Ronaldo Aguilar, Fiberline Coordinator at Veracel. “We asked for production stabilization in the bleach plant and we got it with OPP. I’m very satisfied with the results so far.” Sindus Andritz is working with Veracel to optimize the automated control of each process area. The first target was the bleach plant – where production was increased by 2-3%.
Ronaldo Aguilar is, admittedly, not an expert in control theory. His expertise is in pulp production and leading a world-class production team.
So Aguilar, Fiberline Coordinator at Veracel Celulose’s 900,000+ t/a pulp mill in Brazil, has little interest in the inner workings of the Optimization of Process Performance (OPP) software. He is, however, very interested in the results that OPP achieved for his bleaching line – in just one week.
“I don’t need to know how it works, I just need to know that it does work!” ex-claims Aguilar. “We’re getting a two to three percent increase in our OEE now.” OEE, Aguilar explains, stands for Overall Equipment Efficiency, which impacts the quality, performance, and cost of production.
A 2-3% production increase in OEE may not seem like much until you consider the production capacity of the Veracel mill. On its best day, the mill produced a record of nearly 3800 admt/d. The OPP service helped Veracel establish a new quality record in October, producing 99.94% prime pulp. The economids of OPP, according to Aguilar, are quite favorable.
“We are optimizing each area in terms of reducing costs and stabilizing production,” Aguilar says. With all the production systems at Veracel operating near full capacity to keep up with market demand, the bleaching line was chosen as the first target for OPP because it is currently the “bottleneck” (if running above design capacity can be called a bottleneck).
OPP is a new service offered by Sindus Andritz (see side story on page 23) to its pulp and paper clients in Brazil.
Sindus Andritz maintains all the instrumentation and control systems at Veracel and 11 other Brazilian mills. Five mills have OPP systems installed.
Daniel Schuck , Automation Consultant for Sindus Andritz, is one of the main developers of OPP and is the product champion within the company. It took about two years of internal development before the service was ready for use, Schuck says.
“OPP is a combination of special software and our in-house knowledge about instrumentation and control systems, plus our knowledge of our client’s process,” Schuck says. “Our OPP service reduces process variability by analyzing control loops and strategies. Production is stabilized, costs are reduced, and maintenance is improved.”
OPP software collects information from distributed control systems and programmable logic controllers about each control loop in the process. It uses a standard language for automation data interchange (OLE for Process Control).
“A modern mill has thousands of control loops,” Schuck says. “We included sophisticated statistical tools into OPP that give us an overview of all the loops that are not performing well. On its own, the OPP software would be of little value to our clients. It’s when we combine it with our knowledge of the process that we achieve results.”
To understand what OPP does, it first helps to know a little bit about controllers, loops, and tuning. Fortunately, Schuck can explain these terms in such a way that is understandable to non-technical listeners.
“A control loop,” Schuck explains, “consists of the process, a measurement, a controller, and an actuator (valve or other device). The sensor measures a process variable (like temperature, pressure, level, or flow) and sends this measurement to the controller.
“The controller is designed to eliminate the need for continuous operator attention. The thermostat in your house is a common example. You set a comfortable temperature (the setpoint) and the thermostat automatically controls the heating system to hold the temperature at the desired level.”
Very different controller behaviors can be considered “good” depending upon the process. “One process may be best operated with a fast and aggressive control action,” Schuck says, “while another may be better suited for a slow and gentle response.”
In your home, it would not be a life-threatening event if the thermostat overshot or undershot the temperature by a degree or two. But in a mill, there are many areas where the control must be very precise for safety, quality, or environmental reasons. This is where “tuning” comes in.
Schuck explains, “The control response of each controller can be programmed to tell it how far to move the actuator (gain), how fast (time constant), and how much delay to allow between actions (dead time).”
The “traditional” way of tuning loops is by trial and error. An engineer or technician enters a new setpoint and observes the process response to the change. This requires a skilled intuitive understanding, and only experienced people are able to achieve good setpoint response this way.
OPP software automatically identifies process gain and dead time and gives recommended tuning. “Then we have the potential for getting closer to what the process is actually capable of,” Schuck says. “However, we at Sindus Andritz never take a tuning action remotely ourselves. The mill operators must be involved and take the action as they are the owners of the process.”
Ultimately, the best control is based upon knowledge and experience. “Things we take into account include what the process is physically able to achieve, how this fits into the bigger safety and profitability picture, the impact on downstream processes, and what management has planned,” Schuck says. “OPP has tools built-in to help with this where we, together with the mill, rate the importance of each loop in terms of its contribution to safety and economics.”
In Veracel’s case, Schuck used the OPP software to analyze about 150 control loops in the bleach plant in one week.
Fiberline team operators at Veracel were involved every step of the way. “When Sindus Andritz identified the problem loops, we had team meetings with our operators to get their input and to involve them in finding the correct solutions,” Aguilar explains.
“To get optimal control, all the components must work together properly,” Schuck says. “For the problem loops, we had to verify that each component – control device, sensor, actuator, controller, and software – was operating properly and that the design was appropriate. We worked side-by-side with the operators to do this.”
According to Aguilar, “For some loops, it was a simple fix like repairing a sticking valve or replacing a bad transmitter. Others required analyzing the loop’s overall objective. For example, we found that one controller was cycling a valve 100 times per hour. By slowing the loop to reduce noise as well as interaction with other control loops we were able to achieve smoother, more coordinated production.”
With the help of OPP, Schuck and his team from Sindus Andritz identified 12 valves with problems, 32 control loops that required tuning, and eight loops that needed a different control strategy.
Schuck explains, “One problem that Veracel was having was unstable high pressure in the feed to the DD washers, which triggered interlocks that shut down the system. After analyzing the loops with OPP, we felt that the location of the measurement points and the programmed action of the controllers was contributing to the problem. We discussed this with Veracel and with the Andritz equipment experts and reconfigured the control logic. Things improved dramatically.”
According to Aguilar, the next targets for OPP are oxygen delignification and the screening areas. “We hope to make similar improvements in each area as we move down the line,” he says.
There is much to be gained by optimizing control loops. It has been estimated that 80% of process control loops are causing more variability running in automatic mode than in manual. An often quoted study by EnTech showed that some 30% of all loops oscillate because of poor controller tuning.
With a poorly tuned loop, an upset in the process results in waste – causing off-spec production or over-consuming expensive raw materials.
The OPP service also provides data to Veracel about how tuning changes with time. That’s important, Schuck stresses, because “one of the most critical things we do is to stay apprised of the variability in process conditions. This is not a one-shot deal, but an ongoing service we offer. I can have a one-screen view of overall equipment efficiency and, seconds later, drill down to the current tuning parameters of a specific control loop.”
The OPP software also includes online tools for alarms, motors, valves, and analyzer management. An interesting offline tool, according to Schuck, is the ability to simulate electrical maneuvers. “This is a very useful tool for safety, training, and documentation of the mill’s electrical systems.”
As an added bonus, OPP can be accessed remotely. Schuck demonstrates this capability from his desktop in Sindus Andritz’s Porto Alegre office (2500 km from the Veracel mill). A few keystrokes to establish a secure connection and he is viewing real-time status information from the mill’s DCS. Two more mouse clicks and he has a summary of the loops that are in need of attention. One more click and he has the tuning data on a troubled loop. Within seconds, he is on the phone to a technician in the mill’s control room, and the optimization continues…
In June 2007, Andritz acquired a 50% stake in Sindus Human Technology, a company specializing in maintenance services for pulp, paper, and other industries in Brazil. “We see many ways to cooperate and there are many synergies,” says Luis Binotto, President of Sindus Andritz.
An electronics engineer and chemical technician by training, Binotto was working at a corrosion laboratory near his home in Porto Alegre (in the southernmost state of Brazil). It was 1987 when he, four other engineers, and a brother-in-law with an administrative background ventured out on their own and formed Sindus Human Technology.
“Our initial business was the maintenance of complex laboratory equipment such as chromatographs, spectrometers, etc.,” Binotto explains. “Later, we developed microprocessor-based systems for controlling the lab equipment.”
In 1989, Sindus was asked by a customer to develop a gas analyzer for environmental monitoring – the TRS analyzer. Today, it is the only manufacturer of TRS analyzers in Brazil and has around 100 units operating.
Sindus saw an opportunity and moved into maintenance outsourcing of instrumentation and control systems in 1992. Outsourced maintenance, Binotto explains, is different than maintenance contracting or providing technical assistance. “When a customer outsources their maintenance to you, they are relying on you to manage the maintenance, not just provide manpower,” he says. “They ask us to recruit and train the specialists, schedule the work, and perform it. Sometimes they turn their staff over to us and their people become Sindus employees.
“This is one of our key advantages of interest to Andritz,” Binotto says. “We are in the mills all the time doing maintenance and we are always on-call for emergency repairs. Our customers know us and trust us – and we know their needs and priorities very well.”
Sindus Andritz today has about 650 employees. 15 key site managers are located close to customers all around Brazil. The company has 12 major contracts in force and works with all the leading pulp and paper producers. In general, pulp and paper accounts for about 80% of Sindus Andritz’s annual revenue. The rest comes from other process industries such as steel, chemicals, and mining.