Method and apparatus for controlling at least one quality feature of a running material web

Abstract

A method and apparatus for controlling at least one quality feature of a running material web, in particular a fibrous web, during its production, including a measurement system, an electronic control and/or an evaluation unit. The measurement system measures the quality feature on a running material web repeatedly over the entire width of the material web. The electronic control and/or evaluation unit determines the variability of the quality feature through a variance component analysis, which is conducted on the basis of the profile measure values of a measured value set including several consecutive CD profiles in the machine direction, the set having been recorded during a selectable previous time window. The result of this variance component analysis is also drawn on to control the quality feature.

1 . A method for controlling at least one quality feature of a running fibrous web during its production, the method comprising the steps of:
 repeatedly measuring a quality feature in the running fibrous web over an entire width of the fibrous web by way of a measurement system;   determining a variability of the quality feature through a variance component analysis by way of an electronic unit, said determining step including the step of conducting said variance component analysis on a basis of a plurality of profile measured values of a measured value set, said measured value set including a plurality of consecutive cross direction (CD) profiles in a machine direction, said measured value set having been recorded during a selectable previous time window; and   controlling the at least one quality feature dependant on said variance component analysis.   
     
     
         2 . The method of  claim 1 , wherein said conducting step further comprising the step of dividing said variability of the quality feature by way of said variance component analysis into a CD variability component in the cross direction, into a machine direction (MD) variability component in the machine direction and into at least one of residual variability components and confidence intervals for at least one of said CD profile and a MD profile calculated from determined variance values, said controlling step controlling the at least one quality feature being dependent upon at least one of said variability components and said confidence intervals. 
     
     
         3 . The method of  claim 2 , wherein individual measured values of said measured value set are arranged by said electronic unit in a matrix having said CD profile measured values in one of rows and columns of said matrix and said MD profile measured values in one of said rows and said columns. 
     
     
         4 . The method of  claim 3 , wherein said profile measured values in a respective one of said measured value set are variously weighted respective to their contribution to the variability of the quality feature at least partially according to how far back they lie in time as compared to a moment when said determining step is carried out. 
     
     
         5 . The method of  claim 4 , wherein said profile measured values that are farther back in time from said moment are weighted relatively more lowly than those said profile measure values that are less far back in time from said moment. 
     
     
         6 . The method of  claim 4 , wherein said conducting step is performed in a recursive manner. 
     
     
         7 . The method of  claim 4 , wherein said variance component analysis is repeated when a selectable number of new profile measured values are measured by way of said measurement system. 
     
     
         8 . The method of  claim 4 , wherein at least one result of said variance component analysis is drawn on to change a behavior of at least one control system to control the quality feature. 
     
     
         9 . The method of  claim 8 , wherein at least one of a control parameter and a filter parameter of said at least one control system is variably adjusted dependant on said at least one result of said variance component analysis. 
     
     
         10 . The method of  claim 4 , wherein said variance component analysis is based on at least one algorithm, said at least one algorithm including one of a Technical Association of the Pulp and Paper Industry (TAPPI) Technical Information Paper (TIP) 1101, TAPPI T585 and DAHLIN exact variance analysis. 
     
     
         11 . The method of  claim 10 , further comprising the step of calculating short-term deviations with said at least one algorithm used as a basis for said variance component analysis. 
     
     
         12 . The method of  claim 4 , wherein said CD variability component and said MD variability component are each divided into a controllable fraction and a non-controllable fraction. 
     
     
         13 . The method of  claim 8 , further comprising the step of determining if said residual variability component is relatively large compared to one of a current CD variability component, a current MD variability component, a controllable CD fraction and a MD controllable fraction and if said residual variability component is relatively large then using said control system to control the quality feature at least one of slower and more weakly during the production than if said residual variability component was not relatively large. 
     
     
         14 . The method of  claim 8 , further comprising the step of determining if said residual variability component is relatively large compared to one of a current CD variability component, a current MD variability component, a controllable CD fraction and a MD controllable fraction and if said residual variability component is relatively large then intensively filtering said residual variability component prior to a calculation of a respective set-point variable used by said control system to control the quality feature. 
     
     
         15 . The method of  claim 2 , wherein said confidence intervals for one of said CD profile and said MD profile are calculated from determined variance values, using said confidence intervals a distinction is drawn between a profile deviation based on an existing cause and a deviation based on substantially only said residual variability component. 
     
     
         16 . The method of  claim 15 , further comprising the step of intervening in the production of the fibrous web by a control system if in the light of at least one of said confidence intervals a deviation is based on an existing cause. 
     
     
         17 . The method of  claim 16 , further comprising the step of intervening in the production of the fibrous web by said control system one of relatively weakly and not at all if in the light of at least one of said confidence intervals a deviation is based substantially only on said residual variability component. 
     
     
         18 . The method of  claim 2 , wherein at least one of said variability components and said confidence intervals are drawn on for assessing a quality of a process model saved in said electronic unit. 
     
     
         19 . The method of  claim 18 , wherein any deviations of default values of said process model from recorded profile measured values is determined by way of said electronic unit, any of said deviations that are determined to be at least one of a relatively high variability of the quality feature and a relatively low confidence is at least one of weighted relatively more lowly and is filtered relatively more intensively prior to further use by said electronic unit than if said deviations are determined to not be one of a relatively high variability of the quality feature and a relatively low confidence. 
     
     
         20 . The method of  claim 18 , wherein at least one of said calculated variability components and said confidence intervals are used as input to change said process model. 
     
     
         21 . The method of  claim 20 , wherein at least one of said calculated variability components and said confidence intervals are used as input to change said process model by way of adaptation parameters. 
     
     
         22 . The method of  claim 20 , wherein said process model is corrected by way of the electronic unit more slowly if there is at least one of a relatively higher variability of the quality feature and a relatively lower confidence of said profile measured values. 
     
     
         23 . The method of  claim 2 , wherein at least one of a MD variance and a CD variance is not explicitly determined in said variance component analysis. 
     
     
         24 . The method of  claim 1 , wherein the quality feature is at least one of moisture, thickness, gsm substance and filler material content of the fibrous web. 
     
     
         25 . An apparatus for controlling at least one quality feature of a running fibrous web during its production, the apparatus comprising:
 a measurement system for repeated measurement of the at least one quality feature over an entire width of the fibrous web; and   one of an electronic control and an evaluation unit for determining a variability of the at least one quality feature by way of a variance component analysis dependent on a plurality of profile measured values of a measured value set, said measured value set including several consecutive cross direction (CD) profiles in a machine direction (MD), said measured value set having been recorded during a selectable previous time window, one of said electronic control and said evaluation unit controlling the quality feature intervenes in the production of the fibrous web at an intensity determined by said variance component analysis.   
     
     
         26 . The apparatus of  claim 25 , wherein one of said electronic control and said evaluation unit controls the variability of the quality feature by way of said variance component analysis that includes a CD variability component in said CD, and a MD variability component in said MD and at least one of a residual variability component and confidence intervals for at least one of said CD profile and a MD profile calculated from determined variance values such that the control of the quality feature is dependent upon at least one of said variability components and said confidence intervals. 
     
     
         27 . The apparatus of  claim 26 , wherein at least one of a MD variance and a CD variance is not explicitly determined in said variance component analysis. 
     
     
         28 . The method of  claim 25 , wherein the quality feature is at least one of moisture, thickness, gsm substance and filler material content of the fibrous web.