论文部分内容阅读
高斯过程是新近发展的一种机器学习方法,对处理复杂非线性问题具有很好的适应性。采煤工作面瓦斯涌出量与其影响因素之间存在着复杂的非线性关系,针对传统预测方法的局限性,提出了瓦斯涌出量预测的高斯过程机器学习模型。通过对少量学习样本的学习,采用该模型可建立瓦斯涌出量与其影响因素之间的复杂非线性映射关系。将模型应用于工程实例,研究结果表明,瓦斯涌出量预测的高斯过程机器学习方法是科学可行的,具有预测精度高、适用性强、参数自适应化且易于实现的优点。
Gaussian process is a newly developed machine learning method that has good adaptability to complex nonlinear problems. There is a complicated nonlinear relationship between gas emission and its influencing factors in coal mining face. In view of the limitations of traditional prediction methods, a Gaussian process machine learning model predicting gas emission is proposed. Through the study of a small number of learning samples, the model can be used to establish the complex nonlinear mapping relationship between the amount of gas emission and its influencing factors. The model is applied to the engineering example. The results show that the Gaussian process machine learning method of gas emission prediction is scientific and feasible, with the advantages of high prediction accuracy, applicability, parameter adaptation and easy to implement.