基于Bayes算法的煤矿井下突水水源判识系统的设计与实现

System design and implementation of water source identification of mine water inrush based on the Bayes Algorithm

矿井突水是严重影响煤矿安全生产的矿井灾害,快速准确地判识矿井突水水源是矿井突水灾害防治的重要基础。基于Bayes算法设计开发的煤矿井下突水水源判识系统,具有通用性强、操作简单、判识效率高的优点。系统可借助标准水样数据进行自主训练学习,根据标准水样中因子的判别能力自动对因子进行选入或剔除。通过在安徽省祁东煤矿的实际应用,系统判识正确率达91%以上,判识结果与实际水文情况符合良好,证明了基于Bayes算法设计实现的判识系统可为煤矿井下突水水源的快速准确判识提供有力的工具。

Mine water inrush is one of the most serious geological disasters impacting the safe production in a coal mine. Quick and accurate identification of water source of underground water inrush is the important foundation of preventing and controlling mine water inrush. A water source identification system is designed and developed based on the Bayes algorithm. The system has advantages including strong universality,and is simple to operate with high identification efficiency. It can train and learn autonomously with the help of standard water samples. At the same time it selects or rejects factors by their capacity of identification. The system is used in the Qidong coal mine of Anhui and its accuracy of identification exceeds 91%. The results show that the water source identification system based on the Bayes algorithm provides a powerful instrument to quickly and accurately identify the water source of underground water inrush.