Control system using process model

Abstract

A control system is provided for controlling a process for making paper or paper pulp. The process has a process product output at an end of the process. The controls include a process variable sensor input configured to receive a process variable related to the process. The controller is configured to provide a control signal to control the process. A process model has a model product output which is a model representation of the process product output. The model process output is a function of the sensed process variable and a product output setpoint representative of a desired process product output. The control signal is a function of the product output setpoint and the modeled product output.

Description

BACKGROUND OF THE INVENTION

[0001] The present invention relates to industrial process control. More specifically, the present invention relates to controlling processes which are used to manufacture paper or paper pulp.

[0002] In a control system, including controls systems which are used to control paper and paper pulp manufacturing, the state of a process is monitored by measuring “process variables.” A process variable refers to a variable which is related to the condition of the process such as a flow rate, a level, a temperature, a pressure, etc. Based upon the measured process variable, a controller operates a control element in order to bring the process variable within a target range. For example, in order to achieve a desired flow rate (process variable) a controller adjusts a valve (control element) accordingly.

[0003] One type of process control observes the quality of the product output to adjust the various process variables within the process. The quality measurement is a process variable which is measured at the output of the process. It can be difficult to control paper and pulp manufacturing processes using this technique because there is both a physical distance between the process output and a significant lag time due to the time it takes for a product to move through the process. The physical separation and the separation in time can make controlling the process to achieve a desired quality difficult.

SUMMARY OF THE INVENTION

[0004] A control system is provided for controlling a process for making paper or paper pulp. The process has a process product output at an end of the process. The controls include a process variable sensor input configured to receive a process variable related to the process. The controller is configured to provide a control signal to control the process. A process model has a model product output which is a model representation of the process product output. The model process output is a function of the sensed process variable and a product output setpoint representative of a desired process product output. The control signal is a function of the product output setpoint and the modeled product output.

BRIEF DESCRIPTION OF THE DRAWINGS

[0005] FIG. 1 is a simplified diagram showing a portion of a paper manufacturing process.

[0006] FIG. 2 is a simplified block diagram of a control system in accordance with one embodiment of the invention.

[0007] FIG. 3 is a simplified block diagram showing a microprocessor used to implement the control system of FIG. 2.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Typical process control techniques in paper or paper pulp manufacturing rely on the feedback of the product quality to adjust the physical setpoints of the process variables within the manufacturing process. Such measurements are made at the output of the process. Although there are other process variable sensors located at various points along the process, they are typically only used for controlling an immediate process variable, such as a level, to thereby regulate the setpoints for a physical parameter as determined by a process control loop.

[0009] With this configuration, it is difficult to control product quality which is typically measured at the output of the process. The process output may be hundreds of feet away from the rest of the process and corrections can only be made after the fact. Any product output which is out of specification is discarded. Further, due to the dead time associated with the disparate locations between sensors and actuators, the control system cannot tolerate high gain in the control loop. With the present invention, a model is used to model the output of the process based upon one or more process variables measured at one or more locations along the process. The output of the model can be indicative of a quality measurement which related to the quality of the output product which will be obtained if the current process setpoints or control algorithms are not changed. Thus, the model is used to provide an inferred measurement related to a modeled product output. Based upon this inferred measurement related to the “quality” of the process output, the up stream process setpoints or control algorithms can be adjusted accordingly. In a more advanced embodiment, the information from the model can be used to coordinate distributed control algorithms which communicate on local process control loops in which control elements are controlled based upon local surrogate measurements.

[0010] The present invention can reduce the dead time between the control element and the surrogate or inferred measurements such that increased loop gain can be used in the control algorithms to improve the performance of the control loop without destabilizing the process. Further, a deviation in the product quality can be detected prior to actually producing the product and the control algorithms or setpoints can be adjusted to correct and compensate for the variation before the product exits the process. Integration of feed forward control, integrated sensors and inferred measurements with this feedback control technique provide improved control of the process and the quality of the output.

[0011] FIG. 1 is a simplified diagram showing one aspect of a paper sheet fabrication process. In this simplified embodiment, a vacuum pump 12 is coupled to a reservoir 14 through an adjusting valve 16. A receiving chamber 18 couples to the reservoir 14 through a fast acting valve 20 and is configured to receive a supply of paper sheets 22 through a manufacturing process. A drain valve 24 is configured to drain the reservoir 18. A pressure sensor 30, flow sensor 32 and temperature sensor 34 are configured to measure the pressure, flow rate and temperature associated with chamber 18. Information from sensors 30, 32 and 34 are provided to control system 40 which operates in accordance with the present invention.

[0012] In operation, control system 40 has at least one process variable input configured to receive a process variable from a process variable sensor such as sensors 30, 32 or 34. Sensors 30, 32 and 34 can be process variable transmitters coupled to a process control loop such as a two wire process control loop. As discussed below, control system 40 includes a process model and controls the quality of the paper produce manufactured by the process based upon an output from the model and a setpoint. This control is achieved by controlling a control element such as drain 24, fast acting valve 20 or adjusting valve 16. In the process illustrated in FIG. 1, paper sheets are received in receiving chamber 18 and are dried through the operation of drain 24 and vacuum supplied by vacuum pump 12. Excess liquid is collected in reservoir 14. The particular setpoints of the process used to dry the paper 22 can detrimentally effect the quality of the paper produced by the process.

[0013] Pursuant to one embodiment of the invention, FIG. 2 is a simplified block diagram of control system 40 used, for example, in the process illustrated in FIG. 1. The control system can couple to process devices (sensors or control elements) through one or more process control loops 41. Example control loops include two-wire loops such as 4-20 mA loops, loops in accordance with the Fieldbus standards, HART standards and others. Control system 40 includes a controller 60 which receives a setpoint 62 and provides a control signal output to a control element, such as one of the control elements illustrated in FIG. 1. In accordance with the present invention, control system 40 includes a process model 70 which receives a process variable sensor input 72 and provides a modeled product output 74 to controller 60. As discussed above, controller 60 controls one or more control elements in the process based upon the modeled product output 74 and the setpoint 62. The controller 60 can operate using any control algorithm including neural networks, regressive learning techniques, fuzzy logic, rules, any type of proportional integral and/or derivative control, etc. The present invention is not limited to the particular algorithm or technique implemented in a controller which is used to process the modeled product output. Controller 60 can receive other setpoints 76 and one or more process variables 78 and the control signal output 64 can be a function of these additional inputs. Model 70 can provide the modeled product output 74 as a function of other process signals 80. Examples of other process signals include other process variables, process setpoints, control signals, information related to the type or operation of a particular process sensor or control element, date or time information or any other information related to the operation of the process. The model can be preconfigured or can be generated using learning techniques such as that provided through the use of a neural network.

[0014] FIG. 3 is a simplified block diagram of control system 40 showing one example physical configuration of the control system. In the example of FIG. 3, control system 40 utilizes a microprocessor 100. Microprocessor 100 is configured to receive setpoint 62 which can be stored in memory, such as memory 102. Memory 102 can also contain permanent and/or temporary values such as used for programming or for storing program instructions or temporary storage. The process variable input can be through any type of physical layer such as through standard communication protocols. Example protocols include two-wire based protocols such as HART or Fieldbus. However, the control system can receive the process variable information, or other data, using any appropriate technique. Similarly, the control signal output can be implemented in any type of hardware including hardware which operates in accordance with process control industry standards. Example standards include two-wire based standards such as HART or Fieldbus. However, the control signal output 64 can be in accordance with any appropriate technique.

[0015] Although the present invention has been described with reference to preferred embodiments, workers skilled in the art will recognize that changes may be made in form and detail without departing from the spirit and scope of the invention. The present invention can be implemented in process devices located in the field, or in process devices or controls located at a remote location such as a control room or distant facility. Any appropriate hardware and/or software can implement the present invention including personal computers, microprocessors at various locations or configurations, etc. The invention can be implemented in any appropriate process device 40 including the illustrated control system or other devices which have access to process variables such as transmitters or controllers. The device can be at a remote location or in close physical proximity to a process variable sensor.

Claims

1. A control system for controlling a process for making paper or paper pulp having a process product output at an end of the process, comprising: a process variable sensor input configured to receive a process variable related to the process for making paper or pulp paper sensed by a process variable sensor which provides a process variable output; a control element output configured to provide a control signal to a control element which controls the process in response to the control signal; a process model having model product output which is a model representation of the process product output in response to the sensed process variable; a product output setpoint representative of a desired process product output; and a controller configured to compare the product output setpoint to the modeled product output and responsively provide the control signal to the control element.

2. The control system of claim 1 wherein the process product output is controlled to achieve a desired product quality.

3. The control system of claim 1 including a plurality of process variable sensors and wherein the model product output is a function of process variable outputs from the plurality of process variable sensors.

4. The control system of claim 1 wherein the model comprises a plurality of rules which model the process.

5. The control system of claim 1 wherein the model comprises a neural network which models the process.

6. The control system of claim 1 wherein the model comprises fuzzy logic which models the process.

7. The control system of claim 1 wherein the process variable sensor is in a first control loop, the control element is in a second control loop which is down stream from the first loop, and wherein the controller corrects for process variations in the first control loop by adjusting the control element to achieve a desired product quality in the process product output.

8. The control system of claim 1 wherein the controller provides feed forward control as a function of the model product output.

9. The control system of claim 1 wherein the model is implemented in a microprocessor based system.

10. The control system of claim 1 wherein the model is implemented in a process device in physical proximity to the process variable sensor.

11. The control system of claim 1 wherein the process variable sensor transmits sensed process variables over a two wire process control loop.

12. The control system of claim 1 wherein the process model is coupled to the process variable sensor over a control loop which operates in accordance with Fieldbus.

13. The control system of claim 1 wherein the model is a function of an observed process history.

14. A method for controlling a process for making paper or paper pulp, the process of the type having a process product output at the end of the process, comprising: obtaining a process variable related to the process; modeling the process product output as a function of at least the obtained process variable; obtaining a setpoint related to a desired process product output; and generating a control signal as a function of the modeled process product output and the setpoint.

15. The method of claim 14 wherein the process product output is controlled to achieve a desired product quality.

16. The method of claim 14 wherein the modeling is based upon a plurality of process variables.

17. The method of claim 14 wherein the modeling is based upon a plurality of rules.

18. The method of claim 14 wherein the modeling is based upon a neural network.

19. The method of claim 14 wherein the modeling is based upon fuzzy logic.

20. The method of claim 14 wherein the control signal is generated to correct for potential future deviations in the process product output.

21. The method of claim 14 wherein the modeling is a function of an observed process history.

22. The method of claim 14 wherein obtaining a process variable comprises receiving a process variable from a two-wire process control loop.