Ford Motor Company Limited

Ford Global Technolgoies, LLC

Ford Global Tech.inc

Kenneth A Marko

John V James

Bruce David Bryant

James M Dosdall

Lee Albert Feldkamp

Gintaras Vincent Puskorius

Timothy Mark Feldkamp

Vasanth Krishnaswami

Richard Joseph Hampo

Doug Thornburg

Raoul Tawel

Nathan R Soderborg

Nazeeh Aranki

Jimi Sauw-Yoeng Tjong

Shane Rachedi

Danll Valentinovich Prokhorov

George Halow

David Mansfield Mathias

A diagnostic/prognostic system monitors performance of a vehicle or other apparatus wherein the vehicle has a plurality of operational components. Each operational component has a predetermined nominal operating state and generates respective electrical signals pursuant to its operation. A data collection memory in the vehicle stores samples of the electrical signals in a rolling buffer. An analyzer in the vehicle is responsive to the electrical signals for detecting a trigger event indicative of at least a potential variance of an operational component from its nominal operating state. A computation center located remotely from the vehicle has a database storing representations of electrical signals for classifying nominal and irregular operating states of the operational components. A transmitter is activated by the trigger event to transmit at least some of the stored samples in the rolling buffer at the time of the trigger event to the computation center. The computation center receives the transmitted samples and classifies them according to the nominal or irregular operating states.

Remote diagnostics and prognostics methods for complex systems

An engine diagnostic system includes a bit-serial based recurrent neuroprocessor for processing data from an internal combustion engine in order to diagnose misfires in real-time and reduces the number of neurons required to perform the task by time multiplexing groups of neurons from a candidate pool of neurons to achieve the successive hidden layers of the recurrent network topology.

Automotive engine misfire detection system including a bit-serial based recurrent neuroprocessor

An engine misfire detection system (,) for detecting engine misfire. System (,) includes a conventional controller (,) having a memory unit (,) and a plurality of sensors (,). Controller (,) includes a plurality of neural networks, which are trained by system (,), and which determine whether a firing event is a misfire based upon events occurring before the firing event and events occurring after the firing event. The neural networks are adaptively trained to compensate for the effects of engine variability and aging.

Method for adaptive detection of engine misfire

A diagnostic envelope is used to analyze measurements from an internal combustion engine. Measurements may be from a partially assembled engine during a cold test to detect subtle mechanical anomalies, or from a fully assembled engine during a cold test or a hot test (with active fuel and spark). The measurements represent vibrational amplitudes corresponding to various crankshaft angular positions of the internal combustion engine. The first step in generating the diagnostic envelope includes filtering the measurements to enhance local variations. Filtered measurements representing a number of engine cycles are then combined to reduce random cycle-to-cycle variations. The diagnostic envelope is then generated by subtracting corresponding measurements obtained from engines with known operating characteristics. The diagnostic envelope is used in analyzing and/or classifying the engine. A fuzzy logic classifier is used to detect subtle mechanical anomalies while principal component analysis is utilized in analyzing engine performance, detecting less subtle mechanical defects, and in designing and validating engines.

Apparatus and method for processing engine measurements

A predictive model is derived for subsequent use in a misfire detector in an internal combustion engine. Data collected using a test engine is curve-fitted using NARMAX techniques to obtain a computational formula for the predictive model which relates deviant acceleration at the accessible crankshaft section to deviant acceleration at the inaccessible crankshaft section. Thus, the model can be used during actual misfire detection to reduce the effects of the torsional oscillations.

Process for deriving predictive model of crankshaft rotation of a

A method of automating the calibration of lookup tables containing correction values to be used in an on-board vehicle system is disclosed. The method includes training a neural network to model engine behavior by outputting cylinder specific crankshaft acceleration correction values in response to any engine speed and load input conditions. The correction values generated are stored in a memory device. The training takes place off-board the vehicle, using a data set previously obtained from operating a representative engine under normal operating conditions.

Method of generation correction tables for misfire detection using

Irregularities in crankshaft velocity introduced when measuring crankshaft rotation at a section of a crankshaft in an internal combustion engine that is less damped to torsional oscillations than is another more accessible crankshaft section are corrected using a NARMAX-generated model to predict rotation measurements that would have been obtained at the inaccessible section based on data actually collected at the accessible crankshaft section. Thus, the effects of torsional oscillations in the crankshaft are substantially filtered away, resulting in crankshaft acceleration values that form the basis of a misfire detector having nearly maximum signal-to-noise performance.

Misfire detector with torsional oscillation filtering

A method for identifying engine combustion failure of an internal combustion engine having a plurality of cylinders, a crankshaft and a crankshaft position sensor includes the steps of operating the internal combustion engine to rotate the crankshaft, measuring rotational quantities of the crankshaft corresponding to events created by each of the plurality of cylinders during operation of the internal combustion engine, correcting the rotational quantities measured to remove periodic position irregularities to generate a corrected temporal signal, generating an acceleration signal of the crankshaft using the corrected temporal signals, and identifying combustion failures as a function of the acceleration signal. A time-lagged recurrent neural network utilizes the acceleration signal, along with other engine parameters to identify the cylinder-specific misfire events.

Method for identifying misfire events of an internal combustion engine

A misfire detection assembly, used in conjunction with a motor vehicle having a fuel injector, an internal combustion engine and at least one output sensor, includes: a correlator which is electrically connected to the fuel injector and the output sensor wherein the correlator correlates firing events created by the fuel injector with an output signal created by the output sensor; a filter electrically connected to the correlator to receive the output signal and to compare the output signal against a variable threshold signal created by the filter; and a recorder, electrically connected to the filter, to record the firing events which exceed the variable threshold signal and those which do not exceed the variable threshold signal.

Misfire detection assembly

Misfiring of individual cylinders in an internal combustion engine 70 is detected based on simultaneous use of two or more different methods of misfire detection which examine different combustion-related parameters of engine operation, such as crankshaft acceleration and exhaust gas pressure. An analyzer 76 produces a final misfire determination M.sub.f (i) based on the simultaneously obtained misfire determinations in order to provide greater coverage of the speed-load range of engine 70 and to provide increased confidence and accuracy in calling misfires in the portion of the speed-load range where both misfire methods are presumed to function.

Fuzzy logic method for detecting misfiers in internal combustion engines

A crankshaft based misfire detector for an internal combustion engine detects and identifies misfire over a broad range of engine operating conditions by establishing correction factors to compensate for crankshaft torsional oscillations and static wheel profile errors which corrupt the acceleration signals used to detect misfire. The correction factors are adaptively determined during in-use engine operation and are updated in one manner for most firings but in another manner for any cylinder firing which had the relatively lowest corrected acceleration within an engine cycle.

Adaptive correction of torsional and wheel profile position

A system and method compensates for systematic irregularities in measured engine velocities caused by torsional flexing in the crankshaft during rotation to improve the accuracy of misfire detection. A set of generated correction factors is stored in a table that are derived empirically using a test engine. The correction factors are adjusted during on-board operation of the engine whenever a safe adaptation time is determined.

Engine diagnostic monitor with on-board adjustment of correction factors

A system and method for processing test measurements collected from an internal combustion engine that is cold-tested for diagnostic purposes. Test measurements are collected from an engine. The test measurements are 'pre-processed' by filtering and subsampling techniques so that Principle Component Analysis can be applied to condense the quantity of test measurements while still retaining a statistically accurate indication of a majority of the original measurements. The pre-processed test measurements are then passed through one or more classifiers including: a Neural Network classifier, a Fuzzy Logic classifier, a cluster-based classifier (or 'Spherical' classifier) and a Genetic Program classifier. Results from these classifiers can be used to obtain a verdict about an engine (i.e., whether the engine is 'normal' or 'faulty').

System and method for processing test measurements collected from an

Misfiring of individual cylinders in an internal combustion engine is detected based on sensing exhaust pressure during each operating cycle of the engine. A pressure signal is sampled and digitized during each cycle to form an input vector for a pattern classifier. Predetermined internal coefficients of the trained classifier enable the classifier to discriminate between firing and misfiring cylinders in real time with a high degree of accuracy. Overall system cost is reduced by sharing a pressure transducer with an exhaust gas recirculation system on the internal combustion engine.

Misfire detection in an internal combustion engine using exhaust pressure

A method and apparatus are provided for reliably detecting misfires in an internal combustion engine during actual driving conditions by monitoring fluctuations in engine speed between consecutive firing intervals. An average acceleration over a series of firing intervals is removed from an individual acceleration measurement corresponding to a particular firing interval to give an acceleration deviation which is then normalized to an expected torque to provide a power loss measurement. The power loss measurement indicates the occurrence of a misfire with a high signal-to-noise ratio. The information derived from misfire detection can be used to prevent destruction of a catalytic converter by unburned fuel mixtures and can be used to improve vehicle driveability by correcting for or indicating needed service for the vehicle. The misfire detection is implemented using existing sensors and processors and is adapted to be done on-board a vehicle in real time. Noise present in the acceleration data is measured and a no-test condition is recognized depending on the magnitude of the measured noise.

Growing community of inventors

Ann Arbor, MI, United States of America

Kenneth A Marko