Abstract:
A classifier capable of predicting if cylinder valves of an engine commanded to activate or deactivate failed to activate or deactivate respectively. In various embodiments, the classifier can be binary or multi-class Logistic Regression, or a Multi-Layer Perceptron (MLP) classifier. The variable displacement engine can operate in cooperation with a variable displacement engine using cylinder deactivation (CDA) or skip fire, including dynamic skip fire and/or multi-level skip fire.
Abstract:
A variety of methods and arrangements for detecting misfire and other engine-related errors are described. In one aspect, a window is assigned to a target firing opportunity for a target working chamber. There is an attempt to fire a target working chamber during the target firing opportunity. A change in an engine parameter (e.g., crankshaft angular acceleration) is measured during the window. A model (e.g., a pressure model) is used to help determine an expected change in the engine parameter during the target firing opportunity. Based on a comparison of the expected change and the measured change in the engine parameter, a determination is made as to whether an engine error (e.g., misfire) has occurred.
Abstract:
A variety of methods and arrangements for detecting misfire and other engine-related errors are described. In one aspect, a window is assigned to a target firing opportunity for a target working chamber. There is an attempt to fire a target working chamber during the target firing opportunity. A change in an engine parameter (e.g., crankshaft angular acceleration) is measured during the window. A model (e.g., a pressure model) is used to help determine an expected change in the engine parameter during the target firing opportunity. Based on a comparison of the expected change and the measured change in the engine parameter, a determination is made as to whether an engine error (e.g., misfire) has occurred.
Abstract:
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.
Abstract:
In one aspect, a skip fire engine controller is described. The skip fire engine controller includes a skip fire module arranged to determine an operational firing fraction and associated cylinder load for delivering a desired engine output. The skip fire engine controller also includes a firing controller arranged to direct firings in a skip fire manner that delivers the selected operational firing fraction. Various methods, modules, lookup tables and arrangements related to the selection of a suitable operational firing fraction are also described.
Abstract:
Methods and systems are described for detecting valve actuation faults in internal combustion engines operating in a skip fire operational mode. In one aspect, a torque model is used to estimate an expected net torque during a selected operating window. The torque model considers an expected torque contribution from each of the cylinders and accounts for the effects of specific skip fire firing decisions that affect the expected torque contribution from each cylinder. A parameter indicative of the actual engine torque is also measured. Valve actuation faults can then be identified based at least in part on a comparison of the measured parameter to an expected parameter value that is based at least in part on the expected net torque. With the described approaches, the occurrence of the valve actuation fault can be made within one engine cycle of the initial occurrence of the fault.
Abstract:
In one aspect, a skip fire engine controller is described. The skip fire engine controller includes a skip fire module arranged to determine an operational firing fraction and associated cylinder load for delivering a desired engine output. The skip fire engine controller also includes a firing controller arranged to direct firings in a skip fire manner that delivers the selected operational firing fraction. Various methods, modules, lookup tables and arrangements related to the selection of a suitable operational firing fraction are also described.
Abstract:
Various methods and arrangements for determining a combustion control parameter for a working chamber in an engine are described. In one aspect, an engine controller includes a firing counter that stores a firing history for the working chamber. A combustion control module is used to determine a combustion control parameter, which is used to help manage combustion in the working chamber. The combustion control parameter is determined based at least in part on the firing history.
Abstract:
Methods and devices are described that utilize skip fire techniques to rapidly meet requests for transitory changes in the output of an engine. Specifically, the fraction or percentage of the working cycles that are fired can be changed during a transitory event so that the engine delivers the desired transitory engine output. Once the transitory event is over, normal engine operation may be restored. The described techniques are useful in a variety of applications that require a relatively quick, but transitory, reduction in engine output to meet vehicle control requirements. One particularly useful application is during transmission shift events. Other representative applications include: loss of traction events, stability control events, wheel hop prevention events, etc.
Abstract:
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.