5 specifications), and unconverted cooking oil (UCO). The four products that are the main focus of this study are LPG, gasoline, kerosene, and diesel, as they make up more than 95% of the total production. Figure 1 and other figures pre - sented in this study show a comparison between the model predictions and actual values of resulting product yields. The average absolute deviations (AADs) are calculated using the following equation:
H make up Nm /Sm of fresh HC feed
480 490
470
460
450
440
n
1 n ∑ (|Y pred. - Y act. |) –
AADs =
430
i=1
420
for the 16 runs expressed in the following graphs for prod - uct property value.⁸ These AAD values help in evaluating the degree of variation between actual and simulation cases. This represents the prediction of the model on each product yield rather than the overall yield, which is the revenue key of the refinery. In contrast, absolute deviation shows how the model affects the estimation of the refinery profit by consid - ering deviations in the same scale towards overall production. Figure 4 compares the predicted gasoline yield (dash lines) and actual gasoline yield (solid lines) from studied test runs. It is evident that the predicted values are very close to the actual values in the range of 380-400°C. This is the recom - mended operating window of the commercial catalyst used in this study, as stated by the catalyst manufacturer and pro - cess unit licensor. Table 1 shows the calculated overall AADs for each studied product yield and stream property value. Market and economic analysis Crude oil prices are the most determining factor for fuel costs. Available crude oil price trends show gradual ascending logic behaviour, with two sharp declines at the end of 2018 and the beginning of 2020. The first decline was caused by polit - ical negotiations about production quantities ruled by the Organization of the Petroleum Exporting Countries (OPEC), especially the Kingdom of Saudi Arabia and its share in the oil market competing with Russia and Iran. The second decline was caused by the COVID-19 pan - demic. Due to the pandemic, there were restrictions on move - ments, especially across countries, leading to a decrease in fuel demand and hence a reduction in crude oil prices. In response, the oil sector adjusted by reducing fuel produc - tion to achieve price balance. There was also a decrease in movement restrictions, the production of new vaccines, and an increase in the number of vaccinated people to achieve herd immunity or community immunity. These rapid changes in crude oil prices were interesting to study. In this section, a brief economic analysis studies how the business model of the hydrocracking unit changed over the previous year.⁷
410 400
370
380
390
400
410
420
430
440
450
Temp. (˚C)
Mix 2 Pred. (VGO 80% + WLO 20%) Mix 4 Pred. (VGO 70% + WLO 20% + WCO 10%)
Mix 1 Pred. (VGO 80% + WCO 20%) Mix 3 Pred. (VGO 80% + WLO 10% + WCO 10%)
In contrast, looking at the long-term behaviour of the oil market, supply is ensured since new deposits are continu - ously discovered. However, despite the steady supply, the demand for oil is not expected to increase as environmental restrictions become more stringent in developed countries. For example, the International Maritime Organization (IMO) lowered the limit for sulphur content in marine fuel from 3.5% to 0.5% in 2020. When reviewing the prices for VGO, WCO, and WLO, it has been noticed that there is a narrow margin between the selling price of ultra-low-sulphur diesel, jet fuel, and gasoline and feedstock price. This indicates that the fuel market for fossil fresh feed and waste recycle feed is highly compet - itive. As a result, selecting the capacity of the processing plant needs to be done with great care to ensure profitability. This study has already selected the process capacity, as the research work depends on the commercial catalyst used in the existing plant, while the operating conditions of this plant are used in building the simulation model. It is worth mentioning that research work has confirmed the availability of WCO and WLO quantities in the local mar - ket. This ensures a stable supply chain and stable production of the hydrocracking unit under the studied feed mixtures and pre-selected unit processing capacity. Based on this data, the capacity of the hydrocracking unit Figure 2 Predicted make-up of hydrogen from hydrocracking simulation model at different reaction temperatures and feed mixtures
AAD values at different temperatures for each studied product yield and product property value
AAD @380 °C
AAD @400 °C
AAD @420 °C
AAD @440°C
AAD 0.32 0.98 0.71 0.07
Gasoline yield Kerosene yield
0.43 0.99 0.79 0.06
0.18 0.98 0.41 0.03
0.28 0.98 0.68 0.06
0.38 0.98 0.94 0.12
Diesel yield
(Kerosene + diesel) yield
Table 1
77
PTQ Q1 2024
www.digitalrefining.com
Powered by FlippingBook