Block Recursive MPCA and its application in batch process monitoring
暂无分享,去创建一个
A new approach, Block Recursive MPCA, to monitor batch process based on the process variables trajectories is proposed. It is cumbersome to overcome the traditional MPCA's need of estimating or filling in the unknown part of the process variable trajectory deviations from the current time to the end. To tackle the problem, Block Recursive MPCA method involves a sequential of PCA models and forgetting factors among them to analyze the 3-dimension history data. In addition, a method for calculating the distance between PCA models is proposed to evaluate the forgetting factor. Application in industrial fermentation batch process reveals that Block Recursive MPCA describes the process more accurately and objectively than traditional MPCA.
[1] Manabu Kano,et al. Statistical process monitoring based on dissimilarity of process data , 2002 .
[2] J. Macgregor,et al. Monitoring batch processes using multiway principal component analysis , 1994 .
[3] Manabu Kano,et al. A new multivariate statistical process monitoring method using principal component analysis , 2001 .
[4] Weihua Li,et al. Recursive PCA for adaptive process monitoring , 1999 .
[5] Weihua Li,et al. Recursive PCA for Adaptive Process Monitoring , 1999 .