Data processing: generic control systems or specific application – Generic control system – apparatus or process – Optimization or adaptive control
Reexamination Certificate
2000-06-09
2003-08-19
Patel, Ramesh (Department: 2121)
Data processing: generic control systems or specific application
Generic control system, apparatus or process
Optimization or adaptive control
C700S028000, C700S031000, C700S040000, C700S046000, C700S047000, C700S048000, C706S014000, C706S015000, C706S016000, C701S002000, C701S003000, C701S011000, C701S023000
Reexamination Certificate
active
06609036
ABSTRACT:
FIELD OF THE INVENTION
The instant invention relates generally to a system and method for process parameter estimation using operating mode partitioning and, in particular, to a system and method for performing high sensitivity surveillance of an asset such as a process and/or apparatus preferably having at least two distinct modes of operation wherein surveillance is performed using an operating mode partitioned parameter estimation model of the asset.
BACKGROUND OF THE INVENTION
Conventional process surveillance schemes are sensitive only to gross changes in the mean value of a process signal or to large steps or spikes that exceed some threshold limit value. These conventional methods suffer from either a large number of false alarms (if thresholds are set too close to normal operating levels) or from a large number of missed (or delayed) alarms (if the thresholds are set too expansively). Moreover, most conventional methods cannot perceive the onset of a process disturbance or sensor signal error that gives rise to a signal below the threshold level or an alarm condition. Most conventional methods also do not account for the relationship between a measurement by one sensor relative to another sensor.
Recently, improved methods for process surveillance have developed from the application of certain aspects of artificial intelligence technology. Specifically, parameter estimation methods have been developed using either statistical, mathematical or neural network techniques to learn a model of the normal patterns present in a system of process signals. After learning these patterns, the learned model is used as a parameter estimator to create one or more virtual signals given a new observation of the actual process signals. Further, high sensitivity surveillance methods have been developed for detecting process and signal faults by analysis of a mathematical comparison between the actual process signal and its virtual signal counterpart.
Parameter estimation based surveillance schemes have been shown to provide improved surveillance relative to conventional schemes for a wide variety of assets including industrial, utility, business, medical, transportation, financial, and biological systems. However, parameter estimation based surveillance schemes have in general shown limited success when applied to complex processes. Applicant recognizes and believes that this is because the parameter estimation model for a complex process must characterize the entire operating state space of the process to provide effective surveillance. Moreover, a review of the known prior-art discloses that virtually all such systems developed to date utilize a single model of the process to span the entire set of possible operating modes. Hence, a significant shortcoming of the known prior-art is that, inter alia, statistically derived models become extremely large and neural network models become difficult or impractical to train when the process operating state space is complex. The implication for statistically derived models is that the parameter estimation method and system becomes computationally expensive to operate thereby limiting the utility of the method for on-line or real-time surveillance. An alternative for statistically derived models is to constrain the size of the model; however this constraint limits the accuracy of the parameter estimation method and thereby limits the sensitivity of the surveillance method. The implication for mathematical and neural network models is simply that the parameter estimation method and system becomes less accurate thereby degrading the sensitivity of the surveillance method.
Many attempts to apply multivariate state estimation techniques, mathematical modeling techniques and neural network techniques to assets such as industrial, utility, business, medical, transportation, financial, and biological processes have met with poor results in part because the parameter estimation models used were expected to characterize the entire operating state space of the process. In one example, a multivariate state estimation technique (MSET) based surveillance system for the Space Shuttle Main Engine's telemetry data was found to produce numerous false alarms when the learned MSET parameter estimation model was constrained to a size suitable for on-line, real-time surveillance. In this case, the surveillance system false alarm rate could be reduced by desensitizing the surveillance threshold parameters; however, the missed alarm rates then became too high for practical use in the telemetry data monitoring application.
Moreover, current multivariate state estimation techniques, mathematical modeling techniques and neural network techniques for surveillance of assets such as industrial, utility, business, medical, transportation, financial, and biological processes fail to recognize the surveillance performance limitations that occur when it becomes necessary to trade-off decision processing speed against decision accuracy. This may be attributed, in part, to the relative immaturity of the field of artificial intelligence and computer-assisted surveillance with regard to real-world process control applications. Additionally, a general failure to recognize the specific limitations of trading off decision processing speed against decision accuracy for computer-assisted surveillance is punctuated by an apparent lack of known prior art teachings that address potential methods to overcome this limitation. In general, the known prior-art teaches computer-assisted surveillance solutions that are either applied globally to all operating modes of an asset or applied only to a single predominant operating mode, for example, applied only to steady state operations while neglecting all transient operating states of the asset.
For the foregoing reasons, there is a need for a surveillance system and method that overcomes the significant shortcoming of the known prior-art as delineated hereinabove.
SUMMARY OF THE INVENTION
The instant invention is distinguished over the known prior art in a multiplicity of ways. For one thing, the instant invention provides a surveillance system and method that partitions parameter estimation models of an asset for overcoming a performance limiting trade-off between decision processing speed and decision accuracy that has been generally unrecognized by the known prior-art. Additionally, the instant invention can employ any one of a plurality of parameter estimation methods and the process models used therewith for improving surveillance performance. Furthermore, the instant invention provides a surveillance system and method that provides an operating mode partitioned parameter estimation model that can be accomplished by observation and analysis of a time sequence of process signal data and by a combination of a plurality of techniques.
Moreover, the instant invention provides a surveillance system and method that provides an operating mode partitioning of the parameter estimation model which enables different parameter estimation methods, thresholds and decision logic to be used for surveillance within each individual operating mode of an asset. This ability enables surveillance to be performed by the instant invention with lower false alarm rates and lower missed alarm rates than can be achieved by the known prior-art methods.
Hence, the instant invention provides a surveillance system and method that performs its intended function much more effectively by enabling higher decision processing speed without a concomitant reduction in decision accuracy. Conversely, the instant invention alternately enables improved decision accuracy without a concomitant reduction in decision processing speed. Additionally, these competing criteria may be traded-off to achieve the optimal performance solution for a specific surveillance application. Further, parameter estimation methods, thresholds and decision logic may be individually tailored for each operating mode of the asset thereby providing additional capability to reduce decision error rates for th
DeBoo Dennis A.
Patel Ramesh
LandOfFree
Surveillance system and method having parameter estimation... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Surveillance system and method having parameter estimation..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Surveillance system and method having parameter estimation... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3129048