Looping the Loop

Digitalisation and automation are great for control loop optimisation - but we still need humans, says Kevin Starr

The world of control loop optimisation has changed beyond recognition. What once involved binders of numbers, pneumatic actuation, chart recorders and paper tapes now involves computers, technology and what can feel like an almost-infinite amount of data. What used to take up the space on a wall is now not even a pixel on a computer screen.

Name almost any industrial process and it will involve PID control loops, from paper mills to petrochemicals. According to a McKinsey & Company report published in 2018, the global petrochemical industry experienced more than 15 years of strong volume growth, along with a 4% compound annual growth rate since 2005 in terms of value. That level of growth has brought with it the need for the industry to be more efficient than ever. That efficiency is driven by improving processes.

Every petrochemical plant relies on process control systems for its operation - the majority of which are based on proportional-integral (PI) and proportional-integral-derivative (PID) feedback control technologies. Once upon a time, these loops may have been relatively easy to tune to ensure optimisation, but the vast numbers required in petrochemicals plants makes control loop optimisation in this context a Herculean task, even for the experts. For a customer with several thousand I/O points, the effort to optimise control components is a full-time job. Add to this the nonlinear behaviour of some petrochemical plants and the potentially catastrophic effects if something does go wrong, and control loop optimisation becomes more important than ever.

The benefits of big data

For any industry in 2020, automation has become key. Processing the huge amounts of data associated with modern industrial operations into an amount that can be managed by onsite staff not only allows them to cope with the seemingly impossible task, but effectively allows them to expand their sphere of influence. As we know, the paper industry, while not immediately considered by some as being within the chemical industry, is very much a chemicals business – 90% chemistry, 10% paper drying. With this in mind, our Ability Performance Optimisation for Quality Control System (QCS), a solution widely implemented within the paper industry, identifies, classifies and helps prioritise opportunities to improve product quality, measurement and control health, and control utilisation in a plant. Performance optimisation for QCS service uses non-stop data collection and analysis to transform raw data into actionable information, quickly pinpointing issues to ensure optimal performance of equipment and process.

For any industry in 2020, automation has become key. Processing the huge amounts of data associated with modern industrial operations into an amount that can be managed by onsite staff not only allows them to cope with the seemingly impossible task, but effectively allows them to expand their sphere of influence

Automating production requires processing large amounts of data. As complexity increases, so does the volume of data. We help paper producers make productive use of this data to identify and address production, quality and cost issues that can inhibit peak performance, in turn improving return on capital.

For one Southeast Asian paper mill with 13 paper machines, the use of such data helps mill managers constantly measure, monitor and amend the performance of each paper machine to reduce variations and ensure the exacting standards of quality and thickness. This helps to keep productivity high, maintains predictable service costs and improve process performance.

In an environment like this, data is collected from paper machines (transition times, sheet breaks, weight, moisture, speed and additional DCS and QCS control loop parameters) and used to calculate key performance indicators (KPIs) that measure times and conditions. These KPIs can then be addressed to optimise transition times, resulting in higher production.

The Southeast Asian paper mill, known for using leading-edge technology to produce a wide range of high-quality pulp and paper products, now uses a wide variety of our remote-enabled, software-assisted services, which are all managed centrally in so-called collaborative operations centres, with teams based in Singapore.

This helps to maximise business value by turning data insights into direct action: for example, at the mill, reduced cycle times on product grade changes are resulting in higher production. It is also stabilising moisture as well as additive levels using multivariable predictive controls. Specific improvements include increased production due to higher equipment availability, fewer sheet breaks, lower chemical costs, and reduced paper quality variation.

As a result, the mill has been able to reduce the time to complete variation testing by 70% and reduce overall loop tuning effort by 75%.

Standards or no standards?