"It is easy to understand that with such a large number of different control loops, it is feasible that they are influencing one another," explains Bernd Schuhmann of ABB’s Service sales department. Moreover, suboptimal control parameters may cause the process to fluctuate, which in turn has negative consequences for throughput and yield. Optimizing control loops eliminates weak points and bottlenecks, thereby extending the durability of components. Minor changes to the process can help operators run their plants more closely to their specification limits. This boosts output and energy efficiency without reducing product quality.
Following the first performance analysis, ABB’s experts were able to propose possible courses of action. Changes that were easy to implement included increasing the awareness of plant operators for loops that work less efficiently than normal, and inspecting individual final control elements. In addition, some controllers needed to be changed into cascaded loop control structures to achieve better control performance.
Before implementing more expensive enhancements, the costs and potential improvements should always be weighed. To do so, it is a good idea to calculate the energy efficiency and conduct quality checks by adding customer KPIs to the ServicePort database.
When working with plants like Wacker’s that have developed over years, it is not uncommon to reveal issues that remained undetected for years. Once people get used to these issues, they consider them natural and start to ignore them. When ServicePort first detected signal spikes without an obvious root cause, a closer inspection of the underlying Freelance project was performed. The findings included signals that are no longer valid but are still being used inadvertently. The analysis also revealed missing setpoint tracking values, or prioritizations of individual processing steps (tasking), both jeopardizing proper execution under certain conditions. In addition, redundant parts from the control system, such as obsolete program modules, were identified. These elements not only burden the control system unnecessarily, but may also cause unwanted effects.
These findings led to suggestions on how to improve the Freelance project, and new ways of using the process control system. A second analysis report was compiled for Wacker’s ketene plant at the end of 2014, concluding the data collection period for the time being. Based on this report, further measures for improvement had been discussed.