李建林, 王路扬, 李松营,等.基于灰色-多步马尔科夫模型的矿井涌水量预测[J].河南理工大学学报（自然科学版）,2023,42(5):66-71.

LI J L, WANG L Y, LI S Y,et al.Forecast of mine water inflow based on grey-multi-step Markov model[J].Journal of Henan Polytechnic University(Natural Science) ,2023,42(5):66-71.

基于灰色-多步马尔科夫模型的矿井涌水量预测

李建林1,2, 王路扬1, 李松营3, 王冲4, 张孟佼1

1.河南理工大学 资源环境学院，河南 焦作  454000;2.河南理工大学 煤炭安全生产与清洁高效利用省部共建协同创新中心，河南 焦作  454000;3.河南能源化工集团研究总院有限公司，河南 郑州 450018;4.天津华北地质勘查局 核工业二四七大队，天津 301800

摘要:为提高矿井涌水量预测精度，以洛阳龙门矿为例，基于2011年1月—2020年12月正常涌水量序列，提出灰色-多步马尔科夫模型，并用2021年1—4月实测正常涌水量验证模型。结果表明：灰色-多步马尔科夫模型预测精度达99.35%，明显高于单一灰色GM（1，1）模型和灰色-马尔科夫模型；灰色-多步马尔科夫模型对波动较大的非平稳序列拟合效果良好，可在一定程度上减少预测主观性，预测精度较高，是进行矿井涌水量预测的一种有效方法。

关键词:矿井涌水量;灰色-多步马尔科夫模型;非平稳序列;残差修正

doi:10.16186/j.cnki.1673-9787.2021060039

基金项目:国家自然科学基金资助项目（41972254）；河南省高校重点科研项目（22A170009）

收稿日期:2021/06/09

修回日期:2021/09/18

出版日期:2023/09/25

Forecast of mine water inflow based on grey-multi-step Markov model

LI Jianlin1,2, WANG Luyang1, LI Songying3, WANG Chong4, ZHANG Mengjiao1

1.Institute of Resources & Environment，Henan Polytechnic University，Jiaozuo  454000，Henan，China;2.Collaborative Innovation Center of Coal Work Safety and Clean High Efficiency Utilization，Henan Polytechnic University，Jiaozuo  454000，Henan，China;3.Henan Energy and Chemical Industry Group Research Institute Co.，Ltd.，Zhengzhou  450018，Henan，China;4.The Nuclear Industry 247 Brigade of Tianjin North China Geological Exploration Bureau，Tianjin  301800，China

Abstract:To improve the prediction accuracy of mine water inflow，taking Longmen Mine in Luoyang as an example，a grey multi-step Markov model was proposed based on the normal water inflow sequence from January 2011 to December 2020.And the model was validated by using the measured water inflow from January to April 2021.The results showed that the prediction accuracy of the gray multi-step Markov model reached 99.35%，which was significantly higher than the prediction accuracy of a single gray model and gray-Markov model.The greymulti-step Markov model had a good fitting effect on non-stationary series with large fluctuations，reduced certain subjectivity and improved predict accuracy，which could provide an effective method for mine water inflow prediction.

Key words:mine water inflow;grey-multi-step Markov model;non-stationary series;residual correction

基于灰色-多步马尔科夫模型的矿井涌水量预测_李建林.pdf