Application of Action Dependent Heuristic Dynamic Programming to Control an Industrial Waste Incineration Plant
Stephan Völker, F. Wintrich, Klaus Debes · 2004
In this paper, we describe our application of a neurocontrol ler based on Action Dependent Heuristic Dynamic Programming (ADHDP) to optimize the comb ustion-process for an industrial hazardous waste incineration plant. This ADHDP -controller originally was designed for online learning. That implies, that this contr oller starts with a randomly initialized policy and improves its performance while inte racting with the process. This learning scheme could not be used in our case, since the plant operators would not allow long training periods with inevitable poor performance. Wedescribe, how this problem can be solved by a modified training procedure for the action n etwork. Finally, we present first and promising results for the optimization of action ne t output in order to improve the waste incineration process with respect to the targets d by the plant operator.