Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Mechanical measurement system
Reexamination Certificate
1999-12-15
2002-02-26
Assouad, Patrick (Department: 2857)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Mechanical measurement system
C702S181000
Reexamination Certificate
active
06351713
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention generally relates to mechanical diagnostic systems, and particularly to a distributed stress wave analysis system for detecting structure borne sounds caused by friction.
2. Description of the Prior Art
All diagnostic techniques are selected to detect discrepant components and monitor their rate of degradation up to the end of their useful life. Thus, they are closely related to inherent reliability and to the rate at which components degrade from fault initiation to loss of intended function.
Since diagnostic techniques are often used during scheduled inspections, the aircraft maintenance and inspection concept forms a critical interface with diagnostics. Diagnostics is the “gate” through which basic reliability, failure progression intervals, and the maintenance/inspection concept must interface to produce the availability, safety, and component removal rates for the overall system.
One of the key relationships in the detection of failures is the one between the failure progression interval and the inspection interval. This relationship determines how often the diagnostic technique will have a chance to detect a discrepancy during the progression of a failure. The concept of this relationship, is what has caused frequent inspections on army aircraft, for example, in order to provide maximum probability of detection.
Army helicopters diagnostic techniques are typically optimized to assure minimum accident rate. However, this has been accomplished at the cost of mission reliability (unnecessary aborts) and mean time between removals (“MTBR”) (due to incorrect or premature removals). Over the past several decades, many millions of dollars have been spent to improve the inherent reliability and failure progression intervals related to historically significant failure modes. However, in many cases, the major MTBR and mission reliability benefits of a higher mean time between failures (“MTBF”) and slower failure progression cannot be realized without some improvement to the diagnostics gate.
There are four basic parameters that define the accuracy and effectiveness of any diagnostic technique. These parameters are defined as follows:
P
1
—the probability of calling a good part good;
P
2
—the probability of calling a good part bad (this parameter represents false indications and P
2
=1−P
1
);
P
3
—the probability of calling a bad part good (these are undetected levels of degradation P
3
=1−P
4
); and
P
4
—the probability of calling a bad part bad (these are correctly identified degraded parts).
The effectiveness of any diagnostic technique depends upon the detection threshold employed to indicate a degraded condition. For each possible detection threshold, there is an associated set of P
1
and P
4
effectiveness values. Variations of the detection threshold invariably produce divergent changes in P
1
and P
4
.
The effectiveness parameters P
1
and P
4
are also related to the number of times that a diagnostic technique is used during the progression of a fault from “initial discrepancy” to “end of useful life”. The more often the test is made, the better the chance of detecting the failure process. While the “instantaneous” probability of detecting a fault is a function of the detection threshold, the “cumulative” probability is a function of the technique utilization interval and the rate of failure progression. The following equation defines how instantaneous probabilities convert to cumulative probabilities of fault detection.
P
4c
=1−(1−P
4d
) Equation A
Where:
P
4c
=The cumulative probability of detection after “n” uses of the diagnostic technique.
P
4d
=The probability of detection after each use of the diagnostic technique (a “decision cycle”).
n=The number of times the diagnostic technique is used during the failure progression interval.
The MTBR and diagnostic technique cost effectiveness are inseparable elements in setting the indication thresholds for any current or proposed technique. Except for the cost savings attributable to accidents prevented by diagnostic/prognostic indications, there is no other area where significant cost savings can be achieved in environments such as the army aviation environment. The MTBR is an expression of the rate at which all component removals occur, regardless of whether or not the removals were justified. Thus, incorrect removals due to false diagnostic indications are a contributing factor to the overall MTBR. Accordingly, it is vital to set indication thresholds that will (a) reliably detect the presence of degradation early in the failure progression interval, and (b) have a very low probability of false indication when used to test healthy components for a period of time that is greater than the component's inherent MTBF. This same type of analysis also applies to the mean time between precautionary landings, mission aborts, and maintenance actions for diagnostic techniques and indications that result in these events.
It is to the effective resolution to achieve these accurate indications that the present invention is directed.
SUMMARY OF THE INVENTION
The present invention provides a distributed stress wave analysis system for detecting structure borne sounds cause by friction. The detected information is processed using feature extraction and polynomial network artificial intelligence software. The system consists of stress wave sensors, interconnect cables, and preferably three modules: (1) distributed processing units (“DPU”), (2) maintenance advisory panel (“MAP”), and (3) laptop computer (“LTC”).
Where the system is applied to helicopter drive train components, the sensors, DPU and MAP can be airborne components, while the LTC can be ground based. The DPU can have a serial interface for integration into an airborne Flight Data Recorder or Health Usage Monitoring System (“HUMS”).
The stress wave analysis (“SWAN”) portion of the system is a high frequency acoustic sensing and signal conditioning technology that provides a time history of friction and shock events in a machine, such as a helicopter drive train. The SWAN portion of the system is similar to the stress wave analysis described and shown in U.S. Pat. No. 5,852,793, issued to Board et al. (the '793 Patent), the disclosure of which is incorporated herein by reference. A derived stress wave pulse train (“SWPT”) is independent of background levels of vibration and audible noise. The SWPT preferably is digitized and used to extract signature features, which when processed by neural networks of polynomial equations, characterize the mechanical health of the monitored components.
The system includes an adjustable data fusion architecture to optimize indication thresholds, maximize fault detection probability, and minimize false alarms. System testing preferably indicates a 100% probability of detecting gear or bearing damage within one hour of operation with a discrepant condition, and less than a one tenth of one percent chance of a false alarm during 1000 hours of operation with healthy components. In addition, to accurately detecting faults, the software used by the system will locate a fault, isolate its cause to either a gear or bearing source, display the percent degradation, and estimate its remaining useful life.
The application of artificial intelligence techniques for classification of SWPT features advances current technology to achieve accurate, real time, diagnostic capability at all flight power levels. However, it should be recognized that the hardware and operating system software of the present invention are readily adaptable to numerous other mobile and fixed based applications, and all applications are considered within the scope of the invention.
Accordingly, it is an object of the invention to provide a system that performs distributed stress wave analysis on one or more components of a machine or equipment.
It is another object of the invention to provide a system for reliably detecting the prese
Andrews Sylvia
Board David B.
Caulfield Michael
Hughes Craig
Assouad Patrick
Malin Haley & DiMaggio, P.A.
Swantech, L.L.C.
LandOfFree
Distributed stress wave analysis system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Distributed stress wave analysis system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Distributed stress wave analysis system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2967046