论文部分内容阅读
通过分析采煤工作面煤与瓦斯涌出量与地质构造指标的对应关系,应用支持向量机(SVM)方法对煤与瓦斯涌出类型及涌出量进行分析。建立两类突出识别的SVM模型、多类型突出识别的H-SVMs模型以及预测瓦斯涌出量的支持向量回归模型。研究结果表明:SVM方法能够很好地对煤与瓦斯突出模式进行识别,所建立的采煤工作面瓦斯涌出量预测模型的精度高于应用BP神经网络预测精度;SVM理论基础严谨,决策函数结构简单,泛化能力强,并且决策函数中的法向量W可以反映突出模式识别的地质结构指标的权重。
By analyzing the correspondence between coal and gas emission and geological structure index in coal mining face, the types of coal and gas emission and the amount of emission are analyzed by support vector machine (SVM). Two kinds of SVM model with prominent recognition, H-SVMs model with multi-type highlight recognition and support vector regression model with gas emission forecasting are established. The results show that the SVM method can well identify the coal and gas outburst modes, and the accuracy of the prediction model of gas emission from the coal mining face is higher than that of the BP neural network. The SVM theory is rigorous and the decision-making function Simple structure and strong generalization ability, and the normal vector W in the decision function can reflect the weight of the geological structure index that highlights the pattern recognition.