System and method for processing test measurements collected from an internal combustion engine for diagnostic purposes
First Claim
1. A method for cold testing an internal combustion engine comprising the steps of:
- cranking the engine at a predetermined speed for a predetermined number of revolutions;
monitoring an operational parameter of the engine during said predetermined number of revolutions;
producing an electrical signal representing said operational parameter;
preprocessing said electrical signal to produce a principle component value vector signal representing said operational parameter; and
analyzing said principle component value vector signal to produce a classification signal indicating, based on the operational parameter, whether the engine is normal or faulty.
2 Assignments
0 Petitions
Accused Products
Abstract
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").
127 Citations
20 Claims
-
1. A method for cold testing an internal combustion engine comprising the steps of:
-
cranking the engine at a predetermined speed for a predetermined number of revolutions; monitoring an operational parameter of the engine during said predetermined number of revolutions; producing an electrical signal representing said operational parameter; preprocessing said electrical signal to produce a principle component value vector signal representing said operational parameter; and analyzing said principle component value vector signal to produce a classification signal indicating, based on the operational parameter, whether the engine is normal or faulty. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A method for preprocessing diagnostic test measurements collected from a multi-cylinder internal combustion engine, comprising the steps of:
-
cranking the engine at a predetermined speed for a predetermined number of revolutions; receiving a diagnostic cold-test measurement signal from the engine; passing the cold-test measurement signal through a median filter to produce a median filter vector signal having a plurality of vector points; passing the median filter vector through an averaging filter to produce a subsampled signal; partitioning the subsampled signal into segments corresponding to each cylinder event of an engine cycle to produce N subsampled signals, wherein N is the number of cylinders; passing each of the N subsampled signals through a mean filter to produce a mean cylinder signal equal to a mean of the N subsampled signals; passing each of the N subsampled signals through a standard deviation filter to produce a standard deviation vector signal equal to the standard deviation of the N subsampled signals; and passing the standard deviation vector signal and the mean cylinder signal through a principle-component-analysis filter to convert the standard deviation vector signal and the mean cylinder signal into a principle component value vector signal. - View Dependent Claims (12, 13, 14, 15)
-
-
16. A classification system for cold testing an internal combustion engine to determine whether the engine is faulty or normal, comprising:
-
means for monitoring an operational parameter of the engine while the engine is cranked at a predetermined speed for a predetermined number of revolutions, and for producing an electrical signal representing said operational parameter; and neural network means, coupled to said monitoring and producing means, for producing a classification signal indicating whether the engine is normal or faulty. - View Dependent Claims (17, 18, 19)
-
-
20. A classification system for testing an internal combustion engine to determine whether the engine is faulty or normal, comprising:
-
means for monitoring an operational parameter of the engine while the engine is cranked at a predetermined speed for a predetermined number of revolutions, and for producing an electrical signal representing said operational parameter; cluster-based classifier means, coupled to said monitoring and producing means, the plotting a value of said electrical signal on a plot, wherein the origin of said plot represents a predetermined non-faulty engine, and for determining the distance of value from the origin, and for producing a distance signal representing the distance of said value from the origin, wherein the distance signal provides an indication whether the engine is normal or non-faulty.
-
Specification