WO2022105286A1 - 一种基于递进式识别来监测异常状态的系统及方法 - Google Patents

一种基于递进式识别来监测异常状态的系统及方法 Download PDF

Info

Publication number
WO2022105286A1
WO2022105286A1 PCT/CN2021/107897 CN2021107897W WO2022105286A1 WO 2022105286 A1 WO2022105286 A1 WO 2022105286A1 CN 2021107897 W CN2021107897 W CN 2021107897W WO 2022105286 A1 WO2022105286 A1 WO 2022105286A1
Authority
WO
WIPO (PCT)
Prior art keywords
target device
signal
abnormal state
monitoring terminal
identification
Prior art date
Application number
PCT/CN2021/107897
Other languages
English (en)
French (fr)
Inventor
郭春林
郭尔富
Original Assignee
华北电力大学
北京大地纵横科技有限公司
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
Application filed by 华北电力大学, 北京大地纵横科技有限公司 filed Critical 华北电力大学
Publication of WO2022105286A1 publication Critical patent/WO2022105286A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • the present invention relates to the technical field of equipment detection, and more particularly, to a system and method for monitoring abnormal state based on progressive identification.
  • Health status monitoring and diagnosis are key technologies for equipment operation and maintenance.
  • Existing monitoring technologies all directly diagnose the state of equipment based on signal characteristics. This method has the disadvantages of low diagnostic accuracy and inability to determine the cause of abnormality.
  • equipment in normal operation is rarely damaged, so it is time-consuming and inefficient to inspect all aspects of equipment every time. Due to the low rate of failures identified during normal operation, it is necessary to first determine whether there is an abnormality in the equipment in certain cases. Comprehensive testing is not required to determine the absence of anomalies.
  • the present application proposes a system and method for monitoring abnormal state based on progressive identification, firstly diagnosing the health state accurately, and determining the cause, location, components, etc. of damage only when the abnormal health state is determined.
  • a system for monitoring abnormal state based on progressive identification comprising:
  • At least one monitoring terminal used for real-time monitoring of the running state of the target device to be monitored, and setting each monitoring terminal in the at least one monitoring terminal at a different position of the target device;
  • Each of the monitoring terminals includes:
  • At least one signal sensor for monitoring the device signal of the target device in operation
  • the sampling unit samples the information monitored by each signal sensor to obtain the equipment signal
  • a control unit according to the control instructions, to control the start-up operation, stop operation, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode of each monitoring terminal;
  • an information processing server which communicates with at least one monitoring terminal through a wired communication link and/or a wireless communication link, and receives device signals from at least one monitoring terminal based on the wired communication link and/or wireless communication link;
  • a dedicated identification component associated with the target device from a plurality of dedicated identification components, and use the dedicated identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result;
  • the device signal is identified by using a common identification component to determine the abnormal information of the target device in the abnormal state.
  • a system for monitoring abnormal state based on progressive identification comprising:
  • At least one monitoring terminal used for real-time monitoring of the running state of the target device to be monitored, and setting each monitoring terminal in the at least one monitoring terminal at a different position of the target device;
  • Each of the monitoring terminals includes:
  • At least one signal sensor for monitoring the device signal of the target device in operation
  • the sampling unit samples the information monitored by each signal sensor to obtain the equipment signal
  • a communication unit that communicates with the information processing server through a wired communication link and/or a wireless communication link, and when it is determined that the target device is in an abnormal state, sends the device signal to the information processing via the wired communication link and/or the wireless communication link server;
  • a control unit according to the control instructions, to control the start-up operation, stop operation, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode of each monitoring terminal;
  • the identification unit uses a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result;
  • the information processing server communicates with at least one monitoring terminal through a wired communication link and/or a wireless communication link, receives device signals from at least one monitoring terminal based on the wired communication link and/or wireless communication link, and uses a common identification component to The device signal is identified to determine the abnormal information of the target device in the abnormal state.
  • the information processing server and/or each monitoring terminal calculates, sets, fits, and adjusts based on the measured data, historical data, statistical data, rated parameters or/and basic parameters of multiple devices of each device type within a predetermined range Or train to generate common recognition components;
  • the common identification component is capable of identifying abnormal information associated with each device type that is in an abnormal state
  • the information processing server and/or each monitoring terminal performs calculation, setting, fitting, Tune or train to generate specialized recognition components associated with devices of a particular device type;
  • a dedicated identification component associated with a device of a particular device type can determine whether the device of a particular device type is in an abnormal state.
  • the information processing server and/or each monitoring terminal generates a specific identification of the target device based on the calculation, setting, fitting, adjustment or training of the measured data, historical data, statistical data, rated parameters or/and basic parameters of the target device components;
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • the information processing server and/or each monitoring terminal generates a dedicated identification component of the target device based on training the measured data and/or historical data of the target device;
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • It also includes identifying the device signal with a common identification component to determine the abnormality degree, abnormality type, abnormality component, abnormality cause, abnormality location, health status and/or remaining life of the target device in the abnormal state.
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection , breakdown, pipe burst, burst, cable rack collapse, cable drop, digging, shovel, impact, landslide, foreign object intrusion, foreign object collision or partial discharge;
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection Lines, flashovers, partial discharges, tower collapses, loose tower foundations, loose connectors, broken connectors, galloping, icing or impact from foreign objects.
  • the special identification component is used to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result, including:
  • the special identification component performs feature extraction on the device signal of the collected target device, so as to obtain the device feature data of the collected target device;
  • the special identification component is used to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result, including:
  • the special identification component performs feature extraction on the device signal of the collected target device, thereby obtaining the device feature data of the collected target device, and separates the collected device feature data of the target device with the device feature data when the target device is in a normal state.
  • the comparison is performed, so that the comparison result is determined as the identification result, and whether the target device is in an abnormal state is determined based on the identification result.
  • the device signal is identified by the public identification component to determine the abnormal information of the target device in the abnormal state, including:
  • the public identification component performs feature extraction on the collected device signal of the target device, thereby obtaining the collected device feature data of the target device;
  • the abnormality information of the target device in the abnormal state is determined according to the identification result.
  • the device signal is identified by the public identification component to determine the abnormal information of the target device in the abnormal state, including:
  • the public identification component performs feature extraction on the collected device signal of the target device, thereby obtaining the collected device feature data of the target device;
  • the abnormality information of the target device in the abnormal state is determined according to the identification result.
  • the abnormal information of the target device whose abnormal state is determined according to the identification result includes:
  • the matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is used as the identification result
  • the abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • It also includes determining the existence range of the abnormal state, and determining the range of the target equipment that needs to be inspected and/or repaired according to the existence range of the abnormal state.
  • the determining whether the target device is in an abnormal state based on the identification result includes:
  • the collected device characteristic data of the target device and the device characteristic data when the target device is in a normal state are respectively identified, so as to The identification result is determined, and whether the target device is in an abnormal state is determined based on the identification result.
  • the determining whether the target device is in an abnormal state based on the identification result includes:
  • the abnormal information of the target device whose abnormal state is determined according to the identification result includes:
  • the collected equipment characteristic data of the target equipment are respectively compared with the existing equipment characteristic data associated with each abnormal information. Identifying to obtain the matching degree of the collected device feature data of the target device with the existing device feature data associated with each abnormal information;
  • the matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is used as the matching result
  • the abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • the device feature data includes at least one of a time domain feature, a frequency domain feature, a time-frequency domain feature, a power spectrum feature, a spectral envelope feature, a differential feature, an integral feature, and an artificial intelligence feature.
  • the determining whether the target device is in an abnormal state based on the identification result includes:
  • Whether or not the target device is in an abnormal state is determined according to whether at least one of the time domain feature, frequency domain feature, time-frequency domain feature, power spectrum feature, spectral envelope feature, differential feature, integral feature, and artificial intelligence feature exceeds a set threshold state.
  • Each signal sensor includes at least one voiceprint sensor, and the voiceprint sensor is arranged on the target device in a fit manner to monitor the device signal of the target device in operation;
  • the monitored equipment signals are voiceprint signals, video signals, infrared video signals, weather signals, position signals, electrical signals and/or electromagnetic wave signals.
  • the target device is an overhead transmission line, or the target device is a power transformer.
  • the target equipment is an underground cable, a power distribution cabinet, a ring network cabinet, a switch cabinet, a circuit breaker, an isolating switch, a grounding switch, a reactor, a capacitor, a resistor, a wind turbine or a solar cell device.
  • the information processing server can display or play device signals, the operating status of the target device and/or the abnormal characteristics of the target device.
  • the information processing server sequentially displays or plays the device signal, the running state of the target device and/or the abnormal feature of the target device according to the set sequence and time interval.
  • a method for monitoring an abnormal state based on progressive identification comprising: using at least one monitoring terminal to monitor the running state of a target device to be monitored in real time, wherein the at least one monitoring terminal is Each monitoring terminal in the terminal is set at different positions of the target device; at least one signal sensor of each monitoring terminal is used to monitor the device signal of the target device in operation; the sampling unit of each monitoring terminal is used to pair The information monitored by each signal sensor is sampled to obtain equipment signals; the communication unit of each monitoring terminal is used to communicate with the information processing server through a wired communication link and/or a wireless communication link; using the communication unit of each monitoring terminal The control unit controls the start-up operation, stop-operation, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode of each monitoring terminal according to the control instructions; use the information processing server through wired communication link and/or wireless communication link to communicate with at least one monitoring terminal, receiving device signals from at least one monitoring terminal based on wired communication link
  • a method for monitoring an abnormal state based on progressive identification comprising: using at least one monitoring terminal to monitor the running state of a target device to be monitored in real time, wherein the at least one monitoring terminal is Each monitoring terminal in the terminal is set at different positions of the target device; at least one signal sensor of each monitoring terminal is used to monitor the device signal of the target device in operation; the sampling unit of each monitoring terminal is used to pair The information monitored by each signal sensor is sampled to obtain equipment signals; the communication unit of each monitoring terminal is used to communicate with the information processing server through a wired communication link and/or a wireless communication link; using the communication unit of each monitoring terminal The control unit controls the start-up operation, stop operation, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode of each monitoring terminal according to the control instructions; using the identification unit of each monitoring terminal, The device signal is identified by a dedicated identification component to determine whether the target device is in an abnormal state based on the identification result; the information processing server
  • generating a common identification component based on the calculation, setting, fitting, adjustment or training of measured data, historical data, statistical data, rated parameters or/and basic parameters of a plurality of devices of each device type within a predetermined range;
  • the common identification component is capable of identifying abnormal information associated with each device type that is in an abnormal state
  • a dedicated identification component associated with the device based on the calculation, setting, fitting, adjustment or training of measured data, historical data, statistical data, rated parameters and/or basic parameters of multiple devices of a specific device type within a predetermined range to generate a specific device type.
  • a dedicated identification component associated with the device based on the calculation, setting, fitting, adjustment or training of measured data, historical data, statistical data, rated parameters and/or basic parameters of multiple devices of a specific device type within a predetermined range to generate a specific device type.
  • a dedicated identification component associated with a device of a particular device type can determine whether the device of a particular device type is in an abnormal state.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • It also includes identifying the device signal with a common identification component to determine the abnormality degree, abnormality type, abnormality component, abnormality cause, abnormality location, health status and/or remaining life of the target device in the abnormal state.
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection , breakdown, pipe burst, burst, cable rack collapse, cable drop, digging, shovel, impact, landslide, foreign object intrusion, foreign object collision or partial discharge;
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection Lines, flashovers, partial discharges, tower collapses, loose tower foundations, loose connectors, broken connectors, galloping, icing or impact from foreign objects.
  • a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device.
  • Feature data Identify the collected device feature data of the target device based on the artificial intelligence discrimination subcomponent and/or the neural network discrimination subcomponent to determine the identification result, and determine whether the target device is in an abnormal state based on the identification result.
  • the use of a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device. Feature data, compare the collected device feature data of the target device with the device feature data when the target device is in a normal state, so as to determine the comparison result as the identification result, and determine whether the target device is abnormal based on the identification result. state.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device Data; based on the artificial intelligence identification sub-component and/or the neural network identification sub-component, identify the collected device characteristic data of the target device to determine the identification result, and determine the abnormal information of the target device in an abnormal state according to the identification result.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device data; based on the device identification of the target device, obtain the existing device feature data associated with each abnormal information in all abnormal information of the target device from the database; respectively, compare the collected device feature data of the target device with each abnormal information The associated existing equipment feature data is compared, and the comparison result is used as the identification result;
  • the abnormality information of the target device in the abnormal state is determined according to the identification result.
  • Determining the abnormal information of the target device in the abnormal state according to the identification result includes: comparing the collected device feature data of the target device with the existing device feature data associated with each abnormal information, so as to obtain the collected data of the target device.
  • the matching degree between the device feature data of the target device and the existing device feature data associated with each abnormal information; the matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is taken as The identification result; the abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • It also includes determining the existence range of the abnormal state, and determining the range of the target equipment that needs to be inspected and/or repaired according to the existence range of the abnormal state.
  • the determining whether the target device is in an abnormal state based on the identification result includes:
  • the collected device characteristic data of the target device and the device characteristic data when the target device is in a normal state are respectively identified, so as to The identification result is determined, and whether the target device is in an abnormal state is determined based on the identification result.
  • the determining whether the target device is in an abnormal state based on the identification result includes:
  • the abnormal information of the target device whose abnormal state is determined according to the identification result includes:
  • the collected equipment characteristic data of the target equipment are respectively compared with the existing equipment characteristic data associated with each abnormal information. Identifying to obtain the matching degree of the collected device feature data of the target device with the existing device feature data associated with each abnormal information;
  • the matching degree of the collected device feature data of the target device and the existing device feature data associated with each abnormal information is used as the matching result; the abnormal information with the largest matching degree with the collected device feature data of the target device is determined as being in Abnormal information of the target device in abnormal state.
  • the device feature data includes at least one of a time domain feature, a frequency domain feature, a time-frequency domain feature, a power spectrum feature, a spectral envelope feature, a differential feature, an integral feature, and an artificial intelligence feature.
  • the determining whether the target device is in an abnormal state based on the identification result includes: according to time domain features, frequency domain features, time-frequency domain features, power spectrum features, spectrum envelope features, differential features, integral features and artificial intelligence features. Whether at least one exceeds a set threshold value determines whether the target device is in an abnormal state.
  • Each signal sensor includes at least one voiceprint sensor, and the voiceprint sensor is arranged on the target device in a fit manner to monitor the device signal of the target device in operation; the monitored device signal is a voiceprint signal, video signal, infrared video signal, weather signal, position signal, electrical signal and/or electromagnetic wave signal.
  • the target device is an overhead transmission line, or the target device is a power transformer.
  • the target equipment is an underground cable, a power distribution cabinet, a ring network cabinet, a switch cabinet, a circuit breaker, an isolating switch, a grounding switch, a reactor, a capacitor, a resistor, a wind turbine or a solar cell device.
  • the information processing server can display or play device signals, the operating status of the target device and/or the abnormal characteristics of the target device.
  • the information processing server sequentially displays or plays the device signal, the running state of the target device and/or the abnormal feature of the target device according to the set sequence and time interval.
  • the technical solution according to the present invention can be widely applied to overhead transmission lines, underground cables, power transformers, power distribution cabinets, ring network cabinets, switch cabinets, circuit breakers, isolating switches, grounding switches, inductors, capacitors, resistors, wind power generation Machines, solar cell devices and other equipment, with good application value.
  • FIG. 1 is a schematic structural diagram of a system for monitoring abnormal states based on progressive identification according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a system for monitoring abnormal states based on progressive identification according to another embodiment of the present invention.
  • FIG. 3 is a schematic diagram of monitoring a target device according to an embodiment of the present invention.
  • FIG. 4 is a flowchart of a method for monitoring an abnormal state based on progressive identification according to an embodiment of the present invention
  • FIG. 5 is a flowchart of a method for monitoring an abnormal state based on progressive identification according to another embodiment of the present invention.
  • a signal eg, a device signal, an input signal, and/or an environmental signal, etc.
  • a signal should be understood broadly, or broadly, as a directly acquired signal or the result after it has been processed.
  • FIG. 1 is a schematic structural diagram of a system for monitoring an abnormal state based on progressive identification according to an embodiment of the present invention.
  • the system includes at least one monitoring terminal and an information processing server.
  • At least one monitoring terminal includes monitoring terminal 1, monitoring terminal 2, . . . , monitoring terminal n.
  • Each of the at least one monitoring terminal is used for real-time monitoring of the running state of the target device to be monitored.
  • each monitoring terminal in the at least one monitoring terminal is set at different positions of the target device.
  • each of the at least one monitoring terminal includes: at least one signal sensor, a sampling unit, a communication unit and a control unit.
  • Each signal sensor is disposed at the target device in a fit manner for monitoring the device signal of the target device in operation.
  • the target device may be a device randomly selected from a plurality of devices to be monitored/selected according to a preset rule.
  • the target device is an overhead transmission line, or the target device is a power transformer.
  • the target equipment is underground cables, power distribution cabinets, ring main units, switch cabinets, circuit breakers, disconnectors, earthing switches, reactors, capacitors, resistors, wind turbines or solar cell installations.
  • the device signals may be: device signals communicated in the target device, input signals communicated in the target device, device signals propagated in the medium, input signals propagated in the medium, and ambient signals.
  • the device signal transmitted in the target device is a signal generated by the components included in the target device and can be transmitted in the target device.
  • the input signal transmitted in the target device is a signal generated by a preset input device through contact/collision/knock with the target device.
  • a device signal propagating in a medium is a signal that is generated by components included in the target device and that can propagate in the medium in the vicinity of the target device.
  • the input signal propagating in the medium is a signal generated by a preset input device through contact/collision/knock with the target device and the like and can be transmitted in the medium near the target device.
  • the input device is, for example, a metal hammer.
  • Ambient signals are signals propagating in the medium near the target device, such as sound signals produced by lightning strikes.
  • the medium is, for example, air, nitrogen, or the like.
  • the sampling unit samples the information monitored by each signal sensor to obtain the device signal.
  • the communication unit communicates with the information processing server through a wired communication link and/or a wireless communication link.
  • the control unit controls each monitoring terminal to start running, stop running, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode according to the control instructions.
  • the information processing server communicates with the at least one monitoring terminal through a wired communication link and/or a wireless communication link, and receives device signals from the at least one monitoring terminal based on the wired communication link and/or the wireless communication link.
  • the information processing server and/or each monitoring terminal can select a dedicated identification component associated with the target device from a plurality of dedicated identification components, and use the dedicated identification component to identify the device signal, so as to determine whether the target device is abnormal based on the identification result state.
  • the information processing server and/or each monitoring terminal can use the common identification component to identify the device signal to determine abnormal information of the target device in the abnormal state.
  • the information processing server and/or each monitoring terminal may perform calculation, setting, fitting, calculation, and fitting based on measured data, historical data, statistical data, rated parameters or/and basic parameters of multiple devices of each device type within a predetermined range. Tune or train to generate common recognition components.
  • the common identification component is capable of identifying anomaly information associated with a device in an anomalous state for each device type.
  • the common identification component is capable of identifying anomalous information associated with devices in an anomalous state for each device type based on artificial intelligence or neural networks.
  • the information processing server and/or each monitoring terminal may calculate, set, fit, and adjust based on measured data, historical data, statistical data, rated parameters and/or basic parameters of multiple devices of a specific device type within a predetermined range Or train to generate specialized recognition components associated with devices of a specific device type.
  • a dedicated identification component associated with a device of a particular device type can determine whether the device of a particular device type is in an abnormal state.
  • the information processing server and/or each monitoring terminal can generate a specific target device based on the calculation, setting, fitting, adjustment or training of the measured data, historical data, statistical data, rated parameters or/and basic parameters of the target device. Identify components.
  • a dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • Dedicated recognition components are able to identify whether the target device is in an abnormal state based on artificial intelligence or neural networks.
  • the information processing server and/or each monitoring terminal may generate a dedicated identification component of the target device based on training on measured data and/or historical data of the target device.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state. It also includes identifying device signals using a common identification component to determine the degree of anomaly, the type of anomaly, the component of anomaly, the cause of the anomaly, the location of the anomaly, the state of health, and/or the remaining life of the target device in an anomalous state.
  • a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device.
  • Feature data Identify the collected device feature data of the target device based on the artificial intelligence discrimination subcomponent and/or the neural network discrimination subcomponent to determine the identification result, and determine whether the target device is in an abnormal state based on the identification result.
  • the use of a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device. Feature data, compare the collected device feature data of the target device with the device feature data when the target device is in a normal state, so as to determine the comparison result as the identification result, and determine whether the target device is abnormal based on the identification result. state.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device Data; based on the artificial intelligence identification sub-component and/or the neural network identification sub-component, identify the collected device characteristic data of the target device to determine the identification result, and determine the abnormal information of the target device in an abnormal state according to the identification result.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device data. Based on the device identification of the target device, the existing device characteristic data associated with each abnormality information of all the abnormality information of the target device is obtained from the database. The collected device feature data of the target device is compared with the existing device feature data associated with each abnormal information, and the comparison result is used as the identification result. The abnormality information of the target device in the abnormal state is determined according to the identification result.
  • Determining the abnormal information of the target device in the abnormal state according to the identification result includes: comparing the collected device feature data of the target device with the existing device feature data associated with each abnormal information, so as to obtain the collected data of the target device.
  • the matching degree between the device feature data of the target device and the existing device feature data associated with each abnormal information; the matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is taken as The identification result; the abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • Determining whether the target device is in an abnormal state based on the identification result includes: using theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis to separate the collected device characteristic data of the target device from the target device.
  • the device characteristic data when the device is in a normal state is identified to determine the identification result, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • the abnormality information includes at least: abnormality type, abnormality degree, abnormality location and/or abnormality component.
  • the abnormal types include: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection, shock Wear, burst, burst, cable rack collapse, cable drop, digging, shovel, impact, landslide, foreign object intrusion, foreign object collision or partial discharge.
  • the abnormal types include: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection, Flashover, partial discharge, tower collapse, loose tower base, loose connector, broken connector, galloping, icing or impact from foreign objects.
  • Determining whether the target device is in an abnormal state based on the identification result includes: identifying the collected device characteristic data of the target device based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, Thereby, the identification result is determined, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • the abnormal information of the target device in the abnormal state determined according to the identification result includes: based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, the collected device characteristic data of the target device are divided into Identify the existing equipment characteristic data associated with each abnormal information to obtain the matching degree of the collected equipment characteristic data of the target equipment and the existing equipment characteristic data associated with each abnormal information; The matching degree between the device feature data of the target device and the existing device feature data associated with each abnormal information is used as the matching result; the abnormal information with the largest matching degree with the collected device feature data of the target device is determined as the target in the abnormal state Device exception information.
  • the device feature data includes at least one of a time domain feature, a frequency domain feature, a time-frequency domain feature, a power spectrum feature, a spectral envelope feature, a differential feature, an integral feature, and an artificial intelligence feature.
  • Determining whether the target device is in an abnormal state based on the identification result includes: according to at least one of time domain features, frequency domain features, time-frequency domain features, power spectrum features, spectral envelope features, differential features, integral features, and artificial intelligence features Whether the set threshold is exceeded to determine whether the target device is in an abnormal state.
  • Each signal sensor includes at least one voiceprint sensor, and the voiceprint sensor is arranged on the target device in a fit manner to monitor the device signal of the target device in operation; the monitored device signal is a voiceprint signal, video signal, infrared video signal, weather signal, position signal, electrical signal and/or electromagnetic wave signal.
  • the information processing server and/or each monitoring terminal can display or play the device signal, the operating status of the target device and/or the abnormal characteristics of the target device.
  • the information processing server and/or each monitoring terminal sequentially displays or plays the device signal, the running state of the target device and/or the abnormal characteristics of the target device according to the set sequence and time interval.
  • FIG. 2 is a schematic structural diagram of a system for monitoring abnormal state based on progressive identification according to another embodiment of the present invention.
  • the system includes at least one monitoring terminal and an information processing server.
  • At least one monitoring terminal includes monitoring terminal 1, monitoring terminal 2, . . . , monitoring terminal n.
  • Each of the at least one monitoring terminal is used for real-time monitoring of the running state of the target device to be monitored.
  • each monitoring terminal in the at least one monitoring terminal is set at different positions of the target device.
  • each of the at least one monitoring terminal includes: at least one signal sensor, a sampling unit, a communication unit and a control unit.
  • Each signal sensor is disposed at the target device in a fit manner for monitoring the device signal of the target device in operation.
  • the target device may be a device randomly selected from a plurality of devices to be monitored/selected according to a preset rule.
  • the target device may be an overhead transmission line, or the target device may be a power transformer.
  • the target equipment can be underground cables, distribution cabinets, ring main units, switch cabinets, circuit breakers, disconnectors, earthing switches, reactors, capacitors, resistors, wind turbines or solar cell installations.
  • the device signals may be: device signals communicated in the target device, input signals communicated in the target device, device signals propagated in the medium, input signals propagated in the medium, and ambient signals.
  • the device signal transmitted in the target device is a signal generated by the components included in the target device and can be transmitted in the target device.
  • the input signal transmitted in the target device is a signal generated by a preset input device through contact/collision/knock with the target device.
  • a device signal propagating in a medium is a signal that is generated by components included in the target device and that can propagate in the medium in the vicinity of the target device.
  • the input signal propagating in the medium is a signal generated by a preset input device through contact/collision/knock with the target device, etc.
  • the input device is, for example, a metal hammer.
  • Ambient signals are signals propagating in the medium near the target device, such as sound signals produced by lightning strikes.
  • the medium is, for example, air, nitrogen, or the like.
  • the sampling unit samples the information monitored by each signal sensor to obtain the device signal.
  • the communication unit communicates with the information processing server through a wired communication link and/or a wireless communication link.
  • the control unit controls each monitoring terminal to start running, stop running, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode according to the control instructions.
  • the identification unit uses a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result.
  • the information processing server communicates with the at least one monitoring terminal through a wired communication link and/or a wireless communication link, and receives device signals from the at least one monitoring terminal based on the wired communication link and/or the wireless communication link.
  • the information processing server and/or each monitoring terminal may select a dedicated identification component associated with the target device from a plurality of dedicated identification components, and use the dedicated identification component to identify the device signal to determine the target based on the result of the identification Whether the device is in an abnormal state.
  • the information processing server and/or each monitoring terminal can use the common identification component to identify the device signal to determine the abnormal information of the target device in the abnormal state.
  • the information processing server and/or each monitoring terminal may perform calculation, setting, fitting, calculation, and fitting based on measured data, historical data, statistical data, rated parameters or/and basic parameters of multiple devices of each device type within a predetermined range. Tune or train to generate common recognition components.
  • the common identification component is capable of identifying anomaly information associated with a device in an anomalous state for each device type.
  • the common identification component is capable of identifying anomalous information associated with devices in an anomalous state for each device type based on artificial intelligence or neural networks.
  • the information processing server and/or each monitoring terminal may calculate, set, fit, and adjust based on measured data, historical data, statistical data, rated parameters and/or basic parameters of multiple devices of a specific device type within a predetermined range Or train to generate specialized recognition components associated with devices of a specific device type.
  • a dedicated identification component associated with a device of a particular device type can determine whether the device of a particular device type is in an abnormal state.
  • the information processing server and/or each monitoring terminal can generate a specific target device based on the calculation, setting, fitting, adjustment or training of the measured data, historical data, statistical data, rated parameters or/and basic parameters of the target device. Identify components.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • Dedicated recognition components are able to identify whether the target device is in an abnormal state based on artificial intelligence or neural networks.
  • the information processing server and/or each monitoring terminal may generate a dedicated identification component of the target device based on training on measured data and/or historical data of the target device.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state. It also includes identifying the device signal with a common identification component to determine the abnormality degree, abnormality type, abnormality component, abnormality cause, abnormality location, health status and/or remaining life of the target device in the abnormal state.
  • the special identification component is used to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result.
  • Feature data Identify the collected device feature data of the target device based on the artificial intelligence discrimination subcomponent and/or the neural network discrimination subcomponent to determine the identification result, and determine whether the target device is in an abnormal state based on the identification result.
  • the use of a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device. Feature data, compare the collected device feature data of the target device with the device feature data when the target device is in a normal state, so as to determine the comparison result as the identification result, and determine whether the target device is abnormal based on the identification result. state.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device Data; based on the artificial intelligence identification sub-component and/or the neural network identification sub-component, identify the collected device characteristic data of the target device to determine the identification result, and determine the abnormal information of the target device in an abnormal state according to the identification result.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device data. Based on the device identification of the target device, the existing device characteristic data associated with each abnormality information of all the abnormality information of the target device is obtained from the database. The collected device feature data of the target device is compared with the existing device feature data associated with each abnormal information, and the comparison result is used as the identification result. The abnormality information of the target device in the abnormal state is determined according to the identification result.
  • Determining the abnormal information of the target device in the abnormal state according to the identification result includes: comparing the collected device feature data of the target device with the existing device feature data associated with each abnormal information, so as to obtain the collected data of the target device.
  • the matching degree between the device feature data of the target device and the existing device feature data associated with each abnormal information; the matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is taken as The identification result; the abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • Determining whether the target device is in an abnormal state based on the identification result includes: using theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis to separate the collected device characteristic data of the target device from the target device.
  • the device characteristic data when the device is in a normal state is identified to determine the identification result, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • the abnormality information includes at least: abnormality type, abnormality degree, abnormality location and/or abnormality component.
  • the abnormal types include: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection, shock Wear, burst, burst, cable rack collapse, cable drop, digging, shovel, impact, landslide, foreign object intrusion, foreign object collision or partial discharge.
  • the abnormal types include: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection, Flashover, partial discharge, tower collapse, loose tower base, loose connector, broken connector, galloping, icing or impact from foreign objects.
  • Determining whether the target device is in an abnormal state based on the identification result includes: identifying the collected device characteristic data of the target device based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, Thereby, the identification result is determined, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • the abnormal information of the target device in the abnormal state determined according to the identification result includes: based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, the collected device characteristic data of the target device are divided into Identify the existing equipment characteristic data associated with each abnormal information to obtain the matching degree of the collected equipment characteristic data of the target equipment and the existing equipment characteristic data associated with each abnormal information; The matching degree between the device feature data of the target device and the existing device feature data associated with each abnormal information is used as the matching result; the abnormal information with the largest matching degree with the collected device feature data of the target device is determined as the target in the abnormal state Device exception information.
  • the device feature data includes at least one of time domain features, frequency domain features, time-frequency domain features, power spectrum features, spectral envelope features, differential features, integral features, and artificial intelligence features. Determining whether the target device is in an abnormal state based on the identification result includes: according to at least one of time domain features, frequency domain features, time-frequency domain features, power spectrum features, spectral envelope features, differential features, integral features, and artificial intelligence features Whether the set threshold is exceeded to determine whether the target device is in an abnormal state.
  • Each signal sensor includes at least one voiceprint sensor, and the voiceprint sensor is arranged on the target device in a fit manner to monitor the device signal of the target device in operation; the monitored device signal is a voiceprint signal, video signal, infrared video signal, weather signal, position signal, electrical signal and/or electromagnetic wave signal.
  • the information processing server and/or each monitoring terminal can display or play the device signal, the operating status of the target device and/or the abnormal characteristics of the target device.
  • the information processing server and/or each monitoring terminal sequentially displays or plays the device signal, the running state of the target device and/or the abnormal characteristics of the target device according to the set sequence and time interval.
  • FIG. 3 is a schematic diagram of monitoring a target device according to an embodiment of the present invention.
  • the signal sensors 202-1, 202-2, 202-3, 202-4, 202-5 and 202-6 are provided at the target device 200.
  • the signal sensors 202-1 to 202-6 may belong to the same monitoring terminal or to different monitoring terminals.
  • monitoring terminal A includes signal sensors 202-1, 202-2, and 202-3; and monitoring terminal B includes signal sensors 202-4, 202-5, and 202-6.
  • the signal sensors 202-1, 202-2, 202-3, 202-4, 202-5, and 202-6 are disposed at the target device 200 in a fitting manner or abutting manner, for monitoring the operation of the target device 200 device signal in .
  • the target device 200 may be a device randomly selected from a plurality of devices to be monitored/selected according to a preset rule.
  • the target device 200 may be an overhead power line, or the target device 200 may be a power transformer.
  • the target equipment can be underground cables, distribution cabinets, ring main units, switch cabinets, circuit breakers, disconnectors, earthing switches, reactors, capacitors, reactors, wind turbines or solar cell installations.
  • the device signal of the target device 200 in operation is a signal generated by the components included in the target device and can be transmitted in the target device.
  • a device signal is a signal that is generated by a component during operation due to a failure and can be passed on in the target device.
  • the device signal of the target device 200 in operation may also be a signal generated by a component included in the target device 200 and capable of being transmitted in a medium near the target device.
  • a device signal is a signal that is generated by a component during operation due to a malfunction and can be transmitted in a medium near the target device 200 .
  • the medium is, for example, air, nitrogen, or the like.
  • FIG. 4 is a flowchart of a method 400 for monitoring an abnormal state based on progressive identification according to an embodiment of the present invention.
  • Method 400 begins at step 401 .
  • step 401 at least one signal sensor of each monitoring terminal is used to monitor the device signal of the target device in operation. It also includes using at least one monitoring terminal to monitor the running state of the target device to be monitored in real time, wherein each monitoring terminal in the at least one monitoring terminal is set at different positions of the target device.
  • step 402 the information monitored by each signal sensor is sampled using the sampling unit of each monitoring terminal, thereby obtaining a device signal.
  • step 403 the communication unit of each monitoring terminal is used to communicate with the information processing server through a wired communication link and/or a wireless communication link.
  • step 404 use the control unit of each monitoring terminal to control the start-up operation, stop operation, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode of each monitoring terminal according to the control instructions .
  • the information processing server communicates with at least one monitoring terminal through a wired communication link and/or a wireless communication link, and receives a device signal from the at least one monitoring terminal based on the wired communication link and/or wireless communication link.
  • step 406 select a dedicated identification component associated with the target device from a plurality of dedicated identification components, and use the dedicated identification component to identify the device signal to determine whether the target device is in an abnormal state based on the identification result; when it is determined that the target device is in an abnormal state In an abnormal state, a common identification component is used to identify the device signal, so as to determine the abnormal information of the target device in the abnormal state.
  • the common identification component is generated based on the calculation, setting, fitting, adjustment or training of measured data, historical data, statistical data, nominal parameters or/and basic parameters of a plurality of devices of each device type within a predetermined range.
  • the common identification component is capable of identifying abnormal information associated with each device type that is in an abnormal state.
  • a dedicated identification component associated with a device A dedicated identification component associated with a device of a particular device type can determine whether the device of a particular device type is in an abnormal state.
  • the specific identification component of the target device is generated based on the calculation, setting, fitting, adjustment or training of the measured data, historical data, statistical data, rated parameters or/and basic parameters of the target device.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • a specific identification component of the target device is generated based on training on measured data and/or historical data of the target device.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state. Also included is the identification of device signals using a common identification component to determine the degree of anomaly, type of anomaly, component of anomaly, cause of anomaly, location of anomaly, state of health and/or remaining life of the target device in an anomalous state.
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection , breakdown, pipe burst, burst, cable rack collapse, cable drop, digging, shovel, impact, landslide, foreign object intrusion, foreign object collision or partial discharge.
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection Lines, flashovers, partial discharges, tower collapses, loose tower foundations, loose connectors, broken connectors, galloping, icing or impact from foreign objects.
  • the use of a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device. characteristic data. Identify the collected device characteristic data of the target device based on the artificial intelligence discrimination subcomponent and/or the neural network discrimination subcomponent to determine the identification result, and determine whether the target device is in an abnormal state based on the identification result.
  • the use of a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device. Feature data, compare the collected device feature data of the target device with the device feature data when the target device is in a normal state, so as to determine the comparison result as the identification result, and determine whether the target device is abnormal based on the identification result. state.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device data. Identify the collected device characteristic data of the target device based on the artificial intelligence identification subcomponent and/or the neural network identification subcomponent to determine the identification result. The abnormality information of the target device in the abnormal state is determined according to the identification result.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device data. Based on the device identification of the target device, the existing device characteristic data associated with each abnormality information of all the abnormality information of the target device is obtained from the database. The collected device feature data of the target device is compared with the existing device feature data associated with each abnormal information, and the comparison result is used as the identification result. The abnormality information of the target device in the abnormal state is determined according to the identification result.
  • Determining the abnormal information of the target device in the abnormal state according to the identification result includes: comparing the collected device feature data of the target device with the existing device feature data associated with each abnormal information, so as to obtain the collected data of the target device. The matching degree between the device feature data of the target device and the existing device feature data associated with each abnormal information. The matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is used as the identification result. The abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • Determining whether the target device is in an abnormal state based on the identification result includes: using theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis to separate the collected device characteristic data of the target device from the target device.
  • the device characteristic data when the device is in a normal state is identified to determine the identification result, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • Determining whether the target device is in an abnormal state based on the identification result includes: identifying the collected device characteristic data of the target device based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, Thereby, the identification result is determined, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • the abnormal information of the target device in the abnormal state determined according to the identification result includes: based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, the collected device characteristic data of the target device are divided into Identifying the existing device feature data associated with each abnormal information to obtain the matching degree of the collected device feature data of the target device with the existing device feature data associated with each abnormal information.
  • the matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is used as the matching result.
  • the abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • the device feature data includes at least one of a time domain feature, a frequency domain feature, a time-frequency domain feature, a power spectrum feature, a spectral envelope feature, a differential feature, an integral feature, and an artificial intelligence feature.
  • Determining whether the target device is in an abnormal state based on the identification result includes: according to at least one of time domain features, frequency domain features, time-frequency domain features, power spectrum features, spectral envelope features, differential features, integral features, and artificial intelligence features Whether the set threshold is exceeded to determine whether the target device is in an abnormal state.
  • Each signal sensor includes at least one voiceprint sensor, and the voiceprint sensor is disposed on the target device in a fit manner to monitor the device signal of the target device in operation.
  • the monitored equipment signals are voiceprint signals, video signals, infrared video signals, weather signals, position signals, electrical signals and/or electromagnetic wave signals.
  • the target device is an overhead transmission line, or the target device is a power transformer.
  • the target equipment is an underground cable, a power distribution cabinet, a ring network cabinet, a switch cabinet, a circuit breaker, an isolating switch, a grounding switch, a reactor, a capacitor, a resistor, a wind turbine or a solar cell device.
  • the information processing server or each monitoring terminal can display or play the device signal, the operating status of the target device and/or the abnormal characteristics of the target device.
  • the information processing server or each monitoring terminal sequentially displays or plays the device signal, the running state of the target device and/or the abnormal characteristics of the target device according to the set sequence and time interval.
  • FIG. 5 is a flowchart of a method 500 for monitoring an abnormal state based on progressive identification according to another embodiment of the present invention.
  • Method 500 begins at step 501 .
  • step 501 at least one signal sensor of each monitoring terminal is used to monitor the device signal of the target device in operation.
  • the running state of the target device to be monitored is monitored in real time by using at least one monitoring terminal, wherein each monitoring terminal in the at least one monitoring terminal is set at different positions of the target device.
  • step 502 the information monitored by each signal sensor is sampled using the sampling unit of each monitoring terminal, thereby obtaining a device signal.
  • the communication unit of each monitoring terminal is used to communicate with the information processing server through a wired communication link and/or a wireless communication link.
  • step 504 use the control unit of each monitoring terminal to control the start-up operation, stop operation, position movement, software update, parameter setting, signal sampling, data processing, abnormal diagnosis and/or working mode of each monitoring terminal according to the control instructions .
  • step 505 using the identification unit of each monitoring terminal, a dedicated identification component is used to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result.
  • the information processing server communicates with the at least one monitoring terminal through a wired communication link and/or a wireless communication link, receives a device signal from the at least one monitoring terminal based on the wired communication link and/or the wireless communication link, and processes the information.
  • the server and/or each monitoring terminal uses the common identification component to identify the device signal, so as to determine the abnormal information of the target device in the abnormal state.
  • the common identification component is generated based on the calculation, setting, fitting, adjustment or training of measured data, historical data, statistical data, nominal parameters or/and basic parameters of a plurality of devices of each device type within a predetermined range.
  • the common identification component is capable of identifying abnormal information associated with each device type that is in an abnormal state.
  • a dedicated identification component associated with a device A dedicated identification component associated with a device of a particular device type can determine whether the device of a particular device type is in an abnormal state.
  • the specific identification component of the target device is generated based on the calculation, setting, fitting, adjustment or training of the measured data, historical data, statistical data, rated parameters or/and basic parameters of the target device.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state.
  • a specific identification component of the target device is generated based on training on measured data and/or historical data of the target device.
  • the dedicated identification component of the target device can determine whether the target device is in an abnormal state. It also includes identifying the device signal with a common identification component to determine the abnormality degree, abnormality type, abnormality component, abnormality cause, abnormality location, health status and/or remaining life of the target device in the abnormal state.
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection , breakdown, pipe burst, burst, cable rack collapse, cable drop, digging, shovel, impact, landslide, foreign object intrusion, foreign object collision or partial discharge.
  • the abnormal information includes abnormal types: short circuit, single-phase short circuit, three-phase short circuit, two-phase short circuit, two-phase short-circuit grounding, disconnection, single-phase disconnection, two-phase disconnection, three-phase disconnection Lines, flashovers, partial discharges, tower collapses, loose tower foundations, loose connectors, broken connectors, galloping, icing or impact from foreign objects.
  • the special identification component is used to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result.
  • characteristic data Identify the collected device characteristic data of the target device based on the artificial intelligence discrimination subcomponent and/or the neural network discrimination subcomponent to determine the identification result, and determine whether the target device is in an abnormal state based on the identification result.
  • the use of a special identification component to identify the device signal, so as to determine whether the target device is in an abnormal state based on the identification result includes: the special identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected equipment of the target device. Feature data, compare the collected device feature data of the target device with the device feature data when the target device is in a normal state, so as to determine the comparison result as the identification result, and determine whether the target device is abnormal based on the identification result. state.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device data. Identify the collected device characteristic data of the target device based on the artificial intelligence identification subcomponent and/or the neural network identification subcomponent to determine the identification result. The abnormality information of the target device in the abnormal state is determined according to the identification result.
  • the use of the common identification component to identify the device signal to determine the abnormal information of the target device in an abnormal state includes: the common identification component performs feature extraction on the collected device signal of the target device, so as to obtain the collected device characteristics of the target device data. Based on the device identification of the target device, the existing device characteristic data associated with each abnormality information of all the abnormality information of the target device is obtained from the database. The collected device feature data of the target device is compared with the existing device feature data associated with each abnormal information, and the comparison result is used as the identification result. The abnormality information of the target device in the abnormal state is determined according to the identification result.
  • Determining the abnormal information of the target device in the abnormal state according to the identification result includes: comparing the collected device feature data of the target device with the existing device feature data associated with each abnormal information, so as to obtain the collected data of the target device. The matching degree between the device feature data of the target device and the existing device feature data associated with each abnormal information. The matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is used as the identification result. The abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • Determining whether the target device is in an abnormal state based on the identification result includes: using theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis to separate the collected device characteristic data of the target device from the target device.
  • the device characteristic data when the device is in a normal state is identified to determine the identification result, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • Determining whether the target device is in an abnormal state based on the identification result includes: identifying the collected device characteristic data of the target device based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, Thereby, the identification result is determined, and based on the identification result, it is determined whether the target device is in an abnormal state.
  • the abnormal information of the target device in the abnormal state determined according to the identification result includes: based on the use of theoretical formulas, experience curves, test curves, measured curves, statistical curves, fitting curves or big data analysis, the collected device characteristic data of the target device are divided into Identifying the existing device feature data associated with each abnormal information to obtain the matching degree of the collected device feature data of the target device with the existing device feature data associated with each abnormal information.
  • the matching degree between the collected device feature data of the target device and the existing device feature data associated with each abnormal information is used as the matching result.
  • the abnormal information with the largest matching degree with the collected device characteristic data of the target device is determined as the abnormal information of the target device in the abnormal state.
  • the device feature data includes at least one of a time domain feature, a frequency domain feature, a time-frequency domain feature, a power spectrum feature, a spectral envelope feature, a differential feature, an integral feature, and an artificial intelligence feature.
  • Determining whether the target device is in an abnormal state based on the identification result includes: according to at least one of time domain features, frequency domain features, time-frequency domain features, power spectrum features, spectral envelope features, differential features, integral features, and artificial intelligence features Whether the set threshold is exceeded to determine whether the target device is in an abnormal state.
  • Each signal sensor includes at least one voiceprint sensor, and the voiceprint sensor is disposed on the target device in a fit manner to monitor the device signal of the target device in operation.
  • the monitored equipment signals are voiceprint signals, video signals, infrared video signals, weather signals, position signals, electrical signals and/or electromagnetic wave signals.
  • the target device is an overhead transmission line, or the target device is a power transformer.
  • the target equipment is an underground cable, a power distribution cabinet, a ring network cabinet, a switch cabinet, a circuit breaker, an isolating switch, a grounding switch, a reactor, a capacitor, a resistor, a wind turbine or a solar cell device.
  • the information processing server or each monitoring terminal can display or play the device signal, the operating status of the target device and/or the abnormal characteristics of the target device.
  • the information processing server or each monitoring terminal sequentially displays or plays the device signal, the running state of the target device and/or the abnormal characteristics of the target device according to the set sequence and time interval.

Abstract

本发明涉及一种基于递进式识别来监测异常状态的系统及方法，其中系统包括：至少一个监测终端，用于对待监测的目标设备的运行状态进行实时监测,其中每个监测终端包括：至少一个信号传感器，用于监测所述目标设备在运行中的设备信号；信息处理服务器，通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号；从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。

Description

一种基于递进式识别来监测异常状态的系统及方法 技术领域
本发明涉及设备检测技术领域，并且更具体地，涉及一种基于递进式识别来监测异常状态的系统及方法。
背景技术
健康状态监测与诊断是对设备进行运行维护的关键技术。现有的监测技术都是根据信号特征直接诊断设备的状态。这种方式存在诊断准确性低、以及无法确定异常原因的不足。实际上，正常运行时的设备很少形成损伤，所以如果每次都对设备进行全方面检测是比较耗时且效率低下的。由于正常运行时的识别的故障率较低，为此需要在特定情况下首先确定设备是否存在异常。在确定不存在异常时不需要进行全面检测。
发明内容
因此，本申请提出一种基于递进式识别来监测异常状态的系统和方法，首先准确地诊断健康状态，并且在确定健康状态异常时才确定损伤产生原因、位置、部件等。
根据本发明的一个方面，提供一种基于递进式识别来监测异常状态的系统，所述系统包括：
至少一个监测终端，用于对待监测的目标设备的运行状态进行实时监测，将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；
其中每个监测终端包括：
至少一个信号传感器，用于监测所述目标设备在运行中的设备信号；
采样单元，对每个信号传感器所监测到的信息进行采样，从而获得设备信号；
通信单元，通过有线通信链路和/或无线通信链路与信息处理服务
器进行通信；
控制单元，根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；
信息处理服务器，通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号；
从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；
当确定目标设备处于异常状态时，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
根据本发明的再一方面，提供一种基于递进式识别来监测异常状态的系统，所述系统包括：
至少一个监测终端，用于对待监测的目标设备的运行状态进行实时监测，将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；
其中每个监测终端包括：
至少一个信号传感器，用于监测所述目标设备在运行中的设备信号；
采样单元，对每个信号传感器所监测到的信息进行采样，从而获得设备信号；
通信单元，通过有线通信链路和/或无线通信链路与信息处理服务器进行通信，当确定目标设备处于异常状态时，经由有线通信链路和/或无线通信链路将设备信号发送给信息处理服务器；
控制单元，根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；
识别单元，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；
信息处理服务器，通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
信息处理服务器和/或每个监测终端基于对预定范围内每种设备类型的多个设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成公共识别组件；
所述公共识别组件能够识别与每种设备类型的处于异常状态的设备相关联的异常信息；
或者，信息处理服务器和/或每个监测终端基于对预定范围内特定设备类型的多个设备的实测数据、历史数据、统计数据、额定参数和/或基本参数进行计算、设定、拟合、调整或训练来生成与特定设备类型的设备相关联的专用识别组件；
与特定设备类型的设备相关联的专用识别组件能够确定特定设备类型的设备是否处于异常状态。
信息处理服务器和/或每个监测终端基于对目标设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成目标设备的专用识别组件；
所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。
信息处理服务器和/或每个监测终端基于对目标设备的实测数据和/或历史数据进行训练来生成目标设备的专用识别组件；
所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。
还包括利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常程度、异常类型、异常部件、异常原因、异常位置、健康状态和/或剩余寿命。
当目标设备为地下电缆时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、击穿、爆管、爆裂、电缆架倒塌、电缆掉落、挖、铲、撞击、塌方、外物入侵、外物碰撞或局部放电；
当目标设备为架空输电线路时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、闪络、局部放电、电塔倒塌、塔基松动、连接件松动、连接件断裂、舞动、覆冰或外物撞击。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：
专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；
基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：
专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据，将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行比对，从而将比对结果确定为识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：
公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；
基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，
根据识别的结果确定处于异常状态的目标设备的异常信息。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：
公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；
基于目标设备的设备标识从数据库中获取与目标设备的所有异常信息中每种异常信息相关联的现有设备特征数据；
将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，将比对结果作为识别的结果；
根据识别的结果确定处于异常状态的目标设备的异常信息。
根据识别的结果确定异常状态的目标设备的异常信息包括：
将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度；
将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为识别的结果；
将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
还包括，确定异常状态的存在范围，并根据异常状态的存在范围确定目标设备的需要检查和/或维修的范围。
所述基于识别的结果确定目标设备是否处于异常状态包括：
基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
所述基于识别的结果确定目标设备是否处于异常状态包括：
基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
根据识别的结果确定异常状态的目标设备的异常信息包括：
基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行识别，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度；
将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为匹配结果；
将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
所述设备特征数据包括时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个。
所述基于识别的结果确定目标设备是否处于异常状态包括：
根据时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个是否超过设定的阈值来确定目标设备是否处于异常状态。
每个信号传感器至少包括一个声纹传感器，将所述声纹传感器以贴合方式设置在目标设备上用于监测所述目标设备在运行中的设备信号；
所述监测到的设备信号为声纹信号、视频信号、红外视频信号、气象信号、位置信号、电气信号和/或电磁波信号。
所述目标设备为架空输电线路、或者所述目标设备为电力变压器。
所述目标设备为地下电缆、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电抗器、电容器、电阻器、风力发电机或太阳能电池装置。
所述信息处理服务器能够显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。
所述信息处理服务器按照设置的顺序和时间间隔依次显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。
根据本发明的再一方面，提供一种基于递进式识别来监测异常状态的方法，所述方法包括：使用至少一个监测终端对待监测的目标设备的运行状态进行实时监测，其中将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；使用每个监测终端的至少一个信号传感器来监测所述目标设备在运行中的设备信号；使用每个监测终端的采样单元对每个信号传感器所监测到的信息进行采样，从而获得设备信号；使用每个监测终端的通信单元通过有线通信链路和/或无线通信链路与信息处理服务器进行通信；使用每个监测终端的控制单元根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；利用信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号；从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；当确定目标设备处于异常状态时，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
根据本发明的另一方面，提供一种基于递进式识别来监测异常状态的方法，所述方法包括：使用至少一个监测终端对待监测的目标设备的运行状态进行实时监测，其中将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；使用每个监测终端的至少一个信号传感器来监测所述目标设备在运行中的设备信号；使用每个监测终端的采样单元对每个信号传感器所监测到的信息进行采样，从而获得设备信号；使用每个监测终端的通信单元通过有线通信链路和/或无线通信链路与信息处理服务器进行通信；使用每个监测终端的控制单元根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；使用每个监测终端的识别单元，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
基于对预定范围内每种设备类型的多个设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成公共识别组件；
所述公共识别组件能够识别与每种设备类型的处于异常状态的设备相关联的异常信息；
或者，基于对预定范围内特定设备类型的多个设备的实测数据、历史数据、统计数据、额定参数和/或基本参数进行计算、设定、拟合、调整或训练来生成与特定设备类型的设备相关联的专用识别组件；
与特定设备类型的设备相关联的专用识别组件能够确定特定设备类型的设备是否处于异常状态。
基于对目标设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成目标设备的专用识别组件；
所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。
基于对目标设备的实测数据和/或历史数据进行训练来生成目标设备的专用识别组件；
所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。
还包括利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常程度、异常类型、异常部件、异常原因、异常位置、健康状态和/或剩余寿命。
当目标设备为地下电缆时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、击穿、爆管、爆裂、电缆架倒塌、电缆掉落、挖、铲、撞击、塌方、外物入侵、外物碰撞或局部放电；
当目标设备为架空输电线路时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、闪络、局部放电、电塔倒塌、塔基松动、连接件松动、连接件断裂、舞动、覆冰或外物撞击。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备 是否处于异常状态。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据，将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行比对，从而将比对结果确定为识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，根据识别的结果确定处于异常状态的目标设备的异常信息。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；基于目标设备的设备标识从数据库中获取与目标设备的所有异常信息中每种异常信息相关联的现有设备特征数据；将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，将比对结果作为识别的结果；
根据识别的结果确定处于异常状态的目标设备的异常信息。
根据识别的结果确定异常状态的目标设备的异常信息包括：将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度；将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为识别的结果；将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
还包括，确定异常状态的存在范围，并根据异常状态的存在范围确定目标设备的需要检查和/或维修的范围。
所述基于识别的结果确定目标设备是否处于异常状态包括：
基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别与目标设备 处于正常状态时的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
所述基于识别的结果确定目标设备是否处于异常状态包括：
基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
根据识别的结果确定异常状态的目标设备的异常信息包括：
基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行识别，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度；
将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为匹配结果；将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
所述设备特征数据包括时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个。
所述基于识别的结果确定目标设备是否处于异常状态包括：根据时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个是否超过设定的阈值来确定目标设备是否处于异常状态。每个信号传感器至少包括一个声纹传感器，将所述声纹传感器以贴合方式设置在目标设备上用于监测所述目标设备在运行中的设备信号；所述监测到的设备信号为声纹信号、视频信号、红外视频信号、气象信号、位置信号、电气信号和/或电磁波信号。
所述目标设备为架空输电线路、或者所述目标设备为电力变压器。
所述目标设备为地下电缆、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电抗器、电容器、电阻器、风力发电机或太阳能电池装置。所述信息处理服务器能够显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。所述信息处理服务器按照设置的顺序和时间间隔依次显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。根据本发明的技术方案可以广泛应用到架空输电线路、地下电缆、电力变压器、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电感器、电容器、电阻器、风力发电机、太阳能电池装置等设备，具有良好的应用价值。
附图说明
通过参考下面的附图，可以更为完整地理解本发明的示例性实施方式：
图1为根据本发明实施方式的基于递进式识别来监测异常状态的系统的结构示意图；
图2为根据本发明另一实施方式的基于递进式识别来监测异常状态的系统的结构示意图；
图3为根据本发明实施方式的对目标设备进行监测的示意图；
图4为根据本发明实施方式的基于递进式识别来监测异常状态的方法的流程图；
图5为根据本发明另一实施方式的基于递进式识别来监测异常状态的方法的流程图。
具体实施方式
在本申请中，信号(例如，设备信号、输入信号和/或环境信号等)应当被宽泛地、或概括地理解为直接采集到的信号或者其被处理之后的结果。
图1为根据本发明实施方式的基于递进式识别来监测异常状态的系统的结构示意图。系统包括至少一个监测终端和信息处理服务器。至少一个监测终端包括监测终端1、监测终端2、…..、监测终端n。至少一个监测终端中的每个监控终端用于对待监测的目标设备的运行状态进行实时监测。为了对待监测的目标设备的运行状态进行实时监测，将至少一个监测终端中的每个监测终端设置在目标设备的不同位置处。
如图1所示，至少一个监测终端中的每个监测终端包括：至少一个信号传感器、采样单元、通信单元和控制单元。每个信号传感器以贴合方式被设置在目标设备处，用于监测所述目标设备在运行中的设备信号。目标设备可以是从多个待监测的设备中随机选择的/按照预设规则选择的一个设备。目标设备为架空输电线路、或者所述目标设备为电力变压器。目标设备为地下电缆、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电抗器、电容器、电阻器、风力发电机或太阳能电池装置。
设备信号可以是：目标设备中传递的设备信号、目标设备中传递的输入信号、介质中传播的设备信号、介质中传播的输入信号以及环境信号。目标设备中传递的设备信号是由目标设备所包括的部件所产生的并且能够在目标设备中传递的信号。目标设备中传递的输入信号是由预先设置的输入器件通过与目标设备的接触/碰撞/敲击等所产生的信号。介质中传播的 设备信号是由目标设备所包括的部件所产生的并且能够在目标设备附近的介质中传递的信号。介质中传播的输入信号是由预先设置的输入器件通过与目标设备的接触/碰撞/敲击等所产生的并且能够在目标设备附近的介质中传递的信号。输入器件例如是金属锤。环境信号是在目标设备附近的介质中传播的信号，例如雷击等所产生的声音信号。介质例如是空气、氮气等。
采样单元，对每个信号传感器所监测到的信息进行采样，从而获得设备信号。通信单元，通过有线通信链路和/或无线通信链路与信息处理服务器进行通信。控制单元，根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式。
信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号。信息处理服务器和/或每个监测终端可以从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态。当确定目标设备处于异常状态时，信息处理服务器和/或每个监测终端可以利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
信息处理服务器和/或每个监测终端可以基于对预定范围内每种设备类型的多个设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成公共识别组件。公共识别组件能够识别与每种设备类型的处于异常状态的设备相关联的异常信息。公共识别组件能够基于人工智能或神经网络来识别与每种设备类型的处于异常状态的设备相关联的异常信息。
信息处理服务器和/或每个监测终端可以基于对预定范围内特定设备类型的多个设备的实测数据、历史数据、统计数据、额定参数和/或基本参数进行计算、设定、拟合、调整或训练来生成与特定设备类型的设备相关联的专用识别组件。其中与特定设备类型的设备相关联的专用识别组件能够确定特定设备类型的设备是否处于异常状态。
信息处理服务器和/或每个监测终端可以基于对目标设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成目标设备的专用识别组件。所述目标设备的专用识别组 件能够确定目标设备是否处于异常状态。专用识别组件能够基于人工智能或神经网络来识别目标设备是否处于异常状态。
信息处理服务器和/或每个监测终端可以基于对目标设备的实测数据和/或历史数据进行训练来生成目标设备的专用识别组件。所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。还包括利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常程度、异常类型、异常部件、异常原因、异常位置、健康状态和/或剩余寿命。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据，将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行比对，从而将比对结果确定为识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，根据识别的结果确定处于异常状态的目标设备的异常信息。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于目标设备的设备标识从数据库中获取与目标设备的所有异常信息中每种异常信息相关联的现有设备特征数据。将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，将比对结果作为识别的结果。根据识别的结果确定处于异常状态的目标设备的异常信息。
根据识别的结果确定异常状态的目标设备的异常信息包括：将所采集 的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度；将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为识别的结果；将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
还包括，确定异常状态的存在范围，并根据异常状态的存在范围确定目标设备的需要检查和/或维修的范围。基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
异常信息至少包括：异常类型、异常程度、异常位置和/或异常部件。当目标设备为地下电缆时，异常类型包括：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、击穿、爆管、爆裂、电缆架倒塌、电缆掉落、挖、铲、撞击、塌方、外物入侵、外物碰撞或局部放电。当目标设备为架空输电线路时，异常类型包括：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、闪络、局部放电、电塔倒塌、塔基松动、连接件松动、连接件断裂、舞动、覆冰或外物撞击。
基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
根据识别的结果确定异常状态的目标设备的异常信息包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行识别，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度；将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为匹配结果；将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
设备特征数据包括时域特征、频域特征、时频域特征、功率谱特征、 频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个。基于识别的结果确定目标设备是否处于异常状态包括：根据时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个是否超过设定的阈值来确定目标设备是否处于异常状态。
每个信号传感器至少包括一个声纹传感器，将所述声纹传感器以贴合方式设置在目标设备上用于监测所述目标设备在运行中的设备信号；所述监测到的设备信号为声纹信号、视频信号、红外视频信号、气象信号、位置信号、电气信号和/或电磁波信号。信息处理服务器和/或每个监测终端能够显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。信息处理服务器和/或每个监测终端按照设置的顺序和时间间隔依次显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。
图2为根据本发明另一实施方式的基于递进式识别来监测异常状态的系统的结构示意图。系统包括至少一个监测终端和信息处理服务器。至少一个监测终端包括监测终端1、监测终端2、…..、监测终端n。至少一个监测终端中的每个监控终端用于对待监测的目标设备的运行状态进行实时监测。为了对待监测的目标设备的运行状态进行实时监测，将至少一个监测终端中的每个监测终端设置在目标设备的不同位置处。
如图2所示，至少一个监测终端中的每个监测终端包括：至少一个信号传感器、采样单元、通信单元和控制单元。每个信号传感器以贴合方式被设置在目标设备处，用于监测所述目标设备在运行中的设备信号。目标设备可以是从多个待监测的设备中随机选择的/按照预设规则选择的一个设备。目标设备可以为架空输电线路、或者目标设备为电力变压器。目标设备可以为地下电缆、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电抗器、电容器、电阻器、风力发电机或太阳能电池装置。
设备信号可以是：目标设备中传递的设备信号、目标设备中传递的输入信号、介质中传播的设备信号、介质中传播的输入信号以及环境信号。目标设备中传递的设备信号是由目标设备所包括的部件所产生的并且能够在目标设备中传递的信号。目标设备中传递的输入信号是由预先设置的输入器件通过与目标设备的接触/碰撞/敲击等所产生的信号。介质中传播的设备信号是由目标设备所包括的部件所产生的并且能够在目标设备附近的介质中传递的信号。介质中传播的输入信号是由预先设置的输入器件通过与目标设备的接触/碰撞/敲击等所产生的并且能够在目标设备附近的介质 中传递的信号。输入器件例如是金属锤。环境信号是在目标设备附近的介质中传播的信号，例如雷击等所产生的声音信号。介质例如是空气、氮气等。
采样单元，对每个信号传感器所监测到的信息进行采样，从而获得设备信号。通信单元，通过有线通信链路和/或无线通信链路与信息处理服务器进行通信。控制单元，根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式。识别单元，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态。
信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号。可替换地，信息处理服务器和/或每个监测终端可以从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态。当确定目标设备处于异常状态时，信息处理服务器和/或每个监测终端可以利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
信息处理服务器和/或每个监测终端可以基于对预定范围内每种设备类型的多个设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成公共识别组件。公共识别组件能够识别与每种设备类型的处于异常状态的设备相关联的异常信息。公共识别组件能够基于人工智能或神经网络来识别与每种设备类型的处于异常状态的设备相关联的异常信息。
信息处理服务器和/或每个监测终端可以基于对预定范围内特定设备类型的多个设备的实测数据、历史数据、统计数据、额定参数和/或基本参数进行计算、设定、拟合、调整或训练来生成与特定设备类型的设备相关联的专用识别组件。其中与特定设备类型的设备相关联的专用识别组件能够确定特定设备类型的设备是否处于异常状态。
信息处理服务器和/或每个监测终端可以基于对目标设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成目标设备的专用识别组件。所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。专用识别组件能够基于人工智能或神经网络来识别目标设备是否处于异常状态。
信息处理服务器和/或每个监测终端可以基于对目标设备的实测数据和/或历史数据进行训练来生成目标设备的专用识别组件。所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。还包括利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常程度、异常类型、异常部件、异常原因、异常位置、健康状态和/或剩余寿命。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据，将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行比对，从而将比对结果确定为识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，根据识别的结果确定处于异常状态的目标设备的异常信息。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于目标设备的设备标识从数据库中获取与目标设备的所有异常信息中每种异常信息相关联的现有设备特征数据。将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，将比对结果作为识别的结果。根据识别的结果确定处于异常状态的目标设备的异常信息。
根据识别的结果确定异常状态的目标设备的异常信息包括：将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，以获得将所采集的目标设备的设备特征数据与每种异常信 息相关联的现有设备特征数据的匹配度；将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为识别的结果；将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
还包括，确定异常状态的存在范围，并根据异常状态的存在范围确定目标设备的需要检查和/或维修的范围。基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
异常信息至少包括：异常类型、异常程度、异常位置和/或异常部件。当目标设备为地下电缆时，异常类型包括：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、击穿、爆管、爆裂、电缆架倒塌、电缆掉落、挖、铲、撞击、塌方、外物入侵、外物碰撞或局部放电。当目标设备为架空输电线路时，异常类型包括：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、闪络、局部放电、电塔倒塌、塔基松动、连接件松动、连接件断裂、舞动、覆冰或外物撞击。
基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
根据识别的结果确定异常状态的目标设备的异常信息包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行识别，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度；将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为匹配结果；将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
设备特征数据包括时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个。基于识别的结果确定目标设备是否处于异常状态包括：根据时域特征、频域 特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个是否超过设定的阈值来确定目标设备是否处于异常状态。
每个信号传感器至少包括一个声纹传感器，将所述声纹传感器以贴合方式设置在目标设备上用于监测所述目标设备在运行中的设备信号；所述监测到的设备信号为声纹信号、视频信号、红外视频信号、气象信号、位置信号、电气信号和/或电磁波信号。信息处理服务器和/或每个监测终端能够显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。信息处理服务器和/或每个监测终端按照设置的顺序和时间间隔依次显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。
图3为根据本发明实施方式的对目标设备进行监测的示意图。如图3所示，被设置在目标设备200的信号传感器202-1、202-2、202-3、202-4、202-5和202-6。信号传感器202-1至202-6可以属于同一监测终端或属于不同的监测终端。例如，监测终端A包括信号传感器202-1、202-2和202-3；并且监测终端B包括信号传感器202-4、202-5和202-6。
信号传感器202-1、202-2、202-3、202-4、202-5和202-6以贴合方式或邻接方式被设置在目标设备200处，用于监测所述目标设备200在运行中的设备信号。目标设备200可以是从多个待监测的设备中随机选择的/按照预设规则选择的一个设备。目标设备200可以为架空输电线路、或者目标设备200为电力变压器。目标设备可以为地下电缆、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电抗器、电容器、电抗器、风力发电机或太阳能电池装置。
目标设备200在运行中的设备信号是由目标设备所包括的部件所产生的并且能够在目标设备中传递的信号。例如，设备信号是由部件在运行时由于故障所产生的并且能够在目标设备中传递的信号。目标设备200在运行中的设备信号还可以是由目标设备200所包括的部件所产生的并且能够在目标设备附近的介质中传递的信号。例如，设备信号是由部件在运行时由于故障所产生的并且能够在目标设备200附近的介质中传递的信号。介质例如是空气、氮气等。
图4为根据本发明实施方式的基于递进式识别来监测异常状态的方法400的流程图。方法400从步骤401处开始。
在步骤401，使用每个监测终端的至少一个信号传感器来监测所述目标设备在运行中的设备信号。还包括使用至少一个监测终端对待监测的目 标设备的运行状态进行实时监测，其中将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处。
在步骤402，使用每个监测终端的采样单元对每个信号传感器所监测到的信息进行采样，从而获得设备信号。
在步骤403，使用每个监测终端的通信单元通过有线通信链路和/或无线通信链路与信息处理服务器进行通信。
在步骤404，使用每个监测终端的控制单元根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式。
在步骤405，利用信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号。
在步骤406，从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；当确定目标设备处于异常状态时，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
基于对预定范围内每种设备类型的多个设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成公共识别组件。所述公共识别组件能够识别与每种设备类型的处于异常状态的设备相关联的异常信息。或者，基于对预定范围内特定设备类型的多个设备的实测数据、历史数据、统计数据、额定参数和/或基本参数进行计算、设定、拟合、调整或训练来生成与特定设备类型的设备相关联的专用识别组件。与特定设备类型的设备相关联的专用识别组件能够确定特定设备类型的设备是否处于异常状态。
基于对目标设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成目标设备的专用识别组件。所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。
基于对目标设备的实测数据和/或历史数据进行训练来生成目标设备的专用识别组件。所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。还包括利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常程度、异常类型、异常部件、异常原因、异常 位置、健康状态和/或剩余寿命。
当目标设备为地下电缆时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、击穿、爆管、爆裂、电缆架倒塌、电缆掉落、挖、铲、撞击、塌方、外物入侵、外物碰撞或局部放电。当目标设备为架空输电线路时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、闪络、局部放电、电塔倒塌、塔基松动、连接件松动、连接件断裂、舞动、覆冰或外物撞击。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据，将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行比对，从而将比对结果确定为识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果。根据识别的结果确定处于异常状态的目标设备的异常信息。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于目标设备的设备标识从数据库中获取与目标设备的所有异常信息中每种异常信息相关联的现有设备特征数据。将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，将比对结果作为识别的结果。根据识别的结果确定处于异常状态的目标设备的异常信息。
根据识别的结果确定异常状态的目标设备的异常信息包括：将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度。将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为识别的结果。将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
还包括，确定异常状态的存在范围，并根据异常状态的存在范围确定目标设备的需要检查和/或维修的范围。基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
根据识别的结果确定异常状态的目标设备的异常信息包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行识别，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度。将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为匹配结果。将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
所述设备特征数据包括时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个。基于识别的结果确定目标设备是否处于异常状态包括：根据时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个是否超过设定的阈值来确定目标设备是否处于异常状态。
每个信号传感器至少包括一个声纹传感器，将所述声纹传感器以贴合方式设置在目标设备上用于监测所述目标设备在运行中的设备信号。所述 监测到的设备信号为声纹信号、视频信号、红外视频信号、气象信号、位置信号、电气信号和/或电磁波信号。
所述目标设备为架空输电线路、或者所述目标设备为电力变压器。所述目标设备为地下电缆、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电抗器、电容器、电阻器、风力发电机或太阳能电池装置。信息处理服务器或每个监测终端能够显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。信息处理服务器或每个监测终端按照设置的顺序和时间间隔依次显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。
图5为根据本发明另一实施方式的基于递进式识别来监测异常状态的方法500的流程图。方法500从步骤501处开始。
在步骤501，使用每个监测终端的至少一个信号传感器来监测所述目标设备在运行中的设备信号。使用至少一个监测终端对待监测的目标设备的运行状态进行实时监测，其中将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处。
在步骤502，使用每个监测终端的采样单元对每个信号传感器所监测到的信息进行采样，从而获得设备信号。
在步骤503，使用每个监测终端的通信单元通过有线通信链路和/或无线通信链路与信息处理服务器进行通信。
在步骤504，使用每个监测终端的控制单元根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式。
在步骤505，使用每个监测终端的识别单元，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态。
在步骤506，信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号，信息处理服务器和/或每个监测终端利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
基于对预定范围内每种设备类型的多个设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成公共识别组件。所述公共识别组件能够识别与每种设备类型的处于异常状态的设备相关联的异常信息。或者，基于对预定范围内特定设备类 型的多个设备的实测数据、历史数据、统计数据、额定参数和/或基本参数进行计算、设定、拟合、调整或训练来生成与特定设备类型的设备相关联的专用识别组件。与特定设备类型的设备相关联的专用识别组件能够确定特定设备类型的设备是否处于异常状态。
基于对目标设备的实测数据、历史数据、统计数据、额定参数或者/和基本参数进行计算、设定、拟合、调整或训练来生成目标设备的专用识别组件。所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。
基于对目标设备的实测数据和/或历史数据进行训练来生成目标设备的专用识别组件。所述目标设备的专用识别组件能够确定目标设备是否处于异常状态。还包括利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常程度、异常类型、异常部件、异常原因、异常位置、健康状态和/或剩余寿命。
当目标设备为地下电缆时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、击穿、爆管、爆裂、电缆架倒塌、电缆掉落、挖、铲、撞击、塌方、外物入侵、外物碰撞或局部放电。当目标设备为架空输电线路时，异常信息包括异常类型：短路、单相短路、三相短路、两相短路、两相短路接地、断线、单相断线、两相断线、三相断线、闪络、局部放电、电塔倒塌、塔基松动、连接件松动、连接件断裂、舞动、覆冰或外物撞击。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据，将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行比对，从而将比对结果确定为识别的结果，基于识别的结果确定目标设备是否处于异常状态。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的 目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果。根据识别的结果确定处于异常状态的目标设备的异常信息。
其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据。基于目标设备的设备标识从数据库中获取与目标设备的所有异常信息中每种异常信息相关联的现有设备特征数据。将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，将比对结果作为识别的结果。根据识别的结果确定处于异常状态的目标设备的异常信息。
根据识别的结果确定异常状态的目标设备的异常信息包括：将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行比对，以获得将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度。将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为识别的结果。将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
还包括，确定异常状态的存在范围，并根据异常状态的存在范围确定目标设备的需要检查和/或维修的范围。基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别与目标设备处于正常状态时的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
基于识别的结果确定目标设备是否处于异常状态包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
根据识别的结果确定异常状态的目标设备的异常信息包括：基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析将所采集的目标设备的设备特征数据分别同与每种异常信息相关联的现有设备特征数据进行识别，以获得将所采集的目标设备的设备特征数 据与每种异常信息相关联的现有设备特征数据的匹配度。将所采集的目标设备的设备特征数据与每种异常信息相关联的现有设备特征数据的匹配度作为匹配结果。将与所采集的目标设备的设备特征数据匹配度最大的异常信息确定为处于异常状态的目标设备的异常信息。
所述设备特征数据包括时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个。基于识别的结果确定目标设备是否处于异常状态包括：根据时域特征、频域特征、时频域特征、功率谱特征、频谱包络线特征、微分特征、积分特征和人工智能特征中的至少一个是否超过设定的阈值来确定目标设备是否处于异常状态。
每个信号传感器至少包括一个声纹传感器，将所述声纹传感器以贴合方式设置在目标设备上用于监测所述目标设备在运行中的设备信号。所述监测到的设备信号为声纹信号、视频信号、红外视频信号、气象信号、位置信号、电气信号和/或电磁波信号。
所述目标设备为架空输电线路、或者所述目标设备为电力变压器。所述目标设备为地下电缆、配电柜、环网柜、开关柜、断路器、隔离开关、接地开关、电抗器、电容器、电阻器、风力发电机或太阳能电池装置。信息处理服务器或每个监测终端能够显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。信息处理服务器或每个监测终端按照设置的顺序和时间间隔依次显示或播放设备信号、目标设备的运行状态和/或目标设备的异常特征。

Claims (10)

  1. 一种基于递进式识别来监测异常状态的系统，所述系统包括：
    至少一个监测终端，用于对待监测的目标设备的运行状态进行实时监测，将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；
    其中每个监测终端包括：
    至少一个信号传感器，用于监测所述目标设备在运行中的设备信号；
    采样单元，对每个信号传感器所监测到的信息进行采样，从而获得设备信号；
    通信单元，通过有线通信链路和/或无线通信链路与信息处理服务器进行通信；
    控制单元，根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；
    信息处理服务器，通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号；
    从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；
    当确定目标设备处于异常状态时，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
  2. 一种基于递进式识别来监测异常状态的系统，所述系统包括：
    至少一个监测终端，用于对待监测的目标设备的运行状态进行实时监测，将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；
    其中每个监测终端包括：
    至少一个信号传感器，用于监测所述目标设备在运行中的设备信号；
    采样单元，对每个信号传感器所监测到的信息进行采样，从而获得设备信号；
    通信单元，通过有线通信链路和/或无线通信链路与信息处理服务器进行通信，当确定目标设备处于异常状态时，经由有线通信链路和/或无线通信链路将设备信号发送给信息处理服务器；
    控制单元，根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；
    识别单元，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；
    信息处理服务器，通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
  3. 根据权利要求1或2所述的系统，其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：
    专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；
    基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
  4. 根据权利要求1或2所述的系统，其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：
    公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；
    基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标 设备的设备特征数据进行识别，以确定识别的结果，
    根据识别的结果确定处于异常状态的目标设备的异常信息。
  5. 权利要求4所述的系统，所述基于识别的结果确定目标设备是否处于异常状态包括：
    基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
  6. 一种基于递进式识别来监测异常状态的方法，所述方法包括：
    使用至少一个监测终端对待监测的目标设备的运行状态进行实时监测，其中将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；
    使用每个监测终端的至少一个信号传感器来监测所述目标设备在运行中的设备信号；
    使用每个监测终端的采样单元对每个信号传感器所监测到的信息进行采样，从而获得设备信号；
    使用每个监测终端的通信单元通过有线通信链路和/或无线通信链路与信息处理服务器进行通信；
    使用每个监测终端的控制单元根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；
    利用信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号；
    从多个专用识别组件中选择与目标设备相关联的专用识别组件，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；
    当确定目标设备处于异常状态时，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
  7. 一种基于递进式识别来监测异常状态的方法，所述方法包括：
    使用至少一个监测终端对待监测的目标设备的运行状态进行实时监测，其中将至少一个监测终端中的每个监测终端设置在所述目标设备的不同位置处；
    使用每个监测终端的至少一个信号传感器来监测所述目标设备在运行中的设备信号；
    使用每个监测终端的采样单元对每个信号传感器所监测到的信息进行采样，从而获得设备信号；
    使用每个监测终端的通信单元通过有线通信链路和/或无线通信链路与信息处理服务器进行通信；
    使用每个监测终端的控制单元根据控制指令来控制每个监测终端的启动运行、停止运行、位置移动、软件更新、参数设置、信号采样、数据处理、异常诊断和/或工作模式；
    使用每个监测终端的识别单元，利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态；
    信息处理服务器通过有线通信链路和/或无线通信链路与至少一个监测终端进行通信，基于有线通信链路和/或无线通信链路从至少一个监测终端接收设备信号，利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息。
  8. 根据权利要求6或7所述的方法，其中利用专用识别组件对设备信号进行识别，以基于识别的结果确定目标设备是否处于异常状态包括：
    专用识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；
    基于人工智能判别子组件和/或神经网络判别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
  9. 根据权利要求6或7所述的方法，其中利用公共识别组件对设备信号进行识别，以确定处于异常状态的目标设备的异常信息包括：
    公共识别组件对所采集的目标设备的设备信号进行特征提取，从而获得所采集的目标设备的设备特征数据；
    基于人工智能识别子组件和/或神经网络识别子组件对所采集的目标设备的设备特征数据进行识别，以确定识别的结果，
    根据识别的结果确定处于异常状态的目标设备的异常信息。
  10. 权利要求9所述的方法，所述基于识别的结果确定目标设备是否处于异常状态包括：
    基于使用理论公式、经验曲线、试验曲线、实测曲线、统计曲线、拟合曲线或大数据分析对所采集的目标设备的设备特征数据进行识别，从而确定识别的结果，基于识别的结果确定目标设备是否处于异常状态。
PCT/CN2021/107897 2020-11-18 2021-07-22 一种基于递进式识别来监测异常状态的系统及方法 WO2022105286A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011295083.3 2020-11-18
CN202011295083.3A CN112418053B (zh) 2020-11-18 2020-11-18 一种基于递进式识别来监测异常状态的系统及方法

Publications (1)

Publication Number Publication Date
WO2022105286A1 true WO2022105286A1 (zh) 2022-05-27

Family

ID=74774423

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/107897 WO2022105286A1 (zh) 2020-11-18 2021-07-22 一种基于递进式识别来监测异常状态的系统及方法

Country Status (2)

Country Link
CN (1) CN112418053B (zh)
WO (1) WO2022105286A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115453254A (zh) * 2022-11-11 2022-12-09 浙江万胜智能科技股份有限公司 一种基于专变采集终端的电能质量监测方法及系统
CN116866047A (zh) * 2023-07-18 2023-10-10 山东溯源安全科技有限公司 工业设备网络中恶意设备的确定方法、介质及设备

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112418053B (zh) * 2020-11-18 2023-12-26 华北电力大学 一种基于递进式识别来监测异常状态的系统及方法
CN115865683B (zh) * 2023-03-02 2023-05-23 山东创安交通预警工程有限公司 智慧社区设备管理系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101753992A (zh) * 2008-12-17 2010-06-23 深圳市先进智能技术研究所 一种多模态智能监控系统和方法
US20150181171A1 (en) * 2013-12-24 2015-06-25 Kt Corporation Interactive and targeted monitoring service
CN111504385A (zh) * 2020-05-13 2020-08-07 兰州工业学院 一种适用于机械设备异常状态多参数监测装置及方法
CN111600383A (zh) * 2020-05-12 2020-08-28 合肥中科类脑智能技术有限公司 输电线路智能一体化巡检装置
CN111650922A (zh) * 2020-05-27 2020-09-11 北京金茂绿建科技有限公司 一种智能家居异常检测方法和装置
CN112418053A (zh) * 2020-11-18 2021-02-26 华北电力大学 一种基于递进式识别来监测异常状态的系统及方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101753992A (zh) * 2008-12-17 2010-06-23 深圳市先进智能技术研究所 一种多模态智能监控系统和方法
US20150181171A1 (en) * 2013-12-24 2015-06-25 Kt Corporation Interactive and targeted monitoring service
CN111600383A (zh) * 2020-05-12 2020-08-28 合肥中科类脑智能技术有限公司 输电线路智能一体化巡检装置
CN111504385A (zh) * 2020-05-13 2020-08-07 兰州工业学院 一种适用于机械设备异常状态多参数监测装置及方法
CN111650922A (zh) * 2020-05-27 2020-09-11 北京金茂绿建科技有限公司 一种智能家居异常检测方法和装置
CN112418053A (zh) * 2020-11-18 2021-02-26 华北电力大学 一种基于递进式识别来监测异常状态的系统及方法

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115453254A (zh) * 2022-11-11 2022-12-09 浙江万胜智能科技股份有限公司 一种基于专变采集终端的电能质量监测方法及系统
CN115453254B (zh) * 2022-11-11 2023-05-02 浙江万胜智能科技股份有限公司 一种基于专变采集终端的电能质量监测方法及系统
CN116866047A (zh) * 2023-07-18 2023-10-10 山东溯源安全科技有限公司 工业设备网络中恶意设备的确定方法、介质及设备

Also Published As

Publication number Publication date
CN112418053B (zh) 2023-12-26
CN112418053A (zh) 2021-02-26

Similar Documents

Publication Publication Date Title
WO2022105286A1 (zh) 一种基于递进式识别来监测异常状态的系统及方法
WO2022105287A1 (zh) 一种基于声纹诊断来监测异常状态的系统及方法
EP2082246B1 (en) Cable fault detection
US8174268B2 (en) Protective relay monitoring system and method of comparing behavior patterns
KR101574615B1 (ko) 통계적 위상패턴 기반의 신호검출 방법을 이용한 전력설비 부분 방전 감시진단 시스템
WO2022105285A1 (zh) 一种基于冲击事件来监测异常状态的系统及方法
KR102192609B1 (ko) 순시유도전압센서를 이용한 부분방전 모니터링 감시기능이 탑재된 배전반(고압 배전반, 저압 배전반, 분전반, 모터제어반)
CN111551352B (zh) 一种gis设备的断路器的状态的检测方法及系统
CN108418304B (zh) 变电站二次回路状态监测方法、装置及系统
CN103336228B (zh) 一种发电机定子绝缘故障检修方法和系统
US9188632B1 (en) Self learning radio frequency monitoring system for identifying and locating faults in electrical distribution systems
CN113049920A (zh) 一种开关柜局部放电和温度在线检测系统及其检测方法
KR101843792B1 (ko) 전력설비의 부분방전 판정방법
CN110849419A (zh) 一种开关柜状态监测诊断方法及系统
CN110987169A (zh) 一种gis母线振动情况监测装置及监测方法
CN114113995B (zh) 一种gis用隔离开关机械故障检测方法及系统
CN210952934U (zh) 一种开关柜状态监测诊断系统
Adami et al. New approach to improve high-voltage transmission line reliability
CN114114001A (zh) 一种gis设备隔离开关机械状态监测方法及系统
CN114167119A (zh) 一种监控绝缘子出现爬电、污闪、闪络现象的系统及方法
CN110849611B (zh) 一种利用开关辐射场在线检测断路器机械特性的方法
Achatz et al. Features and benefits of UHF partial discharge monitoring systems for GIS
Neumann et al. The impact of insulation monitoring and diagnostics on reliability and exploitation of service life
CN116338390A (zh) 一种基于局部放电在线监测的环网柜绝缘缺陷评估方法
KR20230078353A (ko) 부분방전 측정 시스템 및 이를 위한 동작 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21893438

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21893438

Country of ref document: EP

Kind code of ref document: A1