Systems and methods for a hybrid diagnostic approach of real time diagnosis of electronic systems

• US 6,519,552 B1
• Filed: 03/10/2000
• Issued: 02/11/2003
• Est. Priority Date: 09/15/1999
• Status: Expired due to Term

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Jackson, Eric, Sampath, Meera, Godambe, Ashok, Mallow, Edward W.
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): KIM, PAUL L
• Application Number:
  US09/522,082
• Time in Patent Office: 1,068 Days
• Field of Search: 702/183, 702/185, 702/182, 714/26
• US Class Current: 702/183
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G05B 23/0278 Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA
• CPC Class Codes: G06F 3/121 Facilitating exception or error detection and recovery, e.g. fault, media or consumables depleted
• CPC Class Codes: G06F 3/1234 Errors handling and recovery, e.g. reprinting G06F3/1261 takes precedence
• CPC Class Codes: G06F 3/1279 Controller construction, e.g. aspects of the interface hardware
• CPC Class Codes: G06F 3/1284 Local printer device
• CPC Class Codes: H04N 1/00002 Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for error detection, error correction or monitoring in digital computers or digital computer components G06F11/00
• CPC Class Codes: H04N 1/00029 Diagnosis, i.e. identifying a problem by comparison with a normal state
• CPC Class Codes: H04N 1/00076 locally

Abstract

Through a combination of a hybrid diagnostic scheme based on qualitative and quantitative technologies, at least component level status information can be obtained about a machine, such as an electronic device. In particular, qualitative model based diagnostic technologies is used in conjunction with quantitative analysis techniques and signature analysis to achieve accurate and reliable diagnosis of components and systems down to the individual component level, for example, down to the consumer replaceable unit level. Thus, the hybrid diagnostic methodology exploits the diagnostic information content in already available system signals via intelligent processing and allows for diagnosis with minimal sensor requirements. Furthermore, the diagnostic methodology allows for a self-diagnosing machine having diagnostic intelligence which in turn can reduce service time, service costs, increased machine up time and improve customer satisfaction.

Claims

1. A hybrid diagnostic system for status prediction and diagnosis comprising:

at least one system sensor;

at least one virtual sensor, wherein the virtual sensor determines status information based on a quantitative analysis of data from a monitored system; and

a diagnostic inference device that performs a qualitative analysis on the information from at least one of the at least one system sensor and the at least one virtual sensor.

2. The system of claim 1, wherein the at least one virtual sensor is at least one of a signature analysis module, a statistical analysis module, a threshold analysis module, a rate analysis module, a timing analysis module, a quantitative state estimator module, a computational module that is based on at least one of linear algebra, differential calculus, qualitative calculus, logical equations, and topological analysis, a pattern recognition module, an image processing module, a control sequence monitoring module, a threshold analysis module, a counter module and a timer module.

3. The system of claim 2, wherein the quantitative state estimator module is based on at least one of an analytical redundancy based scheme, a Luenberger observer, a robust observer, a Kalman filter, a fault detection filter, a parity space approach and a parameter estimation technique.

4. The system of claim 1, wherein the qualitative analysis is based on at least one of model based reasoning, a discrete event systems diagnostic technology, a constraint based programming methodology, a bayesian network, a causal network, a neural network and an expert system.

5. The system of claim 1, further comprising an event generator device that translates information from one or more of the at least one sensor and the at least one virtual sensor into a format appropriate for the diagnostic inference device.

6. The system of claim 5, wherein an output of the event generator device is based on at least one of a control signal, an actuator and a set-point signal, a scheduling and a timing signal, a job and a machine operational mode information, a sensor response to control commands and test commands, process control data information, monitoring data, wherein the monitoring data is at least one of an output from at least one of a special diagnostic sensor, a counter and a sensor signal change, environmental and operational condition data and a virtual sensor output.

7. The system of claim 5, wherein the translation of the sensor information into event information is based on at least one of tracking and translating machine signals into discrete event sequences, discretization of analog data into qualitative values, transformation of quantitative sensor values into qualitative values, sequencing the monitored system and sensor signals and formatting the event information.

8. The system of claim 1, wherein an output of the diagnostic inference device is forwarded to at least one of a user, a system operator, a service provider, a parts supplier, a repair planning system, an autonomous repair agent, a diagnostic server, and a secondary information resource.

9. A hybrid diagnostic system for prediction and diagnosis of system status in a document processing system comprising:

a document processing system comprising at least one of at least one system sensor and at least one virtual sensor, wherein the virtual sensor is based on a quantitative analysis of data from one or more components of the document processing system; and

a diagnostic inference engine based on qualitative analysis of the data from the one or more sensor or virtual sensor.

10. The system of claim 9, wherein the at least one virtual sensor is at least one of a signature analysis module, a statistical analysis module, a threshold analysis module, a rate analysis module, a timing analysis module ,a quantitative state estimator module, a computational module that is based on at least one of linear algebra, differential calculus, qualitative calculus, logical equations, and topological analysis, a pattern recognition module, an image processing module, a control sequence monitoring module, a threshold analysis module, a counter module and a timer module.

11. The system of claim 10, wherein the quantitative state estimator module is based on at least one of an analytical redundancy based scheme, a Luenberger observer, a robust observer, a Kalman filter, a fault detection filter, a parity space approach and a parameter estimation technique.

12. The system of claim 9, wherein the qualitative analysis is based on at least one of model based reasoning, a discrete event systems diagnostic technology, a constraint programming methodology, a bayesian network, a causal network, a neural network and an expert system.

13. The system of claim 9, further comprising an event generator device that translates data from one or more of the at least one sensor and the at least one virtual sensor into a format appropriate for the diagnostic inference device.

14. The system of claim 13, wherein an output of the event generator device is based on at least one of a control signal, an actuator and a set-point signal, a scheduling and a timing signal, a job and a machine operational mode information, a sensor response to control commands and test commands, process control data information, monitoring data, wherein the monitoring data is at least one of an output from at least one of a special diagnostic sensor, a counter and a sensor signal change, environmental and operational condition data and a virtual sensor output.

15. The system of claim 13, wherein the translation of the sensor information into event information is based on at least one of tracking and translating machine signals into discrete event sequences, discretization of analog data into qualitative values, transformation of quantitative sensor values into qualitative values, sequencing the monitored system and sensor signals and formatting the event information.

16. The system of claim 9, wherein an output of the diagnostic inference device is forwarded to at least one of a user, a system operator, a service provider, a parts supplier, a repair planning system, an autonomous repair agent, a diagnostic server, and a secondary information resource.

17. A method for diagnosing or predicting failures in an electronic system comprising:

monitoring one or more system sensors and virtual sensors to capture sensor information;

translating the sensor information into event information; and

estimating a state of the system based on the event information.

18. The method of claim 17, wherein the at least one virtual sensor is at least one of a signature analysis module, a statistical analysis module, a threshold analysis module, a rate analysis module, a timing analysis module, a quantitative state estimator module, a computational module that is based on at least one of linear algebra, differential calculus, qualitative calculus, logical equations, and topological analysis, a pattern recognition module, an image processing module, a control sequence monitoring module, a threshold analysis module, a counter module and a timer module.

19. The method of claim 18, wherein the quantitative state estimator module is based on at least one of an analytical redundancy based scheme, a Luenberger observer, a robust observer, a Kalman filter, a fault detection filter, a parity space approach and a parameter estimation technique.

20. The method of claim 17, wherein the state is estimated based on at least one of model based reasoning, a discrete event systems diagnostic technology, a constraint programming methodology, a bayesian network, a causal network, a neural network and a expert system.

21. The method of claim 17, wherein the event information is based on at least one of a control signal, an actuator and a set-point signal, a scheduling and a timing signal, a job and a machine operational mode information, a sensor response to control commands and test commands, process control data information, monitoring data, wherein the monitoring data is at least one of an output from at least one of a special diagnostic sensor, a counter and a sensor signal change, environmental and operational condition data and a virtual sensor output.

22. The method of claim 17, wherein the diagnosis and/or prediction can be performed during at least one of a normal operation of the electronic system, a diagnostic mode and a test mode.

23. The method of claim 17, wherein the translation of the sensor information into event information is based on at least one of tracking and translating machine signals into discrete event sequences, discretization of analog data into qualitative values, transformation of quantitative sensor values into qualitative values, sequencing the machine and sensor signals and formatting the event information.

24. An information storage medium having information that diagnoses or predicts failures in an electronic system comprising:

information that acquires information about the system using at least one system sensor;

information that performs a quantitative analysis on a portion of the information about the system using at least one virtual sensor;

information that performs a qualitative analysis on a portion of the information about the system; and

information that diagnoses a status of the system based on the combined results of the quantitative and the qualitative analysis.

25. The information storage medium of claim 24, wherein the quantitative information is derived from at least one of a signature analysis module, a statistical analysis module, a threshold analysis module, a rate analysis module, a timing analysis module, a quantitative state estimator module, a computational module that is based on at least one of linear algebra, differential calculus, qualitative calculus, logical equations, and topological analysis, a pattern recognition module, an image processing module, a control sequence monitoring module, a threshold analysis module, a counter module and a timer module.

26. The information storage medium of claim 25, wherein the quantitative state estimator module is based on at least one of an analytical redundancy based scheme, a Luenberger observer, a robust observer, a Kalman filter, a fault detection filter, a parity space approach and a parameter estimation technique.

27. The information storage medium of claim 24, wherein the qualitative analysis is based on at least one of model based reasoning, a discrete event systems diagnostic technology, a constraint programming methodology, a bayesian network, a causal network, a neural network and an expert system.

28. The information storage medium of claim 24, wherein the diagnosis and/or prediction can be performed during at least one of a normal operation of the electronic system, a diagnostic mode and a test mode.

29. A method for generating a diagnostic inference engine for predicting or diagnosing the status of an electronic system comprising:

building a model of the system to be diagnosed from at least one of individual component models, controller models, and fault symptom tables, wherein the fault symptom tables list an output of one or more of at least one sensor and at least one virtual sensor for various operational states of the system; and

generating a diagnostic engine from the system model and at least one diagnostic requirement.

30. A method for generating a diagnostic inference device for predicting or diagnosing the status of an electronic system comprising:

generating a diagnostic inference device from at least one of an individual component model, a controller model, a fault symptom table and a diagnostic requirement, wherein the fault symptom table comprises a list of outputs of one or more of at least one sensor and at least one virtual sensor for various operational states of the electronic system.