Electrical anomaly detection method and system

{
  "publication_number": "US2011153236A1",
  "country": "US",
  "kind": "A1",
  "title": "Electrical anomaly detection method and system",
  "abstract": "There are provided a method and a system for real time monitoring of a powered on electrical network in a facility in order to detect an electrical anomaly while the electrical network is in use. The system comprises a sensor data receiver for receiving real time sensor data including at least one measured value measured in real time by at least one sensor installed on a conductor of the electrical network in the facility while the electrical network is in use; an anomaly detector for retrieving an anomaly detection rule from an anomaly detection rule database, the rule having an identification of a required input, a formula, and a threshold reference value, for a detection of an anomaly; receiving the real time sensor data from the sensor data receiver and extracting at least one relevant measured value from the at least one measured value using the identification of the required input; comparing the at least one relevant measured value to the threshold reference value according to the formula to determine one of a presence and an absence of the anomaly in the real time sensor data; an event generator controlled by the processor for retrieving and providing anomaly monitoring data if the anomaly is determined to be present by the processor, the anomaly monitoring data including an indication of a monitoring course of action to be carried out to address the anomaly determined to be present.",
  "claims": [
    "1. A method for real time monitoring of an electrical network in a facility in order to detect an electrical anomaly, the method comprising: use of a processor for carrying out the steps of: receiving real time sensor data including at least one measured value measured in real time by at least one sensor installed on a conductor of said electrical network in said facility, each said at least one measured value being one of a current value and a voltage value; retrieving an anomaly detection rule, said rule having an identification of a required input, a formula, and a threshold reference value, for a detection of an anomaly, said threshold reference value being determined using a nominal value for said input for said relevant element; extracting at least one relevant measured value from said at least one measured value using said identification of said required input; comparing said at least one relevant measured value to said threshold reference value according to said formula to determine one of a presence and an absence of said anomaly in said real time sensor data; and if said anomaly is determined to be present, retrieving and providing anomaly monitoring data, said anomaly monitoring data including an indication of a monitoring course of action to be carried out to address said anomaly determined to be present. 2. The method as claimed in claim 1, further comprising generating an indication that said electrical anomaly is detected and outputting a message including said anomaly monitoring data. 3. The method as claimed in claim 2, further comprising formatting said real time sensor data into formatted data, and wherein said outputting said message comprises including said formatted data with said anomaly monitoring data. 4. The method as claimed in claim 1, wherein said receiving real time sensor data comprises receiving real time sensor data including at least one measured value measured in real time by sixteen current sensors and a voltage sensor. 5. The method as claimed in claim 1, further comprising: retrieving potential sources data for said anomaly determined to be present including a list of potential sources for said anomaly; retrieving installation data for said electrical network; determining a source of said electrical anomaly in said electrical network using said list of potential sources, said real time sensor data and said installation data; and wherein said providing anomaly monitoring data comprises providing an indication of said determined source. 6. The method as claimed in claim 1, further comprising: retrieving type data for said anomaly determined to be present including a list of types for said anomaly and a corresponding detection rule for each of said types; determining a type of said electrical anomaly in said electrical network using said list of types, said corresponding detection rule and said real time sensor data; and wherein said providing anomaly monitoring data comprises providing an indication of said determined type. 7 - 8. (canceled) 9. The method as claimed in claim 1, wherein said anomaly is one of: an erroneous installation of said sensor, an electrical arc, a short circuit, an external intervention on a current limitation device, a bonding, a leakage of current, an overcurrent, a lost load, a harmonic distortion anomaly, a mechanical jam, an unbalanced current, an overload of a load, a repetitive start, a heavy start, a thermal protection anomaly, a power failure, an unbalanced voltage, a frequency value anomaly, a voltage value anomaly, a loss of a neutral conductor, and a temperature anomaly. 10. The method as claimed in claim 9, wherein said anomaly is said electrical arc, said at least one sensor is a current sensor, and said at least one measured value is a current value, said identification of said required input is a value for a current, said formula is performing a standard deviation on a vector of subtractions of said value for said current for a previous cycle by another current value, for a number n of cycles, said another current value being one of a current value for a present cycle and a current value for an average cycle, said threshold reference value is a reference value for said standard deviation; wherein said extracting comprises extracting n current acquired signals from said real time sensor data; and wherein said comparing comprises applying said formula to said n current acquired signals and comparing said standard deviation to said reference value to determine one of a presence and an absence of said electrical arc. 11. (canceled) 12. The method as claimed in claim 10, further comprising determining said current value for an average cycle by accumulating a number m of current values over m cycles and performing an average on said m current values to obtain said current value for said average cycle. 13. The method as claimed in claim 9, wherein said anomaly is said external intervention on said current limitation device, said at least one sensor is a current sensor, and said at least one measured value is two sets of values from consecutive time periods, each set including a current presence value, a voltage presence value and a load presence value, said identification of said required input is a status of a current, a status of a voltage and a status of a load, said formula is subtracting each value from one of said two consecutive sets from a corresponding value in the other one of said two consecutive sets, said threshold reference value is a transition value; wherein said extracting comprises extracting an indication of a status of a current from said current presence value, of a status of a voltage from said voltage presence value and a status of a load from said load presence value from said real time sensor data; wherein said comparing comprises applying said formula and comparing said obtained set of subtracted values to said transition value to determine an occurrence of a abnormal transition if one of said transition for each of said voltage, said current and said load is from a present status to an absent status over said two consecutive time periods; and said transition for said voltage is from a present status to an absent status over said two consecutive time periods and said status of said current and said load is an absent status and said transition for said current and said load is nil; further comprising retrieving a status for other anomalies, said status being one of presence of at least one other anomaly and absence of any other anomaly; and wherein said comparing further comprises if said status for other anomalies is said absence of any other anomaly and said abnormal transition is determined to be occurred, determining a presence of said external intervention on a current limitation device anomaly. 14. The method as claimed in claim 13, further comprising determining said current presence value by comparing a measured current value with a current presence threshold value, determining said voltage presence value by comparing a measured voltage value with a voltage presence threshold value and determining said load presence value by comparing a measured load value with a load presence threshold value. 15. The method as claimed in claim 9, wherein said anomaly is said erroneous installation of said sensor, said at least one sensor is a current sensor having at least two conductor holes, and said at least one measured value is a measured voltage value for each of said at least two conductor holes of said sensor, said identification of said required input is a value for a voltage, said threshold reference value is a threshold voltage value, said formula is subtracting said value for said voltage by said threshold voltage value; wherein said comparing comprises applying said formula to each said measured voltage value and further comprises: if said measured voltage value is greater than said threshold voltage value, determining said corresponding one of said sensor holes to be a line conductor; if said measured voltage value is smaller than said threshold voltage value, determining said corresponding one of said sensor holes to be a neutral conductor; determining a number of holes having been determined to be a neutral conductor to be neutral holes, if said number of neutral holes is greater than one, determining a presence of said erroneous installation anomaly and determining a type of said erroneous installation anomaly to be a number of neutral holes anomaly. 16. The method as claimed in claim 15, wherein said comparing further comprises: determining a number of holes having been determined to be a line conductor to be a number of live holes; if said number of live holes is zero, determining a presence of said erroneous installation anomaly and determining a type of said erroneous installation anomaly to be a number of live holes anomaly. 17. The method as claimed in claim 16, wherein said at least one measured value includes a measured phase value for each of said at least two conductor holes of said sensor, said identification of said required input includes a value for a phase, said threshold reference value includes a single phase threshold phase value and a three-phase threshold phase value, said formula further includes subtracting a phase shift between phases of each line conductors with one of said single phase threshold phase value and said three-phase threshold phase value; wherein said comparing further comprises: if said number of live holes is two, determining a phase shift between said phase measured value of said line conductors, comparing said phase shift with said single phase threshold phase value, if said erroneous installation anomaly is determined to be present, determining a type of said erroneous installation anomaly to be a phase shift anomaly; if said number of live holes is three, determining three phase shifts between said phase measured value of said line conductors, comparing each of said three phase shifts with said three-phase threshold phase value, if said erroneous installation anomaly is determined to be present, determining a type of said erroneous installation anomaly to be a phase shift anomaly; if said number of live holes is four, determining a presence of an erroneous installation anomaly and determining a type of said erroneous installation anomaly to be a live hole anomaly. 18. A system for real time monitoring of an electrical network in a facility in order to detect an electrical anomaly, the system comprising: a sensor data receiver for receiving real time sensor data including at least one measured value measured in real time by at least one sensor installed on a conductor of said electrical network, each said at least one measured value being one of a current value and a voltage value; an anomaly detector for retrieving an anomaly detection rule from an anomaly detection rule database, said rule having an identification of a required input, a formula, and a threshold reference value, for a detection of an anomaly, said threshold reference value being determined using a nominal value for said input for said relevant element; receiving said real time sensor data from said sensor data receiver and extracting at least one relevant measured value from said at least one measured value using said identification of said required input; comparing said at least one relevant measured value to said threshold reference value according to said formula to determine one of a presence and an absence of said anomaly in said real time sensor data; an event generator controlled by said processor for retrieving and providing anomaly monitoring data if said anomaly is determined to be present by said processor, said anomaly monitoring data including an indication of a monitoring course of action to be carried out to address said anomaly determined to be present. 19 - 20. (canceled) 21. The system as claimed in claim 18, further comprising a source locator for retrieving potential sources data for said anomaly determined to be present from said anomaly detection rule database including a list of potential sources for said anomaly; retrieving installation data for said electrical network from a system installation database; and determining a source of said electrical anomaly in said electrical network using said list of potential sources, said real time sensor data and said installation data; wherein said event generator provides an indication of said determined source. 22. The system as claimed in claim 18, further comprising a type determiner for retrieving type data for said anomaly determined to be present from said anomaly detection rule database including a list of types for said anomaly and a corresponding detection rule for each of said types; and determining a type of said electrical anomaly in said electrical network using said list of types, said corresponding detection rule and said real time sensor data; wherein said event generator provides an indication of said determined type. 23 - 24. (canceled) 25. The system as claimed in claim 18, wherein said anomaly is one of: an erroneous installation of said sensor, an electrical arc, a short circuit, an external intervention on a current limitation device, a bonding, a leakage of current, an overcurrent, a lost load, a harmonic distortion anomaly, a mechanical jam, an unbalanced current, an overload of a load, a repetitive start, a heavy start, a thermal protection anomaly, a power failure, an unbalanced voltage, a frequency value anomaly, a voltage value anomaly, a loss of a neutral conductor, and a temperature anomaly. 26. The system as claimed in claim 25, wherein said anomaly is said electrical arc, said at least one sensor is a current sensor, and said at least one measured value is a current value, said identification of said required input is a value for a current, said formula is performing a standard deviation on a vector of subtractions of said value for said current for a previous cycle by another current value, for a number n of cycles, said another current value being one of a current value for a present cycle and a current value for an average cycle, said threshold reference value is a reference value for said standard deviation; wherein said anomaly detector extracts n current acquired signals from said real time sensor data and applies said formula to said n current acquired signals and compares said standard deviation to said reference value to determine one of a presence and an absence of said electrical arc. 27. The system as claimed in claim 25, wherein said anomaly is said external intervention on a current limitation device, said at least one sensor is a current sensor, and said at least one measured value is two sets of values from consecutive time periods, each set including a current presence value, a voltage presence value and a load presence value, said identification of said required input is a status of a current, a status of a voltage and a status of a load, said formula is subtracting each value from one of said two consecutive sets from a corresponding value in the other one of said two consecutive sets, said threshold reference value is a transition value; wherein said anomaly detector extracts an indication of a status of a current from said current presence value, of a status of a voltage from said voltage presence value and a status of a load from said load presence value from said real time sensor data; applies said formula and compares said obtained set of subtracted values to said transition value to determine an occurrence of a abnormal transition if one of said transition for each of said voltage, said current and said load is from a present status to an absent status over said two consecutive time periods; and said transition for said voltage is from a present status to an absent status over said two consecutive time periods and said status of said current and said load is an absent status and said transition for said current and said load is nil; further comprising a system communicator for retrieving a status for other anomalies, said status being one of presence of at least one other anomaly and absence of any other anomaly; and wherein said anomaly detector obtains said status for other anomalies from said system communicator and if said status for other anomalies is said absence of any other anomaly and said abnormal transition is determined to be occurred, determines a presence of said external intervention on a current limitation device anomaly. 28. The system as claimed in claim 25, wherein said anomaly is an erroneous installation of said sensor, said at least one sensor is a current sensor having at least two conductor holes, and said at least one measured value is a measured voltage value for each of said at least two conductor holes of said sensor, said identification of said required input is a value for a voltage, said threshold reference value is a threshold voltage value, said formula is subtracting said value for said voltage by said threshold voltage value; wherein said anomaly detector applies said formula to each said measured voltage value; if said measured voltage value is greater than said threshold voltage value, determines said corresponding one of said sensor holes to be a line conductor; if said measured voltage value is smaller than said threshold voltage value, determines said corresponding one of said sensor holes to be a neutral conductor; determines a number of holes having been determined to be a neutral conductor to be neutral holes, if said number of neutral holes is greater than one, determines a presence of said erroneous installation anomaly and determines a type of said erroneous installation anomaly to be a number of neutral holes anomaly. 29. The system as claimed in claim 28, wherein said anomaly detector: determines a number of holes having been determined to be a line conductor to be a number of live holes; if said number of live holes is zero, determines a presence of said erroneous installation anomaly and determines a type of said erroneous installation anomaly to be a number of live holes anomaly. 30. The system as claimed in claim 29, wherein said at least one measured value includes a measured phase value for each of said at least two conductor holes of said sensor, said identification of said required input includes a value for a phase, said threshold reference value includes a single phase threshold phase value and a three-phase threshold phase value, said formula further includes subtracting a phase shift between phases of each line conductors with one of said single phase threshold phase value and said three-phase threshold phase value; wherein said anomaly detector if said number of live holes is two, determines a phase shift between said phase measured value of said line conductors, compares said phase shift with said single phase threshold phase value, if said erroneous installation anomaly is determined to be present, determines a type of said erroneous installation anomaly to be a phase shift anomaly; if said number of live holes is three, determines three phase shifts between said phase measured value of said line conductors, compares each of said three phase shifts with said three-phase threshold phase value, if said erroneous installation anomaly is determined to be present, determines a type of said erroneous installation anomaly to be a phase shift anomaly; if said number of live holes is four, determines a presence of an erroneous installation anomaly and determines a type of said erroneous installation anomaly to be a live hole anomaly."
  ],
  "description_excerpt": "This invention relates to electrical anomaly detection in electrical distribution and grounding networks and more particularly, to detecting an anomaly and providing course of action data.\n\nElectrical networks are used in industrial applications such as, for example, in production, manufacturing, industrial or agricultural facilities. These networks, when implemented in large scale facilities, may comprise various electrical equipments such as motors, pumps, heaters, and cooling systems to accomplish various industrial tasks. Electrical networks used in a large scale industrial application tend to be difficult to manage, for example, because of the number of electrical elements, their complexity in terms of electrical connections, the efficiency of the network planning, the lifetime of the network, etc. Inappropriate management, replacement or planning of electrical equipments in a large scale industrial application may affect the electrical network and render the electrical equipment or a part of the electrical network unusable, and thus decrease the electrical signal quality.\n\nIn animal livestock facilities, the electrical networks are placed in a very demanding environment of high humidity, ammonia gas, frequent spraying of water and electrical equipment operating at many different frequencies.\n\nContinuous real-time monitoring of electrical networks, with or without loads, is necessary to identify any events or faulty equipment over an entire electrical network. However, this is not currently achieved.",
  "cpc": [
    "G01R 19/2513",
    "G01R 31/088",
    "G01R 31/343",
    "G01R 31/52",
    "H02H 3/00"
  ],
  "ipc": [
    "G06F 19/00",
    "G01R 31/00"
  ],
  "inventors": [
    "Michel Montreuil",
    "Karl Fecteau"
  ],
  "filing_date": "2009-04-14",
  "publication_date": "2011-06-23",
  "priority_date": "2008-04-14",
  "application_number": "US-93775309-A",
  "family_id": "41198747",
  "cited_by_count": 102,
  "citations": [
    "US3911323A",
    "US4296450A",
    "US4398188A",
    "US5123017A",
    "US5250894A",
    "US5452223A",
    "US5680025A",
    "US6308140B1",
    "US5739698A",
    "US5825170A",
    "US5839092A",
    "US6175810B1",
    "US6292717B1",
    "US6963475B2",
    "US6396283B1",
    "US6631063B2",
    "US6989977B2",
    "US7254004B2",
    "US7236338B2",
    "US7161393B1",
    "US7191074B2",
    "US20060176631A1",
    "US20060215335A1",
    "US20070265713A1",
    "US7368918B2",
    "US20080180866A1",
    "US20130060524A1"
  ]
}