Sustainable optimization-based process control II
|Coordinator||KUNGLIGA TEKNISKA HÖGSKOLAN - KTH Avdelningen för Reglerteknik|
|Funding from Vinnova||SEK 4 200 000|
|Project duration||July 2016 - June 2018|
|Venture||Strategic innovation programme for process industrial IT and automation – PiiA|
Purpose and goal
The aim of this project was to develop the next generation quality control systems(QCS) for paper machines. This goal was met by the development of an economic model predictive controller (EMPC). To enable direct control of quality variables which cannot be measured on-line, soft-sensors able to estimate off-line lab-measurements from on-line data were developed and tested. To enable sustained use of the controller concept, novel algorithms for detecting parameter errors in the model were developed as well as methods for performing closed-loop parameter estimation.
Expected results and effects
Paper manufacturing is a highly competitive business with limited profit margins. By including economic optimization into the quality control, the economic performance of the manufacturing process may be improved. Due to the low profit margins, the relative efficiency gains obtained in the process becomes amplified in terms of the profit, hence leading to potentially large increases of the net profit. Such a control system should hence be attractive to manufacturers. Work on including the developed EMPC concept into ABB´s next generation QCS is ongoing.
Planned approach and implementation
This project has been a two-year continuation on an initial one year project. As a consequence, the goals an challenges of this project were relatively well understood already during the initial phases of the project. This has resulted in a project that was able to closely follow the initial project plan. However, during the project, it was decided that the paper-machine on which pilot tests were planned were going to be decommissioned. As this was not foreseen, it unfortunately hindered pilot testing of some of the developed concepts.