中国安全科学学报 ›› 2023, Vol. 33 ›› Issue (7): 127-132.doi: 10.16265/j.cnki.issn1003-3033.2023.07.2235

• 安全工程技术 • 上一篇    下一篇

基于PSO-BPNN的煤自燃危险性预测模型

汪伟1,2(), 梁然1,**(), 祁云1,2, 贾宝山2,3, 武泽伟1   

  1. 1 山西大同大学 煤炭工程学院,山西 大同 037000
    2 辽宁工程技术大学 安全科学与工程学院,辽宁 阜新 123000
    3 矿山热动力灾害与防治教育部重点实验室(辽宁工程技术大学),辽宁 葫芦岛 125000
  • 收稿日期:2023-02-17 修回日期:2023-05-08 出版日期:2023-07-28
  • 通讯作者:
    ** 梁然(1997—),男,河北石家庄人,硕士研究生,主要研究方向为矿井火灾防治。E-mail:
  • 作者简介:

    汪伟 (1991—),男,河北玉田人,博士,讲师,硕士生导师,主要从事矿井灾害防治、安全评价、应急技术与管理等方面的研究。E-mail:

    祁云,副教授

    贾宝山,教授

  • 基金资助:
    山西省基础研究计划(自由探索类)青年项目(202203021222300); 山西省高等学校科技创新计划项目(2022L449); 山西省高等学校科技创新计划项目(2022L448); 国家重点研发计划资助项目(2018YFC0807900); 山西大同大学博士科研启动项目(2020-B-08); 山西大同大学博士科研启动项目(2020-B-18)

Prediction model of coal spontaneous combustion risk based on PSO-BPNN

WANG Wei1,2(), LIANG Ran1,**(), QI Yun1,2, JIA Baoshan2,3, WU Zewei1   

  1. 1 School of Coal Engineering, Shanxi Datong University, Datong Shanxi 037000, China
    2 College of Safety Science and Engineering, Liaoning Technical University, Fuxin Liaoning 123000, China
    3 Key Laboratory of Mine Thermodynamic Disasters and Control of Ministry of Education (Liaoning Technical University), Huludao Liaoning 125000,China
  • Received:2023-02-17 Revised:2023-05-08 Published:2023-07-28

摘要:

为提高采空区遗煤自燃危险性预测的准确性,采用粒子群优化算法(PSO)改进反向传播神经网络(BPNN)的连接权重和阈值,构建一种将PSO算法与BPNN相耦合的煤自燃预测模型(PSO-BPNN模型),并从不同方面对比分析PSO-BPNN模型与BPNN模型和支持向量回归机(SVR)模型的预测结果。研究表明:优化后的平均相对误差、平均绝对误差和均方根误差相较于BPNN模型分别降低了9.35%、0.170 7和0.205 6,判定系数增大了0.116 9;比SVR模型分别降低了5.41%、0.115 2和0.171 5,判定系数增大了0.089 1。证明PSO-BPNN模型具有更高的预测精准性。

关键词: 煤自燃, 粒子群优化算法(PSO), 反向传播神经网络(BPNN), 支持向量回归机(SVR), 预测模型, 采空区

Abstract:

In order to improve the accuracy of the risk prediction of coal spontaneous combustion in the goaf, a PSO algorithm was used to improve the connection weight and threshold values of BPNN. A coal spontaneous combustion risk prediction model (PSO-BPNN model) coupling the PSO algorithm with BPNN was constructed. The PSO-BPNN model was compared and analyzed with the prediction results of the standard BPNN model and SVR model respectively. The results show that the optimized mean relative error, mean absolute error and root mean square error are reduced by 9.35%, 0.170 7 and 0.205 6, respectively, and the coefficient of determination is increased by 0.116 9 compared with the BPNN model. Compared with the SVR model, the optimized mean relative error, mean absolute error and root mean square error are reduced by 5.41%, 0.115 2 and 0.171 5, respectively, and the coefficient of determination is increased by 0.0891. It was demonstrated that the PSO-BPNN model has higher prediction accuracy than the standard BPNN model and SVR model.

Key words: coal spontaneous combustion, particle swarm optimization(PSO), back propagation neural network(BPNN), support vector regression(SVR), prediction model, goaf