An iterative (run-to-run) optimization method was presented for batch processes under input constraints. Generally it is very difficult to acquire an accurate mechanistic model for a batch process.Because support vect...An iterative (run-to-run) optimization method was presented for batch processes under input constraints. Generally it is very difficult to acquire an accurate mechanistic model for a batch process.Because support vector machine is powerful for the problems characterized by small samples,nonlinearity, high dimension and local minima, support vector regression models were developed for the end-point optimization of batch processes. Since there is no analytical way to find the optimal trajectory, an iterative method was used to exploit the repetitive nature of batch processes to determine the optimal operating policy. The optimization algorithm is proved convergent. The numerical simulation shows that the method can improve the process performance through iterations.展开更多
文摘在语音与唇读识别应用中,传统的LDA(linear discriminant analysis)算法一般以音节、半音节、HMM状态等基元为类别进行数据分段,经线性判别分析后获得的特征投影方向与识别率不直接相关,影响了识别率。提出了一种新的基于LDAO(linear discriminant analysis based on object)的唇读特征提取算法,该算法以待识别对象为类别进行线性判别分析,在理论上保证了唇读特征矢量向最具判别能力的方向投影。基于唇读数据库的实验证明,该算法明显优于现有各种唇读特征提取算法,比DCT+LDA算法识别率提高了3%。
基金National Natural Science Foundation of China(No.60504033)
文摘An iterative (run-to-run) optimization method was presented for batch processes under input constraints. Generally it is very difficult to acquire an accurate mechanistic model for a batch process.Because support vector machine is powerful for the problems characterized by small samples,nonlinearity, high dimension and local minima, support vector regression models were developed for the end-point optimization of batch processes. Since there is no analytical way to find the optimal trajectory, an iterative method was used to exploit the repetitive nature of batch processes to determine the optimal operating policy. The optimization algorithm is proved convergent. The numerical simulation shows that the method can improve the process performance through iterations.
