摘要： 将粗糙集与神经网络结合，提出由样本更新、粗糙集预处理、神经网络训练、规则提取4个模块组成的煤矿瓦斯预测知识获取模型，将其应用于实时数据进行实验，结果表明，该模型实时性好、可靠性及精度高，可以较好地解决煤矿瓦斯预测知识获取困难的问题，为煤矿瓦斯预测专家系统知识库的建立奠定基础。

关键词: 知识获取, 煤矿瓦斯预测, 粗糙集, 神经网络

Abstract:

By combining rough sets and neural networks, this paper develops a knowledge acquisition model for colliery gas forecast, which consists of four modules: sample refreshment, rough sets preprocessing, neural network training and rules extraction. It is applied in the real-time data, whose results show that it solves the problem of the knowledge acquisition difficulty of colliery gas forecast, and has good real-time characteristic, high reliability and perfect precision. The model provides foundation for establishing knowledge database of colliery gas forecast.

Key words: knowledge acquisition, colliery gas forecast, rough sets, neural networks

中图分类号: 

• TP39