Using machine learning for cylinder misfire detection in a dynamic firing level modulation controlled internal combustion engine is described. In a classification embodiment, cylinder misfires are differentiated from intentional skips based on a measured exhaust manifold pressure. In a regressive model embodiment, the measured exhaust manifold pressure is compared to a predicted exhaust manifold pressure generated by neural network in response to one or more inputs indicative of the operation of the vehicle. Based on the comparison, a prediction is made if a misfire has occurred or not. In yet other alternative embodiment, angular crank acceleration is used as well for misfire detection.

Methods and systems are provided for user controlled engine revving. In one example, the engine may be revved to a peak speed according to a target revving sequence in response to a user-initiated command. A revving sound may be provided in response to an input device, without demanding manual depression of an accelerator pedal.

A misfire detection device for an internal combustion engine includes a storage device and processing circuitry. The storage device stores mapping data. The mapping data is data specifying a mapping that outputs a misfire variable using a rotation waveform variable as an input. The misfire variable is a variable related to a probability that a misfire has occurred in the internal combustion engine. The rotation waveform variable is a variable based on an instantaneous speed variable corresponding to each of discontinuous rotational angle intervals selected from multiple continuous rotational angle intervals.

The disclosure is directed to an ignition circuit for a combustion engine. The combustion engine has a piston movable between a top dead center and a bottom dead center and, via a connecting rod, drives a crankshaft. Combustion air is apportioned via an intake channel. A control circuit provides an ignition time for the idling situation and an ignition time for the acceleration situation. To adjust an early switch to an ignition time for the acceleration situation, provision is made to monitor the rotational speed of the crankshaft over a crankshaft angle range and to detect the value of a drop in rotational speed. The value of the detected drop in rotational speed is compared to a predetermined value of a drop in rotational speed and, when the predetermined value of the drop in rotational speed is exceeded, a switch is made to the ignition time for the acceleration situation.

A method for managing start up of a four-stroke engine, the method being performed by a controller communicatively connected to the engine. The method includes determining, using a crankshaft sensor, an angular orientation of the crankshaft, the crankshaft being rotated by a starter motor prior to ignition of the engine; determining, using the crankshaft sensor, at least one engine speed variation as the crankshaft rotates through at least one measurement window; and identifying a working cycle phase of the crankshaft including in response to an absolute value of the at least one engine speed variation being above a threshold, determining that the crankshaft is in an ignition revolution of a two revolution working cycle of the engine in the measurement window, subsequent ignition of the engine being based on determination of the angular orientation and the working cycle phase of the crankshaft.

A control device and a control method for variable valve timing mechanism according to the present invention obtains a first measurement of a rotational phase based on a rotational angle of the motor, obtains a second measurement of the rotational phase based on a relative relationship between a rotational angle of the crankshaft and a rotational angle of the camshaft, calibrates the first measurement based on the second measurement, obtains a derivative term proportional to a rate of change in a deviation between the first measurement and a target value, reduces change in derivative term when calibrating the first measurement based on the second measurement, and controls the motor based on a manipulated variable including the derivative term.

An imbalance detection device is provided. An obtainment process includes obtaining a rotation waveform variable based on a detection value of a sensor that detects a rotational behavior of a crankshaft, and an air-fuel ratio detection variable in each of a plurality of first intervals. A calculation process includes calculating an imbalance variable based on an output of a mapping having a value obtained by the obtainment process as an input. The imbalance variable indicates a degree of variations in an air-fuel ratio of the internal combustion engine. The rotation waveform variable indicates a difference between instantaneous speed variables that are variables corresponding to the rotational speed of the crankshaft in each of the second intervals.

A diagnosis apparatus is applied to an in-line three-cylinder internal combustion engine, and includes a diagnosing unit and an angular acceleration derivation unit. The diagnosing unit is configured to execute a diagnosis process for diagnosing whether there is a compression leak in any one of cylinders while the internal combustion engine is in steady operation. The angular acceleration derivation unit is configured to derive an angular acceleration of a crankshaft of the internal combustion engine. The diagnosing unit is configured to, in the diagnosis process, when an amount of change in the angular acceleration per a predetermined period during an expansion stroke of any one of the cylinders is less than or equal to a threshold amount of change, diagnose that there is a compression leak in the any one of the cylinders.

A vehicle includes an exhaust passage including an exhaust cleaner, an engine coupled to the exhaust passage, fuel injection valves, and a control device. The fuel injection valves are respectively provided in cylinders of the engine and inject fuel into the cylinders. The control device performs, after main-injection, post-injection control that causes the fuel injection valves to perform post-injection. The control device determines a fuel injection amount for single-time post-injection performed by each fuel injection valve such that the fuel injection amount for the single-time post-injection is greater than a fuel injection amount when the post-injection is performed in every combustion cycle in all the cylinders. The control device performs the post-injection control such that how many times the post-injection is performed is less than how many times the post-injection is performed in a case in which the post-injection is performed in every combustion cycle in all the cylinders.

Using machine learning for cylinder misfire detection in a dynamic firing level modulation controlled internal combustion engine is described. In a classification embodiment, cylinder misfires are differentiated from intentional skips based on a measured exhaust manifold pressure. In a regressive model embodiment, the measured exhaust manifold pressure is compared to a predicted exhaust manifold pressure generated by neural network in response to one or more inputs indicative of the operation of the vehicle. Based on the comparison, a prediction is made if a misfire has occurred or not. In yet other alternative embodiment, angular crank acceleration is used as well for misfire detection.