部分最小二乘算法的神经元网络实现

部分最小二乘(PLS)算法在多元统计过程监控等领域得到了广泛应用.但常用的求解方法需要多次迭代求解残差矩阵,不利于对算法的理论分析和结论的解释.基于PLS算法的优化函数形式,该文提出一种新的PLS优化目标函数及相应简化算法.在此基础上构造了PLS算法与线性神经元网络之间的自然映射,给出了相应的训练算法及其理论分析.仿真结果验证了所提出算法的有效性,表明该算法可直接从原数据矩阵得到相应的成分及回归系数,并易于对其进行解释.     

时变大纯滞后系统的单神经元预测控制

为对时变大纯滞后系统实现快速有效的实时控制 ,将单神经元与 L evinson预测器相结合采用比例积分微分调节器 (PID)控制方式 ,设计了单神经元预测控制器 ,其权值和可整定参数能够在线自适应调整 ,克服了大滞后对象控制结果不能及时反馈的不足。应用该控制策略对大滞后一阶和二阶对象的仿真研究表明 ,对大滞后时变系统具有很强的适应性和鲁棒性 ,各种控制性能明显优于常规 PID等其它控制     

Multivariate Statistical Process Monitoring Using Robust Nonlinear Principal Component Analysis

The principal component analysis （PCA） algorithm is widely applied in a diverse range of fields for performance assessment, fault detection, and diagnosis. However, in the presence of noise and gross errors, the nonlinear PCA （NLPCA） using autoassociative bottle-neck neural networks is so sensitive that the obtained model differs significantly from the underlying system. In this paper, a robust version of NLPCA is introduced by replacing the generally used error criterion mean squared error with a mean log squared error. This is followed by a concise analysis of the corresponding training method. A novel multivariate statistical process monitoring （MSPM） scheme incorporating the proposed robust NLPCA technique is then investigated and its efficiency is assessed through application to an industrial fluidized catalytic cracking plant. The results demonstrate that, compared with NLPCA, the proposed approach can effectively reduce the number of false alarms and is, hence, expected to better monitor real-world processes.     

Convergence Analysis of Forgetting Gradient Algorithm by Using Martingale Hyperconvergence Theorem

Introduction　　Considerthefollowinglineartime-varyingsystem[1,2],A(t,z-1)y(t)=B(t,z-1)u(t) v(t)(1)where{u(t)}and{y(t)}aretheinputandoutputsequencesofthesystem,respectively,{v(t)}isastochasticnoisesequence,andz-1representstheunitbackwardshiftoperator,i.e.,z-1y(t)=y(t-1),A(t,z-1)andB(t,z-1)aretime-varyingcoefficientpolynomialsintheunitbackwardshiftoperatorz-1,andA(t,z-1)=1 a1(t)z-1 a2(t)z-2 … ana(t)z-na,B(t,z-1)=b1(t)z-1 b2(t)z-1 … bnb(t)z-nb.Definetheinformationvector(t)andthetime-varyin…