Tuning Feedback Controller of Paper Machine for Optimal Process Disturbance Rejection


This paper presents a method for tuning the feedback controller of paper machine cross-directional control systems. The tuning method is based on identification of the process model, identification of the disturbance model and tuning the feedback controller by minimizing the paper property variations. To obtain a longer disturbance realization sequence and increase data available for identification of the process disturbance, 2-dimensional process identification residuals are used for identifying an integrated moving average disturbance model. The identification method is based on the Recursive Extended Least Squares. Based on the identified process and disturbance models, the Dahlin controller and control filter are tuned to minimize a quadratic performance index which includes the process output variance and the incremental control move variance. A penalty for excessive actuator move can be used to minimize the process variation while keeping the control action within acceptable bounds. The proposed method has been implemented in an industrial tuning tool. It has been validated using many sets of paper mill data. Extensive tests have shown that the identification algorithms are capable of 1 Control Systems'98, Porvoo, Finland, September 1998 Finnish Society of Automation, Helsinki, Finland identifying the process model as well as the disturbance model with satisfactory precision. The tool predicts the closed-loop process variance and control variance with a satisfactory degree of accuracy. 1. Problem statement This paper considers the tuning problem of the feedback controller of a commercial paper machine crossdirectional control system. Paper machines produce 2-dimensional paper sheet from the pulp suspension. On a paper machine, the paper properties, such as basis weight, moisture content and caliper, are measured and controlled in two directions, the machine direction (MD), i.e. the direction in which the paper sheet moves, and the cross-direction (CD), i.e., across the paper web. The goal of paper machine control systems is to compensate process variability and maintain the paper properties on target in both MD and CD. CD profiles of paper sheet properties are controlled by various CD actuators. Each type of CD actuator includes a set of the identical actuators located usually at evenly spaced points along the cross-direction. Depending on the application and the actuator type, there can be 20 to 300 individual actuator units in one CD actuator. An example of important CD actuator is the weight actuator, which adjusts the stock distribution across the machine by changing the opening of different sections of the slice lip in the headbox. Sensor measurements are located at a distance down the machine-direction from the actuation. Due to the high cost of sensors, limited number of sensors (1 2) measures only a zigzag portion of the paper sheet. From this limited number of measurements, the entire sheet profile must be estimated at each sampling time for feedback control. This estimation can be performed in a straightforward manner using Kalman filtering and averaging techniques [1]. The control problem is to calculate the actuator moves based on the estimated profile at each sampling instance. For better control of papermaking processes, various advanced control strategies have been proposed and have achieved a certain degree of success [2 5]. However, most paper machine control systems still use some kinds of well established simple controllers, such as PI, PID, Dahlin and so on. Tuning of those controllers plays an important role in reducing impact of the process variability on the product uniformity and in ensuring that the process is operated at the chosen target. Controller auto-tuning has been an important research topic for a long time and various tuning strategies have been proposed in the literature. The Ziegler-Nichols method for tuning PID regulators [6] is a popular and widely accepted scheme. It is based on detection of the critical gain and critical period and a quarter decay criterion for controller parameter design. The Dahlin controller is a well known dead-time compensator and is widely used in industries. Its tuning requires the process model being identified and

Extracted Key Phrases

13 Figures and Tables

Cite this paper

@inproceedings{Zhang2004TuningFC, title={Tuning Feedback Controller of Paper Machine for Optimal Process Disturbance Rejection}, author={Ming-Heng Zhang and Dimitry M. Gorinevsky and Guy Albert Dumont}, year={2004} }