Numerical Estimation of Wiebe Function Parameters Using Artificial Neural Networks in SI Engine Universitat Politècnica de València

Format:

Book

Conference/Event

Author/Creator:

Torregrosa, Antonio J., author.

Contributor:

Aceros, Sebastian

Broatch, Alberto

Olmeda, Pablo

Conference Name:

SAE WCX Digital Summit (2021-04-13 : Live Online, Pennsylvania, United States)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2021

Summary:

In modeling an Internal Combustion Engine, the combustion sub-model plays a critical role in the overall simulation of the engine as it provides the Mass Fraction Burned (MFB). Analytically, the Heat Release Rate (HRR) can be obtained using the Wiebe function, which is nothing more than a mathematical formulation of the MFB. The mentioned function depends on the following four parameters: efficiency parameter, shape factor, crankshaft angle, and duration of the combustion. In this way, the Wiebe function can be adjusted to experimentally measured values of the mass fraction burned at various operating points using a least-squares regression, and thus obtaining specific values for the unknown parameters. Nevertheless, the main drawback of this approach is the requirement of testing the engine at a given engine load/speed condition. Furthermore, the main objective of this study is to propose a predictive model of the Wiebe parameters for any operating point of the tested SI engine. For this purpose, an Artificial Neural Network (ANN) is developed from the experimental data. A criterion was defined to choose the best-trained network. Finally, the Wiebe parameters are estimated with the neural networks for different operating conditions. Moreover, the mass fractions burned generated from the Wiebe functions are compared with the respective experimental values from several operating points measured in the engine test bench. Small differences were found between the estimated and experimental mass fractions burned. Therefore, the effectiveness of the developed ANN model as a prediction tool for the engine MFB is verified

Notes:

Vendor supplied data

Publisher Number:

2021-01-0379

Access Restriction:

Restricted for use by site license