Model based optimization of MSWC process control
conference paper
Optimization of municipal solid waste combustion (MSWC), processes is an important issue due to the ever-lasting need for emission reduction, more optimal use of raw materials and overall cost reduction. The key of the approach of TNO (Netherlands Organisation for Applied Scientific Research) to this optimization problem is the usage of dynamic models, in particular a first-principles model and empirically obtained models, obtained through system identification. This paper discusses these models, how they are obtained and how they can be used for validation and, in particular, for process control. The application of system identification techniques to the Dutch MSWC plant AZN is discussed in order to analyze their process behaviour and to validate the developed dynamic, first principles, process model. The latter has resulted in a validated simulator of the AZN combustion process. Based upon the simulations and results from the system identification, AZN has carried out a large modification project comprising several major adaptations, including a new control concept. Also discussed is how, based upon system identification experiments and the dynamic model, a new control concept has been developed for the Dutch MSWC plant HVC. Finally, advanced control technologies based upon dynamic models like (non-linear) model predictive control and evolutionary control are discussed briefly.
Topics
TNO Identifier
256594
Publisher
TNO
Source title
13th DVV Kolloquium, Thermische Verfahren der Abfallbehandlung, Weimar, Germany, 23-24 September 2002
Files
To receive the publication files, please send an e-mail request to TNO Repository.