When implementing a new Model Predictive Controller (MPC) project or maintaining an existing MPC application, model identification is the most difficult, time-consuming and labor-intensive part of the process. Traditional testing, modeling, integration and commissioning procedures typically take three to four months of around-the-clock work to complete.
BOC Gases (now part of the Linde Group), which wanted to replace the poorly performing MPC on the air separation unit at its Hartford, Ill., site, sought a faster and lower cost method to deploy a new system while maintaining safe and effective plant operation. So, in 2006 the company partnered with Matrikon, Inc., Edmonton, a veteran of 49 conventional MPC implementations at other BOC sites, to apply a radical new methodology using Tai Ji technology in testing and modeling the Hartford plant.
BOC’s Hartford air separation unit makes both liquid and gas products. On average, the plant produces 600 metric tons of gaseous oxygen and 650 metric tons of liquid nitrogen and oxygen. The site also has gaseous oxygen pipelines that feed nearby customers and a gaseous argon/nitrogen (GAN) pipeline.
BOC Gases monitors most of its plants from remote operation centers (ROC); few personnel actually work at the plants. Operators at the ROC are watching five-to-six plants at any given time. To ease the load on operators, BOC decided in October, 2001, to implement an MPC for the plant.
In five years of operation, the performance of this original MPC consistently fell short of specifications. Operators complained of constant controller issues that regularly led to the MPC being switched off and the plant running under operator control. Worse yet, the operators called the plant one of the hardest for them to run.
The process at Hartford is very nonlinear, especially the low pressure column (LPC) purity control, which is extremely critical. Running the plant under automatic (regulatory) control is difficult and often can require operator intervention to maintain purities; even with manual intervention, some products, especially GAN and cryogenic liquid argon (CLAR), can fall below specifications.
With operators burdened with monitoring five or six other sites, reliance on automatic control needed to be reduced to an absolute minimum. It was crucial that MPC be revamped as soon as possible. And, of course, reliable MPC was a must.
The Tai Ji implementation method
BOC and Matrikon decided to use Matrikon Control Performance Monitor Tai Ji to greatly accelerate the commissioning process. Based on Dr. Yucai Zhu’s industry-proven Tai Ji identification technology, it provides an alternative to traditional step testing and model identification methods. BOC saw three clear advantages in the method:
- It’s automatic. Testing and model identification are done automatically rather than manually by engineering personnel.
- It’s multivariable. Multiple manipulated variables (MV) are tested simultaneously, making test time much shorter than for single variable tests.
- It’s closed loop. Tests can be performed closed loop (MPC or PID), resulting in fewer disturbances and operator interventions.
In short, this new method of MPC implementation allows the modeling, step testing, integration and commissioning phases to overlap — at the procedure’s peak all four phases, in fact, are underway simultaneously — while requiring fewer personnel resources, at a greatly reduced risk of process upsets. Step-by-step, the methodology ran as follows:
Closed-loop operating data were obtained from the plant — approximately one month’s worth of data from when there was a lot of movement in the plant. The data were put through Control Performance Monitor Tai Ji to obtain initial models. These then were incorporated into the MPC and the controller was brought online.
The plant was commissioned for two days to ensure that everything was running correctly and that the MV and controlled variables (CV) were functioning as they should. Once the initial commissioning was complete, a Tai Ji step test was performed for three to four days. This test involved moving all the MV simultaneously by perturbing CV target values every minute.
“Using Tai-Ji to step test the plant in a closed loop mode was great,” says Mike Golinsky, ROC engineer for BOC. “It reduced testing time, improved model quality and, most importantly, reduced the risk of a loss-producing event occurring during the testing process.”
While the step test was underway, we conducted modeling passes every 12 to 16 hours to give the implementation team direct feedback on how the test was going. This feedback led to step size changes to the CV targets (higher signal-to-noise ratio) to better develop the models. Toward the end of the step test a lot of key models had converged — so, the decision was made to put them online with the step test still going on. With the new models online and the step test running, commissioning of the new models began. This is the new methodology at its peak, with all four phases — integration, step testing, modeling and commissioning — occurring at the same time. Once the step test was over, commissioning and fine-tuning of the plant continued.
The end result of using this methodology was the commissioning of a reliable grassroots MPC in 12 work weeks rather than the industry standard 3-4 months (Figure 1). The greatest challenge faced in this implementation was neither technical nor operational — it was achieving confidence in the results of the new method. After all, it’s so much easier to see results when step testing is done one MV at a time versus moving all MV together. BOC had a tough time in the beginning believing the results from Tai Ji but eventually confidence grew and it accepted the results.
Figure 1. Overlapping of phases in Tai Ji method significantly shortens implementation time.
Even then there was a challenge in getting operations staff, conditioned by five years working with an under-performing MPC, to trust the new controller. The operators were so accustomed to seeing issues with the old MPC that during the step testing and commissioning phases they always were quick to jump in to take a loop out of MPC and put it into automatic.
Asked why a certain action was taken, an operator replied, “Well, the old MPC would tend to bury this purity, so I don’t want to take any chances and have an upset.” An upset could mean anywhere from four to eight hours of lost production. It took a while but the operators finally developed enough confidence in the new MPC.
Immediate benefits and future plans
When you add up the savings in engineering coverage, operator interventions and time necessary both for testing and analyzing data — all cut by approximately 50% — the real bottom-line benefit of implementing an MPC in this compressed time frame is enormous. This is in addition to the well-known industry-proven advantages of having a reliable well-modeled MPC. The Hartford site gained a number of important benefits:
- increased LPC stability, which leads to better CLAR recovery, reduced operator intervention, and less downtime or fewer loss-producing events;
- the ability to change production rates very quickly, without upsetting the plant and the key purities;
- optimization of the evaporation tower;
- improved control of power demand — the MPC is set up such that it can drive production while also maintaining the power demand targets set by the power company;
- better load-following on pipelines — the MPC can quickly adjust plant production to meet changes in pipeline demand;
- enhanced constraint handling; and
- improved ramping — overall average ramp rate improvement of 220 to 650%.
“LMPC [Linear Model Predictive Control] has done a much better job than our previous APC [Advanced Process Control] system in maintaining the ever-so-critical low pressure column purity control,” says Golinsky. “The LMPC system has been in place five months now, and our downtime on argon has been reduced by over 75%. The plant will run for weeks or longer without operator intervention being necessary on the LPC purity.”
The old underperforming MPC didn’t provide any of these benefits. While specific numbers for the Hartford plant aren’t yet available, Golinsky estimates a $20,000/year savings from reduction in upsets alone. Industry experience shows that a solid MPC implementation will provide $80,000 to $85,000 annual benefit to a plant.
After the demonstrated success at Hartford, BOC is implementing this methodology at other eight plants. We have finished two already and are currently are in the process of doing three more.
Zul Bandali is an advanced controls engineer for Matrikon in Edmonton, Alberta. Reach him via e-mail at Zul.firstname.lastname@example.org