汽油替代物辛烷值预测与组分比例确定模型

基于响应曲面法(response surface methodology, RSM),建立了用于预测甲苯掺比燃料(toluene reference fuel, TRF)替代汽油燃料的辛烷值和确定其组分比例的数学模型,并应用该模型的一阶模型(first order model, FOM)、二阶模型(second order model, SOM)和三阶模型(third order model, TOM)预测TRF辛烷值和确定TRF组分比例.结果表明:在已知组分比例下,用FOM模型和TOM模型在预测对应TRF法辛烷值和马达法辛烷值时,预测结果远优于FOM模型;在已知TRF法辛烷值和马达法辛烷值条件下, TOM模型预测TRF构成比例的相对误差在8%以内,优于SOM模型,表明基于RSM建立的TOM模型能够准确的预测TRF燃料辛烷值和确定TRF燃料的构成比例. 

Prediction Model for Octane Number and Component Proportions of Gasoline Surrogate Fuel

The response surface method (RSM) was used to build models for predicting the octane number and determining the component proportions of a gasoline surrogate fuel, which is synthesized using toluene, is-octane and n-heptane and referred to as toluene reference fuel (TRF). The models, including the first-order model (FOM), the second-order model (SOM) and the third-order model (TOM), are compared in predicting the octane number of gasoline surrogate fuel with predefined component proportions of TRF. As a result, both the SOM and TOM show much better performances than the FOM in predicting the research and motor octane number, with predefined component proportions of TRF. Furthermore, the TOM shows greater accuracy than the SOM in determining the component proportions of TRF with predefined octane number, and the relative error is less than 8%. Therefore, the TOM based on RSM can accurately predict the octane number and determine the component proportions of TRF.