Run-to-run convergence analysis of model-based policy iteration algorithms for experimental optimization of batch processes

CRISTALDI M., CRISTEA, S., MARTÍNEZ, E.

Ischia, Naples, Italy

Simposio; ESCAPE 2010 (European Symposium on Computer Aided Process Engineering); 2010

The Italian Association of Chemical Engineering

Convergence analysis of iterative identification-optimization schemes is a key issue in modeling for optimization of batch processes. In this work, it is formally shown that for convergence is sufficient to guarantee that parametric uncertainty is increasingly  reduced on a run-to-run basis. Convergence of a policy iteration algorithm to an optimal policy which satisfies the Hamilton-Jacobi-Bellman equation is thus assured as long as parametric uncertainty is iteratively reduced such that the performance prediction mismatch is driven to zero. The integration of global sensivity analysis with confidence  interval boostrapping in the design of a convergent algorithm for model-based policy iteration is proposed. A simple bioprocess is used to exemplify run-to-run improvement. reduced on a run-to-run basis. Convergence of a policy iteration algorithm to an optimal policy which satisfies the Hamilton-Jacobi-Bellman equation is thus assured as long as parametric uncertainty is iteratively reduced such that the performance prediction mismatch is driven to zero. The integration of global sensivity analysis with confidence interval boostrapping in the design of a convergent algorithm for model-based policy iteration is proposed. A simple bioprocess is used to exemplify run-to-run improvement. sufficient to guarantee that parametric uncertainty is increasingly  reduced on a run-to-run basis. Convergence of a policy iteration algorithm to an optimal policy which satisfies the Hamilton-Jacobi-Bellman equation is thus assured as long as parametric uncertainty is iteratively reduced such that the performance prediction mismatch is driven to zero. The integration of global sensivity analysis with confidence interval boostrapping in the design of a convergent algorithm for model-based policy iteration is proposed. A simple bioprocess is used to exemplify run-to-run improvement. sufficient to guarantee that parametric uncertainty is increasingly  reduced on a run-to-run basis. Convergence of a policy iteration algorithm to an optimal policy which satisfies the Hamilton-Jacobi-Bellman equation is thus assured as long as parametric uncertainty is iteratively reduced such that the performance prediction mismatch is driven to zero. The integration of global sensivity analysis with confidence interval boostrapping in the design of a convergent algorithm for model-based policy iteration is proposed. A simple bioprocess is used to exemplify run-to-run improvement.