On-line performance diagnostics for internal combustion engines

An online diagnostics algorithm and measurement system for automotive engines is described. Both front-end and flywheel engine speeds are measured digitally by time interval detection between encoder pulses, using a single-board computer for data collection and processing. Based on a lumped parameter crankshaft model, the torque contribution of each cylinder is computed. Those cylinders contributing less than a predetermined threshold value are identified as faulty. A series of tests with various engines and fault conditions have shown that the diagnostic system provides for reliable online determination of cylinder fault locations, as well as an estimate of the fault severity, which is determined from the magnitude of the performance index. An estimate of the maximum engine torque is derived from snap acceleration tests. The diagnostics system can be applied to all engines of the same type without requiring training data