PTQ Q3 2022 Issue

Daily Recommendations (-24H)

Recommended Process Conditions (-24H)

Expected Performance Improvement (-24H)

WAIT Change [˚C]

H2/HC Ratio Change [mol fr]

Catalyst Circulation Rate Change [...

Aromatic Yeild Delta [wt%]

24H Aromatic Production Delta (wt. basis) [ton]

Aromatic Yield Delta (wt. basis)

Aromatic Yield Delta (wt. basis)

wt%

kg/h

wt%

kg/h

wt%

kg/h

wt%

wt%

kg/h

wt%

kg/h

wt%

avg current

avg current

Aromatic Production Delta (wt%)

Aromatic Yield Delta (wt%)

Optimised Process Parameters Over Selected Time Period (indicated on the top right)

Aromatic Yield : Optimised vs. Operation

Delta WAIT vs. Delta Aromatic Yield

wt%

˚C

wt%

wt%

˚C

wt%

wt%

˚C

wt%

wt%

˚C

wt%

wt%

˚C

wt%

wt%

˚C

wt%

wt%

˚C

wt%

avg

current

avg

current

ExReactor Yield: Aromatics (wt%) (calc.) Optomised Aromatic Yield

Delta WAIT Delta Aromatic Yield [Optimised - Measured] (right-y)

Figure 3 Optimiser targeting maximum aromatics production

process data – the compositions required for near real-time modelling. The following cases are from aromatic complexes currently in operation. For example, in the naphtha splitter, LN detailed carbon breakdown can be continuously determined to estimate the total naphtha feed TBP curve based on carbon PONA composition. The splitter LN/HN operating cut point can be calculated, and a set of optimal conditions is proposed in real-time to operators in order to adjust the LN swing cut flow rates that can be directed to HN. Aromatics are increased by maximising benzene precur- sors in the CCR unit inlet while checking continuously for potential bottlenecks in downstream units such as the aro- matic extraction unit. By applying this strategy in a 1 Mtpy paraxylene production complex, the achievable margin gain is estimated to be ~$4M/year, based on the incremental value of LN conversion to benzene vs LN conversion to liq- uefied petroleum gas (LPG). Figure 2 shows the real-time status of an operating CCR. Actual figures have been masked for proprietary reasons, but the process performance overview includes unit capac- ity usage, reactor temperature and pressure, hydrogen production, aromatics yield, FG/LPG ratio, material balance calculation, and coke on spent catalyst, among many other units of measure. Alert functions attract the viewer’s attention to operation

within design specifications (in green), outside design speci - fications but within operating constraints (in orange), or out - side operating constraints (in red). Each parameter can be viewed as a function of time – graph mode – or in an ‘odome- ter view’ mode. A separate screen displays the impact of the CCR operating conditions on the reformate splitter opera- tion. A ‘what if’ tool calculates performance prediction as a function of a new set of operating parameters, allowing the operator to evaluate the impact of changes in feed composi- tion, temperature, pressure and so on the overall yields, cata- lyst consumption, and heater duty and performance. Reducing plant production costs and energy consumption is achieved by implementing monitoring for equipment such as heaters or feed effluent exchangers. For instance, alerts in case of consumption anomalies will be triggered on the dif- ferent dashboards, and calculated key performance indica- tors (KPIs) are pinpointing where it is possible to improve the efficiency of different assets while checking their integrity. Thanks to this permanent monitoring, traditional preven- tive maintenance can be reduced considerably and correc- tive maintenance almost eliminated, thereby significantly dropping associated costs. For example, tracking precisely excess air at the arch provides insights for heater efficiency optimisation; applied to CCR heaters, this enables an inlet process flow/total fuel gas flow ratio increase by a factor of +3%, decreasing CO 2 emissions accordingly.

17

PTQ Q3 2022

www.digitalrefining.com

Powered by