential voltage and ICS-Elman neural networks[J]. Acta Energiae Solaris Sinica,2023,44(12):433-443.
[19]	韩会宾. 基于目标PSO算法的工程项目优化研究[J]. 西南大学学报(自然科学版),2023,45(12):167-177. HAN Huibin. Research on engineering project optimization based on target PSO algorithm[J]. Journal of Southwest University(Natural Science Edition),2023,45(12):167-177.
[20]	焦龄霄,周凯,张子熙,等. 基于mRMR−IPSO的短期负荷预测双阶段特征选择 [J/OL]. 重庆大学学报,1-13[2023-08-17]. http://kns.cnki.net/kcms/detail/50.1044.N.20231206.1636.002.html. JIAO Lingxiao,ZHOU Kai,ZHANG Zixi,et al. Dual-stage feature selection for short-term load forecasting based on mRMR-IPSO[J/OL]. Journal of Chongqing University,1-13[2023-08-17]. http://kns.cnki.net/kcms/detail/50.1044.N.20231206.1636.002.html.
[21]	吴以童,李斌,刘春,等. 融合GMM和PSO的储罐超声检测机器人避障研究[J]. 计算机仿真,2023,40(11):408-414. doi: 10.3969/j.issn.1006-9348.2023.11.078 WU Yitong,LI Bin,LIU Chun,et al. Research on obstacle avoidance of storage tank ultrasonic inspection robot integrating GMM and PSO[J]. Computer Simulation,2023,40(11):408-414. doi: 10.3969/j.issn.1006-9348.2023.11.078