Engine Control Using Torque Estimation

On-Board Indicated Pressure and Torque Estimation in Engines With a High Number of Cylinders

Design, Application, Performance and Emissions of Modern Internal Combustion Engine Systems and Components ◽

10.1115/icef2002-537 ◽

2002 ◽

Author(s):

Enrico Corti ◽

Davide Moro

Keyword(s):

Control Strategy ◽

Engine Speed ◽

Fuel Injection ◽

Control Strategies ◽

Engine Control ◽

Cylinder Pressure ◽

Signal Processing Algorithm ◽

Operating Condition ◽

Torque Estimation ◽

Speed Fluctuations

In recent years engine control development focused the attention on torque-based models, that allow improving driveability and implementing traction control strategies. The design of such a torque-based engine control strategy requires the knowledge of the torque produce by the engine, which depends on fuel injection time, spark advance, throttle opening, EGR command, … In the actual engine control strategies this is mainly done by means of static maps stored in the ECU memory. The real engine torque production under every operating condition can be evaluated by means of the in-cylinder pressure estimation, thus allowing a torque based closed loop control strategy. Many approaches are present in the literature showing the possibility of on-board estimating the actual torque produced by the engine not simply by using static maps, but estimating it through other measured signals. Most of the methodologies that do not require a specific sensor placed on the engine are based either on the engine speed fluctuations (measured by a pick-up facing the flywheel teeth) or on the engine block vibrations (measured by the knock sensor), performing better for engines with a low number of cylinders. The paper presents an original methodology based on the instantaneous engine speed fluctuations, that has been usefully applied to engines with higher number of cylinders. The methodology is based on the observation of the speed fluctuations in a crankshaft window inside the expansion stroke and on the hypothesis that there exists a strong correlation between these engine speed fluctuations and pressure inside the selected cylinder. This relationship has been characterized using Frequency Response Functions (FRF) for each steady-state engine operating condition. In the following the FRFs have been used to perform in-cylinder pressure and then indicated torque estimation under every operating condition, and a specific signal processing algorithm has been developed in order to apply the procedure during speed and load engine transients. The experimental tests have been conducted mounting a six-cylinder turbo-charged spark-ignited engine in a test cell. The application on-board a vehicle of the same methodology seems to be feasible due to the quickness of the algorithm employed and the presence on-board of all the sensors required for the implementation.

Download Full-text

Torque estimation of spark ignition engines via cylinder pressure measurement

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering ◽

10.1177/095440700321700906 ◽

2003 ◽

Vol 217 (9) ◽

pp. 809-817 ◽

Cited By ~ 10

Author(s):

Seungbum Park ◽

Myoungho Sunwoo

Keyword(s):

Engine Management System ◽

Engine Control ◽

Spark Ignition Engines ◽

Cylinder Pressure ◽

Load Torque ◽

Engine Torque ◽

Estimation Algorithms ◽

Torque Estimation ◽

Feasible Alternative ◽

Vehicle Dynamics Control

Indicated torque estimation and load torque observation algorithms are presented and appear to be a feasible alternative to the use of the engine torque maps in a modern torque-based engine management system. The proposed method, which uses a cylinder pressure sensor, has advantages of simplicity from the elimination of the requirement for a complex indicated torque model. Moreover, the proposed algorithms are accurate and robust to the variations in the environmental factors that affect the torque production procedure. The indicated torque is estimated from the peak pressure and its location, and the load torque is observed on the basis of the estimated indicated torque. The proposed torque estimation algorithms may provide new ideas for many application areas such as engine diagnostics, torque-based engine control, traction control via engine control, and vehicle dynamics control.

Download Full-text