PTQ Q3 2025 Issue

600,000

500,000

400,000

300,000

200,000

100,000

Time (min)

Pressure (barg)

Sump level (%)

Relief load (kg/hr)

Reboiler duty (MW)

Figure 2 Depropaniser loss of condenser dynamic relief results

system response aggravated the relief load, sensitivities were performed on controller input parameters as well as the initial liquid level to ensure the worst-case relief load was reported. In summary, the dynamic model uncovered a higher relief load compared to the UBH approach due to its ability to account for interactions between the dynamic behaviour of the process and the control system response. While the UBH calculation could incorporate this effect by conserv- atively assuming a maximum reboiler duty, the transient surge in liquid level was not an expected consequence of

the initiating event. Thus, its impact was not accounted for in the initial UBH calculation. Case study 2: alky fractionator total power failure The initial relief load analysis for an alkylation unit (alky) frac- tionator column was performed using the UBH approach (see Figure 3 ). The governing relief scenario for the alky fractionator was found to be partial power failure (PPF), but the total power failure (TPF) scenario was governing the overall flare system design due to simultaneous relief loads from other sources. The basis for the UBH calculation of the TPF relief load can be summarised as follows:  Column feed and product flows are set to zero due to the loss of electric feed pumps. v Overhead condenser duty is set to zero due to the loss of electric cooling water pumps. w Upper reboiler (LP steam) continues to operate during relief, with credit taken for reduced LMTD as column tem- perature rises. x Lower reboiler (fired heater) stops due to loss of electric reboiler pumps, with continued residual heat input assumed at 25% of normal duty (typical). The relief load by UBH was calculated to be 440,000 kg/ hr for a 25,000 BPD alkylation unit. As part of a flare load mitigation study, a dynamic sim - ulation was performed for the TPF scenario. The dynamic relief load was employed to estimate the site-wide flare load for this scenario and evaluate potential mitigations. The basis for the dynamic simulation can be summarised as follows: u Column feed stops immediately due to the loss of elec- tric feed pumps. Liquid products driven by electric pumps also stop immediately. v Cooling water to overhead condenser stops immediately due to the loss of electric cooling water pumps. w Upper reboiler (LP steam) continues to operate during relief, with credit taken for reduced LMTD as column tem- perature rises. x Lower reboiler (fired heater) firing stops due to the loss of electric reboiler pumps, with continued residual heat

Relief

Cooling water

Overhead condenser

FIC

Overhead drum

Overhead liquid to depropaniser

Alky fractionator

Overhead pumps

Alky reactor euent feed

Mixed butane side draw

FIC

LP steam

Condensate

Upper reboiler

Alkylate product

Reboiler pumps

Lower reboiler

Figure 3 Alky fractionator system

PTQ Q3 2025