We propose an approach for modeling of multi-phase reactor systems under the hypothesis of local equilibrium. The main idea is to write the mass balances based on elements (atoms) or functional groups that are not produced or destroyed with chemical reaction. Thus we represent our state space in terms of elemental mole concentrations instead of compounds. This reduces the size of state space as number of compounds is much higher than the elements involved and we are able to explore the idea of a minimal state space representation for a thermodynamic process. The idea of a minimal state plays an important role in control theory as it is closely related to the concepts of controllability and observability.

The modeling approach is to represent the process as a combination of various sub-systems called modules. Each module represents a physical phenomena or entity like transport, thermodynamics or phase. These modules interact with each other in terms of flow of mass and/or information to simulate the whole process. This approach is especially useful for large scale and complex dynamic systems such as multi-phase reactors in metallurgical industry. As in these systems, the chemistry involved is highly complex and it is desirable to have a separate module, designed specifically to handle the thermodynamics.

The main focus of the paper is to show how stability theory and thermodynamics can be used to design algorithms and methods for fast and stable process simulation of large scale and complex dynamic systems. We are in particular interested in methods that can be used to design algorithms that can be implemented in distributed web-based simulation environments.

The approach is applied to two industrial systems: Vapor Recovery Reactor (VRR) in Carbothermic aluminum production process and the gasification reactor in Integrated Gasification Combined Cycle (IGCC). Both systems are in design phase and are of considerable interest in the respective industries. We present sensitivity analysis results which can be used for studying the interaction between design and process control at an early stage in the design process.