[go: up one dir, main page]

WO2003019377A2 - Systeme et procede de diagnostic multiniveau echelonnable a distance et maintenance conditionnelle - Google Patents

Systeme et procede de diagnostic multiniveau echelonnable a distance et maintenance conditionnelle Download PDF

Info

Publication number
WO2003019377A2
WO2003019377A2 PCT/IB2002/003409 IB0203409W WO03019377A2 WO 2003019377 A2 WO2003019377 A2 WO 2003019377A2 IB 0203409 W IB0203409 W IB 0203409W WO 03019377 A2 WO03019377 A2 WO 03019377A2
Authority
WO
WIPO (PCT)
Prior art keywords
machine
node
anomaly
diagnosis
tool
Prior art date
Application number
PCT/IB2002/003409
Other languages
English (en)
Other versions
WO2003019377A3 (fr
Inventor
Sascha Nick
Original Assignee
Idtect
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US09/934,000 external-priority patent/US20030046382A1/en
Application filed by Idtect filed Critical Idtect
Priority to EP02767748A priority Critical patent/EP1419442A2/fr
Priority to AU2002330673A priority patent/AU2002330673A1/en
Publication of WO2003019377A2 publication Critical patent/WO2003019377A2/fr
Publication of WO2003019377A3 publication Critical patent/WO2003019377A3/fr

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
    • G05B23/0283—Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL]

Definitions

  • the present invention relates generally to predictive maintenance, and more particularly relates to diagnosing processes and machines at remote locations.
  • Time-based preventive maintenance is one of the popular techniques currently employed by the manufacturing industry for reducing the number of unscheduled shut downs of a manufacturing line.
  • time-based preventive maintenance components are inspected and/or replaced at periodic intervals. For example, a bearing rated for so many hours of operation is always replaced after a set number of operational hours regardless of its condition.
  • Figure 7a shows typical failure probability charts for a variety of components.
  • Curve 1000 illustrates the failure probability of components subject to dominant age-related failure and "infant mortality" (i.e., high initial failure rates decreasing over time to a stable level).
  • Curve 1002 illustrates the failure probability of components having a dominant age- related failure mode only.
  • Curves 1004, 1006 illustrate the failure probability of components subject to failure fatigue.
  • Curve 1008 illustrates the failure probability of complex electromechanical components without a dominant failure mode and electromechanical components that are not subject to an excessive force.
  • Curve 1010 illustrates the failure probability of electronic components (e.g., controllers, sensors, actuators, drives, regulators, displays, Places, computers).
  • electronic components e.g., controllers, sensors, actuators, drives, regulators, displays, Places, computers.
  • Time based preventive maintenance decreases failures for components that exhibit a failure probability illustrated in curves 1000, 1002.
  • These components which comprise a low percentage of approximately four to six percent of installed equipment, include complex mechanical equipment subject to premature failures (e.g., gearboxes and transmissions) and mechanical equipment with a dominant age-related failure mode (e.g., pumps, valves, pipes).
  • Preventive maintenance does not decrease or increase failures for components that exhibit a failure probability similar to the failure probability illustrated in curves 1004 - 1008. However, if some other component is disrupted during the maintenance, the failure rate of these components actually increases with time based preventive maintenance.
  • Time based preventive maintenance actually increases the failure rate of electronic components by prematurely shutting down a manufacturing line for scheduled maintenance and introducing "infant mortality" in what is an otherwise stable system.
  • Curve 1100 in figure 7b illustrates the increased failure probability due to "infant mortality" when electronic components are replaced due to preventive maintenance and curve 1102 illustrates the failure probability with no preventive maintenance performed.
  • predictive maintenance monitors the condition of operating parameters on a machine over a period of time. Predictions are generated of when a component should be replaced based on detected changes in the operating parameters. The changes can also be used to indicate specific faults in the system being monitored. Techniques for predictive maintenance that are available today, however, are either poorly matched to the particular circumstances and, therefore, less than completely effective or they are so expensive as to be prohibitive in all but the most expensive manufacturing settings.
  • Predictive maintenance systems have had only a limited acceptance by the manufacturing industry. It has been estimated that these systems are being used today in less than one percent of the total maintenance market. Many predictive maintenance systems are expensive, require local experts, and are often unstable or unreliable. These systems require continuous monitoring of operating parameters and conditions. This continuous monitoring results in an enormous amount of data that, in turn, requires significant processing power. As a result, predictive maintenance is often cost-prohibitive. Due to the expense of the installation and maintenance of these predictive systems, manufacturers either limit the number of systems installed in a manufacturing site, limit the number of components at the site that are monitored, or perform time sampling of components instead of continuous monitoring. The reduced monitoring reduces the effectiveness of the system and ultimately results in its unreliable performance.
  • Another problem is with the use of experts that analyze the data. Locally based experts may be difficult to find. Transmitting all data to an expert for analysis requires bandwidth. Additionally, experts are expensive and often become a bottleneck in the process.
  • Another problem is the reliability of sensor signals used to monitor the system. It has been estimated that fifty percent of monitoring problems are a direct result of sensor failure, sensor obstruction (e.g., oil, dust, or other particles), and severed or damaged sensor cables. The present monitoring systems typically do not monitor the health of the sensors used to monitor the system.
  • the invention provides a method for remotely monitoring and diagnosing operations of a device, machine, or system (hereinafter called "machine") and for performing predictive maintenance on a machine.
  • a signal model of the machine is created based on sensed signals during normal operation of the machine. Signals representative of the machine's operating and condition parameters are sensed and compared to the signal model locally maintained proximate to the device in order to detect anomalies. Once an anomaly is detected, information describing each anomaly is transmitted to a location remote from the machine. The information is diagnosed at the remote location.
  • the signal model is adapted to work with the remaining sensors if a failed sensor is detected.
  • the diagnosis includes an initial analysis of the information by diagnostic tools maintained at the remote location.
  • the diagnostic tools include a library of patterns comprising information describing systemic anomalies and a library of patterns comprising information describing component anomalies. The information is compared to patterns in the library describing systemic anomalies and component anomalies for a match. If a match is found, a diagnosis is made.
  • the initial analysis fails to provide a diagnosis
  • a subsequent analysis of the information by diagnostic tools maintained elsewhere is performed.
  • a final analysis by a team of humans aided by a collaborative environment is performed if the initial and subsequent analyses fail to provide a diagnosis.
  • the diagnosis of the anomaly is reported to a location capable of attending to repair of the machine.
  • Each new diagnosis is added to the appropriate pattern library for analysis of future anomalies, which improves the diagnostic capability of the system.
  • FIG. 1 is a block diagram generally illustrating an exemplary environment in which the present invention operates
  • FIG. 2 is a flow chart of a method of diagnosing failures of components in accordance with the present invention
  • FIG. 3 is a block diagram of an exemplary end user plant in which part of the present invention operates according to one embodiment of the present invention
  • FIG. 4 is a block diagram of an embodiment of a local detector in accordance with the present invention.
  • FIG. 5 a is a flow chart of an exemplary process performed in level 202 of the flow chart of FIG. 2;
  • FIG. 5b is a flow chart of an exemplary process performed in level 204 of the flow chart of FIG. 2;
  • FIG. 5c is a flow chart of an exemplary process performed in level 206 of the flow chart of FIG. 2;
  • FIG. 5d is a flow chart of an exemplary process performed in level 208 of the flow chart of FIG. 2;
  • FIG. 6 is a block diagram illustrating the step of auto-configuring a communications link in accordance with the present invention.
  • Figure 7a and 7b are charts above discussed which illustrate typical failure probabilities.
  • the invention is illustrated as being implemented in a suitable operating environment.
  • the invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network.
  • program modules may be located in both local and remote memory storage devices.
  • FIG. 1 illustrates an example of a suitable operating environment 100 in which the invention may be implemented.
  • the operating environment 100 is only one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality of the invention.
  • the operating environment 100 includes one or more end users 102 in communication with an OEM server 108 via a network 106.
  • Each end user 102 comprises a location where one or more machines or devices are located.
  • an end user 102 may be a manufacturing plant, a remote station or machine, a business, a home, a vehicle, or any other place where reliability of equipment is a concern.
  • the end users 102 are connected to the network 106 via proxy/gateways 104.
  • the network 106 in one embodiment is the Internet.
  • the network 106 may be a virtual private network, a dedicated network, a public switched network, a wireless network, a satellite link, or any other type of communication link.
  • the OEM servers 108 communicate with each other in a peer-to-peer network 110.
  • the network 110 may be other type of networks such as a virtual private network, a dedicated network, or any other type of communication link.
  • a directory server 1 12 maintains a list of all OEM servers 108 and, as described hereinbelow, is used to aid OEM servers find other OEM servers.
  • the directory server 112 also communicates with the expert network server 114.
  • the expert network 114 maintains a list of available experts located in a collaborative network 116 that can be used to solve particular problems.
  • Level 202 includes the end user 102 and proxy/gateway 104.
  • the equipment 300 being monitored is located in the end user location (see FIG. 3).
  • a detector 302 that monitors the machine 300 with sensors 304 is located proximate to the machine 300.
  • Each detector 302 is in communication with the proxy/gateway 104 via a wireless LAN 306 and sends data to an OEM server 108 if a problem is detected.
  • the detector 302 communicates with the proxy/gateway through a powerline carrier for signal transmission.
  • level 204 includes an OEM server 108.
  • the OEM server 108 hosts an expert system that analyzes the data received from the detector 302 and diagnoses the problem. If the OEM server 108 is unable to diagnose the problem, the data is sent to other OEM servers 108 that are selected by the directory server 112 in level 206.
  • Level 206 includes the OEM servers 108 in the network 110 and the directory server 112.
  • the selected OEM servers 108 attempt to diagnose the problem.
  • the diagnosis and solution are returned to the OEM server 108 in level 204. If the selected OEM servers 108 are unable to diagnose the problem, the data is sent to the expert network server 114.
  • Level 208 includes the expert network server 114 and the collaborative network 116.
  • experts are chosen to diagnose the problem.
  • the experts are located throughout the world and the collaborative network 116 allows the experts to diagnose the problem without having to travel from their home locations.
  • the solution is returned to the OEM server 108 in level 204.
  • the invention may be practiced with other computer system configurations, including hand-held devices, multi-processor systems, embedded devices, microprocessor based or programmable consumer electronics and consumer appliances, network PCs, minicomputers, mainframe computers, and the like.
  • the invention will be described in terms of monitoring a machine.
  • the present invention can be used on any type of installation or device where reliability is a concern and in any location (e.g., inside an installation, in an automobile or truck, in an outdoor environment, etc.)
  • the detector 302 includes a power supply module 400, analog sensor input module 402, reset/relearn button 404, indicator 406, communication module 408, and a central processing unit (CPU) 410.
  • the primary functions of the detector 302 are sensor data collection and buffering, data transformation using fast Fourier transforms (FFTs) or other transformation techniques, statistical model generation, real time model data calculation, real time decision making, sensor health monitoring, communication with the proxy/gateway 104, and local indication of machine status.
  • FFTs fast Fourier transforms
  • the power supply module 400 provides power to the other components of the detector 302.
  • the analog sensor input module 402 receives and processes signals from sensors mounted on or proximate to the machine being monitored.
  • the sensors are connected to the analog sensor input module 402 by point-to-point wire connections, a sensor bus, or a wireless connection.
  • the sensors are used to monitor the machine's operating and condition parameters.
  • the operating and condition parameters include parameters such as vibration, speed, rotor position, oil temperatures, inlet and outlet temperatures, bearing temperature, pressure, power draw, flow rates, harmonic content, etc.
  • the sensors 304 include vibration sensors, temperature sensors, speed/position sensors, electrical parameter sensors (e.g., voltage and current), pressure sensors, flow rate sensors, and status inputs.
  • the analog sensor input module 402 performs filtering and other signal conditioning when necessary. For example, vibration sensor signals typical require high pass filtering to filter out undesirable low frequency noise and at least one gain stage to optimize signal levels. Those skilled in the art will recognize that many functions of the analog sensor input module 402 may be integrated into individual sensors as sensor technology improves.
  • the reset/relearn button 404 is used to reset the CPU 410 and put the CPU 410 into the learning mode as will be describe below.
  • the indicator 406 comprises one or more LEDs to indicate whether or not the machine 300 is operating normally or whether an anomaly has occurred.
  • the communication module 408 is used to communicate with the proxy/gateway 104.
  • the communication module 408 may be an Ethernet card, a wireless LAN card using a protocol such as 802.1 lb, Bluetooth, any other wireless communication protocol, or wired communication such as a powerline carrier signal.
  • the CPU 410 monitors the machine 300 and detects small but statistically significant signal deviations relative to normal operating conditions using statistical modeling techniques as known by those skilled in the art. The signal deviations may be indicative of future machine or component failure.
  • the CPU 410 also monitors sensor health and excludes inputs from failed sensors, adapting the model to work with the remaining sensors. Alternatively, the CPU 410 generates replacement sensor signals for failed sensors and inputs it into the model.
  • the detector 302 may be a stand-alone unit or integrated with other components of an installation, including operating as a software object on any processor or distributed processors having sufficient processing capability .
  • FIG. 5a-5d the steps taken to monitor and diagnose a machine are shown.
  • the proxy/gateway 104 performs an auto-configuration of the communications link (step 502).
  • FIG. 6 shows one embodiment of an auto-configuration sequence.
  • the proxy/gateway 104 senses all available communication access modes that are active (step 600). This step is repeated periodically and when transfer errors occur.
  • the modes include LANs 700, dial-up modems 702, wireless devices 704, satellites 706, and other modes 708.
  • the proxy/gateway 104 establishes a data connection, finds the OEM server 108 (step 602 ), and establishes a secure connection (step 606).
  • the establishment of the secure connection utilizes hardware and software authentication keys, authorization levels, 128 bit data encryption, data integrity checks, and data traceability.
  • the proxy/gateway 104 tests the effective transmission speed (step 606) and establishes a hierarchy of connection modes (step 608).
  • the hierarchy lists the available connections in order of preference. The preference is established using parameters such as transmission speed, mode reliability, and cost. Once the hierarchy is established, the non-permanent connections such as the dial-up modem are disconnected to reduce cost (step 610).
  • the detector 302 generates a statistical signal model for the machine (step 504). This step is performed by the detector 302 entering into a learning mode to learn how the sensor signals correlate with each other during normal operation.
  • the detector 302 enters into the learning mode during installation and start-up and whenever the detector 302 is commanded to enter the learning mode.
  • the command to enter the learning mode is transmitted remotely or locally,
  • the reset/relearn button 404 is pressed to enter the learning mode locally.
  • the remote command is received through the communication module 408.
  • the detector 302 obtains representative data (i.e.
  • the detector 302 then fits the best reference curve(s) through the training data points as known in the art to generate the statistical model.
  • Those skilled in the art will recognize that there are a wide variety of methods that can be used to fit the curve and a wide variety of optimization points that may be chosen. Additionally, there are a number of different types of curves that may be used (e.g., higher order curves such as second order, third order, fourth order, etc. or multiple- segment linear curves). As statistical modeling techniques improve or are developed, the detector 302 is updated with the new/improved techniques.
  • the detector 302 monitors the operation of the machine. In this phase of operation, the detector 302 obtains the processed data and performs an FFT or other transformation algorithm on the data (step 506).
  • the detector 302 has enough memory to hold a working data buffer for the processed data (i.e., the sensor data in which filtering, amplification, integration, A/D conversion and similar operations have been applied) . For example, in one embodiment, five minutes of data for ten sensors with 16 bit resolution at a 5 kHz sampling rate requires a storage capacity of approximately 30 MB.
  • the detector 302 also maintains an incident archive and a context archive. Each archive contains 120 FFT images of all sensor data for relevant high sampling rate sensors.
  • the incident archive contains one FFT per minute for two hours.
  • the incident archive is cyclically rewritten so that after two hours, each data entry is deleted. Before deletion, one FFT per hour (i.e., two FFTs from the entire incident archive) is moved into the context archive and kept for five days (i.e., 120 hours).
  • the data in the incident archive and context archive is not analyzed by the detector 302. In the event that sensor data does not fit the model as described below (i.e., an anomaly), the incident and context archives are transmitted to the OEM server 108 in level 204, where it is compared to the systemic pattern library.
  • the data in the incident and context archive is transmitted to level 208 and utilized by human experts.
  • the memory required for each archive is approximately 240 kB. It should be noted that the size (i.e., number of samples) and sampling rate of the incident and context archives can be reconfigured.
  • the detector 302 compares the actual sensor data to the statistical model to determine if the sensor data changes relative to the statistical model in a similar manner (step 508). This step is performed by calculating the distance between the model reference curve and each actual data point. These distance points are analyzed over a period of time. If the distance remains small and random (i.e., the sensor data fits the model), the machine 300 is operating normally (step 510) and steps 506 and 508 are repeated. A signal is sent periodically to the OEM server 108 to indicate that the machine operation is normal.
  • the detector 302 transmits the sensor data to the OEM server 108 (step 512), provides a visual or audio alert by changing the status of the indicator LED 406, and continues monitoring the machine 300 by repeating steps 506 -512.
  • the sensor data is compressed prior to transmission (for faster and more cost-effective transmission) and sent to the OEM server 108 via the proxy/gateway 104. If the anomaly persists, the detector 302 periodically transmits transformed data in batches to the OEM server 108 in order to avoid OEM server saturation and excessive transmission costs.
  • the detector 302 does not fit a reference curve through the training data points.
  • the detector 302 selects a relevant subset of the training data that is representative of normal machine operation and compares the actual sensor data to the subset of training data as described above. The distance between the selected training data points and actual data points is used and analyzed over a period of time.
  • virtual sensors are created for a select number of real sensors by maintaining a weighted moving average of sensor data and comparing the actual sensor data to the weighted moving average over a period of time.
  • the detector 302 also monitors the health of sensors 304.
  • the health is monitored by first calculating an estimated sensor signal from other sensor signals and the statistical model. The difference between the estimated sensor signal and actual sensor signal is compared. If the difference is not small and random, an alert is provided that the sensor has failed. The failed sensor is excluded from further model calculation until it is repaired or replaced. After a failed sensor has been repaired or replaced, the detector 302 waits until it enters the learning mode before it uses the sensor in the model calculation. The sensor health monitoring is repeated periodically for each sensor at an appropriate period of time. For most sensors, a time period of once per second is adequate.
  • the OEM server 108 in level 204 receives the sensor data transmitted by the proxy/gateway 104 and decompresses the data.
  • the OEM server 108 hosts an expert system that has a component pattern library and a systemic pattern library.
  • the OEM server 108 or its components may be integrated with other components of an installation, including operating as a software object on any processor or distributed processors having sufficient processing capability.
  • the component pattern library contains known component specific failure patterns.
  • the component pattern library may contain failure patterns for ball bearings, motors, gearboxes, cams, etc.
  • the systemic pattern library contains systemic patterns as diagnosed by human experts. This library is updated each time an expert identifies and classifies a new pattern. The patterns can be characterized either as normal operation or as a specific failure situation.
  • the expert system automatically generates a model of a machine's systemic behavior each time a pattern is added to the systemic pattern library.
  • the OEM server 108 compares the sensor data with known systemic patterns in the systemic pattern library using a model of systemic behavior (step 520). If there is a match between the sensor data and a specific failure pattern in the systemic pattern library (step 522), the OEM server 108 performs a failure report operation (step 528).
  • the sensor data analyzed for comparison is typically the transformed FFT data. Alternatively, the sensor data is a single sample of raw data (i.e., the sensor signals prior to signal processing) or a time- series set of data. The time-series set of data contains data sets that correspond to a point of time in a time line.
  • the last data set (i.e., the last point of data in the time line) is used to select a possible failure pattern as a hypothesis.
  • the hypothesis is compared to the other elements of the time-series set using an appropriate statistical tool to determine if the hypothesis is the likely cause of failure.
  • the failure report operation includes generating an action alert, generating a report, transmitting the action alert to selected maintenance individuals or to an enterprise asset management, an enterprise resource planning program, or any other maintenance management software operated by the party responsible for maintenance.
  • the report is added to a machine-specific database.
  • the action alert is provided to the party responsible for maintenance of the machine so that appropriate action may be taken.
  • the action alert includes a machine identification, a time stamp, an identification of the component that is likely to fail or that has failed, an estimated time of failure, and a recommended action (i.e., replace, align, check, clean, etc.)
  • the report added to the machine- specific database includes the action alert information and a portion of the sensor data for long term machine monitoring (e.g., historical data to see changes over time).
  • the sensor data is compared with known component patterns (step 524). If the sensor data matches a component pattern (step 526), the failure report operation (step 528) is performed. If there is no match, a component ID is assigned and transmitted to the directory server 1 12 in level 206 (step 530).
  • the component ID is a reference number uniquely describing a machine component, such as a ball bearing, motor or gearbox, etc..
  • the directory server 1 12 searches for OEM servers using the same component with the same component ID sent by the OEM server 108 in level 204 (i.e., the requesting OEM server) (step 540). If a component ID matches (step 542), the directory server 112 sends the server ID of one of the OEM servers with a matching component ID. The requesting OEM server and OEM server with a matching component ID establish a peer-to-peer connection and the data is sent to the OEM server with matching component ID for analysis (step 546). The OEM server with matching component ID compares the sensor data with the system and component pattern libraries (step 548).
  • the OEM server with matching component ID transmits the diagnosis and component pattern associated with the sensor data to the requesting OEM server 108 in level 204 (step 552).
  • the requesting OEM server 108 receives the information and performs the failure report operation (step 528).
  • steps 540 to 550 are repeated with other OEM servers 108 with matching component ID until either a match occurs or no further OEM servers 108 with matching component IDs are found.
  • peer-to-peer connections are established with several OEM servers with matching component IDs so that the OEM servers can perform the sensor data comparison in parallel. If no further OEM servers with matching component IDs are found (i.e., the sensor data does not match any known patterns), the directory server 112 informs the requesting OEM server and establishes a connection with expert network server 1 14 and transmits the sensor data to the expert network server 1 14 (step 544).
  • the expert network server 114 receives the sensor data and determines which experts to use.
  • the network server 114 identifies a lead expert from a group of experts that will become responsible for solving the problem and establishes a work session with the lead expert (step 560).
  • the group of experts is identified by matching the expertise of the experts with the type of machine that the detector 302 is monitoring.
  • the lead expert is selected based upon a list of criteria.
  • the list of criteria includes availability of the expert, cost, and urgency of the matter.
  • the group of experts may be narrowed down to those experts that are in an appropriate time zone to start the project (e.g., if the machine problem occurred in the middle of the night in the United States, the lead expert may be chosen from the group of experts residing in that part of the world where the working day is just starting).
  • the lead expert analyses the data and identifies specialists to solve the problem (step 564).
  • the specialists work together sharing the same information in a collaborative environment to solve the problem (step 564).
  • the collaborative environment allows the specialists to work together from remote locations.
  • the collaborative environment is a network that provides the specialists and experts with shared access to sensor and machine data, shared access to pattern libraries, document sharing, secure (and non-secure) communications, and the ability to track individual contributions.
  • the communications between the specialists can be voice, video, e-mail, instant messaging, co-browsing, etc. If the specialists chosen are unable to solve the problem (step 566), the lead expert selects other specialists to see if they are able to solve the problem and step 564 is repeated. The lead expert and selected specialists continue to work on the problem until the problem is solved.
  • the lead expert validates the solution and determines a failure diagnostic description for placing in the database of the OEM server 108 in level 204 (step 568).
  • the system and component patterns and diagnosis are transmitted to the OEM server 108 in level 204 (step 570).
  • the system and component patterns are transmitted to all of the OEM servers that have a component ID matching the component ID sent by the requesting OEM server.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Test And Diagnosis Of Digital Computers (AREA)
  • Debugging And Monitoring (AREA)

Abstract

L'invention concerne un système et un procédé de diagnostic à distance et de maintenance conditionnelle de dispositifs, machines, et systèmes. Le système détecte des signaux d'un ou plusieurs paramètres opérationnels et conditionnels et compare les signaux détectés à un modèle de signal conservé localement par rapport à l'emplacement du dispositif, de la machine, ou du système dans le but de détecter des anomalies. Les informations décrivant chaque anomalie sont transmises à un emplacement à distance à partir du dispositif, de la machine ou du système en vue d'être diagnostiqué. Le diagnostic comprend une analyse initiale des informations au moyen d'outils de diagnostic conservés dans un emplacement à distance, une analyse ultérieure des informations au moyen des outils de diagnostic conservés ailleurs si l'analyse initiale échoue afin de fournir un diagnostic et une analyse finale effectuée par un groupe de personnes aidées par un environnement collaboratif si les analyses initiales et ultérieures échouent afin de fournir un diagnostic. Ce diagnostic est transmis à un service de maintenance afin de réparer le dispositif, la machine ou le système.
PCT/IB2002/003409 2001-08-21 2002-08-07 Systeme et procede de diagnostic multiniveau echelonnable a distance et maintenance conditionnelle WO2003019377A2 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP02767748A EP1419442A2 (fr) 2001-08-21 2002-08-07 Systeme et procede de diagnostic multiniveau echelonnable a distance et maintenance conditionnelle
AU2002330673A AU2002330673A1 (en) 2001-08-21 2002-08-07 System and method for scalable multi-level remote diagnosis and predictive maintenance

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US09/934,000 US20030046382A1 (en) 2001-08-21 2001-08-21 System and method for scalable multi-level remote diagnosis and predictive maintenance
US09/934,000 2001-08-21
FR0116995A FR2828945B1 (fr) 2001-08-21 2001-12-28 Systeme et procede multi-niveaux de maintenance predictive et de diagnostic a distance extensible a un tres grand nombre de machines
FR01/16995 2001-12-28

Publications (2)

Publication Number Publication Date
WO2003019377A2 true WO2003019377A2 (fr) 2003-03-06
WO2003019377A3 WO2003019377A3 (fr) 2003-09-25

Family

ID=26213316

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2002/003409 WO2003019377A2 (fr) 2001-08-21 2002-08-07 Systeme et procede de diagnostic multiniveau echelonnable a distance et maintenance conditionnelle

Country Status (3)

Country Link
EP (1) EP1419442A2 (fr)
AU (1) AU2002330673A1 (fr)
WO (1) WO2003019377A2 (fr)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1536339A3 (fr) * 2003-11-25 2006-11-29 Hewlett-Packard Development Company, L.P. Support logiciel multi niveaux et multi sources utilisant des transfers de données de diagnostique automatiques
CN102023100A (zh) * 2010-04-19 2011-04-20 东莞市罗尔机电科技有限公司 设备故障预警系统及方法
US9069666B2 (en) 2003-05-21 2015-06-30 Hewlett-Packard Development Company, L.P. Systems and methods for controlling error reporting and resolution
EP3133451A1 (fr) * 2015-08-20 2017-02-22 Siemens Aktiengesellschaft Systeme de commande, de surveillance et de reglage et procedes de fonctionnement d'un tel systeme
US9893952B2 (en) 2015-01-09 2018-02-13 Microsoft Technology Licensing, Llc Dynamic telemetry message profiling and adjustment
CN110570052A (zh) * 2019-10-22 2019-12-13 常州机电职业技术学院 一种船用离心泵预测性维护方法
US11061395B2 (en) 2016-02-04 2021-07-13 Hewlett Packard Enterprise Development Lp Real-time alerts and transmission of selected signal samples under a dynamic capacity limitation
IT202100013157A1 (it) * 2021-05-21 2021-08-21 Diskover S R L Sistema di identificazione delle anomalie in ambito automotive

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB0427694D0 (en) 2004-12-17 2005-01-19 Ncr Int Inc A method of determining the cause of an error state in an apparatus

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5311562A (en) * 1992-12-01 1994-05-10 Westinghouse Electric Corp. Plant maintenance with predictive diagnostics
JP3147586B2 (ja) * 1993-05-21 2001-03-19 株式会社日立製作所 プラントの監視診断方法
WO1997015009A1 (fr) * 1995-10-18 1997-04-24 Systemsoft Corporation Systeme et procede de diagnostic pour processeurs de donnees numeriques
US5764509A (en) * 1996-06-19 1998-06-09 The University Of Chicago Industrial process surveillance system
EP0965092A4 (fr) * 1997-03-04 2002-10-30 Emerson Electric Co Systeme de diagnostic distribue
FI108678B (fi) * 1998-06-17 2002-02-28 Neles Controls Oy Kenttälaitteiden hallintajärjestelmä
US6594620B1 (en) * 1998-08-17 2003-07-15 Aspen Technology, Inc. Sensor validation apparatus and method

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7334028B2 (en) 2003-05-21 2008-02-19 Hewlett-Packard Development Company, L.P. Tiered multi-source software support using automated diagnostic-data transfers
US9069666B2 (en) 2003-05-21 2015-06-30 Hewlett-Packard Development Company, L.P. Systems and methods for controlling error reporting and resolution
EP1536339A3 (fr) * 2003-11-25 2006-11-29 Hewlett-Packard Development Company, L.P. Support logiciel multi niveaux et multi sources utilisant des transfers de données de diagnostique automatiques
CN102023100A (zh) * 2010-04-19 2011-04-20 东莞市罗尔机电科技有限公司 设备故障预警系统及方法
US9893952B2 (en) 2015-01-09 2018-02-13 Microsoft Technology Licensing, Llc Dynamic telemetry message profiling and adjustment
EP3133451A1 (fr) * 2015-08-20 2017-02-22 Siemens Aktiengesellschaft Systeme de commande, de surveillance et de reglage et procedes de fonctionnement d'un tel systeme
CN106468910A (zh) * 2015-08-20 2017-03-01 西门子公司 用于控制、监控和调节的系统和运行这种系统的方法
US11061395B2 (en) 2016-02-04 2021-07-13 Hewlett Packard Enterprise Development Lp Real-time alerts and transmission of selected signal samples under a dynamic capacity limitation
US11899444B2 (en) 2016-02-04 2024-02-13 Hewlett Packard Enterprise Development Lp Real-time alerts and transmission of selected signal samples under a dynamic capacity limitation
CN110570052A (zh) * 2019-10-22 2019-12-13 常州机电职业技术学院 一种船用离心泵预测性维护方法
IT202100013157A1 (it) * 2021-05-21 2021-08-21 Diskover S R L Sistema di identificazione delle anomalie in ambito automotive

Also Published As

Publication number Publication date
WO2003019377A3 (fr) 2003-09-25
AU2002330673A1 (en) 2003-03-10
EP1419442A2 (fr) 2004-05-19

Similar Documents

Publication Publication Date Title
US20030046382A1 (en) System and method for scalable multi-level remote diagnosis and predictive maintenance
US6317701B1 (en) Field device management system
US6298308B1 (en) Diagnostic network with automated proactive local experts
RU2417393C2 (ru) Система представления данных для предотвращения нестандартной ситуации на производственном предприятии
EP2193413B1 (fr) Système de conservation et d'affichage de données de contrôle d'un processus associées à une situation anormale
EP1808768B1 (fr) Surveillance à distance automatique et système de diagnostic ainsi que procédé de communication pour la communication entre un automate programmable et une unité centrale
CN110830943A (zh) 一种基于边缘计算和大数据分析的设备状态监测系统
JP2006522386A (ja) プロセスコントロールシステムと使用のためのオープンネットワークベースのデータの収集、集約処理及び最適化
CN119059385A (zh) 一种基于边缘计算和算网融合的电梯智能管理系统
WO2004072749A1 (fr) Installations exterieures pour assurer le diagnostic et les services de maintenance a distance vers une installation de traitement
WO2008116966A2 (fr) Procédé et appareil pour surveiller l'état de machines électriques
KR100354786B1 (ko) 네트워크를 이용한 공조 및 냉동/냉장 설비의통합관리시스템 및 그 제어방법
WO2003019377A2 (fr) Systeme et procede de diagnostic multiniveau echelonnable a distance et maintenance conditionnelle
Ucar et al. E-maintenance in support of e-automated manufacturing systems
WO2008087191A1 (fr) Supervision de réseau avec des systèmes de commande
KR101996237B1 (ko) 분산처리를 통한 IoT 센서 디바이스 및 플랫폼
WO2003040882A2 (fr) Procede et appareil de controle, de surveillance et d'analyse de systemes independants dans une unite de production
KR20220122922A (ko) Ai기반 데이터 로거 시스템
CN119043498A (zh) 一种空预器的红外成像监测系统
CN118659992A (zh) 基于开源鸿蒙的plc网关设备运行状态监测系统及方法
KR20080050696A (ko) 원격상태 감시기반 엔진운전 및 유지보수 관리 시스템
WO2001050099A1 (fr) Reseau diagnostique avec module experts locaux proactifs informatises
KR100928291B1 (ko) 발전기용 엔진 원격 제어방법
Tse et al. Web and virtual instrument based machine remote sensing, monitoring and fault diagnostic system
CN116700197B (zh) 一种工控监测分析预警系统及分析预警处理方法

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ OM PH PL PT RO RU SD SE SG SI SK SL TJ TM TN TR TT TZ UA UG UZ VC VN YU ZA ZM ZW

Kind code of ref document: A2

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BY BZ CA CH CN CO CR CU CZ DE DM DZ EC EE ES FI GB GD GE GH HR HU ID IL IN IS JP KE KG KP KR LC LK LR LS LT LU LV MA MD MG MN MW MX MZ NO NZ OM PH PL PT RU SD SE SG SI SK SL TJ TM TN TR TZ UA UG UZ VC VN YU ZA ZM

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): GH GM KE LS MW MZ SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR IE IT LU MC NL PT SE SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

Kind code of ref document: A2

Designated state(s): GH GM KE LS MW MZ SD SL SZ UG ZM ZW AM AZ BY KG KZ RU TJ TM AT BE BG CH CY CZ DK EE ES FI FR GB GR IE IT LU MC PT SE SK TR BF BJ CF CG CI GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2002767748

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 2002767748

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: JP

WWW Wipo information: withdrawn in national office

Country of ref document: JP