基于HMM的煤矿底板突水短时预测模型  

Short-term Prediction Model of Water Inrush from Coal Mine Floor Based on HMM

作　　者：杨坤 陈伯辉[1,2] 施式亮 YANG Kun;CHEN Bohui;SHI Shiliang(College of Environment and Resources,Fuzhou University,Fuzhou350108,China;School of Resource&Environment and Safety Engineering,Hunan University of Science and Technology,Xiangtan411201,China)

机构地区：[1]福州大学环境与资源学院,福建福州350108 [2]湖南科技大学资源环境与安全工程学院,湖南湘潭411201

出　　处：《安全与环境工程》2020年第5期190-196,共7页Safety and Environmental Engineering

基　　金：国家自然科学基金项目(51774135)。

摘　　要：为解决煤矿底板突水预测在实际应用中工作量大、数据不足等问题,结合突水系数法及其衍生方法,对其进行以突水系数阈值回归分析处理为主的方法改进,并基于隐马尔科夫模型分别构建以单位涌水量为观测对象的突水系数阈值预测模型和实际突水系数预测模型。结合两种模型,通过应用基于条件概率的相关算法,实现了基于单位涌水量的煤矿底板突水短时预测,该模型可用于煤矿底板突水预警工作中监测数据有限条件下的模糊预警。In order to solve the problems of large workload and insufficient data in the practical application of prediction about water inrush from coal mine floor,this paper adopts the method of water inrush coefficient and its derivative method,this paper improves the analysis method based on the treatment of regression analysis of water inrush coefficient threshold.Based on the hidden Markov model,the paper constructs the prediction model of water inrush coefficient threshold and prediction model of actual water inrush coefficient respectively with unit water inflow as observation object.Combined with the two models,the shortterm prediction of water inrush from coal mine floor based on unit water inflow is realized by using the relevant algorithm based on conditional probability.The model can be used for the fuzzy early-warning under the condition of limited monitoring data in the coal mine floor water inrush early-warning work.

关 键 词：煤矿底板突水 突水系数阈值 单位涌水量 隐马尔科夫模型 短时预测 

分 类 号：X936[环境科学与工程—安全科学] T745.21[一般工业技术]