摘要
低阶煤低温干馏生成的煤焦油是重要有机化工原料,焦油加工精炼已作为近代煤化工的一个重要分支.为了建立煤焦油产率预测技术方法,服务于煤焦油地质储量评价,本文在文献调研基础上,遵循“地质约束测井、岩心刻度测井”原则,开展研究区煤焦油产率敏感因素分析,厘定与煤焦油产率相关地质参数,利用统计分析方法确定与煤焦油产率敏感的测井信息;利用多元回归分析方法和神经网络技术首次开展了煤焦油产率测井预测方法探索.研究表明,低阶煤的焦油产率普遍高于中高阶煤,对煤焦油产率敏感的地质因素主要有煤化程度(镜质体反射率)、温度和压力、煤质灰分与固定碳含量;对煤焦油产率敏感的测井信息主要包括补偿密度、自然伽马、电阻率等;分区开展多元回归建模可实现利用密度、自然伽马和电阻率测井响应对煤焦油产率预测;将神经网络技术引入煤焦油产率分区预测可行.研究成果对做好煤焦油地质储量评价具有重要的技术支撑作用.
The coal tar generated from low temperature carbonization of low rank coal is an important organic chemical raw material of organic chemical industry,and tar processing and refining has become an important branch of modern coal chemical industry.In order to establish a technical method for predicting coal tar yield and serve the evaluation of coal tar geological reserves,this paper,on the basis of literature research,follows the principle of"geologically constrained logging,core calibration logging",carries out the analysis of coal tar yield sensitive factors in the study area,determines the geological parameters related to coal tar yield,and uses statistical analysis methods to determine the logging information sensitive to coal tar yield.The multi regression analysis method and neural network technology were used to explore the logging prediction method of coal tar yield for the first time.The research shows that the tar yield of low rank coal is generally higher than that of medium and high rank coal,and the geological factors sensitive to the tar yield are mainly the degree of coalification(vitrinite reflectance),temperature and pressure,coal ash content and fixed carbon content;Logging information sensitive to coal tar yield mainly includes compensation density,natural gamma,resistivity,etc.Multivariate regression modeling in different areas can realize the prediction of coal tar yield using density,natural gamma and resistivity log responses;It is feasible to apply neural network technology to the partition prediction of coal tar yield.The research results play an important role in technical support for the evaluation of coal tar geological reserves.
作者
赵军龙
闫和平
王金锋
王诗聪
ZHAO JunLong;YAN HePing;WANG JinFeng;WANG SiCong(School of Earth Sciences and Engineering,Xi'an Shiyou University,Xi'an 710065,China;Shaanxi Key Laboratory of Petroleum Accumulation Geology,Xi'an Shiyou University,Xi'an 710065,China;Shaanxi Province 194 Coal Geological Co.,Ltd.,Tongchuan 727007,China)
出处
《地球物理学进展》
CSCD
北大核心
2023年第4期1702-1712,共11页
Progress in Geophysics
基金
陕西省煤田地质集团有限公司2019年科研项目“富油煤测井识别技术研究”(SMDZ-2019CX-9)资助。
关键词
低阶煤
煤焦油产率
多元回归分析
BP神经网络
测井预测
Low-rank coal
Coal tar production rate
Multiple regression analysis
BP Neural network
Logging prediction
