• 首页 | 期刊简介 | 编委会 | 投稿须知 | 出版道德规范 | 下载专区 | English
秦楠*,葛强,梁忠豪,孙嘉彬,王永岩.高温对砂岩宏细观损伤及BP神经网络单轴强度预测研究[J].实验力学,2021,36(1):105~113
高温对砂岩宏细观损伤及BP神经网络单轴强度预测研究
Experimental study on macro/micro damage of sandstone caused by high temperature and prediction of uniaxial strength by BP neural network
投稿时间:2020-03-11  修订日期:2020-06-19
DOI:10.7520/1001-4888-20-049
中文关键词:  砂岩  温度  单轴抗压强度  BP神经网络
英文关键词:sandstone  temperature  uniaxial compressive strength  BP neural network
基金项目:国家自然科学基金(51674149);山东省自然科学基金(ZR2018PEE005);煤矿安全高效开采省部共建教育部重点实验室开放研究基金(JYBSYS2018204);煤炭资源与安全开采国家重点实验室开放研究基金(SKLCRSM19KF015);矿山灾害预防控制教育部重点实验室开放研究基金(MDPC201915);青岛科技大学博士科研启动金(210-010022667)
作者单位
秦楠* 青岛科技大学 机电工程学院 山东青岛 266061 
葛强 青岛科技大学 机电工程学院 山东青岛 266061 
梁忠豪 青岛科技大学 机电工程学院 山东青岛 266061 
孙嘉彬 青岛科技大学 机电工程学院 山东青岛 266061 
王永岩 青岛科技大学 机电工程学院 山东青岛 266061 
摘要点击次数: 496
全文下载次数: 1132
中文摘要:
      为了研究高温后砂岩的力学特性和宏细观损伤变化,对高温作用后的砂岩进行单轴压缩试验、声波损伤检测、X射线衍射试验、扫描电镜试验,分析应力-应变曲线、峰值应力、峰值应变、弹性模量、质量损失率、X射线衍射成像和电镜扫描图像,得到砂岩的细观损伤变化对其单轴抗压强度的影响。利用BP神经网络模型对不同物理量进行训练,预测不同高温作用后砂岩单轴抗压强度。研究结果表明:随着温度升高,砂岩峰值应力和弹性模量均降低,峰值应变、质量损失率和体积均增大,砂岩的外观颜色由黄色过渡到棕红色直至呈土灰色;微缺陷(微裂隙和孔洞)的发育明显,晶体结构破坏加剧,内部生成CaO和CO2,孔隙率、热损伤程度增大,声速减小,强度降低。建立BP神经网络模型,利用文献数据验证模型可行性,模型预测值与试验值最大误差8.25%,可靠度较高。
英文摘要:
      In order to study the mechanical properties and macro/micro damage of sandstone after high temperature treatment, the uniaxial compression test, acoustic damage detection, X-ray diffraction test and scanning electron microscope tests are conducted. The influence of micro damage variation on the uniaixal compression strength of sandstone is obtained by analyzing the stress-strain curve, peak stress, peak strain, elastic modulus, mass loss rate, X-ray diffraction images and scanning electron microscope images. Based on the BP neural network model, different physical quantities can be trained, and the uniaxial compressive strength of sandstone under different high temperatures can be predicted. The results show that as the temperature increases, the uniaxial compressive strength and elastic modulus of sandstone decrease, while the peak strain, mass loss rate and volume increase, and the appearance color of sandstone gradually turns red with grey intensified. Furthermore, the development of micro defects (micro cracks and holes) is obviously obtained, the crystal structure is damaged dramatically, the internal CaO and CO2 are generated, the porosity and the degree of thermal damage increase, and the sound velocity and intensity decrease. The BP neural network model is established, and the feasibility of the model is verified based on the literature data. The maximum error between the predicted value and the tested value is about 8.25%, indicating the high reliability of the model.
查看全文  下载PDF阅读器
关闭

版权所有:《实验力学》编辑部
您是本站第 65194387 位访问者,今日一共访问1218次,当前在线人数: 0
技术支持:本系统由北京勤云科技发展有限公司设计