WO2020057402A1 - 系统状态监视方法、装置和存储介质 - Google Patents

系统状态监视方法、装置和存储介质 Download PDF

Info

Publication number
WO2020057402A1
WO2020057402A1 PCT/CN2019/105232 CN2019105232W WO2020057402A1 WO 2020057402 A1 WO2020057402 A1 WO 2020057402A1 CN 2019105232 W CN2019105232 W CN 2019105232W WO 2020057402 A1 WO2020057402 A1 WO 2020057402A1
Authority
WO
WIPO (PCT)
Prior art keywords
operation mode
operating
state
mode
current
Prior art date
Application number
PCT/CN2019/105232
Other languages
English (en)
French (fr)
Inventor
周林飞
梁潇
李晶
施尼盖斯·丹尼尔
Original Assignee
西门子（中国）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 西门子（中国）有限公司 filed Critical 西门子（中国）有限公司
Priority to US17/277,770 priority Critical patent/US11847619B2/en
Priority to EP19861456.2A priority patent/EP3839743B1/en
Publication of WO2020057402A1 publication Critical patent/WO2020057402A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
    • G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
    • G05B19/0428—Safety, monitoring
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
    • G06F11/0706—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
    • G06F11/0766—Error or fault reporting or storing
    • G06F11/0772—Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
    • G06F11/0766—Error or fault reporting or storing
    • G06F11/0787—Storage of error reports, e.g. persistent data storage, storage using memory protection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3065—Monitoring arrangements determined by the means or processing involved in reporting the monitored data
    • G06F11/3072—Monitoring arrangements determined by the means or processing involved in reporting the monitored data where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting
    • G06F11/3079—Monitoring arrangements determined by the means or processing involved in reporting the monitored data where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting the data filtering being achieved by reporting only the changes of the monitored data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/32—Monitoring with visual or acoustical indication of the functioning of the machine
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G06F16/9024—Graphs; Linked lists
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/40—Data acquisition and logging
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
    • H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
    • H02J13/00006—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
    • H02J13/00016—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment using a wired telecommunication network or a data transmission bus
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
    • H02J13/00006—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
    • H02J13/00022—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment using wireless data transmission
    • 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
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
    • Y02B90/20—Smart grids as enabling technology in buildings sector
    • 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
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/30—State monitoring, e.g. fault, temperature monitoring, insulator monitoring, corona discharge
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof

Definitions

  • the present invention relates to the field of information processing, and in particular, to a method, a device, and a storage medium for monitoring a system state.
  • the monitoring mechanism of the system status usually collects the operating parameter values of the system, determines whether the operating parameter values belong to the normal range, and gives the monitoring results of whether the system status is normal according to the judgment results.
  • the system operation law may change, such as the performance degradation due to the aging of equipment and the change of system operation law caused by process changes. Such changes also need to be discovered in time in order to adjust related supporting services, such as equipment maintenance, energy allocation, and so on.
  • the present invention provides a method, a device, and a storage medium for monitoring a system state, which can detect a change in a running rule of the system.
  • Embodiments provide a system status monitoring method, which may include:
  • the reference operating mode includes a plurality of operating states of the system in a unit time period
  • this method determines the usual multiple operating states of the monitored system as the reference operating mode, and can detect changes in the operating law of the system, thereby facilitating the timely adjustment of the monitored system or the peripheral cooperation mechanism and improving the performance of the monitored system.
  • the reference operation mode includes a plurality of first state distribution patterns corresponding to the plurality of operation states, and the current operation mode includes at least one second state distribution pattern;
  • determining whether the system is in the reference operation mode by comparing the current operation mode and the reference operation mode may include:
  • the comparison between the current operation mode and the reference operation mode is converted into a comparison of the similarity of the two sets of graphics.
  • Various graphics processing methods can be used to determine whether the system is in the reference operation mode, which is relatively simple to implement.
  • determining a similarity metric value of the first graphic group and the second graphic group may include:
  • using the metric value set to determine a similarity metric value of the first graphic group and the second graphic group includes one of the following:
  • the calculation amount can be further reduced, and the calculation efficiency can be improved.
  • the method may further include at least one of the following:
  • determining whether the system is in the reference operation mode by comparing the current operation mode and the reference operation mode includes:
  • using the set of metric values and the first weight and / or the second weight to determine whether the system is in a baseline operation mode includes one of the following:
  • the method may further include one of the following:
  • Determining a time distribution law of a system parameter value of the system Determining a time distribution law of a system parameter value of the system, determining a time distribution situation of the system parameter value according to the current operating data, and determining whether the system is in a state of time according to whether the time distribution situation conforms to the time distribution law
  • the reference operating mode Determining a time distribution law of a system parameter value of the system, determining a time distribution situation of the system parameter value according to the current operating data, and determining whether the system is in a state of time according to whether the time distribution situation conforms to the time distribution law The reference operating mode.
  • the method may further include:
  • determining a reference operating mode of the system and determining the current operating mode of the system based on the current operating data of the system include:
  • determining the baseline operating mode of the system may include one of the following:
  • Clustering a plurality of instances of operating states corresponding to the unit time period in the historical operating data of the system to obtain a plurality of clusters, and calculating an operating state for each of the plurality of clusters as the complex number One of the operating states;
  • a matrix is constructed by using a plurality of sets of running state instances corresponding to the unit time in the historical running data of the system, and the matrix is reduced in dimension by using singular value decomposition or principal component analysis to obtain the plurality of running states.
  • Embodiments also provide a system status monitoring device, which may include:
  • a reference mode determination module configured to determine a reference operation mode of the system, wherein the reference operation mode includes a plurality of operation states of the system in a unit time period;
  • a current mode determining module configured to determine a current running mode of the system according to the current running data of the system
  • a determination module is configured to determine whether the system is in the reference operation mode by comparing the current operation mode and the reference operation mode.
  • the system status monitoring device of each embodiment determines a plurality of usual operating states of the monitored system as the reference operation mode, and compares the current operation mode of the system with the reference operation mode to determine whether the system is in the reference operation mode, which can be detected. Changes in the system's operating laws facilitate adjustments to the monitored system or its peripheral cooperation mechanisms in a timely manner and improve the performance of the monitored system.
  • the determination module may include:
  • a similarity determining unit configured to use a plurality of first state distribution patterns corresponding to the plurality of operation states as a first graph group, and use at least one second state distribution pattern corresponding to the current operation mode as a second graph. Group to determine a similarity metric value of the first graphics group and the second graphics group;
  • a determining unit is configured to determine whether the system is in the reference operating mode according to the similarity metric value.
  • the system status monitoring device can use various graphic processing methods to determine whether the system is in the reference operation mode, which is relatively simple to implement.
  • the similarity determining unit is configured to:
  • the determination unit is configured to perform one of the following:
  • the calculation amount of the system state monitoring device can be further reduced, and the calculation efficiency can be improved.
  • the reference mode determination module may be further configured to determine a first weight of each of the plurality of first state distribution graphics according to the historical running data; or, the current mode determination module further It may be used to determine a second weight of each second state distribution pattern in the at least one second state distribution pattern according to the current running data.
  • the determination module may be further configured to determine, for each second state distribution pattern in the second pattern group, a third state distribution pattern that is most similar to the second state distribution pattern in the first pattern group.
  • the judgment result of the judgment module can be made more accurate.
  • the determination module may perform one of the following:
  • the determination module is configured to perform one of the following:
  • Determining a time distribution law of a system parameter value of the system Determining a time distribution law of a system parameter value of the system, determining a time distribution situation of the system parameter value according to the current operating data, and determining whether the system is in a state of time according to whether the time distribution situation conforms to the time distribution law
  • the reference operating mode Determining a time distribution law of a system parameter value of the system, determining a time distribution situation of the system parameter value according to the current operating data, and determining whether the system is in a state of time according to whether the time distribution situation conforms to the time distribution law The reference operating mode.
  • system state monitoring device may further include:
  • the storage module includes a first storage space and a second storage space, where the first storage space is larger than the second storage space;
  • a data acquisition module configured to store the operating data of the system in the first storage space and the second storage space respectively; when the first storage space is full, stop sending data to the first storage space To store operating data in the medium and trigger the reference mode determining module to determine the reference operating mode using the operating data in the first storage space; when the second storage space is full and the reference operating mode exists, Triggering the current mode determination module to determine the current operation mode by using the operation data in the second storage space;
  • the determination module is further configured to: when it is determined that the system is not in the reference operation mode, issue an alarm message and clear the first storage space.
  • the reference mode determination module may be configured to perform one of the following:
  • Clustering a plurality of instances of operating states corresponding to the unit time period in the historical operating data of the system to obtain a plurality of clusters, and calculating an operating state for each of the plurality of clusters as the complex number One of the operating states;
  • a matrix is constructed by using a plurality of sets of running state instances corresponding to the unit time in the historical running data of the system, and the matrix is reduced in dimension by using singular value decomposition or principal component analysis to obtain the plurality of running states.
  • Embodiments also provide a system status monitoring apparatus, including: a processor and a memory, where the memory stores computer-readable instructions that can cause the processor to execute the methods of the embodiments.
  • the system status monitoring device of each embodiment determines a plurality of usual operating states of the monitored system as the reference operation mode, and compares the current operation mode of the system with the reference operation mode to determine whether the system is in the reference operation mode. Changes in the system's operating laws facilitate adjustments to the monitored system or its peripheral cooperation mechanisms in a timely manner and improve the performance of the monitored system.
  • Embodiments also provide a computer-readable storage medium, in which computer-readable instructions are stored, which can cause a processor to execute the methods of the embodiments.
  • the storage medium of each embodiment can enable the processor to determine a plurality of usual operating states of the monitored system as the reference operation mode, and compare the current operation mode of the system with the reference operation mode to determine whether the system is in the reference operation mode. Changes in the operating rules of the system are detected, which facilitates timely adjustments to the monitored system or peripheral cooperation mechanisms and improves the performance of the monitored system.
  • FIG. 1 is a schematic diagram of an application scenario according to an embodiment of the present application.
  • FIG. 2A is a flowchart of a system status monitoring method according to an embodiment of the present application.
  • 2B and 2C are schematic diagrams of obtaining various operating state curves according to historical operating data in the embodiment of the present application.
  • 3A and 3B are flowcharts of a state monitoring method according to an embodiment of the present application.
  • FIGS. 4A and 4B are schematic diagrams of a system state monitoring device according to an embodiment of the present application.
  • FIG. 5 is a schematic diagram of an application scenario according to an embodiment of the present application.
  • FIG. 6 is a processing flowchart of a system state monitoring device according to an embodiment of the present application.
  • FIG. 7 is a schematic diagram of a method for storing data by a system state monitoring device in an embodiment of the present application.
  • FIG. 1 is a schematic diagram of an application scenario according to an embodiment of the present application.
  • This application scenario is only an example of various possible application scenarios, and the technical solutions of the embodiments can also be applied in other scenarios.
  • the scenario may include a system status monitoring device 12, a network 11, a sensor 13, and a monitored system 14.
  • the network 11 can implement communication between the system state monitoring device 12, the sensor 13, and the monitored system 14.
  • the network 11 may be a wired or wireless network using any communication protocol.
  • the monitored system 14 may be a collection of devices or components that operate to achieve a certain function.
  • the monitored system 14 may include one or more devices (such as mechanical equipment, chemical equipment, etc.), and may also include one or more components, such as motors, bearings, and the like.
  • the sensor 13 may be various contact or non-contact detection devices.
  • the sensor 13 may be selected from various devices for detecting parameters in various fields such as sound, light, electricity, magnetism, heat, force, such as a voltage sensor, a current sensor, a displacement sensor, a rotation speed sensor, and the like.
  • the system status monitoring device 12 can obtain the operating data of the monitored system 14 provided by the sensor 13 and / or the monitored system 14 through the network 11 and analyze the operating data by using the methods of the embodiments to determine the operating status of the monitored system 14 Is it normal?
  • the system status monitoring device 12 may include a processor 121 and an operation database 124.
  • the operation database 124 may store operation data of the monitored system 14.
  • the processor 121 may be one or a plurality of processors, and may be disposed in one or a plurality of physical devices.
  • the processor 121 may analyze the operation data to determine whether the operation condition of the monitored system 14 is normal.
  • FIG. 2A is a flowchart of a system status monitoring method according to an embodiment of the present application. This method can be executed by the system state monitoring device 12 shown in FIG. 1. The method may include the following steps.
  • step S21 a reference operating mode of the system is determined, wherein the reference operating mode includes a plurality of operating states of the system in a unit time period.
  • Step S22 Determine the current operating mode of the system according to the current operating data of the system.
  • Step S23 Determine whether the system is in the reference operation mode by comparing the current operation mode with the reference operation mode.
  • the system here can be any one or a collection of multiple devices, for example, it can be the monitored system 14 shown in FIG. 1.
  • the operating state refers to the performance of the system when it is running, and can be determined according to the value of one or more parameters when the system is running.
  • the operating state can be the value of a certain parameter of the system, such as voltage, current, speed, and so on.
  • the operating state may be a value calculated according to the values of a plurality of parameters, such as comprehensive energy consumption, equipment overall efficiency (OEE), and the like.
  • the unit time period can be a time period of any length, and can be set as needed, for example, several hours, one day, several days, one week, several weeks, one month, several months, and so on.
  • the reference operating mode refers to the operating law exhibited by the system during normal operation.
  • the reference operating mode can be derived from the system's historical operating data, or it can be set manually.
  • the reference operating mode may include a plurality of different operating states, such as multiple operating states when the system is operating at different loads, and so on.
  • Each operating state can be represented by charts, functions, or other forms.
  • an operating state may be represented by a distribution curve chart, a line chart, a histogram, etc. of the values of the parameters of the system in a unit time period.
  • a running state may be represented by a combination of one or a plurality of functions through function fitting and the like. Other examples can also use other forms to indicate the operating status.
  • the plurality of different operating states included in the reference operating mode may be obtained by performing various analyses on the historical operating data of the system. For example, multiple instances of operating states corresponding to a unit time period in historical operating data can be clustered to obtain a plurality of clusters, and an operating state is calculated for each of the plurality of clusters as a plurality of operating states in the benchmark operating mode. One of the operating states.
  • the running status instance refers to the actual running status including the collected running data.
  • the options of the clustering algorithm can be K-means, K-medoids, hierarchical clustering, etc.
  • a matrix can be constructed by using multiple sets of running state instances corresponding to unit time in historical running data, and the matrix can be reduced in dimension by using singular value decomposition or principal component analysis to obtain a plurality of running states in the reference running mode.
  • Other embodiments may also adopt other methods, which are not repeated here.
  • FIGS. 2B and 2C are schematic diagrams of obtaining various operating state curves according to historical operating data in the embodiment of the present application.
  • days are used as a unit time period, where each curve is a distribution of a day's running state in a day in historical running data, and is called an example of running state.
  • FIG. 2C shows several operation state curves obtained by analyzing and processing the operation state curves in FIG. 2B, which are used as reference operation modes.
  • the current operating data refers to the operating data of the system collected in a data collection period that is closest to the time when the current operating mode is determined.
  • Various embodiments can set a fixed or variable data collection cycle as required, such as several days, one week, several weeks, one month, several months, etc., and use the data collected during the end of each data collection cycle.
  • the operating data determines the current operating mode.
  • the current operating mode may include one or more operating states. For example, when the data collection cycle includes several days, these days may correspond to the same operating state, or there may be a plurality of days corresponding to the first operating state, a plurality of days corresponding to the second operating state, a plurality of days corresponding to the third operating state, and so on.
  • the form and determination method of the running state in the current running mode are similar to the form and determination method of the running state in the above-mentioned reference running mode, and are not repeated here.
  • comparing the current operation mode with the reference operation mode involves comparison of two sets of operation states, and the purpose is to determine whether the current operation mode conforms to the reference operation mode. If the current operating mode meets the reference operating mode, the system is in the reference operating mode; otherwise, it means that the system is not in the reference operating mode, that is, the system is out of its usual operating mode and is in an operating state with a large contrast from usual. .
  • the monitored system 14 is a factory with three production lines.
  • the system status monitoring device 12 is installed in a power supply company.
  • the sensor 13 is installed in the power distribution equipment that supplies power to the monitored system 14 and collects operating parameter values of the power distribution equipment.
  • the system state monitoring device 12 is provided with a reference operating mode based on the monitored system 14, including several previous operating states of the monitored system 14, such as a first operating state of a factory in a busy period and a second operating state in a normal period. And the third operating state during idle periods.
  • each operating state may be a distribution rule of operating parameter values of the power distribution equipment within a unit time period.
  • the system state monitoring device 12 determines the current operating mode of the monitored system 14 according to the operating data of the monitored system 14 collected in the previous data collection cycle.
  • the system status monitoring device 12 can determine that the operation status in the current operation mode is not the same as the three operation modes in the reference operation mode. It is determined that the monitored system 14 is not in the reference operation mode.
  • the power supply company can handle the situation according to the monitoring result (such as alarm information) of the system status monitoring device 12, for example, it can increase the power supply to the factory, replace the power distribution equipment with higher power, etc. Helps the monitored system 14 improve performance.
  • the embodiments determine whether the system is in the reference operation mode by determining a plurality of usual operating states of the monitored system as the reference operation mode, and comparing the current operation mode of the system with the reference operation mode to determine whether the system is in the reference operation mode, and the system operation can be detected. Regular changes, so as to facilitate timely adjustment of the monitored system or peripheral cooperation mechanism, and improve the performance of the monitored system.
  • the operating status of the system may be represented by a chart, a function, or other forms.
  • the reference operating mode may include a plurality of first state distribution graphics corresponding to a plurality of operating states
  • the current operating mode may include at least one second state distribution graphic.
  • the state distribution graph is a graphical representation of the distribution of the operating parameter values of the system in a unit time period.
  • the first state distribution pattern and the second state distribution pattern are only for distinguishing the respective state distribution patterns of the reference operation mode and the current operation mode, and "first" and "second” have no substantial meaning.
  • the other "first", “second", “third”, etc. below are similar and will not be described again.
  • a plurality of first state distribution figures may be used as the first figure group, and at least one second state distribution figure may be used as the second figure group, and a similarity measure between the first figure group and the second figure group may be determined. Value and determine whether the system is in the above reference operating mode based on the similarity metric value.
  • the similarity metric value is used to indicate the degree of similarity between the first graphic group and the second graphic group.
  • the similarity of the two sets of figures can be calculated, and the similarity is used as the similarity measure value.
  • the degree of difference between the two sets of figures can be calculated as the similarity measure.
  • the similarity metric value may be compared with a preset threshold, and whether the system is in a reference operation mode may be determined according to the comparison result. For example, when the similarity metric value is similarity, when the similarity metric value is greater than the threshold value, it can be determined that the system is in the baseline operation mode; when the similarity metric value is the degree of difference, when the similarity metric value is less than the threshold value, it can be determined The system is in baseline operation mode.
  • the comparison between the current operation mode and the reference operation mode is converted into the similarity comparison of the two sets of graphics, which facilitates the use of various graphics processing methods to determine whether the system is in the reference operation mode and is relatively simple to implement.
  • various graphic distance algorithms can be used, such as Euclidean distance, Hausdorff distance, Frechet distance, and the like.
  • the current operation mode is compared with the reference operation mode, the purpose is to determine whether the current operation mode conforms to the reference operation mode. Therefore, when determining the similarity metric value of the first graphics group and the second graphics group, for each second state in the second graphics group, the graphics can be found in the first graphics group, which is the most similar figure, that is, A third state distribution pattern that is most similar to the second state distribution pattern in the first pattern group is determined, and a graph similarity measurement value of the second state distribution pattern and the third state distribution pattern is obtained.
  • the measurement value set represents the current operating mode.
  • the closeness of each running state graph to the reference running mode reduces the amount of calculation and facilitates subsequent processing.
  • Various embodiments may use various analysis methods to process the metric value set, so as to obtain a similarity metric value of the first graphic group and the second graphic group.
  • an extreme value for example, a maximum value, a minimum value, etc.
  • the metric value set may be selected as the similarity metric value of the first graphic group and the second graphic group. For example, when the maximum value of the similarity corresponding to each running state graph in the metric value set (that is, the similarity to the most similar running state graph in the first graph group), the minimum value (that is, The degree of least dissimilarity) is used as a measure of similarity between the first graphics group and the second graphics group.
  • the maximum value is used as a measure of similarity between the first graphics group and the second graphics group.
  • a plurality of values may be selected from the metric value set, and the average value of the plurality of values is used as the similarity metric value of the first graph group and the second graph group.
  • a plurality of values may be selected from a set of metric values, and the sum of the plurality of values is used as a similarity metric value of the first graphic group and the second graphic group.
  • the values in the metric value set can be sorted according to a preset sorting method, and the plurality of values can be selected in order. For example, the N values that are ranked in the top N bits can be selected, or Select several values that are greater than or less than a preset threshold, and so on.
  • the calculation amount can be further reduced, and the calculation efficiency can be improved.
  • the number and frequency of occurrences of various operating states may be different, that is, some operating states are more common, and some operating states are relatively rare. Therefore, when determining the reference operation mode and / or the current operation mode, the weight of each operation state can be determined, and the weight of the operation state can be taken into account when comparing the reference operation mode and the current operation mode.
  • 3A and 3B are flowcharts of a state monitoring method according to an embodiment of the present application, which may be executed by a system state monitoring device 12.
  • FIG. 3A shows an embodiment in which the weights of each operation state in the reference operation mode are considered. As shown in FIG. 3A, the method may include the following steps.
  • Step S301 Determine a first weight of each of the plurality of first state distribution patterns.
  • the first weight may be determined according to the occurrence probability of the running state corresponding to the first state distribution graph.
  • the historical operating data of the system can be analyzed, such as clustering historical operating states, and obtaining the first state distribution graphs and their first weights based on the number of running state instances in each cluster cluster.
  • Step S302 For each second state distribution pattern in the second pattern group, determine a third state distribution pattern that is most similar to the second state distribution pattern in the first pattern group, and calculate the second state distribution pattern and the third state distribution pattern. A graph similarity measurement value of the state distribution graph, and a metric value set corresponding to the second graph group is obtained according to the graph similarity measurement value;
  • Step S303 Use the metric value set and the first weight to determine whether the system is in a reference operation mode.
  • figure 1 is the most similar to figure 3 in the first figure group, and the similarity is 0.9.
  • the weight of 3 is 0.6; the figure 2 is the most similar to the figure 4 in the first graphics group, the similarity is 0.3, and the weight of figure 3 is 0.2; then the weighted figure of graphic 1 has a similarity of 0.48 and the weight of figure 2
  • the similarity of the graph is 0.06. Therefore, when determining whether the system is in the reference operation mode, although the agreement between the graph 2 and the reference operation mode is low, the influence of the graph 2 on the final result is much smaller than the graph 1 because the weight of the graph 4 is small.
  • the current operation state can be made similar to the weighted operation state in the reference operation state
  • the effect of the degree on the result is greater than the similarity with the low-weight operating state in the reference operating state, thereby making the judgment result more accurate.
  • FIG. 3B shows an embodiment that considers the weights of each operating state in the current operating mode. As shown in FIG. 3B, the method may include the following steps.
  • Step S311 Determine a second weight of each second state distribution pattern in the at least one second state distribution pattern according to the current running data.
  • Step S312 For each second state distribution pattern in the second pattern group, determine a third state distribution pattern that is most similar to the second state distribution pattern in the first pattern group, and calculate the second state distribution pattern and the third state distribution. A graph similarity measurement value of the graph, and according to the graph similarity measurement value, a measurement value set corresponding to the second graph group is obtained;
  • step S313 it is determined whether the system is in a reference operation mode by using the metric value set and the second weight.
  • figure 1 is the most similar to figure 3 in the first figure group, and the similarity is 0.9.
  • the weight of 1 is 0.7; the figure 2 is the most similar to the figure 4 in the first figure group, the similarity is 0.3, and the weight of figure 2 is 0.3, then the weighted figure of graph 1 has a similarity of 0.63 and the weight of figure 2
  • the similarity of the graph is 0.09. Therefore, when determining whether the system is in the reference operation mode, although the agreement between the graph 2 and the reference operation mode is low, the influence of the graph 2 on the final result is much smaller than that of the graph 1 due to its smaller weight.
  • the first weight and the second weight may also be used to determine whether the system is in a reference operation mode.
  • the first weight and / or the second weight may be used to weight each value in the metric value set to obtain a weighted value for each value, and determine whether the system is in a reference operating mode according to an extreme value in the weighted value. For example, when the graph similarity metric value is similarity, it can be determined whether the system is in the reference operating mode according to the weighted minimum value; when the graph similarity metric value is the degree of difference, it can be determined whether the system is in the weighted maximum value. Baseline operation mode.
  • a plurality of values may be selected from a set of metric values, a weighted average of the plurality of values may be calculated according to the first weight and / or a second weight, and whether the system is in a baseline operating mode may be determined according to the weighted average.
  • a plurality of values may be selected from a set of metric values, a weighted sum of the plurality of values may be calculated according to the first weight and / or a second weight, and whether the system is in a baseline operating mode may be determined according to the weighted sum.
  • the values in the metric value set can be sorted according to a preset sorting method, and the plurality of values can be selected in order. For example, the N values that are ranked in the top N bits can be selected, or Several values selected according to a preset number ratio, etc.
  • the above extreme value, or the weighted average value, or the weighted sum may be compared with a preset threshold, and whether the system is in a reference operation mode may be determined according to the comparison result.
  • the occurrence of the operating state has a strong time rule, so the time of the current operating mode can be used to further verify whether the system is in the reference operating mode.
  • time information corresponding to each of a plurality of operating states in the reference operating mode may be determined, and whether the system is in the reference operating mode may be determined by comparing state characteristics and time information of the current operating mode with the plurality of operating states.
  • State characteristics refer to the information used to distinguish each state of the operation, such as the state of the state graph, the fitting function of the state of the state, and so on.
  • the time information corresponding to the running state is information characterizing the time when the system is in the running state.
  • the time information may be time period information, such as January to March, May 10 to July 20, and so on.
  • the time information can be an array or a curve, which indicates the probability of a running state occurring over a period of time. The above is just an example.
  • the time information may be expressed in other ways.
  • the time information is an array
  • the dimension of the array is determined by the statistical period and the change unit time period.
  • the time information can be a 12-dimensional array, where each value represents The occurrence probability of this running state in the corresponding month.
  • the statistical period and time period of the change unit can be set according to the actual situation.
  • the statistical period can be several months
  • the change unit time period can be several days, and so on.
  • the time information is a curve
  • the curve is a continuous curve in the time dimension
  • the first dimension of the curve is a statistical period (such as several months)
  • the second dimension is the appearance probability of the operating state.
  • the first operating state when comparing the time information of the current operating mode with the plurality of operating states, for each operating state in the current operating mode (hereinafter referred to as the first operating state), find the most similar operation in the reference operating mode. State (hereinafter referred to as the third operating state), comparing the time information corresponding to the first operating state and the third operating state, and judging the degree of consistency between the two time information; if they do not match, the current operating mode of the system and the reference operating mode can be determined Time does not match.
  • the third operating state State
  • the time information is a time period
  • the time period corresponding to the current first operation state enters the time period corresponding to the third operation state
  • a period of a corresponding time may be intercepted from the time information corresponding to the third running state according to the time information corresponding to the current first running state, and then compared.
  • the current time information corresponding to the first running state is September 15 to November 15, and the array or curve corresponding to September 15 to November 15 is intercepted from the time information corresponding to the third running state.
  • the comparison of the data or the similarity comparison of the curves is then used to determine the consistency of the time information.
  • the results of the similarity comparison of the state characteristics of the running state can be combined to determine whether the system is in the baseline running mode according to a preset strategy.
  • the preset strategy may be that the state characteristics of each running state in the current running mode are consistent with a running state in the reference running mode (for example, the graphic similarity measure meets a preset condition), and the time information and The running states are consistent (for example, the degree of time matching is greater than a preset threshold, etc.).
  • the preset strategy may be that the degree of conformity of the state characteristics corresponding to each operating state in the current operating mode and the degree of conformance of the time information are weighted according to preset weights, and whether the system is in the benchmark according to the calculation result Run mode.
  • the above is only an example, and other embodiments may also adopt other methods as required.
  • a preset reference operation mode corresponding to each time period may be determined, and whether the system is in the reference operation mode is determined by comparing the current operation mode with a reference operation mode corresponding to the same time period.
  • a reference operation mode corresponding to each time period may be determined according to a pre-divided time period (such as several days, weeks, and months). According to the time period corresponding to the current operation mode, the reference operation mode and the current operation mode in the same time period are compared according to the embodiments, so as to determine whether the state characteristics and time information of the current operation mode are consistent with the reference operation mode. If they match, it is determined that the system is in the reference operation mode.
  • the time distribution of a system parameter value of the system can be determined, the time distribution of the system parameter values can be determined according to the current operating data, and whether the system is in the baseline operation mode according to whether the time distribution conforms to the time distribution rule. For example, an extreme value or mean value of a system parameter value within a unit time period may be determined, and a distribution rule of the extreme value or mean value in time may be determined as a time distribution rule of a system parameter value, such as a curve or an array.
  • the time distribution of the parameter values of the system in the current operating mode is determined, and the time distribution is compared with the above time distribution rule to determine the degree of consistency of the time information. After that, the results of the similarity comparison of the state characteristics of the running state can be combined to determine whether the system is in the reference running mode according to a preset strategy. The specific method is as described above, and is not repeated here.
  • the reference operating mode of the system may be determined according to the historical operating data of the system, and the current operating mode may be determined according to the operating data of the system in the recent period of time (ie, current operating data).
  • the time span of historical running data is generally large.
  • the current operation mode reflect the current operation of the system, the time span of the current operation data is smaller. That is, the first time period for collecting historical operation data is generally longer than the time period for collecting current operation data.
  • historical running data and current running data may be collected separately according to a preset collection duration.
  • the collection period of historical operation data can be several months, years, etc.
  • the collection period of current operation data can be several days, weeks, months, etc.
  • pre-allocated data storage space can be used to control the amount of data of historical running data and current running data.
  • the operating data of the system may be acquired, and the operating data may be stored in the first storage space and the second storage space, respectively.
  • the first storage space is larger than the second storage space.
  • the first storage space may be hard disk space, cache space, and so on.
  • the first storage space is used to store historical running data. When the first storage space is full, stop storing the running data in the first storage space, and use the running data in the first storage space to determine the reference operating mode; when it is determined that the system is not When in the benchmark operation mode, an alarm message is issued and the first storage space is cleared.
  • the second storage space may be a hard disk space, a cache space, and the like.
  • the first storage space is used to store the current running data.
  • the running data in the second storage space can be used to determine the current running mode.
  • the system state monitoring method of each embodiment can be executed by the system state monitoring device 12.
  • 4A and 4B are schematic diagrams of a system state monitoring device 12 according to an embodiment of the present application. As shown in FIG. 4A, the system state monitoring device 12 may include a reference mode determination module 27, a current mode determination module 28, and a determination module 29.
  • the reference mode determination module 27 may determine a reference operation mode of the system, where the reference operation mode includes a plurality of operation states of the system in a unit time period.
  • the current mode determining module may determine the current running mode of the system according to the current running data of the system.
  • the determination module 29 may determine whether the system is in the reference operation mode by comparing the current operation mode and the reference operation mode.
  • the system state monitoring device 12 of each embodiment determines a plurality of usual operating states of the monitored system as the reference operation mode, and compares the current operation mode of the system with the reference operation mode to determine whether the system is in the reference operation mode. To the changes in the system's operating rules, so as to facilitate the timely adjustment of the monitored system or the peripheral cooperation mechanism, and improve the performance of the monitored system.
  • the determination module 29 may include: a similarity determination unit 291 and a determination unit 292.
  • the similarity determination unit 291 may determine a first pattern by using a plurality of first state distribution patterns corresponding to a plurality of operation states as a first pattern group, and using at least one second state distribution pattern corresponding to a current operation mode as a second pattern group. The measure of similarity between the group and the second graphic group.
  • the determination unit 292 may determine whether the system is in the reference operation mode according to the similarity metric value. In this way, the comparison between the current operation mode and the reference operation mode is converted into a comparison of the similarity of the two sets of graphics.
  • the system status monitoring device 12 can use various graphic processing methods to determine whether the system is in the reference operation mode, which is relatively simple to implement.
  • the similarity determination unit 291 may determine, for each of the second state distribution graphics in the second graphics group, A third state distribution pattern that is most similar to the second state distribution pattern, to obtain a graph similarity measurement value of the second state distribution pattern and the third state distribution pattern, and to obtain a measurement value set corresponding to the second graph group according to the graph similarity measurement value; A similarity measurement value of the first graphic group and the second graphic group is determined by using the measurement value set.
  • the determining unit 292 when processing the metric value set, may perform one of the following:
  • a plurality of values are selected from the metric value set, and the sum of the plurality of values is used as a similarity metric value of the first graphic group and the second graphic group.
  • the calculation amount of the system state monitoring device 12 can be further reduced, and the calculation efficiency can be improved.
  • the weight of each operating state may be taken into account when comparing the baseline operating mode with the current operating mode.
  • the reference mode determination module 27 may also determine a first weight of each of the first state distribution patterns in the plurality of first state distribution patterns according to historical running data.
  • the current mode determination module 28 may also determine a second weight of each second state distribution pattern in the at least one second state distribution pattern according to the current running data.
  • the determination module 29 may determine, for each second state distribution pattern in the second pattern group, a third state distribution pattern that is most similar to the second state distribution pattern in the first pattern group, and calculate the second state distribution pattern and the third state.
  • the graph similarity measurement value of the distribution graph is obtained according to the graph similarity measurement value, and the metric value set corresponding to the second graph group is obtained; the metric value set and the first weight and / or the second weight are used to determine whether the system is in a reference operation mode.
  • the determination result of the determination module 29 can be made more accurate.
  • the determination module 29 may perform one of the following:
  • a plurality of values are selected from a set of metric values, a weighted sum of the plurality of values is calculated according to the first weight and / or a second weight, and whether the system is in a reference operation mode is determined according to the weighted sum.
  • the time of the current operation mode may be used to further verify whether the system is in the reference operation mode.
  • the determination module 29 may perform one of the following:
  • the system status monitoring device 12 may further include a storage module 24 and a data acquisition module 26.
  • the storage module 24 may include a first storage space 241 and a second storage space 242.
  • the first storage space 241 is larger than the second storage space 242.
  • the data acquisition module 26 may store the operating data of the system in the first storage space 241 and the second storage space 242, respectively; when the first storage space 241 is full, stop storing the operation data in the first storage space 241, and trigger
  • the reference mode determination module uses the running data in the first storage space 241 to determine the reference running mode; when the second storage space 242 is full and the reference running mode exists, the current mode determination module is triggered to use the running data in the second storage space 242 Determine the current operating mode.
  • the determining module 29 may also issue an alarm message when it is determined that the system is not in the reference operation mode, and clear the first storage space 241.
  • the reference mode determination module 27 may perform one of the following:
  • Cluster the multiple running state instances corresponding to the unit time period in the historical running data of the system to obtain a plurality of clusters, and calculate an operating state for each of the plurality of clusters as one of the plurality of operating states. ;or
  • the matrix is constructed by using multiple sets of operating state instances corresponding to the unit time in the historical operating data of the system, and the matrix is reduced by singular value decomposition or principal component analysis to obtain a plurality of operating states.
  • the system status monitoring device 12 may further include a processor 21, a memory 20, and a communication module 22.
  • the communication module 22 is configured to communicate with other devices through any communication network.
  • the memory 20 may include an operating system 23, a monitoring module 25, and a storage module 24.
  • the monitoring module 25 includes computer-readable instructions corresponding to the aforementioned modules.
  • the processor 21 may execute computer-readable instructions in the memory 20 to implement the methods of the embodiments.
  • FIG. 5 is a schematic diagram of an application scenario according to an embodiment of the present application.
  • the system status monitoring device 12 of each embodiment is applied to a power supply enterprise (monitored system 14) to monitor the power consumption of each power consumption enterprise, thereby ensuring the efficient and stable operation of the power grid.
  • the sensor 13 can be installed on the power distribution equipment (such as a transformer, etc.) that supplies power to various power companies, continuously collecting the operating data (such as voltage, current, power, etc.) of the power distribution equipment and sending the data to the system status monitoring device 12.
  • the system state monitoring device 12 includes a processor 51 and a buffer 52 (for example, a RAM).
  • the processor 51 preprocesses time series data (for example, data cleaning, data normalization, etc.), and stores the preprocessed data in the database 53 to form time series data, that is, data arranged in time sequence.
  • the processor 51 uses a predefined mathematical model and data in the database 53 for modeling, and the model is stored in the cache 52.
  • the system state monitoring device 12 sends a signal to the alarm device 56, and the alarm device 56 may send the alarm information to the relevant personnel by email, SMS message, or other possible means.
  • the system state monitoring device 12 may create a log file and store it in a log database 55.
  • FIG. 6 is a processing flowchart of the system state monitoring device 12 according to the embodiment of the present application.
  • step S61 the system state monitoring device 12 preprocesses the time series data, and stores the preprocessed data into a first storage space and a second storage space in the database 53, respectively.
  • step S62 the system state monitoring device 12 performs modeling using data in the first storage space to obtain a reference operation model (that is, a mathematical model of the reference operation mode).
  • a reference operation model that is, a mathematical model of the reference operation mode
  • Step S63 Use the data in the second storage space to perform modeling to obtain a current running model (that is, a mathematical model of a current running mode).
  • step S64 the system state monitoring device 12 calculates the similarity between the current running model and the reference running model.
  • step S65 the system state monitoring device 12 determines whether the similarity is less than a preset threshold; if the similarity is less than the threshold, step S66 is performed; otherwise, the process returns to step S63.
  • step S66 the system state monitoring device 12 sends a signal to the alarm device 56, writes a log to the log database 55, clears the first storage space, and returns to step S62.
  • FIG. 7 is a schematic diagram of a method for storing data by the system state monitoring device 12 in the embodiment of the present application.
  • the first storage space is used to store time series data used to establish a reference running model
  • the second storage space 82 is used to store time series data used to establish a current running model.
  • step S81 the system state monitoring device 12 attempts to write to the first storage space and the second storage space.
  • the system state monitoring device 12 determines whether the first storage space is full at step S82, and if it is not full, writes the first storage space at step S83; if it is full, then determines at step S84 A reference running model exists, and if it exists, no processing is performed; if it does not exist, then the time series data in the first storage space is used to establish a reference running model in step S85.
  • the system state monitoring device 12 determines whether the second storage space is full at step S86, and if it is not full, writes time series data to the second storage space at step S87; if it is full, then at step S88 Determines whether a baseline running model currently exists. If the baseline running model does not currently exist, the stored old data is updated with the newly written data at step S810. If a reference running model currently exists, then in step S811, the current running model is established using the time series data in the second storage space.
  • the system state monitoring device 12 determines in step S812 whether the similarity between the current running model and the reference running model is less than a threshold. If the similarity is less than the threshold, then the first storage space is cleared in step S813; otherwise, no processing is performed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Computing Systems (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Software Systems (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Automation & Control Theory (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Computer Hardware Design (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Debugging And Monitoring (AREA)

Abstract

一种系统状态监视方法、装置和存储介质。该方法包括：确定系统的基准运行模式，其中，所述基准运行模式包括所述系统在一个单位时间段内的复数种运行状态；根据所述系统的当前运行数据确定所述系统的当前运行模式；通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式。通过将系统的复数种运行状态确定为基准运行模式，可以检测出系统运行规律的变化，从而便于及时对被监视系统或者外围配合机制进行调整，提高系统的性能。

Description

系统状态监视方法、装置和存储介质 技术领域
本发明涉及信息处理领域，特别是一种系统状态监视方法、装置和存储介质。
背景技术
目前，系统状态的监视机制，通常采集系统的运行参数值，判断该运行参数值是否属于正常的范围，并根据判断结果给出系统状态是否正常的监测结果。而现实中，即使运行参数值可能仍处于正常范围内，系统运行规律也可能发生变化，例如由于设备老化导致性能下降、工艺变革造成的系统运行规律变化，等。这种变化也需要及时发现，以便调整相关的配套服务，如设备维护、能源配置，等。
发明内容
有鉴于此，本发明提出了一种系统状态监视方法、装置和存储介质，可以检测出系统运行规律的变化。
各实施例提供了一种系统状态监视方法，可以包括：
确定系统的基准运行模式，其中，所述基准运行模式包括所述系统在一个单位时间段内的复数种运行状态；
根据所述系统的当前运行数据确定所述系统的当前运行模式；
通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式。
可见，该方法将被监视系统惯常的复数种运行状态确定为基准运行模式，可以检测出系统运行规律的变化，从而便于及时对被监视系统或者外围配合机制进行调整，提高被监视系统的性能。
一些实施例中，所述基准运行模式包括与所述复数种运行状态对应的复数个第一状态分布图形，所述当前运行模式包括至少一个第二状态分布图形；
其中，通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式可以包括：
将所述复数个第一状态分布图形作为第一图形组，将所述至少一个第二状态分布图形作为第二图形组，确定所述第一图形组与所述第二图形组的相似性度量值，根据所述相似性度量值确定所述系统是否处于所述基准运行模式。
可见，将当前运行模式与基准运行模式的比较转化为两组图形的相似度比较，可以采用各种图形处理方法来确定系统是否处于基准运行模式，实现起来较为简单。
一些实施例中，确定所述第一图形组与所述第二图形组的相似性度量值可以包括：
针对所述第二图形组中的每个第二状态分布图形，确定所述第一图形组中与所述第二状态分布图形最相似的第三状态分布图形，获得所述第二状态分布图形与所述第三状态分布图形的图形相似性度量值，根据所述图形相似性度量值得到所述第二图形组对应的度量值集合；
利用所述度量值集合确定所述第一图形组与所述第二图形组的相似性度量值。
可见，通过确定第二图形组中各图形在第一图形组中最相似的图形，根据其图形相似性度量值确定当前运行模式中各运行状态图形与基准运行模式的接近程度，可以大幅减少计算量。
一些实施例中，利用所述度量值集合确定所述第一图形组与所述第二图形组的相似性度量值包括以下之一：
选择所述度量值集合中的极值作为所述第一图形组与所述第二图形组的相似性度量值；
从所述度量值集合中选择复数个值，将所述复数个值的均值作为所述第一图形组与所述第二图形组的相似性度量值；
从所述度量值集合中选择复数个值，将所述复数个值的和作为所述第一图形组与所述第二图形组的相似性度量值。
通过从度量值集合中选择部分值来确定第一图形组与第二图形组的相似性度量值，可以进一步减少计算量，提高计算效率。
一些实施例中，该方法还可以包括以下中的至少一个：
确定所述复数个第一状态分布图形中每个第一状态分布图形的第一权重；
根据所述当前运行数据确定所述至少一个第二状态分布图形中每个第二状态分布图形的第二权重；
其中，通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式包括：
针对所述第二图形组中的每个第二状态分布图形，确定所述第一图形组中与所述第二状态分布图形最相似的第三状态分布图形，计算所述第二状态分布图形与所述第三状态分布图形的图形相似性度量值，根据所述图形相似性度量值得到所述第二图形组对应的度量值集合；
利用所述度量值集合和所述第一权重和/或所述第二权重确定所述系统是否处于所述基准运行模式。
通过将各运行状态的权重纳入考虑，可以使得判断结果更准确。
一些实施例中，利用所述度量值集合和所述第一权重和/或所述第二权重确定所述系统是否处于基准运行模式包括以下之一：
利用所述第一权重和/或所述第二权重对所述度量值集合中的每个值进行加权得到每个值的加权值，根据所述加权值中的极值确定所述系统是否处于所述基准运行模式；
从所述度量值集合中选择复数个值，根据所述第一权重和/或所述第二权重计算所述复数个值的加权均值，根据所述加权均值确定所述系统是否处于所述基准运行模式；
从所述度量值集合中选择复数个值，根据所述第一权重和/或所述第二权重计算所述复数个值的加权和，根据所述加权和确定所述系统是否处于所述基准运行模式。
这样，可以在判断系统是否处于基准运行模式时，考虑到当前运行模式和基准运行模式中各运行状态出现的概率，可以使得判断结果更准确。
一些实施例中，该方法还可以包括以下之一：
确定所述复数种运行状态中每种运行状态对应的时间信息，通过比较所述复数种运行状态和所述当前运行模式的状态特征和时间信息确定所述系统是否处于所述基准运行模式；
确定预设的各时段对应的基准运行模式，通过比较所述当前运行模式与对应相同时段的基准运行模式确定所述系统是否处于所述基准运行模式；
确定所述系统的一系统参数值的时间分布规律，根据所述当前运行数据确定所述系统参数值的时间分布情况，根据所述时间分布情况是否符合所述时间分布规律确定所述系统是否处于所述基准运行模式。
通过将当前运行模式的时间信息与基准运行模式中运行状态的时间信息进行比较，可以检测出系统运行状态在时间上的不规律，从而便于对系统的不规律运行进行处理。
一些实施例中，该方法还可以包括：
获取系统的运行数据，将所述运行数据分别存储在第一存储空间和第二存储空间中，所述第一存储空间大于所述第二存储空间；
当确定所述系统不处于所述基准运行模式时，发出告警信息，并清空所述第一存储空间；
其中，确定所述系统的基准运行模式，根据所述系统的当前运行数据确定所述系统的当前运行模式包括：
当所述第一存储空间已满时，停止向所述第一存储空间中存储运行数据，并利用所述第一存储空间中的运行数据确定所述基准运行模式；
当所述第二存储空间已满、且所述基准运行模式存在时，利用所述第二存储空间中的运行数据确定所述当前运行模式。
可见，利用两个不同大小的存储空间分别存储历史运行数据和当前运行数据，并且在其存储空间已满的时候触发各自对应的运行模式的确定过程，不需要使用定时器，实现简单。
一些实施例中，确定系统的基准运行模式可以包括以下之一：
对所述系统的历史运行数据中所述单位时间段对应的多种运行状态实例进行聚类，得到复数个簇，针对所述复数个簇中的每个簇计算一运行状态，作为所述复数种运行状态中的一个运行状态；
利用所述系统的历史运行数据中所述单位时间对应的多组运行状态实例构建矩阵，利用奇异值分解或主成分分析对所述矩阵进行降维，获得所述复数种运行状态。
通过上述处理，可以将大量的运行状态实例简化为少数的几种运行状态，方便后续与当前运行模式进行比较，降低了实现复杂度。
各实施例还提供一种系统状态监视装置，可以包括：
一个基准模式确定模块，用于确定系统的基准运行模式，其中，所述基准运行模式包括所述系统在一个单位时间段内的复数种运行状态；
一个当前模式确定模块，用于根据所述系统的当前运行数据确定所述系统的当前运行模式；
一个判定模块，用于通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式。
可见，各实施例的系统状态监视装置将被监视系统惯常的复数种运行状态确定为基准运行模式，将系统的当前运行模式与基准运行模式进行比较来判断系统是否处于基准运行模式，可以检测到系统运行规律的变化，从而便于及时对被监视系统或者外围配合机制进行调整，提高被监视系统的性能。
一些实施例中，判定模块可以包括：
一个相似度确定单元，用于将与所述复数种运行状态对应的复数个第一状态分布图形作为第一图形组，将所述当前运行模式对应的至少一个第二状态分布图形作为第二图形组，确定所述第一图形组与所述第二图形组的相似性度量值；
一个判定单元，用于根据所述相似性度量值确定所述系统是否处于所述基准运行模式。
这样，将当前运行模式与基准运行模式的比较转化为两组图形的相似度比较，系统状态监视装置可以采用各种图形处理方法来确定系统是否处于基准运行模式，实现起来较为简单。
一些实施例中，相似度确定单元用于，
针对所述第二图形组中的每个第二状态分布图形，确定所述第一图形组中与所述第二状态分布图形最相似的第三状态分布图形，获得所述第二状态分布图形与所述第三状态分布图形的图形相似性度量值，根据所述图形相似性度量值得到所述第二图形组对应的度量值集合；
利用所述度量值集合确定所述第一图形组与所述第二图形组的相似性度量值。
可见，通过确定第二图形组中各图形在第一图形组中最相似的图形，根据其图形相似性度量值确定当前运行模式中各运行状态图形与基准运行模式的接近程度，减少了相似度确定单元的计算量。
一些实施例中，判定单元用于执行以下之一：
选择所述度量值集合中的极值作为所述第一图形组与所述第二图形组的相似性度量值；
从所述度量值集合中选择复数个值，将所述复数个值的均值作为所述第一图形组与所述第二图形组的相似性度量值；
从所述度量值集合中选择复数个值，将所述复数个值的和作为所述第一图形组与所述第二图形组的相似性度量值。
通过从度量值集合中选择部分值来确定第一图形组与第二图形组的相似性度量值，可以进一步减少系统状态监视装置的计算量，提高计算效率。
一些实施例中，基准模式确定模块还可以用于，根据所述历史运行数据确定所述复数个第一状态分布图形中每个第一状态分布图形的第一权重；或者，当前模式确定模块还可以用于，根据所述当前运行数据确定所述至少一个第二状态分布图形中每个第二状态分布图形的第二权重。其中，判定模块还可以用于，针对所述第二图形组中的每个第二状态分布图形，确定所述第一图形组中与所述第二状态分布图形最相似的第三状态分布图形，计算所述第二状态分布图形与所述第三状态分布图形的图形相似性度量值，根据所述图形相似性度量值得到所述第二图形组对应的度量值集合；利用所述度量值集合和所述第一权重和/或所述第二权重确定所述系统是否处于所述基准运行模式。
通过将各运行状态的权重纳入考虑，可以使得判定模块的判断结果更准确。
一些实施例中，判定模块可以执行以下之一：
利用所述第一权重和/或所述第二权重对所述度量值集合中的每个值进行加权得到每个值的加权值，根据所述加权值中的极值确定所述系统是否处于所述基准运行模式；
从所述度量值集合中选择复数个值，根据所述第一权重和/或所述第二权重计算所述复数个值的加权均值，根据所述加权均值确定所述系统是否处于所述基准运行模式；
从所述度量值集合中选择复数个值，根据所述第一权重和/或所述第二权重计算所述复数个值的加权和，根据所述加权和确定所述系统是否处于所述基准运行模式。
这样，可以在判断系统是否处于基准运行模式时，考虑到当前运行模式和基准运行模式中各运行状态出现的概率，可以使得判断结果更准确。
一些实施例中，判定模块用于执行以下之一：
确定所述复数种运行状态中每种运行状态对应的时间信息，通过比较所述复数种运行状态和所述当前运行模式的状态特征和时间信息确定所述系统是否处于所述基准运行模式；
确定预设的各时段对应的基准运行模式，通过比较所述当前运行模式与对应相同时段的基准运行模式确定所述系统是否处于所述基准运行模式；
确定所述系统的一系统参数值的时间分布规律，根据所述当前运行数据确定所述系统参数值的时间分布情况，根据所述时间分布情况是否符合所述时间分布规律确定所述系统是否处于所述基准运行模式。
通过将当前运行模式的时间信息与基准运行模式中运行状态的时间信息进行比较，可以检测出系统运行状态在时间上的不规律，从而便于对系统的不规律运行进行处理。
一些实施例中，系统状态监视装置还可以包括：
存储模块，包括第一存储空间和第二存储空间，所述第一存储空间大于所述第二存储空间；
数据获取模块，用于将所述系统的运行数据分别存储在所述第一存储空间和所述第二存储空间中；当所述第一存储空间已满时，停止向所述第一存储空间中存储运行数据，并触发所述基准模式确定模块利用所述第一存储空间中的运行数据确定所述基准运行模式；当所述第二存储空间已满、且所述基准运行模式存在时，触发所述当前模式确定模块利用所述第二存储空间中的运行数据确定所述当前运行模式；
其中，所述判定模块进一步用于，当确定所述系统不处于所述基准运行模式时，发出告警信息，并清空所述第一存储空间。
可见，利用两个不同大小的存储空间分别存储历史运行数据和当前运行数据，并且在其存储空间已满的时候触发各自对应的运行模式的确定过程，不需要使用定时器，实 现简单。
一些实施例中，基准模式确定模块可以用于执行以下之一：
对所述系统的历史运行数据中所述单位时间段对应的多种运行状态实例进行聚类，得到复数个簇，针对所述复数个簇中的每个簇计算一运行状态，作为所述复数种运行状态中的一个运行状态；
利用所述系统的历史运行数据中所述单位时间对应的多组运行状态实例构建矩阵，利用奇异值分解或主成分分析对所述矩阵进行降维，获得所述复数种运行状态。
通过将当前运行模式的时间信息与基准运行模式中运行状态的时间信息进行比较，可以检测出系统运行状态在时间上的不规律，从而便于对系统的不规律运行进行处理。
各实施例还提供一种系统状态监视装置，包括：处理器和存储器，所述存储器存储有计算机可读指令，可以使所述处理器执行各实施例的方法。
可见，各实施例的系统状态监视装置将被监视系统惯常的复数种运行状态确定为基准运行模式，将系统的当前运行模式与基准运行模式进行比较来判断系统是否处于基准运行模式，可以检测到系统运行规律的变化，从而便于及时对被监视系统或者外围配合机制进行调整，提高被监视系统的性能。
各实施例还提供一种计算机可读存储介质，其中存储有计算机可读指令，可以使一处理器执行各实施例的方法。
可见，各实施例的存储介质可以使处理器将被监视系统惯常的复数种运行状态确定为基准运行模式，将系统的当前运行模式与基准运行模式进行比较来判断系统是否处于基准运行模式，可以检测到系统运行规律的变化，从而便于及时对被监视系统或者外围配合机制进行调整，提高被监视系统的性能。
附图说明
下面将通过参照附图详细描述本发明的优选实施例，使本领域的普通技术人员更清楚本发明的上述及其它特征和优点，附图中：
图1为本申请实施例的一种应用场景的示意图。
图2A为本申请实施例的系统状态监视方法的流程图。
图2B、2C为本申请实施例中根据历史运行数据获得多种运行状态曲线的示意图。
图3A、3B为本申请实施例的状态监视方法的流程图。
图4A、4B为本申请实施例的系统状态监视装置的示意图。
图5为本申请实施例的一个应用场景的示意图。
图6为本申请实施例的系统状态监视装置的处理流程图。
图7为本申请实施例中系统状态监视装置存储数据的方法示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，以下举实施例对本发明进一步详细说明。
本申请实施例提供了一种系统状态监视的技术方案，可以确定系统的基准运行模式，基准运行模式包括系统在一个单位时间段内的复数种运行状态；通过对基准运行模式和当前运行模式进行比较来确定系统是否处于基准运行模式。图1是本申请实施例的一种应用场景的示意图。该应用场景仅为各种可能的应用场景的一个例子，各实施例的技术方案也可以应用在其它的场景中。如图1所示，该场景可以包括系统状态监视装置12、网络11、传感器13和被监视系统14。
网络11可以实现系统状态监视装置12、传感器13和被监视系统14之间的通信。网络11可以是采用任意通信协议的有线或无线网络。
被监视系统14可以是为实现某种功能而运行的设备或部件的集合。被监视系统14可以包括一个或多个设备(如机械设备、化工设备，等)，也可以包括一个或多个部件，如电机、轴承，等。
传感器13可以是各种接触式或非接触式检测装置。传感器13可以选自各种对声、光、电、磁、热、力等各领域的参数进行检测的装置，例如电压传感器、电流传感器、位移传感器、转速传感器，等。
系统状态监视装置12可以通过网络11获得传感器13和/或被监视系统14提供的被监视系统14的运行数据，采用各实施例的方法对运行数据进行分析，从而确定被监视系统14的运行情况是否正常。
一些实施例中，系统状态监视装置12可以包括处理器121和运行数据库124。运行 数据库124可以存储被监视系统14的运行数据。处理器121可以是一个或者复数个处理器，可以设置在一个或复数个物理设备中。处理器121可以对运行数据进行分析，从而确定被监视系统14的运行情况是否正常。
图2A为本申请实施例的系统状态监视方法的流程图。该方法可以由图1所示的系统状态监视装置12执行。该方法可以包括以下步骤。
步骤S21，确定系统的基准运行模式，其中，基准运行模式包括系统在一个单位时间段内的复数种运行状态。
步骤S22，根据系统的当前运行数据确定系统的当前运行模式。
步骤S23，通过比较当前运行模式和基准运行模式确定系统是否处于基准运行模式。
这里的系统可以是任意一个或复数个设备的集合，例如，可以是图1所示的被监视系统14。
运行状态是指系统运行时的表现，可以根据系统运行时的一个或复数个参数的值确定。例如，运行状态可以为系统的某个参数的值，例如电压、电流、转速，等。又例如，运行状态可以是根据复数个参数的值计算得到的值，例如综合能耗、设备综合效率(OEE)，等。
单位时间段可以是任意长度的时间段，可以根据需要设置，例如若干个小时、一天、若干天、一周、若干周、一个月、若干个月，等。
由于系统可能根据订单需求、工况等情况而有复数种不同的运行状态，不同的运行状态对应系统的参数在时间上的不同分布情况。即使参数一直处于正常范围，系统的运行状态也可能处于不断的变化当中。基准运行模式是指该系统正常运行时表现出的运行规律。基准运行模式可以根据系统的历史运行数据得出，也可以由人工设定。
基准运行模式可以包括复数种不同的运行状态，例如系统在以不同的负荷运转时的多种运行状态，等。每种运行状态可以由图表、函数或其它形式表示。例如，一种运行状态可以由系统的参数在单位时间段内的值的分布曲线图、折线图、柱状图，等来表示。又例如，一种运行状态可以通过函数拟合等方式由一个或复数个函数的组合来表示。其它例子也可以采用其它的形式来表示运行状态。
各实施例中，基准运行模式包括的复数种不同的运行状态可以通过对系统的历史运行数据进行各种分析得到。例如，可以对历史运行数据中单位时间段对应的多种运行状态实例进行聚类，得到复数个簇，针对复数个簇中的每个簇计算一运行状态，作为基准运行模式中复数种运行状态中的一个运行状态。运行状态实例是指采集到的运行数据包 括的实际的运行状态。其中，聚类算法的选项可以有K-means，K-medoids，层次聚类等。又例如，可以利用历史运行数据中单位时间对应的多组运行状态实例构建矩阵，利用奇异值分解或主成分分析对该矩阵进行降维，获得基准运行模式中的复数种运行状态。其它实施例也可以采用其它的方法，这里不再赘述。例如，运行状态由图形表示时，图2B、2C为本申请实施例中根据历史运行数据获得多种运行状态曲线的示意图。图2B中以天作为单位时间段，其中每条曲线为历史运行数据中一天的运行状态在一天中的分布情况，称为一个运行状态实例。图2C为对图2B中的运行状态曲线进行分析和处理得到的若干条运行状态曲线，作为基准运行模式。
当前运行数据是指距离确定当前运行模式的时间最近的一个数据采集时间段中采集到的系统的运行数据。各实施例可以根据需要设置固定的或可变的数据采集周期，例如若干天、一周、若干周、一个月、若干个月，等，并在每个数据采集周期结束后利用该周期内采集到的运行数据确定当前运行模式。当前运行模式可以包括一种或复数种运行状态。例如，数据采集周期包括若干天时，可能这若干天都对应同一种运行状态，也可能其中有复数天对应第一运行状态，复数天对应第二运行状态，复数天对应第三运行状态，等。当前运行模式中的运行状态的形式和确定方法与上述的基准运行模式中的运行状态的形式和确定方法相似，这里不再赘述。
这里，比较当前运行模式和基准运行模式涉及两组运行状态的比较，目的是判断当前运行模式是否符合基准运行模式。如果当前运行模式符合基准运行模式，则说明系统处于基准运行模式；反之，则说明系统没有处于基准运行模式，也即，系统脱离了其惯常的运行模式，处于与往常反差较大的运行状态中。
例如，被监视系统14是一具有3条生产线的工厂，系统状态监视装置12设置在供电企业，传感器13设置在为被监视系统14供电的配电设备中，采集配电设备的运行参数值。系统状态监视装置12中设置有根据被监视系统14的基准运行模式，包括被监视系统14以往的几种运行状态，例如工厂在繁忙时期的第一运行状态、在平常时期的第二运行状态，以及空闲时期的第三运行状态。其中，每种运行状态可以是配电设备的运行参数值在单位时间段内的分布规律。系统状态监视装置12根据上一数据采集周期采集到的被监视系统14的运行数据确定被监视系统14的当前运行模式。当工厂为了扩大规模而新增了2条生产线时，通过比较当前运行模式和基准运行模式，系统状态监视装置12可以确定当前运行模式中的运行状态与基准运行模式中的三种运行模式均不匹配，因此确定被监视系统14没有处于基准运行模式。据此，供电企业可以根据系统状态监视装置 12的监视结果(例如告警信息)对该情况进行处理，例如可以加大对该工厂的供电量、更换更大功率的配电设备，等，从而有助于被监视系统14提高性能。
可以看出，各实施例通过将被监视系统惯常的复数种运行状态确定为基准运行模式，将系统的当前运行模式与基准运行模式进行比较来判断系统是否处于基准运行模式，可以检测到系统运行规律的变化，从而便于及时对被监视系统或者外围配合机制进行调整，提高被监视系统的性能。
各实施例中，系统的运行状态可以利用图表、函数或其它形式表示。下面举例说明。例如，当利用图形来表示系统的运行状态时，基准运行模式可以包括与复数种运行状态对应的复数个第一状态分布图形，当前运行模式可以包括至少一个第二状态分布图形。状态分布图形是系统的运行参数值在单位时间段中的分布情况的图形表示。第一状态分布图形和第二状态分布图形仅仅是为了区分基准运行模式和当前运行模式各自的状态分布图形，“第一”、“第二”不具有实质含义。下文其它“第一”、“第二”、“第三”等，情况类似，不再赘述。在上面的步骤S23中，可以将复数个第一状态分布图形作为第一图形组，将至少一个第二状态分布图形作为第二图形组，确定第一图形组与第二图形组的相似性度量值，并根据该相似性度量值确定系统是否处于上述基准运行模式。相似性度量值用于表示第一图形组与第二图形组的相似程度。一些例子中，可以计算两组图形的相似度，以相似度作为上述相似性度量值。另一些例子中，可以计算两组图形的差异度，作为上述相似性度量值。各实施例中，可以将相似性度量值与预设的阈值比较，根据比较结果确定系统是否处于基准运行模式。例如，当相似性度量值为相似度时，当相似性度量值大于阈值时，可以确定系统处于基准运行模式；当相似性度量值为差异度时，当相似性度量值小于阈值时，可以确定系统处于基准运行模式。
这样，将当前运行模式与基准运行模式的比较转化为两组图形的相似度比较，便于采用各种图形处理方法来确定系统是否处于基准运行模式，实现起来较为简单。
确定第一图形组与第二图形组的相似性度量值时，可以采用各种图形距离算法，例如欧式距离、Hausdorff距离、Frechet距离，等。
如前所述，比较当前运行模式和基准运行模式，目的是判断当前运行模式是否符合基准运行模式。因此，确定第一图形组与第二图形组的相似性度量值时，可以针对第二图形组中的每个第二状态分布图形，在第一图形组中查找与其最相似的图形，即，确定 第一图形组中与该第二状态分布图形最相似的第三状态分布图形，并获得第二状态分布图形与第三状态分布图形的图形相似性度量值。这样，利用各第二状态分布图形与其对应的第三状态分布图形的图形相似性度量值，可以得到第二图形组对应的度量值集合，从而利用该度量值集合确定第一图形组与第二图形组的相似性度量值。
可见，通过确定第二图形组中各图形在第一图形组中最相似的图形，根据其图形相似性度量值得到第二图形组对应的度量值集合，该度量值集合就表示当前运行模式中各运行状态图形与基准运行模式的接近程度，减少了计算量，且方便后续处理。
各实施例可以采用各种分析方法来处理该度量值集合，从而得到第一图形组与第二图形组的相似性度量值。
一些实施例中，可以选择度量值集合中的极值(例如最大值、最小值，等)作为第一图形组与第二图形组的相似性度量值。例如，当度量值集合中为各运行状态图形对应的相似度最大值(即与第一图形组中最相似的运行状态图形的相似度)时，可以从度量值集合中选择最小值(也即最不相似的程度)作为第一图形组与第二图形组的相似性度量值。又例如，当度量值集合中为各运行状态图形对应的差异度最小值(即与第一图形组中最相似的运行状态图形的差异度)时，可以从度量值集合中选择最大值(也即最不相似的程度)作为第一图形组与第二图形组的相似性度量值。
一些实施例中，可以从度量值集合中选择复数个值，将该复数个值的均值作为第一图形组与第二图形组的相似性度量值。
一些实施例中，可以从度量值集合中选择复数个值，将该复数个值的和作为第一图形组与第二图形组的相似性度量值。
从度量值集合中选择复数个值时，可以按照预设的排序方式对度量值集合中的值进行排序，按照顺序选取复数个值，例如可以选取排在前N位的N个值，或者可以选取大于或小于预设阈值的若干个值，等。
以上仅为举例，其它实施例还可以采用其它的策略来确定第一图形组与第二图形组的相似性度量值。
通过从度量值集合中选择部分值来确定第一图形组与第二图形组的相似性度量值，可以进一步减少计算量，提高计算效率。
各实施例中，各种运行状态出现的次数、频率会有差异，即有的运行状态比较多见，有的运行状态比较少见。因此，可以在确定基准运行模式和/或当前运行模式时，确定各运行状态的权重，并在比较基准运行模式和当前运行模式时将运行状态的权重考虑进去。图3A、3B为本申请实施例的状态监视方法的流程图，可以由系统状态监视装置12执行。图3A示出了考虑基准运行模式中各运行状态的权重的实施例。如图3A所示，该方法可以包括以下步骤。
步骤S301，确定复数个第一状态分布图形中每个第一状态分布图形的第一权重。
一些例子中，可以根据第一状态分布图形对应的运行状态出现的概率来确定第一权重。例如，可以对系统的历史运行数据进行分析，例如对历史运行状态进行聚类，根据每一聚类簇中运行状态实例的数量，等方法，得到各第一状态分布图形及其第一权重。
步骤S302，针对第二图形组中的每个第二状态分布图形，确定第一图形组中与所述第二状态分布图形最相似的第三状态分布图形，计算第二状态分布图形与第三状态分布图形的图形相似性度量值，根据该图形相似性度量值得到第二图形组对应的度量值集合；
步骤S303，利用该度量值集合和第一权重确定系统是否处于基准运行模式。
例如，以图形之间的相似度作为图形相似性度量值时，假设第二图形组中包括图形1和图形2，图形1与第一图形组中的图形3最相似，相似度为0.9，图形3的权重为0.6；图形2与第一图形组中的图形4最相似，相似度为0.3，图形3的权重为0.2,；则图形1的加权后的图形相似度为0.48，图形2的加权后的图形相似度为0.06。从而，确定系统是否处于基准运行模式时，虽然图形2与基准运行模式的相符度较低，但是由于图形4的权重较小，图形2对最终结果的影响比图形1小很多。
这样，可以在判断系统是否处于基准运行模式时，考虑到当前运行状态在基准运行状态中对应的运行状态出现的概率，则可以使得当前运行状态中与基准运行状态中权重高的运行状态的相似度对结果的影响大于与基准运行状态中权重低的运行状态的相似度，从而使得判断结果更准确。
图3B示出了考虑当前运行模式中各运行状态的权重的实施例。如图3B所示，该方法可以包括以下步骤。
步骤S311，根据当前运行数据确定至少一个第二状态分布图形中每个第二状态分布图形的第二权重。
步骤S312，针对第二图形组中的每个第二状态分布图形，确定第一图形组中与第二 状态分布图形最相似的第三状态分布图形，计算第二状态分布图形与第三状态分布图形的图形相似性度量值，根据图形相似性度量值得到第二图形组对应的度量值集合；
步骤S313，利用度量值集合和第二权重确定系统是否处于基准运行模式。
例如，以图形之间的相似度作为图形相似性度量值时，假设第二图形组中包括图形1和图形2，图形1与第一图形组中的图形3最相似，相似度为0.9，图形1的权重为0.7；图形2与第一图形组中的图形4最相似，相似度为0.3，图形2的权重为0.3,；则图形1的加权后的图形相似度为0.63，图形2的加权后的图形相似度为0.09。从而，确定系统是否处于基准运行模式时，虽然图形2与基准运行模式的相符度较低，但是由于其权重较小，图形2对最终结果的影响比图形1小很多。
这样，可以在判断系统是否处于基准运行模式时，考虑到当前运行状态中各运行状态出现的概率，则可以使得当前运行状态中权重高的运行状态的相似度对结果的影响大于权重低的运行状态的相似度，从而使得判断结果更准确。
各实施例中，也可以同时使用第一权重和第二权重确定系统是否处于基准运行模式。
一些例子中，可以利用第一权重和/或第二权重对度量值集合中的每个值进行加权得到每个值的加权值，根据加权值中的极值确定系统是否处于基准运行模式。例如，当图形相似性度量值为相似度时，可以根据加权后的最小值确定系统是否处于基准运行模式；当图形相似性度量值为差异度时，可以根据加权后的最大值确定系统是否处于基准运行模式。
一些例子中，可以从度量值集合中选择复数个值，根据第一权重和/或第二权重计算复数个值的加权均值，根据加权均值确定系统是否处于基准运行模式。
一些例子中，可以从度量值集合中选择复数个值，根据第一权重和/或第二权重计算复数个值的加权和，根据加权和确定系统是否处于基准运行模式。
从度量值集合中选择复数个值时，可以按照预设的排序方式对度量值集合中的值进行排序，按照顺序选取复数个值，例如可以选取排在前N位的N个值，或者可以按照预设数目比例选取的若干个值，等。可以将上述极值、或加权均值、或加权和与预设的阈值进行比较，根据比较结果确定系统是否处于基准运行模式。
这样，可以在判断系统是否处于基准运行模式时，考虑到当前运行模式和基准运行模式中各运行状态出现的概率，可以使得判断结果更准确。
一些实施例中，运行状态的出现有较强的时间规律，因此可以利用当前运行模式的时间来进一步验证系统是否处于基准运行模式。
一些例子中，可以确定基准运行模式的复数种运行状态中每种运行状态对应的时间信息，通过比较当前运行模式与该复数种运行状态的状态特征和时间信息确定系统是否处于基准运行模式。状态特征是指用于区别各运行状态的信息，例如运行状态图形、运行状态的拟合函数，等。运行状态对应的时间信息是表征系统处于该运行状态的时间的信息。一些例子中，时间信息可以是时间段的信息，如1月到3月、5月10日至7月20日，等。一些例子中，时间信息可以是数组或曲线，表示一运行状态在一段时间内出现的概率。以上仅为举例，其它实施例中，时间信息可以采用其它方式表示。当时间信息为数组时，数组维度由统计周期和变化单位时间段确定，例如统计周期为1年，变化单位时间段为1月时，则时间信息可以是12维的数组，其中每个值表示该运行状态在相应月份的出现概率。统计周期和变化单位时间段可以根据实际情况设置，例如统计周期可以为若干个月，变化单位时间段可以为若干天，等。当时间信息为曲线时，该曲线是在时间维度上连续的一条曲线，曲线的第一维度为统计周期(如若干月)，第二维度为运行状态的出现概率。
一些例子中，比较当前运行模式与该复数种运行状态的时间信息时，可以针对当前运行模式中的每种运行状态(下称第一运行状态)，找到其在基准运行模式中最相似的运行状态(下称第三运行状态)，比较第一运行状态和第三运行状态对应的时间信息，判断二者时间信息的相符程度；如果不相符，则可以确定系统的当前运行模式与基准运行模式的时间不符。
例如，当时间信息为时间段时，如果当前的第一运行状态对应的时间段落入第三运行状态对应的时间段，则可以确定第一运行状态与基准运行模式的时间相符；如果没有落入，则可以确定第一运行状态与基准运行模式的时间不符。
又例如，比较时间信息时，可以根据当前的第一运行状态对应的时间信息，从第三运行状态对应的时间信息中截取相应时间的一段，再进行比较。例如，当前的第一运行状态对应的时间信息是9月15日至11月15日，则从第三运行状态对应的时间信息中截取9月15日至11月15日对应的数组或曲线，再利用数据的比较或者曲线的相似性比较来确定时间信息的相符程度。
确定时间信息的相符程度后，可以结合运行状态的状态特征的相似性比较结果，根 据预设的策略来确定系统是否处于基准运行模式。例如，预设的策略可以是，当前运行模式中的每种运行状态的状态特征均与基准运行模式中的一运行状态相符(例如，图形相似性度量值满足预设条件)、且时间信息与该运行状态相符(例如，时间相符程度大于预设阈值，等)。例如，预设的策略可以是，当前运行模式中的每种运行状态对应的状态特征的符合程度、以及时间信息的符合程度分别按照预设的权重进行加权，根据计算结果来确定系统是否处于基准运行模式。以上仅为举例，其它实施例还可以根据需要采用其它的方式。
另一些例子中，可以确定预设的各时段对应的基准运行模式，通过比较当前运行模式与对应相同时段的基准运行模式确定系统是否处于基准运行模式。例如，可以先按照预先划分的时间段(如若干天、若干周、若干月，等)，确定各时间段对应的基准运行模式。在根据当前运行模式对应的时间段，将同一时间段的基准运行模式与当前运行模式按照各实施例的方式进行比较，从而确定当前运行模式的状态特征和时间信息是否与基准运行模式相符，若相符，则确定系统处于基准运行模式。
还有一些例子中，可以确定系统的一系统参数值的时间分布规律，根据当前运行数据确定系统参数值的时间分布情况，根据时间分布情况是否符合时间分布规律确定系统是否处于基准运行模式。例如，可以确定一系统参数值在单位时间段内的极值或均值，并确定该极值或均值在时间上的分布规律作为系统参数值的时间分布规律，例如曲线或者数组，等。以相似的方式确定当前运行模式中该系统参数值的时间分布情况，将该时间分布情况与上述时间分布规律进行比较，从而确定时间信息的相符程度。之后，可以结合运行状态的状态特征的相似性比较结果，根据预设的策略来确定系统是否处于基准运行模式。具体方法如前所述，这里不再赘述。
通过将当前运行模式的时间信息与基准运行模式中运行状态的时间信息进行比较，可以检测出系统运行状态在时间上的不规律，从而便于对系统的不规律运行进行处理。
各实施例中，可以根据系统的历史运行数据确定系统的基准运行模式，根据系统近一段时间的运行数据(即当前运行数据)确定当前运行模式。为了使得基准运行模式更准确，历史运行数据的时间跨度一般比较大。而为了使得当前运行模式能够体现系统当前的运行情况，当前运行数据的时间跨度则较小一些。也即，采集历史运行数据的第一时长一般大于采集当前运行数据的时长。
一些例子中，历史运行数据和当前运行数据可以根据预设的采集时长分别进行采集。 例如，历史运行数据的采集时长可以为若干个月、若干年，等，当前运行数据的采集周期可以为若干天、若干周、若干月，等。
另一些例子中，可以利用预先分配的数据存储空间来控制历史运行数据和当前运行数据的数据量大小。例如，可以获取系统的运行数据，将运行数据分别存储在第一存储空间和第二存储空间中。其中，第一存储空间大于第二存储空间。第一存储空间可以是硬盘空间、缓存空间，等。第一存储空间用于存储历史运行数据，当第一存储空间已满时，停止向第一存储空间中存储运行数据，并利用第一存储空间中的运行数据确定基准运行模式；当确定系统不处于基准运行模式时，发出告警信息，并清空第一存储空间。类似地，第二存储空间可以是硬盘空间、缓存空间，等。第一存储空间用于存储当前运行数据，当第二存储空间已满、且基准运行模式存在时，可以利用第二存储空间中的运行数据确定当前运行模式。
可见，利用两个不同大小的存储空间分别存储历史运行数据和当前运行数据，并且在其存储空间已满的时候触发各自对应的运行模式的确定过程，不需要使用定时器，实现简单。
各实施例的系统状态监视方法可以由系统状态监视装置12执行。图4A、4B为本申请实施例的系统状态监视装置12的示意图。图4A所示，系统状态监视装置12可以包括：一个基准模式确定模块27、一个当前模式确定模块28，及一个判定模块29。
基准模式确定模块27可以确定系统的基准运行模式，其中，基准运行模式包括系统在一个单位时间段内的复数种运行状态。
当前模式确定模块可以根据系统的当前运行数据确定系统的当前运行模式。
判定模块29可以通过比较当前运行模式和基准运行模式确定系统是否处于基准运行模式。
可见，各实施例的系统状态监视装置12将被监视系统惯常的复数种运行状态确定为基准运行模式，将系统的当前运行模式与基准运行模式进行比较来判断系统是否处于基准运行模式，可以检测到系统运行规律的变化，从而便于及时对被监视系统或者外围配合机制进行调整，提高被监视系统的性能。
一些实施例中，如图4B所示，判定模块29可以包括：一个相似度确定单元291和一个判定单元292。相似度确定单元291可以将与复数种运行状态对应的复数个第一状 态分布图形作为第一图形组，将当前运行模式对应的至少一个第二状态分布图形作为第二图形组，确定第一图形组与第二图形组的相似性度量值。判定单元292可以根据相似性度量值确定系统是否处于基准运行模式。这样，将当前运行模式与基准运行模式的比较转化为两组图形的相似度比较，系统状态监视装置12可以采用各种图形处理方法来确定系统是否处于基准运行模式，实现起来较为简单。
一些实施例中，确定第一图形组与第二图形组的相似性度量值时，相似度确定单元291可以针对第二图形组中的每个第二状态分布图形，确定第一图形组中与第二状态分布图形最相似的第三状态分布图形，获得第二状态分布图形与第三状态分布图形的图形相似性度量值，根据图形相似性度量值得到第二图形组对应的度量值集合；利用度量值集合确定第一图形组与第二图形组的相似性度量值。可见，通过确定第二图形组中各图形在第一图形组中最相似的图形，根据其图形相似性度量值确定当前运行模式中各运行状态图形与基准运行模式的接近程度，减少了相似度确定单元291的计算量。
一些实施例中，处理该度量值集合时，判定单元292可以执行以下之一：
选择度量值集合中的极值作为第一图形组与第二图形组的相似性度量值；
从度量值集合中选择复数个值，将复数个值的均值作为第一图形组与第二图形组的相似性度量值；
从度量值集合中选择复数个值，将复数个值的和作为第一图形组与第二图形组的相似性度量值。
通过从度量值集合中选择部分值来确定第一图形组与第二图形组的相似性度量值，可以进一步减少系统状态监视装置12的计算量，提高计算效率。
一些实施例中，在比较基准运行模式和当前运行模式时将可以将各运行状态的权重考虑进去。例如，基准模式确定模块27还可以根据历史运行数据确定复数个第一状态分布图形中每个第一状态分布图形的第一权重。又例如，当前模式确定模块28还可以根据当前运行数据确定至少一个第二状态分布图形中每个第二状态分布图形的第二权重。判定模块29可以针对第二图形组中的每个第二状态分布图形，确定第一图形组中与第二状态分布图形最相似的第三状态分布图形，计算第二状态分布图形与第三状态分布图形的图形相似性度量值，根据图形相似性度量值得到第二图形组对应的度量值集合；利用度量值集 合和第一权重和/或第二权重确定系统是否处于基准运行模式。通过将各运行状态的权重纳入考虑，可以使得判定模块29的判断结果更准确。
一些实施例中，使用第一权重和/或第二权重确定系统是否处于基准运行模式时，判定模块29可以执行以下之一：
利用第一权重和/或第二权重对度量值集合中的每个值进行加权得到每个值的加权值，根据加权值中的极值确定系统是否处于基准运行模式；
从度量值集合中选择复数个值，根据第一权重和/或第二权重计算复数个值的加权均值，根据加权均值确定系统是否处于基准运行模式；
从度量值集合中选择复数个值，根据第一权重和/或第二权重计算复数个值的加权和，根据加权和确定系统是否处于基准运行模式。
这样，可以在判断系统是否处于基准运行模式时，考虑到当前运行模式和基准运行模式中各运行状态出现的概率，可以使得判断结果更准确。
一些实施例中，可以利用当前运行模式的时间来进一步验证系统是否处于基准运行模式。判定模块29可以执行以下之一：
确定复数种运行状态中每种运行状态对应的时间信息，通过比较复数种运行状态和当前运行模式的状态特征和时间信息确定系统是否处于基准运行模式；
确定预设的各时段对应的基准运行模式，通过比较当前运行模式与对应相同时段的基准运行模式确定系统是否处于基准运行模式；
确定系统的一系统参数值的时间分布规律，根据当前运行数据确定系统参数值的时间分布情况，根据时间分布情况是否符合时间分布规律确定系统是否处于基准运行模式。
通过将当前运行模式的时间信息与基准运行模式中运行状态的时间信息进行比较，可以检测出系统运行状态在时间上的不规律，从而便于对系统的不规律运行进行处理。
一些实施例中，如图4B所示，系统状态监视装置12还可以包括：一个存储模块24和一个数据获取模块26。
存储模块24可以包括第一存储空间241和第二存储空间242。其中，第一存储空间241大于第二存储空间242。
数据获取模块26可以将系统的运行数据分别存储在第一存储空间241和第二存储空间242中；当第一存储空间241已满时，停止向第一存储空间241中存储运行数据，并触发基准模式确定模块利用第一存储空间241中的运行数据确定基准运行模式；当第二存储空间242已满、且基准运行模式存在时，触发当前模式确定模块利用第二存储空间242中的运行数据确定当前运行模式。
其中，判定模块29还可以在确定系统不处于基准运行模式时，发出告警信息，并清空第一存储空间241。
可见，利用两个不同大小的存储空间分别存储历史运行数据和当前运行数据，并且在其存储空间已满的时候触发各自对应的运行模式的确定过程，不需要使用定时器，实现简单。
一些实施例中，确定基准运行模式时，基准模式确定模块27可以执行以下之一：
对系统的历史运行数据中单位时间段对应的多种运行状态实例进行聚类，得到复数个簇，针对复数个簇中的每个簇计算一运行状态，作为复数种运行状态中的一个运行状态；或
利用系统的历史运行数据中单位时间对应的多组运行状态实例构建矩阵，利用奇异值分解或主成分分析对矩阵进行降维，获得复数种运行状态。
通过上述处理，可以将大量的运行状态实例简化为少数的几种运行状态，方便后续与当前运行模式进行比较，降低了实现复杂度。
一些实施例中，系统状态监视装置12还可以包括处理器21、存储器20、通信模块22。通信模块22用于通过任意通信网络与其它设备进行通信。
存储器20可以包括操作系统23、监视模块25和存储模块24。监视模块25包括上述各模块对应的计算机可读指令。
处理器21可以执行存储器20中的计算机可读指令，从而实现各实施例的方法。
可以看出，各实施例的技术方案也可以体现为计算机可读存储介质中的计算机可读指令。
各实施例的系统状态监视方案可以应用于各种领域。例如，图5为本申请实施例的 一个应用场景的示意图。该应用场景中，各实施例的系统状态监控装置12应用在供电企业(被监视系统14)中，对各用电企业的用电情况进行监视，从而确保电网的有效和稳定运行。传感器13可以安装在为各用电企业供电的配电设备(如变压器等)上，不断采集配电设备的运行数据(如电压、电流、功率，等)，并将数据发送到系统状态监视装置12。
系统状态监视装置12包括处理器51和缓存52(例如，RAM)。处理器51对时间序列数据进行预处理(例如，数据清理、数据归一化，等)，将预处理后的数据将存储在数据库53中，形成时间序列数据，即按照时间顺序排列的数据。处理器51使用预定义的数学模型和数据库53中的数据进行建模，模型存储在缓存52中。当确定系统没有处于基准运行模式时，系统状态监视装置12向告警设备56发送信号，告警设备56可以通过电子邮件、SMS消息或其他可能的方式向有关人员发出告警信息。系统状态监视装置12可以创建日志文件并将其存储在日志数据库55中。
图6为本申请实施例的系统状态监视装置12的处理流程图。
步骤S61，系统状态监视装置12对时间序列数据进行预处理，将预处理后的数据分别存储到数据库53中的第一存储空间和第二存储空间。
步骤S62，系统状态监视装置12利用第一存储空间中的数据进行建模，得到基准运行模型(即基准运行模式的数学模型)。
步骤S63，利用第二存储空间的数据进行建模，得到当前运行模型(即当前运行模式的数学模型)。
步骤S64，系统状态监视装置12计算当前运行模型和基准运行模型的相似度。
步骤S65，系统状态监视装置12判断相似度是否小于预设的阈值；若相似度小于阈值，则执行步骤S66；否则，返回步骤S63。
步骤S66，系统状态监视装置12向告警设备56发送信号，向日志数据库55写入日志，清空第一存储空间，并返回步骤S62。
图7为本申请实施例中系统状态监视装置12存储数据的方法示意图。
第一存储空间用于存储用于建立基准运行模型的时间序列数据，第二存储空间82用于存储用于建立当前运行模型的时间序列数据。
在步骤S81产生新的时间序列数据时，系统状态监视装置12均尝试写入第一存储空 间和第二存储空间。
对于第一存储空间，系统状态监视装置12在步骤S82判断第一存储空间是否已满，如果未满，则在步骤S83写入第一存储空间；如果已满，则在步骤S84判断当前是否已存在基准运行模型，若存在，则不执行任何处理；若不存在，则在步骤S85利用第一存储空间中的时间序列数据建立基准运行模型。
对于第二存储空间，系统状态监视装置12在步骤S86判断第二存储空间是否已满，如果未满，则在步骤S87将时间序列数据写入第二存储空间；如果已满，则在步骤S88判断当前是否存在基准运行模型。如果当前不存在基准运行模型，则在步骤S810用新写入的数据更新存储的旧数据。如果当前存在基准运行模型，则在步骤S811利用第二存储空间中的时间序列数据建立当前运行模型。
系统状态监视装置12在步骤S812判断当前运行模型和基准运行模型的相似度是否小于阈值，如果相似度是否小于阈值，则在步骤S813清空第一存储空间；否则，不执行任何处理。
以上仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (12)

  1. 一种系统状态监视方法，其特征在于，包括：
    确定系统的基准运行模式，其中，所述基准运行模式包括所述系统在一个单位时间段内的复数种运行状态；
    根据所述系统的当前运行数据确定所述系统的当前运行模式；
    通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式。
  2. 根据权利要求1所述的方法，其特征在于，所述基准运行模式包括与所述复数种运行状态对应的复数个第一状态分布图形，所述当前运行模式包括至少一个第二状态分布图形；
    其中，通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式包括：
    将所述复数个第一状态分布图形作为第一图形组，将所述至少一个第二状态分布图形作为第二图形组，确定所述第一图形组与所述第二图形组的相似性度量值，根据所述相似性度量值确定所述系统是否处于所述基准运行模式。
  3. 根据权利要求2所述的方法，其特征在于，
    进一步包括以下中的至少一个：
    确定所述复数个第一状态分布图形中每个第一状态分布图形的第一权重；
    根据所述当前运行数据确定所述至少一个第二状态分布图形中每个第二状态分布图形的第二权重；
    其中，通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式包括：
    针对所述第二图形组中的每个第二状态分布图形，确定所述第一图形组中与所述第二状态分布图形最相似的第三状态分布图形，计算所述第二状态分布图形与所述第三状态分布图形的图形相似性度量值，根据所述图形相似性度量值得到所述第二图形组对应的度量值集合；
    利用所述度量值集合和所述第一权重和/或所述第二权重确定所述系统是否处于所述基准运行模式。
  4. 根据权利要求1-3中任一权利要求所述的方法，其特征在于，进一步包括以下之一：
    确定所述复数种运行状态中每种运行状态对应的时间信息，通过比较所述复数种运行状态和所述当前运行模式的状态特征和时间信息确定所述系统是否处于所述基准运行模式；
    确定预设的各时段对应的基准运行模式，通过比较所述当前运行模式与对应相同时段的基准运行模式确定所述系统是否处于所述基准运行模式；
    确定所述系统的一系统参数值的时间分布规律，根据所述当前运行数据确定所述系统参数值的时间分布情况，根据所述时间分布情况是否符合所述时间分布规律确定所述系统是否处于所述基准运行模式。
  5. 根据权利要求1-4中任一权利要求所述的方法，其特征在于，进一步包括：
    获取系统的运行数据，将所述运行数据分别存储在第一存储空间和第二存储空间中，所述第一存储空间大于所述第二存储空间；
    当确定所述系统不处于所述基准运行模式时，发出告警信息，并清空所述第一存储空间；
    其中，确定所述系统的基准运行模式，根据所述系统的当前运行数据确定所述系统的当前运行模式包括：
    当所述第一存储空间已满时，停止向所述第一存储空间中存储运行数据，并利用所述第一存储空间中的运行数据确定所述基准运行模式；
    当所述第二存储空间已满、且所述基准运行模式存在时，利用所述第二存储空间中的运行数据确定所述当前运行模式。
  6. 根据权利要求1-5中任一权利要求所述的方法，其特征在于，确定系统的基准运行模式包括以下之一：
    对所述系统的历史运行数据中所述单位时间段对应的多种运行状态实例进行聚类，得到复数个簇，针对所述复数个簇中的每个簇计算一运行状态，作为所述复数种运行状态中的一个运行状态；
    利用所述系统的历史运行数据中所述单位时间对应的多组运行状态实例构建矩阵，利用奇异值分解或主成分分析对所述矩阵进行降维，获得所述复数种运行状态。
  7. 一种系统状态监视装置，其特征在于，包括：
    一个基准模式确定模块(27)，用于确定系统的基准运行模式，其中，所述基准运行模式包括所述系统在一个单位时间段内的复数种运行状态；
    一个当前模式确定模块(28)，用于根据所述系统的当前运行数据确定所述系统的当前运行模式；
    一个判定模块(29)，用于通过比较所述当前运行模式和所述基准运行模式确定所述系统是否处于所述基准运行模式。
  8. 根据权利要求7所述的装置，其特征在于，所述判定模块(29)包括：
    一个相似度确定单元(291)，用于将与所述复数种运行状态对应的复数个第一状态分布图形作为第一图形组，将所述当前运行模式对应的至少一个第二状态分布图形作为第二图形组，确定所述第一图形组与所述第二图形组的相似性度量值；
    一个判定单元(292)，用于根据所述相似性度量值确定所述系统是否处于所述基准运行模式。
  9. 根据权利要求8所述的装置，其特征在于，
    所述基准模式确定模块(27)进一步用于，根据所述历史运行数据确定所述复数个第一状态分布图形中每个第一状态分布图形的第一权重；或
    所述当前模式确定模块(28)进一步用于，根据所述当前运行数据确定所述至少一个第二状态分布图形中每个第二状态分布图形的第二权重；
    其中，所述判定模块(29)进一步用于，
    针对所述第二图形组中的每个第二状态分布图形，确定所述第一图形组中与所述第二状态分布图形最相似的第三状态分布图形，计算所述第二状态分布图形与所述第三状态分布图形的图形相似性度量值，根据所述图形相似性度量值得到所述第二图形组对应的度量值集合；
    利用所述度量值集合和所述第一权重和/或所述第二权重确定所述系统是否处于所述基准运行模式
  10. 根据权利要求7-9中任一权利要求所述的装置，其特征在于，所述判定模块(29)用于执行以下之一：
    确定所述复数种运行状态中每种运行状态对应的时间信息，通过比较所述复数种运行状态 和所述当前运行模式的状态特征和时间信息确定所述系统是否处于所述基准运行模式；
    确定预设的各时段对应的基准运行模式，通过比较所述当前运行模式与对应相同时段的基准运行模式确定所述系统是否处于所述基准运行模式；
    确定所述系统的一系统参数值的时间分布规律，根据所述当前运行数据确定所述系统参数值的时间分布情况，根据所述时间分布情况是否符合所述时间分布规律确定所述系统是否处于所述基准运行模式。
  11. 一种系统状态监视装置，其特征在于，包括：处理器(21)和存储器(20)，所述存储器(20)存储有计算机可读指令，可以使所述处理器(21)执行如权利要求1-6中任一权利要求所述的方法。
  12. 一种计算机可读存储介质，其特征在于，存储有计算机可读指令，可以使一处理器执行如权利要求1-6中任一权利要求所述的方法。
PCT/CN2019/105232 2018-09-20 2019-09-10 系统状态监视方法、装置和存储介质 WO2020057402A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17/277,770 US11847619B2 (en) 2018-09-20 2019-09-10 System-state monitoring method and device and storage medium
EP19861456.2A EP3839743B1 (en) 2018-09-20 2019-09-10 System-state monitoring method and device and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811103227.3 2018-09-20
CN201811103227.3A CN110928741B (zh) 2018-09-20 2018-09-20 系统状态监视方法、装置和存储介质

Publications (1)

Publication Number Publication Date
WO2020057402A1 true WO2020057402A1 (zh) 2020-03-26

Family

ID=69855574

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/105232 WO2020057402A1 (zh) 2018-09-20 2019-09-10 系统状态监视方法、装置和存储介质

Country Status (4)

Country Link
US (1) US11847619B2 (zh)
EP (1) EP3839743B1 (zh)
CN (1) CN110928741B (zh)
WO (1) WO2020057402A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11847619B2 (en) 2018-09-20 2023-12-19 Siemens Ltd., China System-state monitoring method and device and storage medium

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112306802A (zh) * 2020-10-29 2021-02-02 平安科技（深圳）有限公司 系统的数据获取方法、装置、介质和电子设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101251441A (zh) * 2008-03-20 2008-08-27 桂林电子科技大学 汽车发动机状态在线监测方法及系统
CN103245912A (zh) * 2013-03-23 2013-08-14 中国水利电力物资有限公司 一种风电机组运行状态分析诊断方法和系统
US20160162383A1 (en) * 2013-07-23 2016-06-09 Siemens Aktiengesellschaft Monitoring Redundant Components
CN107203746A (zh) * 2017-05-12 2017-09-26 同济大学 一种道岔故障识别方法
CN107814288A (zh) * 2017-09-30 2018-03-20 南京市特种设备安全监督检验研究院 一种电梯制动器智能监测预警的方法

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4607346A (en) * 1983-03-28 1986-08-19 International Business Machines Corporation Apparatus and method for placing data on a partitioned direct access storage device
US8566549B1 (en) * 2008-12-31 2013-10-22 Emc Corporation Synchronizing performance requirements across multiple storage platforms
JP5468837B2 (ja) * 2009-07-30 2014-04-09 株式会社日立製作所 異常検出方法、装置、及びプログラム
WO2014101086A1 (zh) * 2012-12-28 2014-07-03 华为技术有限公司 存储空间的处理方法、装置及非易失性计算机可读存储介质
JP2015075821A (ja) 2013-10-07 2015-04-20 横河電機株式会社 状態診断方法および状態診断装置
EP3097494A4 (en) * 2014-01-23 2017-10-25 Microsoft Technology Licensing, LLC Computer performance prediction using search technologies
JP6353747B2 (ja) * 2014-08-29 2018-07-04 日置電機株式会社 波形記録装置
US10083073B2 (en) * 2015-09-14 2018-09-25 Dynatrace Llc Method and system for real-time causality and root cause determination of transaction and infrastructure related events provided by multiple, heterogeneous agents
CN105372591B (zh) 2015-09-28 2018-02-16 国家电网公司 一种基于暂态过程的水电机组健康状态量化评价方法
US10909016B2 (en) * 2016-02-03 2021-02-02 Hitachi, Ltd. Management computer and method of managing computer to be managed
CN106447206A (zh) 2016-10-09 2017-02-22 国网浙江省电力公司信息通信分公司 基于用电信息采集数据的用电分析方法
CN106779200A (zh) 2016-12-07 2017-05-31 东北大学 基于在历史数据中进行相似搜索的风电机组状态预测方法
WO2018133205A1 (zh) * 2017-01-20 2018-07-26 华为技术有限公司 实现增值服务的方法、装置与云服务器
CN106656368A (zh) * 2017-02-13 2017-05-10 中国联合网络通信集团有限公司 通信系统监控方法及装置
US10425437B1 (en) * 2017-11-27 2019-09-24 Lacework Inc. Extended user session tracking
CN108121640B (zh) * 2017-12-29 2021-08-24 北京奇虎科技有限公司 一种用于自动获取应用的运行数据的方法及系统
CN110928741B (zh) 2018-09-20 2021-09-24 西门子（中国）有限公司 系统状态监视方法、装置和存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101251441A (zh) * 2008-03-20 2008-08-27 桂林电子科技大学 汽车发动机状态在线监测方法及系统
CN103245912A (zh) * 2013-03-23 2013-08-14 中国水利电力物资有限公司 一种风电机组运行状态分析诊断方法和系统
US20160162383A1 (en) * 2013-07-23 2016-06-09 Siemens Aktiengesellschaft Monitoring Redundant Components
CN107203746A (zh) * 2017-05-12 2017-09-26 同济大学 一种道岔故障识别方法
CN107814288A (zh) * 2017-09-30 2018-03-20 南京市特种设备安全监督检验研究院 一种电梯制动器智能监测预警的方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3839743A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11847619B2 (en) 2018-09-20 2023-12-19 Siemens Ltd., China System-state monitoring method and device and storage medium

Also Published As

Publication number Publication date
EP3839743C0 (en) 2023-07-12
EP3839743A4 (en) 2022-05-04
CN110928741B (zh) 2021-09-24
EP3839743B1 (en) 2023-07-12
US20210286695A1 (en) 2021-09-16
US11847619B2 (en) 2023-12-19
EP3839743A1 (en) 2021-06-23
CN110928741A (zh) 2020-03-27

Similar Documents

Publication Publication Date Title
CN108415789B (zh) 面向大规模混合异构存储系统的节点故障预测系统及方法
WO2017008451A1 (zh) 一种面向云计算在线业务的异常负载检测方法
US10437696B2 (en) Proactive information technology infrastructure management
CN113570200B (zh) 一种基于多维信息的电网运行状态监测方法及系统
WO2020057402A1 (zh) 系统状态监视方法、装置和存储介质
CN105871634A (zh) 检测集群异常的方法及应用、管理集群的系统
CN116976707B (zh) 基于用电信息采集的用户用电数据异常分析方法及系统
CN116683588B (zh) 锂离子电池充放电控制方法及系统
CN109685140A (zh) 一种基于主成分分析的dbscan算法岸桥状态分类方法
US20180115465A1 (en) Inferring a network topology
CN114417971A (zh) 一种基于k近邻密度峰值聚类的电力数据异常值检测算法
CN108989083B (zh) 云环境下基于混合策略的故障检测性能优化方法
CN115664038A (zh) 一种用于电气安全管理的智能配电运维监控系统
CN112905671A (zh) 时间序列异常处理方法、装置、电子设备及存储介质
CN116030955B (zh) 基于物联网的医疗设备状态监测方法及相关装置
CN110647086B (zh) 一种基于运行大数据分析的智能运维监控系统
CN111460027A (zh) 一种适用于能源互联网的智能动态监测方法及系统
CN106649034A (zh) 一种可视化智能运维方法及平台
CN115310516A (zh) 自动化设备状态稳定判断方法、系统及可读存储介质
CN113285847A (zh) 一种智能换流站监测系统的通信网络异常检测方法及系统
CN112988529A (zh) 一种基于机器学习进行数据库系统性能预测的方法及系统
CN117094478B (zh) 能量调度管理方法、装置、设备及存储介质
Hairuman et al. Evaluation of machine learning techniques for anomaly detection on hourly basis kpi
Wang et al. Anomaly monitoring in high-density data centers based on gaussian distribution anomaly detection algorithm
CN111400284B (zh) 一种基于性能数据建立动态异常探测模型的方法

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: 19861456

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019861456

Country of ref document: EP

Effective date: 20210319