主成分回归残差神经网络校正算法用于近红外光谱快速测定汽油辛烷值

根据汽油辛值预测体系本身的非线性特点，提出主成分回归残差神经网络校正算法（principal component regression residual artificial neural network,PCRRANN)用于近红外测定汽油辛烷值的预测模型校正，该方法给合了主成分回归算法（PC），与经典的线性校正算法（PLS（Partial Least Square),PCR, 以及非线性PLS（NPLS，Non-linear PLS)等相比，预测明显的改善，文中还讨论了PCR主成分数及训练参数对预则模可能的影响。

Principal Component Regression Residual Artificial Neur al Network Calibration Algorithm Applied in Near Infrared Fast Measurement of Gasoline Octane Number

A novel calibration algorithm, PCRRANN (principal component regression residual artificial neural network) method, was proposed based on the intrinsic non linearity of the prediction of gasoline octane number, and then applied to the calibration of the prediction model of the near infra red measurement of gasoline octane number. The method combined the linear calibration ability of the pricipal component regression (PCR) method and the excellent non linear approximating ability of artificial neuralnetwork using the residual of PCR calibration as target signal and the PCR scores as input signal of the neuralnetwork respectively. Compared with the classical linear algorithms such as the PLS (partial least squares), PCR and NPLS (Non linear PLS), the proposed method showed obvious improvement in prediction ability. The effects of the number of principal components of PCR part and some training parameters on the prediction model were also discussed.