工程科学学报，第40卷，第11期：1309-1316,2018年11月 Chinese Joural of Engineering,Vol.40,No.11:1309-1316,November 2018 DOI:10.13374/j.issn2095-9389.2018.11.004;http://journals.ustb.edu.cn 煤与瓦斯突出多指标耦合预测模型研究及应用 张友谊12)，崔金雷)四，焦向东) 1)中国矿业大学（北京）资源与安全工程学院，北京100083.2)中国平煤神马能源化工集团有限责任公司，平顶山467000 ☒通信作者，E-mail:2638003562@q9.com 摘要针对煤与瓦斯突出预测准确率问题，在分析煤与瓦斯突出瓦斯地质因素的基础上，构建了包含瓦斯指标、煤体指标、 地应力指标3个一级指标和瓦斯压力、构造煤厚度等12个二级指标的预测指标体系，通过综合运用网络分析法和多类别距离 判别法对灰色关联模型中的输入端和输出端进行研究，建立了煤与瓦斯突出多指标耦合预测模型.该模型基于对煤与瓦斯突 出瓦斯地质因素的综合分析，计算预测指标权重，划分煤与瓦斯突出可能性等级，建立了对突出可能性进行判别的2个判别 式.以平煤八矿为例应用该模型对8组预测样本进行了煤与瓦斯突出可能性判断，预测结果与实际符合，为矿井煤与瓦斯突 出防治提供了技术支撑，证明了煤与瓦斯突出多指标耦合预测模型的准确性和适用性 关键词煤与瓦斯突出：瓦斯指标：地质指标：耦合模型：预测 分类号TG142.71 Study of the multi-index coupling forecasting model of coal and gas outburst and its application ZHANG You-yi),CUl Jin-lei,JIAO Xiang-dong) 1)College of Resources and Safety Engineering,China University of Mining and Technology (Beijing),Beijing 100083.China 2)China Pingmei Shenma Energy and Chemical Group Co.,Ltd.,Pingdingshan 467000,China Corresponding author,E-mail:2638003562@qq.com ABSTRACT Coal and gas outburst seriously threatens the safety of underground coal mining.With the increase in coal mining depth,the geological conditions of coal mines become complicated,which adversely affects the safety of coal mining production.Promi- nent prediction is an important link to reduce the damage caused by outburst accidents.This study investigated the prediction model of coal and gas outburst.On the basis of the analysis of the gas geological factors of coal and gas outburst,a prediction system including 3 major indices (i.e.,gas,coal,and crustal stress indices)and 12 minor indices (e.g.,gas pressure and tectonic coal thickness) was established.A multi-index coupling prediction model based on the improved input and output of the gray correlation model was pro- posed.Network analysis and multi-class distance discriminant method were also comprehensive applied to improve the accuracy of coal and gas outburst index prediction.On the basis of the comprehensive analysis of the gas geological factors of coal and gas outburst,the model calculates the weight of the prediction index,classifies the possible grade of the outburst,and derives the discriminant formulas to discriminate the possibility of outbursts.Taking the Pingdingshan No.8 coal mine as an example,the model estimated the probabili- ty of coal and gas outburst in eight sets of prediction samples.The predicted results are consistent with the actual conditions,provide technical support for coal mine and gas outburst prevention,and prove the accuracy and applicability of the multi-index coupling pre- diction model. KEY WORDS coal and gas outburst;gas index;geological index;coupling model;prediction 收稿日期：2018-05-28 基金项目：国家自然科学基金重点资助项目(41430640)：国家自然科学基金资助项目(41172144)工程科学学报,第 40 卷,第 11 期:1309鄄鄄1316,2018 年 11 月 Chinese Journal of Engineering, Vol. 40, No. 11: 1309鄄鄄1316, November 2018 DOI: 10. 13374 / j. issn2095鄄鄄9389. 2018. 11. 004; http: / / journals. ustb. edu. cn 煤与瓦斯突出多指标耦合预测模型研究及应用 张友谊1,2) , 崔金雷1)苣 , 焦向东2) 1) 中国矿业大学(北京)资源与安全工程学院, 北京 100083 2) 中国平煤神马能源化工集团有限责任公司, 平顶山 467000 苣 通信作者, E鄄mail: 2638003562@ qq. com 摘 要 针对煤与瓦斯突出预测准确率问题,在分析煤与瓦斯突出瓦斯地质因素的基础上,构建了包含瓦斯指标、煤体指标、 地应力指标 3 个一级指标和瓦斯压力、构造煤厚度等 12 个二级指标的预测指标体系,通过综合运用网络分析法和多类别距离 判别法对灰色关联模型中的输入端和输出端进行研究,建立了煤与瓦斯突出多指标耦合预测模型. 该模型基于对煤与瓦斯突 出瓦斯地质因素的综合分析,计算预测指标权重,划分煤与瓦斯突出可能性等级,建立了对突出可能性进行判别的 2 个判别 式. 以平煤八矿为例应用该模型对 8 组预测样本进行了煤与瓦斯突出可能性判断,预测结果与实际符合,为矿井煤与瓦斯突 出防治提供了技术支撑,证明了煤与瓦斯突出多指标耦合预测模型的准确性和适用性. 关键词 煤与瓦斯突出; 瓦斯指标; 地质指标; 耦合模型; 预测 分类号 TG142郾 71 收稿日期: 2018鄄鄄05鄄鄄28 基金项目: 国家自然科学基金重点资助项目(41430640); 国家自然科学基金资助项目(41172144) Study of the multi鄄index coupling forecasting model of coal and gas outburst and its application ZHANG You鄄yi 1,2) , CUI Jin鄄lei 1)苣 , JIAO Xiang鄄dong 2) 1) College of Resources and Safety Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China 2) China Pingmei Shenma Energy and Chemical Group Co. , Ltd. , Pingdingshan 467000, China 苣 Corresponding author, E鄄mail: 2638003562@ qq. com ABSTRACT Coal and gas outburst seriously threatens the safety of underground coal mining. With the increase in coal mining depth, the geological conditions of coal mines become complicated, which adversely affects the safety of coal mining production. Promi鄄 nent prediction is an important link to reduce the damage caused by outburst accidents. This study investigated the prediction model of coal and gas outburst. On the basis of the analysis of the gas geological factors of coal and gas outburst, a prediction system including 3 major indices (i. e. , gas, coal, and crustal stress indices) and 12 minor indices (e. g. , gas pressure and tectonic coal thickness) was established. A multi鄄index coupling prediction model based on the improved input and output of the gray correlation model was pro鄄 posed. Network analysis and multi鄄class distance discriminant method were also comprehensive applied to improve the accuracy of coal and gas outburst index prediction. On the basis of the comprehensive analysis of the gas geological factors of coal and gas outburst, the model calculates the weight of the prediction index, classifies the possible grade of the outburst, and derives the discriminant formulas to discriminate the possibility of outbursts. Taking the Pingdingshan No. 8 coal mine as an example, the model estimated the probabili鄄 ty of coal and gas outburst in eight sets of prediction samples. The predicted results are consistent with the actual conditions, provide technical support for coal mine and gas outburst prevention, and prove the accuracy and applicability of the multi鄄index coupling pre鄄 diction model. KEY WORDS coal and gas outburst; gas index; geological index; coupling model; prediction