论文部分内容阅读
为了克服自变量之间的多重共线问题,提高多元回归模型预测的精确性,将主成分分析(PCA)与多元回归分析(MRA)相结合,提出了主成分多元回归分析(PCMRA)模型。利用RBF神经网络对主成分回归分析残差进行拟合预测,最后利用残差预测值对主成分回归分析预测值进行补偿。结果表明:利用RBF神经网络对主成分回归模型进行补偿,将线性拟合算法和非线性拟合算法结合起来用于瓦斯涌出量预测是一种较为优越的算法。
In order to overcome the problem of multicollinearity between independent variables and improve the accuracy of multivariate regression model prediction, principal component analysis (PCA) and multiple regression analysis (MRA) are combined to propose PCMRA model. The RBF neural network is used to predict the residuals of the principal component regression analysis. Finally, the predicted residuals are used to compensate the predicted values of the principal component regression analysis. The results show that using RBF neural network to compensate the principal component regression model, combining the linear fitting algorithm and non-linear fitting algorithm is a superior algorithm for gas emission prediction.