Modern engine control systems need to know the exact air mass flow drawn by engine. For cost reasons, engines of medium and small displacement are not equipped with an air flow meter and so the air mass flow must be estimated by means of a specific algorithm using other sensors’ data based on a simplification of manifold filling dynamics model [1]. In the last years this algorithm has been modified in order to consider the dependency of air mass flow also on the atmospheric pressure. The actual implementation of this algorithm anyhow does not ever guarantee a correct estimation in altitude.
In this paper an algorithm that uses a novel, simple and effective dependency of the air mass flow on the atmospheric pressure will be presented. As a first step, a new data acquisition in altitude has been performed, in order to better understand how the atmospheric pressure variation influences engine functionality and, in particular, the intake efficiency. Intake efficiency has to be intended as the ratio between the current air mass flow and the air mass flow in the reference condition decided in the calibration phase: for this application the reference atmospheric pressure is 1010 mbar. The definition of the intake efficiency is the base of the entire estimation procedure. Using new data, the real trend of this intake efficiency related to the atmospheric pressure changes has been observed. In the same way the error of the actual model has been evaluated and quantified. Furthermore by means of this set of data, the confirmation that the conventional intake efficiency is invariant by atmospheric pressure if a particular pressure is considered, has been proved. The use of a new variable depending on the measured manifold pressure and the atmospheric one is the basic idea for the new algorithm design.
The performance of the new algorithm has been tested in different ambient conditions, and the results are better than the previous implementation's ones. The engine control systems of future Fiat cars will include this version of algorithm.