Ongoing Control Performance Improvements

I caught up with Emerson’s James Beall, a principal process control consultant who helps manufacturers optimize their processes. You may recall James from numerous control performance-related posts. If you ever attend an Emerson Exchange conference, James is one of the dynamic presenters you don’t want to miss.

James shared with me a Control Performance Program with which manufacturers were finding success. For the duration of this program, James establishes a schedule with the process manufacturer for him to spend a week at the plant on a reoccurring basis. Typically, this occurs once every six weeks.

James works with the plant engineers and operations staff to measure the process dynamics and variability to identify opportunities for improvement. These improvements can reduce energy consumption, improve throughput, improve product quality–all which positively impact the bottom line. During this period, issues are identified. The ones that can be immediately addressed such as improved PID tuning constants are addressed. More complex control improvements such as advanced process control using the DeltaV model predictive controller (MPC), can be planned and executed over several of the on-site sessions.

Many times issues with control valve performance, measurement device location, or changes in the automation system’s control strategy are required. These often cannot be immediately addressed during that week. The six-week window between James’ visits gives the operations and maintenance teams a chance to schedule and resolve the issues before his next visit. These ongoing scheduled visits serve as a force function to get the issues resolved so that each visit can focus on optimization opportunities and the identification of new issues that can’t be immediately resolved.

Also, these visits serve to check on the optimization work that has already been performed. James relayed a rule of thumb that he and the other process control consultants observe. Optimization project that are not monitored and maintained typically have a six-month half-life of their benefits. That is, they lose half of their economic benefits every six months if left largely untouched. It’s classic entropy at work of plant performance tending toward disorder away from optimized control.

James noted a recent example on a distillation column. One of the MPC constraint variables was riding at its lower limits and preventing the MPC from reducing the steam usage. A control valve was thought to be the issue, but upon its repair, the issue remained and the problem was found to be a manual block and bypass valve that was leaking cooling water. Once repaired, the MPC constraints were adjusted and the steam savings were calculated at $150,000 USD per year. These savings were in addition to the performance improvements and derived benefits of the original application of MPC to the column.

In between visits, the plant engineers share historical data in areas they suspect may be an opportunity for optimization. This helps focus and prioritize the visit to make sure optimization opportunities are converted. James also highlighted the importance of having the improved performance quantified. With this particular process manufacturer, an engineer fully trained in Six Sigma, leads the quantification efforts by extracting historical information from the automation system and comparing before and after performance.

This not only helps prove out the value of the control performance program but also avoids the six-month half-life reduction of this captured value!

GreenPodcast.gif MP3 | iTunes

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Leave a Reply