该方法由三部分组成：主元分析pca、时间延迟神经网络、软测量模型的在线校正。
It is composed of three elements: PCA, time-delay neural network and model updating, where the offline model is trained through the algorithm GABP.
利用主元分析法，可以在保留原有数据信息特征的基础上，消除变量关联和部分系统干扰，简化分析的复杂程度。
It can eliminate the correlation between variables and disturbance of system and can simplify complicated degree by PCA when reserving enough features of original data information.
主元回归和部分最小二乘方法能克服批次内不同阶段的控制量存在的相关关系从而得到更准确的模型。
The regression analyses to correlate the control actions with the results for various stages of a batch to obtain more accurate models.
主元回归和部分最小二乘方法能克服批次内不同阶段的控制量存在的相关关系从而得到更准确的模型。
The regression analyses to correlate the control actions with the results for various stages of a batch to obtain more accurate models.
应用推荐