...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Fuel >Predictive ability of four statistical models for determining the influence of coal thermophysical properties during the initial phase of coal spontaneous combustion
【24h】

Predictive ability of four statistical models for determining the influence of coal thermophysical properties during the initial phase of coal spontaneous combustion

机译：四种统计模型的预测能力确定煤自燃初期煤热物理特性的影响

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Studying the thermophysical properties of coal is crucial for mitigating coal spontaneous combustion (CSC). We tested four artificial intelligence statistical models, namely multiple linear regression (MLR), bagging, random forest (RF), and back propagation neural networks (BPNNs), to determine the most accurate prediction model of the thermophysical properties of coal simples. Six impact factors, containing moisture, ash, volatility, fixed carbon, oxygen concentration, and temperature, were obtained using grey correlation analysis. A total of 250 sets of 13 types of coal related data were randomly divided to create 225 training sets and 25 verification sets. The results revealed that MLR is not suitable for the prediction of the thermophysical properties of coal. Although favourable training results could be achieved using BPNNs, its prediction results were inferior. Drilling and forecasting predictions of the bagging and RF models were excellent and their goodness of fit (R2) exceeded 0.91. Moreover, the error (RMSE, MAE, and MAPE) of RF model was lower than the bagging, which exhibited optimal performance. Therefore, the RF model is the most suitable for predicting coal thermophysical properties.
机译：研究煤的热物理性质对于减轻煤炭自燃（CSC）至关重要。我们测试了四种人工智能统计模型，即多元线性回归（MLR），袋装，随机森林（RF）和后传播神经网络（BPNNS），以确定煤模型的热物理性质最准确的预测模型。使用灰色相关性分析获得含有水分，灰，挥发性，固定碳，氧气浓度和温度的六个冲击因子。共有250套13种煤炭相关数据随机分为225种培训集和25种验证集。结果表明，MLR不适合于预测煤的热物理性质。虽然可以使用BPNNS实现有利的培训结果，但其预测结果较差。堆垛和RF模型的钻井和预测预测是优异的，它们的拟合良好（R2）超过0.91。此外，RF模型的误差（RMSE，MAE和MAPE）低于袋装，其表现出最佳性能。因此，RF模型最适合于预测煤热物理性质。

著录项

  • 来源
    《Fuel》 |2021年第15期|120348.1-120348.10|共10页
  • 作者

    Zhong Kai-Qi; Xiao Yang; Zhao Xia; Yin Lan; Shu Chi-Min; Tian Yuan;

  • 作者单位

    Xian Univ Sci & Technol Sch Safety Sci & Engn 58 Yanta Mid Rd Xian 710054 Shaanxi Peoples R China|Xian Univ Sci & Technol Shaanxi Key Lab Prevent & Control Coal Fire Xian 710054 Shaanxi Peoples R China;

    Xian Univ Sci & Technol Sch Safety Sci & Engn 58 Yanta Mid Rd Xian 710054 Shaanxi Peoples R China|Xian Univ Sci & Technol Shaanxi Key Lab Prevent & Control Coal Fire Xian 710054 Shaanxi Peoples R China;

    Xian Univ Sci & Technol Sch Geol & Environm 58 Yanta Mid Rd Xian 710054 Shaanxi Peoples R China;

    Xian Univ Sci & Technol Sch Safety Sci & Engn 58 Yanta Mid Rd Xian 710054 Shaanxi Peoples R China|Xian Univ Sci & Technol Shaanxi Key Lab Prevent & Control Coal Fire Xian 710054 Shaanxi Peoples R China;

    Natl Yunlin Univ Sci & Technol Dept Safety Hlth & Environm Engn Touliu 64002 Yunlin Taiwan;

    De Montfort Univ Inst Energy & Sustainable Dev Sch Engn & Sustainable Dev Leicester LE1 9BH Leics England;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    Artificial intelligence statistical model; Impact factor; Grey correlation analysis; Training and verification; Optimal performance;

    机译：人工智能统计模型;影响因子;灰色相关分析;培训和验证;最佳性能;

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号