Predict octane number for gasoline blends

Journal Article · Sat May 01 04:00:00 EDT 1993 · Hydrocarbon Processing; (United States)

OSTI ID:6425626

Zahed, A H; Mullah, S A; Bashir, M D ^[1]

---

1. King Abdulaziz Univ., Jeddah (Saudi Arabia)

A model with five independent variables is used to predict the octane number of gasoline blends with more accuracy than any previous model. Often, it is useful to know the resulting octane number before the gasoline is blended. Clearly, such a model is useful because good predictive models have been few and far between. With high-powered and faster personal computers, regressional analyses are quite easy to perform with many more independent variables. The objective here was to develop an empirical equation using the regressional analyses are quite easy to perform with many more independent variables. The objective here was to develop an empirical equation using the regression analysis technique to predict the octane rating of 16 blends of motor gasoline. Predicted results for the 16 blends of gasolines were compared with experimental results obtained on CFR engines. Predicted results from the proposed empirical model were in agreement with the experimental data with an average deviational error of 0.54%.

• OSTI ID: 6425626
• Journal Information: Hydrocarbon Processing; (United States), Journal Name: Hydrocarbon Processing; (United States) Vol. 72:5; ISSN HYPRAX; ISSN 0018-8190
• Country of Publication: United States
• Language: English

Related Subjects

02 PETROLEUM
022000 -- Petroleum-- Transport
Handling
& Storage
023000* -- Petroleum-- Properties & Composition
CHEMICAL PROPERTIES
CONTROL
ENGINES
FORECASTING
FUELS
GASOLINE
KNOCK CONTROL
LIQUID FUELS
MATHEMATICAL MODELS
MATHEMATICS
MIXING
PETROLEUM PRODUCTS
PHYSICAL PROPERTIES
REGRESSION ANALYSIS
SPECIFICATIONS
STATISTICS