On-line estimation of inlet and outlet composition in catalytic partial oxidation

• CPOX reformer high fidelity process model was developed.
• Process dynamics was analysed to discover the important measurements.
• Model reduction using subspace identification was used.
• Conducted observability analysis with the objective of estimating gas mole fractions.
• Moving Horizon Estimation with solution times of 0.02 s was achieved.

An estimation strategy is presented for determining inlet and outlet composition of catalytic partial oxidation (CPOX) of methane over rhodium catalyst using simple, fast measurements: temperature, and thermal conductivity. A 1-D high fidelity simulation model for CPOX studied in Ref. [1] for a portable fuel cell application is developed and enhanced for transient experiments. Process dynamics are analysed to demonstrate how solid temperatures along the axes of the reactor reflect the endothermic/exothermic interplay of reactions during a process upset. Model reduction is then used to obtain a low complexity model suitable for use in a moving horizon estimator with update rates faster than 0.02 s. System theoretic observability analysis is then conducted to predict the suitability of different measurement designs and the best locations for temperature measurements for estimating both inlet and outlet gas mole fractions for all species. Finally, a Moving Horizon estimator is implemented and simulation experiments are conducted to verify the accuracy of the estimator.

Figure optionsDownload as PowerPoint slide