Grass–root petrochemical plant uses simulation-based engineering analysis approach to test safety control schemes, Reduce flare load by 30%
After months of deliberations, the client asked its detailed engineering consultant to turn to dynamic simulation to help them reach a conclusion on the worst case flare relief load thereby improving safety level and saving capital costs. The stakes were high and dynamic simulation was the only recourse leG for the engineering team.
The modeling team studied the problem and available data. Keeping the time constraint in mind, model scope boundary was drawn, along with a list of assumptions – to keep data requirement from the external sources as low as possible but yet to keep the rigor sufficiently high to draw meaningful conclusions from the analysis.
Modeling the unit
The starting point was to build a sound steady-state model against the licensor’s design data. A few qualitative runs were taken to simulate situations such as reflux pump failure to assess the directional impact.
“….We felt really comfortable after we saw scenarios being configured by Equinox, on-the-fly, predicting trends matching expected behavior from our own plant experience….”
Process Manager, E&C
Next, a dynamic model was constructed by specifying additional engineering details, including pressure-flow relationships and equipment dimensions. All basic controllers were also built, along with safety and ESD interlocks.
Various emergency scenarios were configured, as a sequence of events, to simulate the pressure built-up, relief valve pop-up to calculate the vapor load through the relief valve. Some of the scenarios studied are listed below:
- Reflux pump failure case
- Raffinate column reboiler failure
- Extract and Finishing column reboiler failure
The analysis confirmed that with SIL-III implementation, the worst-case relief load was ~30% lower than the original value as suggested by the licensor!
Challenge:
Design the flare network for the load suggested by the Licensor or use SIL-III schemes to reduce the worst caseload.
Solution:
Apply a model-based engineering analysis approach on the Xylene section that contributes maximum load to the flare header.
Learning:
A high fidelity engineering model based on dynamic simulation was needed to quantify the relief loads under various scenarios.
Benefits:
- Licensor’s flare load was validated
- Configured and tested SIL-III control schemes
- Various emergency scenarios of partial power failures were simulated
- Worst case relief load was ~30% lower than Licensor’s worst case, with SIL-III implementation
