Statistical models of petrol engines vehicles dynamics

CAR2017 International Congress of Automotive and Transport Engineering - Mobility Engineering and Environment

工业技术;运输工程

Ilie, C.O.^1 ; Marinescu, M.^1 ; Alexa, O.^1 ; Vilu, R.^1 ; Grosu, D.^1

Dept of Military Automotive Engineering and Transportation, Military Technical Academy, Bd. G Cosbuc nr. 39-49, sect 5, Bucharest, Romania^1

关键词: Autoregressive moving average;    Constant coefficients;    Exogenous input;    Gasoline engines;    Intake manifold;    On-board sensors;    System identification methods;    Testing programs;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/252/1/012091/pdf
DOI  :  10.1088/1757-899X/252/1/012091

学科分类：工业工程学

来源: IOP

【 摘 要 】

This paper focuses on studying statistical models of vehicles dynamics. It was design and perform a one year testing program. There were used many same type cars with gasoline engines and different mileage. Experimental data were collected of onboard sensors and those on the engine test stand. A database containing data of 64thtests was created. Several mathematical modelling were developed using database and the system identification method. Each modelling is a SISO or a MISO linear predictive ARMAX (AutoRegressive-Moving-Average with eXogenous inputs) model. It represents a differential equation with constant coefficients. It were made 64thequations for each dependency like engine torque as output and engine's load and intake manifold pressure, as inputs. There were obtained strings with 64 values for each type of model. The final models were obtained using average values of the coefficients. The accuracy of models was assessed.