Refinery simulation: visualize the impact of a given decision
Simulation may be considered as a method of problem solving by trial and error. It consists in imitating, through calculations, the behavior of a physical system, and allows, in a way, virtual experimentation. The fundamental idea is to make it possible to visualize the impact of a given decision, through a series of calculations, rather than in reality. With this definition, we can imagine simulation as a succession of manual calculations; the role of computers is simply to speed up the process thanks to it computation power.
Although simulation can be performed with very simple tools such as Excel spreadsheets, it becomes clear that beyond a certain level of complexity more elaborate tools become necessary.
There are 2 typical situations which give rise to complex behaviors in real life:
- The case where there are complex relations between the different stages of a system: this is the case when there is a large number of degrees of freedom at all stages of the system, and the possible evolution (and thereby the right decisions) at some point is strongly dependent on decisions made at the other stages
- The case where some level of uncertainty exists in the system; here the complexity comes from the interaction between the deterministic and the random parts of the system
PRINCEPS’ software deals with this 2nd sub-type of complex systems.
Both situations exist in refinery simulation
If we focus on the first kind of complexity, i.e. the complexity of physical systems can also be subdivided into 2 sources:
Complex behavior of each component of the system
Simple behavior of each component but complex interactions due to the numerous interrelations in the overall structure
In refinery simulation, depending on the objective of the simulation, we are confronted with one or the other type of complexity. If we are concerned with accurate refinery process units simulation, then we are confronted with the first case of complex components. If we are rather interested in refinery simulation as a whole, and are more concerned with an overall calculation of the flow of materials, tank levels, crude and product movements then we are more in the presence of the 2nd type of complexity.
PRINCEPS approach to refinery simulation
Refinery simulation is considered as one particular application of the dynamic systems simulation engine developed by PRINCEPS.
PRINCEPS simulation engine is a generic one. It handles both discrete and continuous time simulation, and allows local constraint solving and relations definition and propagation.
Within this context, refinery simulation is made possible through a “refining toolbox”. The generic capabilities implemented in the simulation engine are adapted through the refining toolbox and turn into specialized features such as:
- Automatic adjustment of events duration in order to keep tank levels within the limits
- Maintaining consistency between different events through relation definition and propagation. In the context of refinery simulation, this can be insuring consistency between types of feeds or operating parameters to various processes, to take an example