Study on the NN Decoupling Control System of Air-cushioned Headbox

Abstract

The headbox is the key hinge to link the pulp supply system with the sheet forming in the papermaking process. The primary parameters include the total pressure and the stock level which couple each other in the box, and they decide the distribution of the web cross-directional basis weight and influence the paper forming quality. Here taking widely used air-cushion type headbox as the plant, a new neural network (NN) decoupling control system is proposed to overcome the coupling relation between the total pressure and the stock level, decrease the adjusting time as well as to achieve robustness, fault-tolerance and the self-study ability in different environments. The practice has proved that the headbox control system based on NN decoupling algorithm could satisfy the industrial control requirements in the papermaking process well.

Brief Review and Mathematical Modelling of Air Cushioned Pressurized Paper Machine Headbox

“…The tuning is based on finite frequency response data. A simplified dynamic model for the wet-end in a paper mill was proposed in [48], Artificial Neural Networks based moisture and basis weight control in [49], Advanced control methods and decoupling algorithms [52], Decoupling and Time-delay Control Approach [54], A neural network (NN) based decoupling control technique has been devel-oped in [56], Artificial neural network (ANN) based retention control [58], Adaptive fuzzy controller [59], Various controllers and tuning methods have been studied on paper machine headbox in [60]. Neural network decoupler control system [61], Model Predictive Control for consistency and liquid level control [62], GA based Neural PID decoupling control [63], Advanced prediction based control approach [64], Various PID controller algorithms for consistency control [65], A decoupling control system design [67], Non-fragile bilinear state feedback control [68].…”

“…Based on the literature review discussed in section 3, the identified control techniques for paper machine headbox are listed below: a) Adaptive Control [2], [41], [59] b) Predictive Control [36], [47], [62], [78] c) Robust Control [14], [19], [21], [27], [69], [70], [71], [74] d) Optimal Control [28], [30], [33], [40], [44] e) Multivariable Non-linear Control [15], [35], [72], [77] f) Bilinear Control [11], [68] g) Intelligent Control [49], [58], [75], [76] h) Decoupling Control [12], [52], [54], [56], [63], [67] i) Digital Control [2], [4], [30] j) Internal Model Control [20], [24], [79] As listed above, there have been many control strategies developed for different types of paper machine headbox. However, the efficient control can be ensured only by perfect modeling of a dynamical system.…”

“…Many control methodologies have been developed for headbox control in past few decades. The major parameters required to be controlled in headbox are: Total pressure head, Stock level, Consistency, Jet velocity, and rush/drag [3], [10], [11], [15], [16], [19], [20], [27], [28], [30], [31], [52], [56], [61], [63], [64], [68], [69], [71], [74].…”

Stability analysis of paper machine headbox using a new PI(D) tuning technique

“…al. presented adaptive control methodology for paper machine in [2], while a computer-based design of controller and indeterministic state variable model have been proposed in [3] [5], bilinear control strategy in [6] and systematic decoupling control in [7], Nonlinear predictive control in [8], An Internal model control with reference model was proposed in [12], Robust control through loop-shaping design in [10], An object oriented control using modeling language OMOLA in [11], MIMO digital-linear-quadratic-regulator in [12], The fragility issues related to the controllers and the aspect of robustness that has been neglected in analytical treatments of control system design of paper machine headbox is discussed in [13], Shape optimization and optimal control techniques were developed for numerical control of paper machine headbox flows in [14], Design of CD control of paper machine through multivariable problem, Optimal minimum control effort for a Fourdrinier machine headbox in [16], Spatiallydistributed feedback control technique for CD control of paper machines in [17], A non-smooth bi-objective optimization technique for the design of the shape of a slice channel [18], Interactive multiobjective optimization method NIMBUS [19], Advanced control methods and decoupling algorithms [21], A neural network (NN) based decoupling control technique has been developed in [22], Artificial neural network (ANN) based retention control [23], Adaptive fuzzy controller [24], Various controllers and tuning methods have been. studied on paper machine headbox in [25].…”

“…Headbox in a paper machine is intended to cause a steady state movement of pulp to wire. Important step in the paper making process is the stock material which is uniformly distributed on the wire through headbox [22]. A headbox is categorized as: open type headbox and pressurized headbox.…”