论文部分内容阅读
煤与瓦斯危险性的准确预测一直是矿山安全领域的关键技术难题和重大研究课题。支持向量机是在瓦斯预警中广泛使用的一种技术,以统计学习理论和支持向量机为基础,通过研究基于模糊支持向量机的多类分类方法,对原算法进行改进,采用模糊多类支持向量机,并构造模糊隶属函数,同时使用序列最小最优化算法进行求解,以期提高算法的精度和速度。
Accurate prediction of coal and gas hazards has always been a key technical challenge and a major research topic in the field of mine safety. Support vector machine (SVM) is a widely used technique in gas warning. Based on statistical learning theory and SVM, this paper improves the original algorithm by studying the multi-class classification method based on fuzzy support vector machines, and adopts fuzzy multi-class support Vector machine, and constructs fuzzy membership function. At the same time, it uses sequence minimum optimization algorithm to solve, in order to improve the accuracy and speed of the algorithm.