Method for overall regulation of a former of a paper machine or equivalent

• US 5,825,653 A
• Filed: 03/14/1997
• Issued: 10/20/1998
• Est. Priority Date: 03/14/1997
• Status: Expired due to Fees

• Current Assignee: Valmet Incorporated (Metso Corporation)
• Original Assignee: Valmet Incorporated (Metso Corporation)
• Inventors: Hamalainen, Jari, Nyberg, Petri, Pakarinen, Pekka, Huovila, Jyrki
• Primary Examiner(s): Gordon, Paul P.
• Assistant Examiner(s): Garland, Steven R.
• Application Number:
  US08/818,837
• Time in Patent Office: 585 Days
• Field of Search: 364/148, 364/149, 364/152, 364/150, 364/471.01, 364/471.02, 364/148.03, 364/148.06, 364/578, 162/198, 162/252, 162/253, 162/262, 162/903, 162/DIG. 10, 162/258, 162/DIG. 11
• US Class Current: 700/128
• CPC Class Codes: D21F 1/06 Regulating pulp flow
• CPC Class Codes: D21G 9/0027 controlling the forming section
• CPC Class Codes: H01L 21/6835 using temporarily an auxiliary support
• CPC Class Codes: H01L 2924/01033 Arsenic [As]
• CPC Class Codes: Y10S 162/10 Computer control of paper making variables

Abstract

A method for continuous overall regulation of a former of a paper or board machine or equivalent in which a physical fluid flow model of the drainage and the flow state of the pulp suspension flow is formed. The model is solved by making use of the data on the geometry of the former and initial and boundary conditions necessary for a solution to the model. The difference between the simulated state, which has been obtained as the solution of the physical flow model, and the state adopted as the target is determined as the cost function, wherein the target state is determined by means of the quality requirements of the paper and possibly by means of the costs of operation and the runnability of the paper machine. The cost function is optimized and, thus, optimal regulation values and set values are obtained for the regulation devices and/or actuators of the former in view of the given cost function. After the optimization has been carried out in the preceding stage, the former'"'"'s regulation devices and/or actuators are given new optimal regulation and set values.

Claims

1. A method for optimizing the overall regulation of a former of a paper or board machine in which a pulp suspension flow is drained to produce paper, comprising the steps of:

(a) forming a physical fluid flow model of the drainage and flow state of the pulp suspension flow in the former;

(b) solving the flow model to obtain a simulated flow state based on data on the geometry of the former and initial and boundary conditions related to the former;

(c) obtaining a target flow state based on quality requirements of the paper;

(d) determining a difference between the simulated flow state obtained as the solution of the flow model and the target flow state, the difference constituting a cost function;

(e) optimizing the cost function and then determining optimal regulation values and set values for regulation devices and actuators which affect the drainage and flow state of the pulp suspension flow in the former in view of the cost function; and

(f) providing the optimal regulation values and set values to the regulation devices and actuators of the former such that the regulation devices and actuators operate at the optimal regulation values and set values to thereby realize the optimization of the cost function.

2. The method of claim 1, wherein the step of forming the flow model comprises the steps of:

forming the flow model on the basis of fluid dynamics relationships or on the basis of simplified equations derived from fluid dynamics relationships which are solved numerically by means of a computer, and

utilizing on-line and/or laboratory measurements and information provided by an operator of the paper machine to obtain the initial and boundary conditions of the flow model.

3. The method of claim 1, further comprising the steps of:

computing a plurality of flow states by means of a physical model,

adapting a statistical model from the plurality of flow states to illustrate the flow state of the former and/or the dependency of the cost function on the regulation quantities, or instructing a neural network to illustrate the flow state of the former and/or the dependency of the cost function on the regulation quantities from the plurality of flow states, and

utilizing the statistical model or the neural network for optimizing the cost function.

4. The method of claim 1, wherein the step of solving the flow model comprises the steps of:

using a computer having a high computing and processing capacity and comprising at least one processor, and

directing the computer to use numerical methods of solution.

5. The method of claim 4, further comprising the step of selecting the numerical methods of solution from the group consisting of finite-element, finite-difference and finite-volume methods.

6. The method of claim 1, wherein the step of optimizing the cost function and then determining optimal values for the regulation devices and actuators of the former is carried out in accordance with the following algorithm:

(a.) the necessary flow-rate, stock-grade and geometry data of the former are read, and the optimizing is started from the default values α

.sup.(i), i=0, and the value F.sup.(i) of the cost function is computed by solving the flow model of the former,

(b.) the gradient VF.sup.(i) of the cost function with the regulation values α

.sup.(i) concerned is computed,

(c.) the direction p.sup.(i) is determined by means of the gradient VF.sup.(i), which direction indicates the direction the regulation values must change in order to lower the value of the cost function, the sensitivity of the cost function F in relation to the regulation quantity α

_k being computed by means of the difference quotient ##EQU5## or by solving the adjoint state equation corresponding to the flow model of the former,

(d.) the new values of the regulation quantities (C) are determined by varying them in the direction determined in the preceding step (c.) optimally

space="preserve" listing-type="equation">α

.sup.(i+1) =α

.sup.(i) +λ

.sup.(i) p.sup.(i)

in which connection the step length λ

.sup.(i) is determined so that the cost function receives a lower value than with the preceding iteration, F.sup.(i+1) <

F.sup.(i), and the regulation quantities (C) are admissible, and

(e.) if the value of the cost function obtained in the preceding steps is not yet sufficiently low, an index for iteration step is increased, i=i+b 1, and optimizing is continued from the above stage (b.), and in the contrary case, performing of the algorithm is discontinued, and the optimal values of the regulation devices and actuators are set equal to the optimized regulation quantities (C) and they are transmitted to the regulation devices and actuators of the former.

7. The method of claim 1, wherein the step of optimizing the cost function is determined by an algorithm and is not based on computing the gradient of the cost function.

8. The method of claim 1, further comprising the step of selecting the measurement and regulation quantities of the former to be regulated from the group consisting of the state of the pulp suspension jet coming from a headbox of the paper or board machine, the running speed and tensions of forming wires in the former, the vacuum levels in water drain elements in the former, the location, loading and/or position of the water drain elements in the former, and the covering angle on a former roll in the former about which the web is passed.

9. The method of claim 1, wherein the cost function comprises at least the target profiles of finished paper, the target profiles being the basis weight and fiber-orientation profile, or out of values computed from these profiles.

10. The method of claim 9, wherein besides the target profiles, the cost function also comprises components related to the runnability and/or energy consumption of the paper machine.

11. The method of claim 1, further comprising the step of selecting a quality level for the finished paper and determining the lowest possible energy consumption by minimizing consumption of jet waters and wear of wires and water drain elements.

12. The method of claim 1, further comprising the step of weighting different parts of the cost function, the parts being the quality of the paper, the runnability of the machine and the energy consumption of the machine.

13. The method of claim 12, wherein the quality of the paper is set as the most important part, and the runnability is improved and/or energy saved vis-a-vis the cost function only if a deterioration in the quality of the paper does not occur.

14. The method of claim 1, wherein the cost function includes quality quantities which are expressed by means of the properties of the web after the former.

15. The method of claim 1, wherein the step of solving the flow model comprises the steps of:

using a computer having a high computing capacity and comprising at least one processor, and

increasing the speed in which the computer solves the flow model by performing the computations in parallel.

16. The method of claim 15, wherein the computer comprises a plurality of processors and the speed of the computer is increased by computing each of the sensitivities of changes in the cost function in relation to different regulation devices of the former by means of a separate processor in the computer.

17. The method of claim 16, wherein the sensitivities of the cost function in relation to different regulation devices are determined by means of a sensitivity analysis carried out in an environment of linearized solution so that the flow model is solved by means of a perturbation theory by determining the gradient of the cost function, wherein for the optimization of the cost function, in other respects, a real non-linear model is used such that the optimization proper is based on a real flow model, and any faults arising from the perturbation theory do not have a substantial effect on the final result of the optimization.

18. The method of claim 1, wherein the step of solving the flow model comprises the steps of:

composing the flow model of the former from several sub-models,

simultaneously solving each of the sub-models by means of a respective processor of a computer, and

directing data from each of the processors to other ones of the processors.

19. The method of claim 1, wherein the step of solving the flow model comprises the step of:

using a computer; and

enhanced the ability of the computer by dividing the optimizing into parts so that the regulation quantities that have the greatest effect on the quality of the finished paper are optimized more frequently while reacting to the changes rapidly, whereas other regulation quantities are optimized less frequently, and that for optimizing of the regulation quantities for which the whole flow model is not needed a suitable sub-model is used.

20. The method of claim 1, wherein the former includes forming rolls, an MB-unit, forming shoes and/or suction flatboxes, water drain elements including loadable forming ribs, and a pair of forming wires for carrying a web formed therethrough, further comprising the step of:

regulating at least one of the vacuum levels in the forming rolls, the position of the forming rolls, the operation of the MB-unit, the operation of the forming shoes, the operation of the suction flatboxes, the position of the forming ribs, the loading forces applied to the forming ribs, the position of the water drain elements, the tension of the forming wires, the retention of the pulp suspension jet and the retention profile of the pulp suspension jet in a direct transverse to the flow direction in accordance with the optimization of the cost function.

21. The method of claim 1, further comprising the steps of:

forming the physical fluid flow model for a grade of paper different than the grade being produced from the pulp suspension flow so that optimal regulation values and set values are obtained for the new grade of paper.

22. The method of claim 1, further comprising the steps of:

forming the pulp suspension flow by directing a slice jet from a headbox of the paper or board machine,

regulating the headbox to optimize its operation.

23. The method of claim 1, further comprising the step of:

controlling at least one of the cross-direction basis weight profile and fiber orientation profile of a web produced from the drained pulp suspension flow.

24. The method of claim 1, wherein the target flow state is also obtained based on the costs of operation and runnability of the paper machine.