Search Results

The cycle-to-cycle and cylinder-to-cylinder variations that occur in a spark ignited engine create the opportunity for monitoring combustion in real time to provide useful benefits for engine control. Reduction of variation and operation of the engine at closer-to-optimum conditions is possible if real time feedback of the combustion process is available. An in-cylinder pressure sensor with pressure-based control algorithms is one method of monitoring the combustion process. However, such a solution presents new challenges of an additional cylinder penetration location, sensor packaging and added cost. A substitute for the in-cylinder pressure sensor is a device which measures the flame conductivity, commonly known as an ionization current sensor. It can be integrated with the spark plug in the case of SI engines, or with the glow plug in the case of compression ignition engines.

The intensity of a combustion flame ionization current signal (ionsense) can be used to monitor and control combustion in individual cylinders during a cold engine start. The rapid detection of poor or absence of combustion can be used to determine fuel delivery corrections that may prevent engine stalls. With the ionsense cold start control active, no start failures were recorded even when the initially (prior to ionsense correction) commanded fueling had failed to produce a combustible mixture. This new dimension in fuel control allows for leaner cold start calibrations that would still be robust against the possible use of low volatility gasoline. Consequently, when California Phase 2 fuel is used, cold start hydrocarbon emissions could be lowered without the risk of an engine stall if the appropriate fuel is replaced with a less volatile one.

Certain harmonics of angular crankshaft velocity are indicative of engine imbalance induced by cylinder misfire. Application of the Digital Fourier Transformation (DFT) facilitates the production-feasible calculation of a singular index in the frequency domain indicative either of smooth engine operation or misfire. The phase of that particular index with proper interpretation directly points to a misfiring cylinder. The identification of a misfiring pair, either opposing or a non-opposing in the cylinder bank, requires a bit more sophisticated approach since the phase response of the characteristic index in the frequency domain becomes more complex. The method demonstrated here was successfully applied in real time in four-, six-, and eight-cylinder engines, both SI and Diesel, for the On-Board Diagnostic application with reliability exceeding relevant regulatory requirements.

Combustion quality diagnostic techniques utilizing flame ionization measurement, with the spark plug as a sensor, have been in production for some time. This acquired “Ionsense” signal represents the changes in the electrical conductivity of the flame during each combustion event. The present analog versions of this sensor are used to detect knock and engine misfire, and can be used for cam phasing. However, current methodology has fallen short of unlocking the wealth of combustion thermodynamics information encrypted in the ion sense signal. Digital Signal Processing incorporating Artificial Neural Networks (ANN) is well suited for handling the statistical fluctuations of combustion. However to obtain acceptable accuracy, traditional ANN implementations can require processing resources beyond the capability of current engine controllers.

Onboard diagnostic regulations require performance monitoring of diesel particulate filters used in vehicle aftertreatment systems. Delphi has developed a particulate matter (PM) sensor to perform this function. The objective of this sensor is to monitor the soot (PM) concentration in the exhaust downstream of the diesel particulate filter which provides a means to calculate filter efficiency. The particulate matter sensor monitors the deposition of soot on its internal sensing element by measuring the resistance of the deposit. Correlations are established between the soot resistance and soot mass deposited on the sensing element. Currently, the sensor provides the time interval between sensor regeneration cycles, which, with the knowledge of the exhaust gas flow parameters, is correlated to the average soot concentration.

Signal processing incorporating Artificial Neural Networks (ANN) has been shown to be well suited for modeling engine-related performance indicators [ 1 , 2 , 3 ] that require multi-dimensional parametric calibration space. However, to obtain acceptable accuracy, traditional ANN implementation may require processing resources beyond the capability of current engine controllers. This paper explores the practicality of implementing an ANN-based algorithm performing real-time calculations of the volumetric efficiency (VE) for an engine with variable valve actuation (phasing and lift variation). This alternative approach was considered attractive since the additional degree of freedom introduced by variable lift would be cumbersome to add to the traditional multi-dimensional table-based representation of VE.

Recent advancements in the combustion control of new generation engines can benefit from real time, precise sensing of the cylinder pressure profile to facilitate successful combustion feedback. Currently, even laboratory-grade pressure sensors can deliver pressure traces with insufficient signal-to-noise quality due to electrical or combustion-induced signal interference. Consequently, for example, calculation of compression and expansion polytropic indices may require statistical averaging over several cycles to deliver required information. This lag in the resultant feedback may become a concern when the calculated combustion metric is used for feedback control, especially in the case of transients. The method described in this paper involves a special digital filter offering excellent performance which facilitates reduced-error calculation of individual polytropic indices.