Comparision and Development of Innovative Concepts for Grate Control Systems on the Example of Municipal Waste Incineration Plants

Project Number 271

Traditional most incineration plants possess a conventional two point resp. PID-control system. The mode of operation is based on mathematical relationships between in- and output factors. For processes with linear behaviour the results are satisfactorily. But strong troubles, parameters with time dependence und great times of delay can move the working point away from optimal conditions. Especially incineration plants with difficult fuel and complex unsteady processes have a high potential of improvement.

The disadvantages can partially be compensated with soft-computing control concepts like Fuzzy Control, Neural Networks and Model Predictive Control. The main difference is the way of developing the control strategy.

Fuzzy systems construct their basic on the knowledge and the experience of boiler attendants or engineers. For getting the information which are hidden in the heads of the people intensive and structured audits are necessary. In the second step the individual and with mistakes connected knowledge must be organised, be putted in the context and be formulated. Characteristic for fuzzy control is the conversion from normal language in digital and anlog signals.

Neural networks orientate themselves also on human beings; in this case on the function of the brain. Like a child the system learns from the experience in different situations. To get reliable results a qualified preparation of the operating data and a high number of data is necessary.

Goal of the actual project is the testing of every control strategy named above, determine the strengths and weaknesses to formulate a control strategy which combines the strengths. To retrofit is the result with the implementation of a model of grate and combustion chamber, which includes physical and chemical parameters and allows the view in the future. Differences and similarities between different grate systems have to be detected with the help of operating data. At the end possibilities of application and open points have to be shown.

The project is funded by the German Federal Ministry for Economy and made available by Arbeitsgemeinschaft industrieller Forschungsvereinigungen e. V. (AiF). The project starting from September 2005 is running for 24 months at CUTEC Clausthaler Umwelttechnik-Institut GmbH (Dr. Stefan Vodegel). From the technical point of view, the project is accompanied by the VGB Working Panel "Thermal Waste Utilisation".