An engine misfire detection device such that an engine misfire state can be accurately detected is obtained. The engine misfire detection device includes an engine rotational speed change amount detector that detects an engine rotational speed change amount, an engine rotational speed change amount threshold setter that sets a threshold with respect to the rotational speed change amount, and a misfire detector that compares the rotational speed change amount and the threshold, carries out a misfire determination on the engine when the rotational speed change amount exceeds the threshold, and carries out a misfire detection, wherein the misfire detector prohibits the misfire determination when a switching between a differentially locked state and an unlocked state is carried out.

The present disclosure provides a method for learning emergency injection correction of an injector for preventing misfire. A misfire rate in a multi-stage injection mode is monitored, and it is controlled to be forcibly switched to a single injection mode when the misfire rate by the monitoring is equal to or greater than a specific value, such that injector injection correction learning is performed according to a learning entry condition. Accordingly, it is possible to reduce the deviation between cylinders by the sufficient injection amount deviation correction learning, thereby preventing misfire of the injector.

A misfire detection device includes processing circuitry configured to execute a stopping process stopping combustion control of an air-fuel mixture in one or more cylinders and a determination process determining whether a misfire has occurred based on a value of a determination subject rotation fluctuation amount, that is, a rotation fluctuation amount of a determination subject cylinder for misfire. A comparison subject rotation fluctuation amount is a rotation fluctuation amount corresponding to a crank angle separated by a predetermined angular interval from a crank angle corresponding to the determination subject rotation fluctuation amount. The determination process includes a process determining the misfire based on a value of the determination subject rotation fluctuation amount when the predetermined angular interval equals an angular interval between crank angles at which compression top dead center appears in the one or more of the cylinders and the determination subject cylinder during the stopping process.

A misfire detection device and a misfire detection method for an internal combustion engine, and a memory medium are provided. Cylinders adjacent to the deactivated cylinder include a determined cylinder subject to a determination of whether a misfire has occurred and a cylinder different from the determined cylinder. It is determined that a misfire has occurred in the determined cylinder on condition that a divergence degree between a value of a combustion variable of the cylinder different from the determined cylinder and adjacent to the deactivated cylinder and a value of a combustion variable of the determined cylinder is greater than or equal to a specific amount. Combustion control has been executed in the cylinders adjacent to the deactivated cylinder and the cylinder different from the determined cylinder and adjacent to the deactivated cylinder.

An internal combustion engine condition determination apparatus includes a storage device; and an execution device. The storage device stores mapping data that defines a mapping. The execution device is configured to execute an acquisition process of acquiring an internal combustion engine state variable every time a crankshaft of an internal combustion engine rotates by a predetermined angle, and a determination process of determining a condition of the internal combustion engine based on an output obtained through the mapping using the internal combustion engine state variable as an input. The mapping data is trained by machine learning. The execution device is configured to prohibit the determination process when a rotation speed of the crankshaft is equal to or higher than a predetermined threshold.

Time required by a crankshaft to rotate 30° CA from a compression top dead center is defined as time T30. A CPU calculates a rotation fluctuation amount ΔT30 related to a cylinder subject to determination of a misfire by subtracting a value related to a cylinder in which a compression top dead center occurred immediately before the cylinder subject to the determination from a value related to the subject to the determination. The rotation fluctuation amount ΔT30 that corresponds to a cylinder in which a combustion operation is stopped is used as a reference value ΔT30ref. When a combustion operation is performed, it is determined that there is a misfire if the absolute value of the difference between the rotation fluctuation amount ΔT30 and the reference value ΔT30ref is less than or equal to a determination value Δth.

An engine misfire detection device is mounted on a hybrid electric vehicle that includes an internal combustion engine and a generator. The internal combustion engine has a plurality of cylinders and a crankshaft and is dedicated to power generation. The generator is connected to the crankshaft via a torsional damper. The engine misfire detection device includes a generator rotation angle sensor and a processor. The generator rotation angle sensor detects the rotation angle of the generator rotating shaft. The processor is configured to execute a misfire detection process. The misfire detection process includes a first misfire detection process of determining that the internal combustion engine has misfired when an amplitude correlation value that correlates with the magnitude of amplitude of rotation speed of the generator rotating shaft and is detected by the generator rotation angle sensor is greater than a determination threshold value.

A controller for an internal combustion engine is configured to execute a temperature-increasing process, a misfire detecting process that detects a misfire, a determining process, and a decreasing process. The temperature-increasing process includes increasing a temperature of a catalyst through a partial cylinder fuel cut-off process. The determining process includes determining whether a number of misfires detected by the misfire detecting process in a number of times combustion control has been executed in each of cylinders is greater than or equal to a given value. The decreasing process includes setting an amount of temperature increase in the catalyst to be smaller when the number of misfires is greater than or equal to the given value than when the number of misfires is less than the given value.

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.

A vehicle control device includes a storage device that stores mapping data including data that defines mapping that receives input data based on a plurality of detection values which are detection values of an in-vehicle sensor and which are before or after in time series and outputs a predetermined output value and that is learned by machine learning and an execution device that executes an acquisition process of acquiring the input data from the storage device, a calculation process of calculating the predetermined output value with the input data as an input of the mapping, and a transmission process of transmitting the detection values used when the input data is generated and time series data including data based on one or a plurality of detection values which are before or after the detection value used for the input data in time series to an outside of the vehicle.