An Overview of Model Based Control

Date: 6 July 2008

Date: 6 July 2008

While glass manufacturers have little control over raw material costs or finished product orders, process control is one area where even minor improvements can trim production costs, improve product quality, and ultimately increase profits.

A new model-based controller is outperforming PID controls in glass melter and forehearth applications. The ACSI integrated process controller reduces job change time up to 50% and quickly stabilizes temperature variations in order to provide high quality glass with optimum fuel efficiency.

Currently many manufacturers focus on using thermal gradient efficiency (TGE)  and mass flow temperature (MFT) to evaluate forehearth performance. In  doing so, parameters are individually controlled by an operator who makes  individual zone setpoint adjustments. The operator uses a 9-point grid to  achieve desired temperatures and maintain stability and homogeneity.  Traditional forehearth design places a sensor at the exit outlet of each  forehearth zone to measure glass temperature as it exits the zone. The sensor  relays data to a PID controller, which adjusts the heat to bring the glass  temperature back to setpoint. As the molten glass travels through each  forehearth chamber, respective controllers continue to “play catch up.” The  result tends to be length recovery time during which production values  decrease.

The Problem

Two issues are faced when using traditional forehearth control.

1. Temperature not directly controlled – operator makes the decision

2. PID is used to control

PID control works well under steady state conditions, however, such loops are  extremely difficult to tune and operators typically do not see the type of  reaction time they desire for job changes.

The Solution

Model based control achieves up to 50% reduction in job change time. The  controller regulates the temperature directly, relieving an operator from  making tuning judgments. In addition, forehearth homogeneity is regulated  differently from anything previously seen in the glass industry.


Adaptive Control

The model based controller creates a model of a zone’s response to changes.  During installation, an ACSI engineer teaches the controller what the model  looks like by making minor adjustments and allowing the controller to learn  how the process reacts.

Adaptive modeling begins with an estimated model of the process variable (PV)  response. The engineer then bumps the setpoint, and the controller adapts,  improving its model of the process. Now that the controller is aware of the  improved process model, the engineer bumps the setpoint a second time, and  the results show significantly less overshoot for the process variable.  As each adjustment is made, the controller continues to learn the subtleties of  the process and improves its model. After the model is created, the controller  continues to observe process changes and adapt its model. This is radically  different from traditional PID control, which reacts to error the same way  every time. A model-based controller looks at how well it predicted the  previous response and refines the process for the next time. As a result, the  need for tuning is eliminated.

The result of “training” the controller is a mathematical model of what  happens when increasing and decreasing the heating. The controller can then  predict what will happen with each subsequent change.  As shown in this figure, the final model has adapted to significantly different  dead time and gain when compared to the initial estimate. The controller  provides responsive control with no overshoot, using the final model for the  third and fourth setpoint bumps.

Feedforward Input

The model-based controller can continuously model the effect of feedforward  variables. Feedforward is particularly important for processes with  disturbances that result in substantial load changes. The key benefit is that  control actions can be taken before the process variable is affected.

• Controller continuously models up to 3 feedforward variables.

   Process deviation from setpoint is reduced

Implementing feedforward control with PID is difficult, and having more than  one feedforward with a PID loop is impossible. In contrast, model-based  controllers are capable of analyzing feedforward variables.  An example of a feedforward input is monitoring the incoming raw material  rate in a variable production rate process. The controller measures the  incoming glass temperature and predicts the actions needed to adjust before  the glass travels through the zone. Changes are made to the firing in order to  eliminate a major portion of the potential upset. In contrast, with PID an undesired temperature is not detected until the glass  reaches the zone exit.

Multiple Input, Single Output (MISO) controller allows several feedforward  inputs to influence a single output, as shown above. The MISO controller with  the addition of a patented Dynamic Modeling Technology of process dynamics  significantly reduces system commissioning effort and maintenance costs.  Model Based Control connects to the existing control system via a standard OPC communication interface. It is designed to improve the control of processes  that involve several interacting control loops that cannot be adequately  handled using independent loop controllers. This controller decouples the  interacting variables and provides coordinated regulatory control for these  types of processes. Process dead time is easily managed to provide responsive  control with no overshoot or process cycling.

Reduce Job Change Time

Now that a basis has been established for PID vs. MBC, reducing job change  time can be addressed. It is essential to minimize job change time in order to  maximize profit potential. This requires achieving new glass temperature  setpoints as quickly as possible with minimum overshoot. It is difficult to  achieve both objectives simultaneously with a standard PID controller. One of  two scenarios is likely to occur:

1. glass temperature can be raised quickly, but the temperature  overshoots the optimum set point and must be adjusted back to the  setpoint;


2. glass temperature is achieved with a gradual rise in temperature that  requires a long time to reach the setpoint. Either scenario typically  requires several hours to stabilize glass temperature during which  production operates at less than optimal parameters.

If the current setpoint is 2200°, and a job change requires the new setpoint to  be 2240°. One possibility using PID is that the temperature will slowly rise. A  period of dead time will occur before the temperature begins to wind up.  Eventually the temperature will overshoot the new setpoint and eventually  settle back to steady state. A second possibility is that the temperature will  rise quickly and overshoot the new setpoint. The temperature will begin to  oscillate and may not reach steady state.

With model-based control, the mathematical model does not have to wait for  an error to take action, rather it knows how much to change the valve position  to prevent the deviation. When a setpoint change is made, the controller  accounts for the dead time, which results in the temperature reaching the new  setpoint as quickly as possible with zero overshoot.

The result is easily 50% less time to get to the new setpoint, which translates  to 50% shorter job change times. Essentially, it is similar to taking the best  operator and harnessing the information he/she has about how long to hold the  system at each point. With more experience, he updates the information he  knows, just as the controller learns and adjusts.

Controlling glass temperature is the key to achieving optimum glass viscosity  and gob weight. Temperature variations as slight as one degree or less can  negatively impact the quality of the finished product and result in lost  production time. Job change time and zone temperature modeling offer  opportunities for tighter control.

The ACSI model-based controller is effective in controlling job change and zone  temperatures by modeling the existing process. The controller creates models  for each control/process variable and feed forward input. These ideal models  allow the system to anticipate changes needed to maintain consistent glass  temperature. Once the optimum process is modeled, the ACSI model-based  controller can:

- predict control actions required to drive the glass temperature to setpoint  quickly without overshoot 

- adapt to process and production rate changes automatically for better control without loop tuning 

- model feed forward inputs and update control actions to quickly stabilize temperature variation

When integrated as part of a comprehensive control system, the ACSI modelbased  controller delivers the following production and quality benefits: 

- reduces job change time by as much as 50% 

- reduces temperature variation by as much as 50% 

- improves overall production performance

Improving Homogeneity and Stability

Historically, gob temperatures have been controlled by changing the setpoint.  A 9-point grid has been used as a measure of temperature stability and  homogeneity, and an operator has decided how to adjust and continuously tries  to refine. Temperatures have been related to the control loops. Now with model based control, 9-point grid temperatures can be controlled,  and the interrelationships among the different temperatures can be understood  by the controller. Rather than thinking in terms of zones, the forehearth can  be thought of as one single unit. Temperature readings can be prioritized, in  order to determine which readings are most critical and which can be  sacrificed to achieve the essential readings. As a result of the new method of control, zones no longer “fight” each other to  achieve balance, and the complicated relationships are eliminated.

Overall Benefits of Model Based Control

The improved control of multivariable processes allows each system to be  responsive to disturbances or operating point changes. The benefits of modelbased  control over traditional PID control are clear.

Once the controller has modeled the process, it tracks the process dynamics  and predicts the control actions required to drive the glass temperature to  setpoint quickly without overshoot. Consequently, it adapts to process and  production rate changes automatically for better control without ever requiring loop tuning.

The model-based controller understands dead time and knows how to account  for it in its control strategy. PID and other error based controllers do not  account for dead time, and the result is that they often have to be de-tuned to  avoid overshoot. This results in sluggish response to setpoint or load changes.  Alternately, responsively tuned PIDs will overshoot and oscillate around  setpoint following a load or setpoint change.

Tighter control results in 30-50% reduction in deviations from setpoint vs.  traditional PID. This positively affects the process by increasing product  consistency & quality and reducing energy use, which in turn allows for  operation closer to specifications. Adaptive control minimizes job change time by quickly reacting and stabilizing  temperature variations. Feedforward modeling is possible for up to three variables, resulting in control action being taken BEFORE disturbances push the  PV off setpoint.

The results can be measured in terms of production- and quality-related  benefits. Job change time and temperature variation are both reduced by as  much as 50%, overall production performance is visibly improved, and plant  operation and maintenance are simplified.

600450 An Overview of Model Based Control

See more articles about

Others also read

Measurement area separated from glass sheets with buffer plates
Measurement area in direct thermal contact with specimen
In this paper, we present the development of an intrinsic parameter σQM characterizing the sensitivity of a coating (or configuration) to the quench marks.
The goal of this work is to investigate the mechanical performance of VIG’s subjected to soft body impact and gain insight into the failure mechanisms of the VIG when exposed to dynamic loads.
In the present paper, optical anisotropy effects in architectural glass are evaluated using digital image processing.
There are parts of the window inside the glass that play a crucial role in its performance.



Add new comment