公开(公告)号：US6086237A

公开(公告)日：2000-07-11

申请号：US735073

申请日：1996-10-18

申请人： Dimitry M. Gorinevsky ,  Edwin Michael Gyde Heaven

发明人： Dimitry M. Gorinevsky ,  Edwin Michael Gyde Heaven

IPC分类号： G05B13/04 ,  G06F19/00 ,  G06G7/64 ,  G06G7/66

CPC分类号： G05B13/042

摘要： A process for determining shrinkage and alignment of the web in a sheetmaking machine having a plurality of actuators for controlling web parameters in the cross-direction. The process involves the steps of applying excitation to the actuators, collecting data regarding the change in cross-direction web properties due to the excitation of the actuators to determine a measured actuator response profile. A modeled actuator response profile is established and a best fit of the modeled actuator response profile to the measured actuator response profile is established to allow for determination of alignment and shrinkage parameters of the web based on the best fit modeled actuator response profile. The foregoing approach permits reliable identification of alignment and shrinkage within 10-20 data scans of the web, despite process noise, with improved accuracy when compared with conventional "bump" tests.

摘要翻译： 一种在具有用于控制横向方向上的卷筒纸参数的多个致动器的制片机中确定卷材的收缩和对准的方法。 该过程涉及对致动器施加激励的步骤，由于致动器的激励而收集关于横向幅材特性变化的数据，以确定测量的致动器响应曲线。 建立了建模的执行器响应曲线，并且建立了建模的致动器响应曲线与所测量的致动器响应曲线的最佳拟合，以允许基于最佳拟合建模的致动器响应曲线来确定幅材的对准和收缩参数。 尽管过程噪声，与传统的“碰撞”测试相比，上述方法允许在纸幅的10-20次数据扫描中可靠地识别对准和收缩。

公开(公告)号：US06647354B1

公开(公告)日：2003-11-11

申请号：US09667977

申请日：2000-09-22

申请人： Dimitry M. Gorinevsky

发明人： Dimitry M. Gorinevsky

IPC分类号： G06G748

CPC分类号： G05B13/0265 ,  G05B21/02

摘要： Process control methods and apparatus for controlling batch processes including batch heating processes. The present invention includes iterative learning control (ILC) techniques to provide improved run-to-run process control. One method provides a desired temperature profile over the length of a batch time period. This method gathers historical measured value data over the length of a first batch run which can be converted to deviation or error historical data, as well as a historical output history of outputs provided to control the process for the first run. The historical deviation and output histories from the first run can then be used to generate the output profile for a second batch run. Thus, the outputs from a first batch, such as the output to a local heater control, together with the deviation or error history from the first batch run, can be added to the output value provided during the previous batch to generate new output value to be used to control a second batch run.

摘要翻译： 用于控制批处理过程的过程控制方法和装置，包括批量加热过程。 本发明包括迭代学习控制（ILC）技术，以提供改进的跑步运行过程控制。 一种方法在批次时间段的长度上提供期望的温度分布。 该方法在第一批运行的长度上收集历史测量值数据，该数据可以转换为偏差或错误历史数据，以及为控制第一次运行的过程而提供的输出的历史输出历史。 然后可以使用第一次运行的历史偏差和输出历史来生成第二批次运行的输出配置文件。 因此，可以将来自第一批次的输出（例如输出到本地加热器控制）以及来自第一批次运行的偏差或错误历史记录添加到在前一批次期间提供的输出值以产生新的输出值 用于控制第二批运行。