WO2018105872A1 - Dispositif de détection d'anomalies de capteur météorologique à l'aide d'une détermination composite, procédé associé et support d'enregistrement sur lequel est enregistré un programme informatique - Google Patents

Dispositif de détection d'anomalies de capteur météorologique à l'aide d'une détermination composite, procédé associé et support d'enregistrement sur lequel est enregistré un programme informatique Download PDF

Info

Publication number
WO2018105872A1
WO2018105872A1 PCT/KR2017/011290 KR2017011290W WO2018105872A1 WO 2018105872 A1 WO2018105872 A1 WO 2018105872A1 KR 2017011290 W KR2017011290 W KR 2017011290W WO 2018105872 A1 WO2018105872 A1 WO 2018105872A1
Authority
WO
WIPO (PCT)
Prior art keywords
error
determination result
weather information
error determination
controller
Prior art date
Application number
PCT/KR2017/011290
Other languages
English (en)
Korean (ko)
Inventor
명광민
장석웅
Original Assignee
에스케이테크엑스 주식회사
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 에스케이테크엑스 주식회사 filed Critical 에스케이테크엑스 주식회사
Publication of WO2018105872A1 publication Critical patent/WO2018105872A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W1/00—Meteorology
    • G01W1/18—Testing or calibrating meteorological apparatus
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D9/00—Recording measured values
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W1/00—Meteorology
    • G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W1/00—Meteorology
    • G01W1/10—Devices for predicting weather conditions
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W2201/00—Weather detection, monitoring or forecasting for establishing the amount of global warming

Definitions

  • the present invention relates to an apparatus for detecting an abnormality of a weather sensor through a complex determination, a method thereof, and a recording medium on which a computer program is recorded.
  • the sensing data measured by the weather sensor is classified into a plurality of error analysis groups according to error determination reliability.
  • Sensing data received from a plurality of weather sensors has a variety of error patterns according to the sensor and the observation environment, and there is a limit to error detection with existing simple quality control techniques.
  • the error pattern may not show a big difference from normal data, and there is a limit in quality control by humans due to the rapid increase in sensed data due to the proliferation of Internet of Things (IoT) technology.
  • IoT Internet of Things
  • an error may occur in a forecast or a live situation, and a problem that may be difficult to find the cause may occur.
  • An object of the present invention is an abnormality of the weather sensor through a complex decision to determine whether the error is complex based on a plurality of error analysis methods classified into a plurality of error analysis groups according to the error determination reliability for the measured data measured by the weather sensor.
  • a sensing device, a method thereof, and a computer program are provided to provide a recording medium.
  • Another object of the present invention is to add a daily change pattern analysis, probability distribution analysis, frequency analysis, correlation analysis, etc. to the general QC method, which is an error analysis method, and to remove the error data or to determine the detection data by complex determination of individual analysis results.
  • the present invention provides an apparatus for detecting an abnormality of a weather sensor, a method, and a computer program in which a classification of data grades is used to provide detection data having a grade suitable for a purpose of use.
  • a method of detecting an abnormality of a weather sensor through a complex determination includes collecting, by a collector, weather information for a management area; Confirming, by a controller, a grade of the weather information by sequentially applying the collected weather information to a plurality of error analysis groups stored in a storage; Storing, by the controller, the grade of the confirmed weather information and the weather information by matching the weather information for each grade stored in advance in the storage unit; And performing, by the controller, a control function by utilizing at least one weather information corresponding to a class according to a purpose of use among a plurality of class weather information stored in the storage unit.
  • the plurality of error analysis groups may include, by the control unit, a quality control (QC) method, a daily change pattern method, a peripheral area deviation method, and a prediction-based on a predetermined error determination reliability.
  • QC quality control
  • Observational correlation coefficient method, live / prediction-observation deviation method, observation distribution method, abnormal fluctuation / wave method using FFT and anomaly detection method based on machine learning can be classified into a plurality of error analysis groups.
  • the plurality of error analysis groups classify the quality management method into the first error analysis group, classify the daily change pattern method and the surrounding area deviation method into the second error analysis group, and predict-observ Classify the correlation coefficient method and the live / predictive-observation deviation method into the third error analysis group, and classify the observation distribution method, the abnormal variation / wave method using FFT, and the machine learning-based abnormal detection method into the fourth error analysis group. It may be in one state.
  • the step of determining the grade of the weather information by sequentially applying the collected weather information to the plurality of error analysis groups may be included in the first error analysis group.
  • Calculating a first error determination result by applying the weather information to a quality control scheme When the calculated first error determination result is equal to or greater than a reference range, checking, by the controller, the weather information as an error reporting level; On the basis of the calculated first error determination result, the controller applies the weather information to the daily change pattern method and the surrounding area deviation method included in the second error analysis group, respectively, to determine the second error determination result and the third.
  • Calculating an error determination result By the controller, based on the calculated second error determination result and the third error determination result, the weather information is included in the prediction-observation correlation coefficient method and the live / prediction-observation deviation method included in the third error analysis group. Calculating a fourth error determination result to a sixth error determination result by applying each; And an observation value distribution method included in the fourth error analysis group, an abnormal variation / wave method using FFT, and machine learning based on the calculated fourth error determination result to sixth error determination result by the controller. And applying the weather information to the abnormality detection method, respectively, to calculate the seventh to nineth error determination results.
  • the calculating of the second error determination result and the third error determination result may be performed by the controller when the first error determination result calculated by the first error analysis group is within a reference range.
  • Calculating the second error determination result by applying weather information to the daily change pattern method and the surrounding area deviation method included in the second error analysis group;
  • the controller applies weather information to the daily change pattern method and the surrounding area deviation method included in the second error analysis group, respectively.
  • Calculating the third error determination result and when the calculated third error determination result is greater than or equal to a reference range, checking, by the controller, the weather information as a failure report level.
  • the calculating of the fourth error determination result to the sixth error determination result may be performed by the controller when the second error determination result calculated by the second error analysis group is within a reference range.
  • Calculating the fourth error determination result by applying weather information to the prediction-observation correlation coefficient method and the live / prediction-observation deviation method included in the third error analysis group, respectively;
  • the calculated fourth error determination result is greater than or equal to a reference range, confirming, by the controller, the weather information as a failure report level;
  • the control unit predicts and observes correlation included in the third error analysis group.
  • Calculating the fifth error determination result by applying weather information to a counting method and a running / predicting-observing deviation method, respectively; When the calculated fifth error determination result is greater than or equal to a reference range, confirming, by the controller, the weather information as a failure report level; And when the third error determination result calculated by the second error analysis group is no error, by the controller, the prediction-observation correlation coefficient method and the running / prediction-observation deviation method included in the third error analysis group. And applying the weather information to each to calculate the sixth error determination result.
  • the calculating of the seventh error determination result to the ninth error determination result may be performed by the controller when the fourth error determination result calculated by the third error analysis group is within a reference range. Calculating the seventh error determination result by applying the weather information to an observation value distribution method included in the fourth error analysis group, an abnormal fluctuation / wave method using an FFT and an abnormal detection method based on machine learning, respectively.
  • the controller may further include checking, by the controller, the weather information to a normal level.
  • a computer program for performing the method according to the above-described embodiments may be stored in a recording medium on which a computer program according to an embodiment of the present invention is recorded.
  • An apparatus for detecting an abnormality of a weather sensor through a complex determination includes a storage unit for storing a plurality of error analysis groups including a plurality of error analysis methods classified based on error determination reliability; Collecting unit for collecting weather information for the management area; And sequentially applying the collected weather information to a plurality of error analysis groups stored in the storage to check the grade of the weather information, and to predefine the grade and the weather information of the confirmed weather information to the storage. It may include a controller for matching and storing a plurality of stored weather information for each grade.
  • the controller may perform a control function by utilizing at least one weather information corresponding to a grade according to a purpose of use among a plurality of grades of weather information stored in the storage.
  • the controller may include a quality control method included in a first error analysis group, a daily change pattern method and a surrounding area deviation method included in a second error analysis group, and a third one.
  • the weather information may be sequentially applied to the method, and the result of the error analysis method included in the error analysis group of the next step may be changed according to the determination result of the error analysis method of the previous group.
  • the present invention provides a high reliability of the collected weather information by complex determination of errors based on a plurality of error analysis methods classified into a plurality of error analysis groups according to the error determination reliability of the measured data measured by the weather sensor. It is possible to evaluate the quality and to check the error data precisely and accurately to provide reliable sensing data.
  • the present invention adds the daily variation pattern analysis, probability distribution analysis, frequency analysis, correlation analysis, etc. to the general QC method, which is an error analysis method, and removes the error data by complex determination of individual analysis results or data for the sensing data.
  • the general QC method which is an error analysis method
  • the analysis is performed using the reliable sense data, and the suspicious data is referred to simply so that the grade data can be utilized according to the purpose. It has the effect of using quality properly.
  • the present invention has the effect of providing accurate and reliable detection data through the complex determination in response to the abnormal climate by changing and optimizing the complex determination criteria for each error analysis according to the user's administration settings.
  • FIG. 1 is a block diagram illustrating a configuration of an apparatus for detecting an abnormality of a weather sensor through a complex determination according to an exemplary embodiment of the present invention.
  • FIG. 2 is a diagram illustrating a correlation analysis data source for prediction and observation of a wind speed model according to an embodiment of the present invention.
  • FIG. 3 is a diagram illustrating a change in prediction-observation correlation according to an embodiment of the present invention.
  • FIG. 4 is a diagram illustrating a change in prediction-observation deviation according to an embodiment of the present invention.
  • 5 to 7 is a view showing a state according to the distribution of the wind speed frequency for each point according to an embodiment of the present invention.
  • FIG. 8 is a flowchart illustrating a method of detecting an abnormality of a weather sensor according to an exemplary embodiment of the present invention.
  • FIG. 9 is a diagram illustrating an example of applying a multi-step error analysis group according to an exemplary embodiment of the present invention.
  • first and second used in the present invention may be used to describe components, but the components should not be limited by the terms. The terms are used only to distinguish one component from another.
  • first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component.
  • FIG. 1 is a block diagram illustrating a configuration of an apparatus 10 for detecting abnormality of a weather sensor through a complex determination according to an exemplary embodiment of the present invention.
  • the apparatus 10 for detecting abnormality of the weather sensor through the complex determination includes the collecting unit 100, the communication unit 200, the storage unit 300, the display unit 400, and the voice output unit 500. And a control unit 600. Not all components of the abnormality detection device 10 of the weather sensor through the composite determination illustrated in FIG. 1 are not essential components, and more components than the components shown in FIG. The abnormality sensing device 10 may be implemented, or the abnormality sensing device 10 of the weather sensor through the complex determination may be implemented by fewer components.
  • the collection unit 100 collects (or measures) weather information (or sensing data) for a plurality of management areas, respectively. Subsequently, the controller 600 sequentially applies the weather information to the plurality of error analysis groups that have been previously classified with respect to the collected weather information, and checks (or determines) a grade of the weather information. Thereafter, the controller 600 matches the weather information for each grade based on the identified grade of the weather information and stores the matched weather information in the storage 300. Thereafter, the controller 600 performs a control function by utilizing at least one weather information corresponding to a grade according to a purpose of use among the plurality of grades of weather information stored in the storage 300.
  • the abnormality detection device 10 includes a smart phone, a portable terminal, a mobile terminal, a personal digital assistant (PDA), a portable multimedia player (PMP) terminal, and telematics.
  • PDA personal digital assistant
  • PMP portable multimedia player
  • Terminals Navigation Terminals, Personal Computers, Notebook Computers, Slate PCs, Tablet PCs, Ultrabooks, Wearable Devices (e.g., Watches) Smartwatch, Glass Glass, HMD (Head Mounted Display, etc.), Wibro Terminal, IPTV (Internet Protocol Television) Terminal, Smart TV, Digital Broadcasting Terminal, Television,
  • the present invention can be applied to various terminals such as 3D televisions, home theater systems, audio video navigation (AVN) terminals, audio / video (A / V) systems, and flexible terminals.
  • a / V audio video navigation
  • the collection unit 100 collects (or measures) weather information (or sensing data) for a plurality of management areas, respectively.
  • the weather information includes temperature, air pressure, wind speed, relative humidity, sunshine, precipitation, rainfall probability, ultraviolet index, unique identification information of a region (or management region), and location information of a corresponding management region.
  • the communication unit 200 communicates with any component inside or any at least one terminal outside through a wired / wireless communication network.
  • any external terminal may include a server (not shown).
  • the wireless Internet technologies include Wireless LAN (WLAN), Wireless Personal Area Network (WPAN), Digital Living Network Alliance (DLNA), Wireless Broadband (Wibro), and WiMAX (World Interoperability for Microwave Access: Wimax, HSDPA (High Speed Downlink Packet Access), HSUPA (High Speed Uplink Packet Access), IEEE 802.16, Long Term Evolution (LTE), Long Term Evolution-Advanced (LTE-A), Broadband
  • WMBS wireless mobile broadband service
  • the communication unit 200 transmits and receives data according to at least one wireless Internet technology in a range including Internet technologies not listed above.
  • near field communication technologies include Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, and Near Field Communication (NFC).
  • RFID Radio Frequency Identification
  • IrDA Infrared Data Association
  • UWB Ultra Wideband
  • ZigBee ZigBee
  • NFC Near Field Communication
  • USB Ultrasound Communication
  • VLC Visible Light Communication
  • Wi-Fi Direct Wi-Fi Direct
  • the wired communication technology may include power line communication (PLC), USB communication, Ethernet, serial communication, serial communication, optical / coaxial cable, and the like.
  • the communication unit 200 may mutually transmit information with an arbitrary terminal through a universal serial bus (USB).
  • USB universal serial bus
  • the communication unit 200 may include technical standards or communication methods (for example, Global System for Mobile communication (GSM), Code Division Multi Access (CDMA), Code Division Multi Access 2000 (CDMA2000), and EV-) for mobile communication.
  • GSM Global System for Mobile communication
  • CDMA Code Division Multi Access
  • CDMA2000 Code Division Multi Access 2000
  • EV- Enhanced Voice-Data Optimized or Enhanced Voice-Data Only (DO), Wideband CDMA (WCDMA), High Speed Downlink Packet Access (HSDPA), High Speed Uplink Packet Access (HSUPA), Long Term Evolution (LTE), LTE-A ( Long Term Evolution-Advanced, etc.) transmits and receives a radio signal to a base station, a server and the like on a mobile communication network.
  • GSM Global System for Mobile communication
  • CDMA Code Division Multi Access
  • CDMA2000 Code Division Multi Access 2000
  • EV- Enhanced Voice-Data Optimized or Enhanced Voice-Data Only (DO)
  • WCDMA Wideband CDMA
  • HSDPA High
  • the communication unit 200 receives weather information for each management region transmitted from a server (not shown) under the control of the control unit 600.
  • the weather information for each management region includes temperature, air pressure, wind speed, relative humidity, sunshine, precipitation, rainfall probability, UV index, unique identification information of the region (or management region), and location information of the management region. .
  • the storage unit 300 stores various user interfaces (UIs), graphical user interfaces (GUIs), and the like.
  • UIs user interfaces
  • GUIs graphical user interfaces
  • the storage unit 300 stores data and programs necessary for the abnormality detecting apparatus 10 to operate.
  • the storage unit 300 may store a plurality of application programs (application programs or applications), data for operating the abnormality detecting apparatus 10, and instructions that are driven by the abnormality detecting apparatus 10. At least some of these applications may be downloaded from an external service providing apparatus through wireless communication. In addition, at least some of these applications may be present on the abnormality detecting device 10 from the time of shipment for basic functions (for example, call forwarding, originating function, message receiving, and transmitting function) of the abnormality detecting device 10. .
  • the application program is stored in the storage unit 300, may be installed in the abnormality detection device 10, the controller 600 may be driven to perform the operation (or function) of the abnormality detection device 10.
  • the storage unit 300 may include a flash memory type, a hard disk type, a multimedia card micro type, and a card type memory (eg, SD or XD memory). Etc.), magnetic memory, magnetic disk, optical disk, random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EPM), PROM It may include at least one storage medium of (Programmable Read-Only Memory).
  • the abnormality detecting apparatus 10 may operate a web storage that performs a storage function of the storage unit 300 on the Internet, or may operate in connection with the web storage.
  • the storage unit 300 stores weather information collected (or measured) by the sensor unit 100 or weather information received by the communication unit 200 under the control of the controller 600.
  • the display unit 400 may display various contents such as various menu screens using a user interface and / or a graphic user interface stored in the storage unit 300 under the control of the controller 600.
  • the content displayed on the display unit 400 includes various text or image data (including various information data) and a menu screen including data such as icons, list menus, combo boxes, and the like.
  • the display unit 400 may be a touch screen.
  • the display unit 400 may include a liquid crystal display (LCD), a thin film transistor liquid crystal display (TFT LCD), an organic light-emitting diode (OLED), and a flexible display (LCD).
  • the display device may include at least one of a flexible display, a 3D display, an e-ink display, and a light emitting diode (LED).
  • the display unit 400 may be configured as a stereoscopic display unit for displaying a stereoscopic image.
  • the stereoscopic display unit may be a three-dimensional display method such as a stereoscopic method (glasses method), an auto stereoscopic method (glasses-free method), a projection method (holographic method).
  • the display unit 400 displays weather information collected (or measured) by the sensor unit 100 or weather information received by the communication unit 200 under the control of the controller 600.
  • the voice output unit 500 outputs voice information included in the signal processed by the controller 600 by a predetermined signal.
  • the voice output unit 500 may include a receiver, a speaker, a buzzer, and the like.
  • the voice output unit 500 outputs the guide voice generated by the controller 600.
  • the voice output unit 500 may, under the control of the controller 600, weather information collected (or measured) by the sensor unit 100 or voice information corresponding to the weather information received by the communication unit 200. Outputs
  • the controller 600 executes an overall control function of the abnormality detecting apparatus 10.
  • the controller 600 executes the overall control function of the abnormality detecting apparatus 10 by using the program and data stored in the storage 300.
  • the controller 600 may include a RAM, a ROM, a CPU, a GPU, a bus, and the RAM, a ROM, a CPU, a GPU, and the like may be connected to each other through a bus.
  • the CPU may access the storage unit 300 to perform booting using the O / S stored in the storage unit 300, and various operations using various programs, contents, data, etc. stored in the storage unit 300. Can be performed.
  • the controller 600 classifies (or groups) a plurality of error analysis methods into a plurality of error analysis groups based on preset error determination reliability.
  • the plurality of error analysis methods include a quality control (QC) method, a daily variation pattern method, a periphery deviation method, a prediction-observation correlation coefficient method, a live / prediction-observation deviation method, an observation value distribution method, and an FFT. Abnormal fluctuation / wave method, machine learning based anomaly detection method, and the like.
  • the controller 600 classifies (or groups) the quality control method into the first error analysis group based on the error determination reliability with respect to the plurality of error analysis methods, and divides the daily change pattern method and the surrounding area deviation method into the second error.
  • Classify the analysis group classify the predictive-observation correlation coefficient method and the live / predictive-observation deviation method into the third error analysis group, and use the observation distribution method, abnormal variation / wave method using FFT, and machine learning-based anomaly detection.
  • Method is classified into a fourth error analysis group.
  • controller 600 stores information about a plurality of classified (or grouped) error analysis groups in the storage 300.
  • the quality control method is a method of detecting an abnormality according to whether the number of errors in general quality management is more than a predetermined reference value.
  • the daily change pattern method is a method of detecting abnormality according to whether the case where the average wind speed in a specific time zone of the day or more occurs over a predetermined value.
  • the peripheral area deviation method is a method of detecting the abnormality according to whether or not a deviation of a predetermined reference value with respect to a point within a predetermined specific radius occurs.
  • the prediction-observation correlation coefficient method performs anomaly detection according to whether the correlation coefficient between the prediction data and the observation data (or measured / collected weather information / detection data) decreases by more than a preset reference value for a predetermined period of time. That's the way.
  • the running / prediction-observation deviation method (or the running and prediction and observation deviation method) is a deviation between the running analysis value and the observation data, or the deviation between the prediction data and the observation data occurs more than a predetermined reference value, and occurs more than a reference number of times. In this case, abnormality detection is performed according to whether or not a specific period of time occurs.
  • the observation distribution method detects abnormality when the distribution pattern (or frequency pattern) of the observation data is out of the normal pattern and shows a specific error pattern (for example, when the wind speed frequency below a certain value is below the reference value for wind speed). This is how it is done.
  • the abnormal fluctuation / fluctuation method using the FFT is a method of detecting an abnormality depending on whether the frequency pattern of the observed data is excessive or understated in a specific frequency component that is outside the preset normal pattern.
  • the abnormality detection method based on the machine learning is a method of performing abnormality detection according to whether or not an abnormal value is generated through abnormality detection of time series data based on machine learning.
  • the controller 600 confirms (or determines) a grade of the weather information by sequentially applying the weather information to a plurality of error analysis groups stored in the storage 300 with respect to the weather information.
  • the grade may be set in plural according to the designer's design, for example, the first grade (or normal grade), the second grade (or suspect grade), the third grade (or fault reporting grade), etc. Can be.
  • the error determination result for each error analysis group may be set in plural according to the designer's design, and may be, for example, divided (or set) into no error, within a reference range, or above a reference range.
  • the reference range for each error analysis method may be all the same, at least a part or the same for the plurality of error analysis methods.
  • a plurality of error analysis groups are formed into four groups, and error determination results for each error analysis group are divided into three (for example, no error, within a reference range, or above a reference range), and the final grade is
  • the description is divided into three categories (normal grade, suspicion grade, and fault reporting grade), but the present invention is not limited thereto, and the number of error analysis groups, the number of error determination results, and the number of final grades vary according to the designer's design. Can be set.
  • the controller 600 sequentially applies the weather information to one or more error analysis methods included in the error analysis group for each error analysis group based on the plurality of error analysis groups stored in the storage 300. At this time, the controller 600 is configured such that the result of the error analysis method included in the error analysis group of the next step is different according to the determination result of the error analysis method in the previous step.
  • the error analysis of the third error analysis group relates to an intelligent correlation analysis, and the embodiment of the present invention does not use only one correlation analysis but applies a plurality of different error analysis methods.
  • the prediction-observation correlation coefficient method (or prediction-observation analysis method), which is one of the error analysis methods of the third error analysis group of the present invention, analyzes the correlation between the prediction and the observation (or the previously collected weather information), If the correlation decreases, it is determined as an error.
  • the controller 600 Based on the corresponding correlation analysis data source, the change in correlation shown in FIG. 3 and the change in deviation shown in FIG. 4 are respectively calculated.
  • observation value distribution method of the fourth error analysis group is an intelligent error determination method using a statistical method as a method of determining when a measurement distribution pattern of observation data (or previously collected weather information) is out of a predetermined normal pattern. .
  • the control unit 600 determines that the stationary state (or a normal point) as shown in FIG. As shown in 6, it is determined to be in a somewhat suspicious state (or abnormality point) because it exists within the reference range, and it is determined to be in a fault reporting state (or error serious point) because it exists in the reference range or more as shown in FIG. .
  • the controller 600 outputs weather information, a grade (or grade information) for the confirmed (or determined) corresponding weather information through the display unit 400 and / or the voice output unit 500.
  • control unit 600 may correspond to the plurality of grades of weather information stored in advance in the storage unit 300 based on the weather information, the grade (or grade information) of the confirmed (or determined) corresponding weather information, and the like. Matches and stores the class corresponding to the weather information.
  • the controller 600 based on the grade according to the purpose of using the weather information, at least one weather information corresponding to the grade according to the purpose of use among a plurality of weather information for each grade stored in advance in the storage unit 300 To perform the control function.
  • the controller 600 may classify the reliability grades for the weather information (or the sensing data), and may appropriately use such grade-specific data to enable selection of the quantity and quality of the weather information.
  • control unit 600 uses only high grade (eg, normal grade) data (or weather information) for sensitive analysis, and high grade data and medium grade (eg, for less sensitive analysis or live situations). Suspicious grade) data, and the lowest grade (e.g., disability reporting grade) data can be deleted (or discarded) or used for failure analysis.
  • high grade eg., normal grade
  • medium grade e.g., for less sensitive analysis or live situations.
  • the lowest grade (e.g., disability reporting grade) data can be deleted (or discarded) or used for failure analysis.
  • the abnormality detecting apparatus 10 may further include an interface unit (not shown) which serves as an interface with all external devices connected to the abnormality detecting apparatus 10.
  • the interface unit may include a wired / wireless headset port, an external charger port, a wired / wireless data port, a memory card port, a port for connecting a device equipped with an identification module, an audio I / O ( Input / Output) port, video I / O (Input / Output) port, earphone port, and the like.
  • the identification module is a chip that stores various information for authenticating the use authority of the abnormality detection device 10, and includes a user identity module (UIM), a subscriber identity module (SIM), and a general user.
  • the device equipped with the identification module may be manufactured in the form of a smart card. Therefore, the identification module may be connected to the abnormality detecting device 10 through a port.
  • Such an interface unit may receive data from an external device or receive power to transmit the data to each component inside the abnormality detecting device 10 or to transmit the data inside the abnormality detecting device 10 to an external device.
  • the interface unit may be a passage for supplying power from the cradle to the abnormality detecting device 10, or various command signals inputted from the cradle by a user may be used. It may be a passage that is delivered to the abnormality detection device 10. Various command signals or corresponding power input from the cradle may be operated as signals for recognizing that the abnormality detecting device 10 is correctly mounted on the cradle.
  • the abnormality detecting device 10 may be an input unit for receiving a command or a control signal generated by an operation such as receiving a button according to a button operation or an arbitrary function selection by a user, or touching / scrolling a displayed screen. It may further include (not shown).
  • the input unit is a means for receiving at least one of a user's command, selection, data, and information, and may include a plurality of input keys and function keys for receiving numeric or text information and setting various functions.
  • the input unit includes a key pad, a dome switch, a touch pad (static pressure / capacitance), a touch screen, a jog wheel, a jog switch, a jog shuttle, and a mouse.
  • a touch pad static pressure / capacitance
  • a touch screen a touch screen
  • jog wheel a jog wheel
  • a jog switch a jog shuttle
  • mouse a mouse.
  • Various devices such as a stylus pen, a touch pen, and the like may be used.
  • the display unit 400 is formed in the form of a touch screen, some or all of the functions of the input may be performed through the display unit 400.
  • each component (or module) of the abnormality detecting apparatus 10 may be software stored on the memory (or the storage 300) of the abnormality detecting apparatus 10.
  • the memory may be an internal memory of the abnormality detecting device 10, and may be an external memory or another type of storage device.
  • the memory may also be a nonvolatile memory.
  • Software stored on the memory may include a set of instructions to cause the abnormality detection apparatus 10 to perform a specific operation when executed.
  • the detection data measured by the weather sensor may be combined to determine whether the error is based on a plurality of error analysis methods classified into a plurality of error analysis groups according to error determination reliability.
  • the general QC method which is an error analysis method, adds a daily variation pattern analysis, a probability distribution analysis, a frequency analysis, a correlation analysis, and removes error data by determining the individual analysis results in a complex manner, or removes the data for the detected data.
  • the classification can be classified and provided to use the detection data of the grade suitable for the purpose of use.
  • the complex determination criteria for each error analysis step may be changed and optimized according to the user's administration setting.
  • FIG. 8 is a flowchart illustrating a method of detecting an abnormality of a weather sensor according to an exemplary embodiment of the present invention.
  • the controller 600 classifies (or groups) a plurality of error analysis methods into a plurality of error analysis groups based on preset error determination reliability.
  • the plurality of error analysis methods include a quality control (QC) method, a daily variation pattern method, a periphery deviation method, a prediction-observation correlation coefficient method, a live / prediction-observation deviation method, an observation value distribution method, and an FFT.
  • QC quality control
  • Abnormal fluctuation / wave method machine learning based anomaly detection method, and the like.
  • the controller 600 classifies (or groups) the quality control method into the first error analysis group based on the error determination reliability with respect to the plurality of error analysis methods, and divides the daily change pattern method and the surrounding area deviation method into the second error.
  • Classify the analysis group classify the predictive-observation correlation coefficient method and the live / predictive-observation deviation method into the third error analysis group, and use the observation distribution method, abnormal variation / wave method using FFT, and machine learning-based anomaly detection.
  • Method is classified into a fourth error analysis group.
  • controller 600 stores information about a plurality of classified (or grouped) error analysis groups in the storage 300.
  • the controller 600 classifies the quality control method (QC) into the first error analysis group based on the error determination reliability, classifies the daily change pattern method and the surrounding area deviation method into the second error analysis group, and predicts the result.
  • QC quality control method
  • -Observation correlation coefficient method and live / prediction-observation deviation method are classified into the third error analysis group, and observation distribution method, abnormal variation / wave method using FFT, and abnormality detection method based on machine learning are the fourth error analysis group.
  • the controller 600 stores information about the plurality of classified error analysis groups in the storage 300 (S810).
  • the collection unit 100 collects (or measures) weather information (or sensing data) for the plurality of management areas, respectively.
  • the weather information includes temperature, air pressure, wind speed, relative humidity, sunshine, precipitation, rainfall probability, ultraviolet index, unique identification information of a region (or management region), and location information of a corresponding management region.
  • the communication unit 200 may receive weather information for each management area provided from a server (not shown).
  • the collection unit 100 collects first weather information about the first management area (S820).
  • the controller 600 checks (or determines) the grade of the weather information by sequentially applying the weather information to the plurality of error analysis groups stored in the storage 300 with respect to the weather information.
  • the grade may be set in plural according to the designer's design, for example, the first grade (or normal grade), the second grade (or suspect grade), the third grade (or fault reporting grade), etc. Can be.
  • the error determination result for each error analysis group may be set in plural according to the designer's design, and may be, for example, divided (or set) into no error, within a reference range, or above a reference range.
  • the reference range for each error analysis method may be all the same, at least a part or the same for the plurality of error analysis methods.
  • a plurality of error analysis groups are formed into four error analysis groups, and error determination results for each error analysis group are divided into three (for example, no error, within a reference range, and a reference range or more), and The three classifications (normal rating, suspicious rating, and failure reporting rating) are explained, but not limited thereto.
  • the number of error analysis groups, the number of error determination results, and the number of final ratings are determined. It can be set differently.
  • the controller 600 sequentially applies the weather information to one or more error analysis methods included in the error analysis group for each error analysis group based on the plurality of error analysis groups stored in the storage 300. At this time, the controller 600 is configured such that the result of the error analysis method included in the error analysis group of the next step is different according to the determination result of the error analysis method in the previous step.
  • the controller 600 outputs weather information, a grade (or grade information) for the confirmed (or determined) corresponding weather information through the display unit 400 and / or the voice output unit 500.
  • control unit 600 may correspond to the plurality of grades of weather information stored in advance in the storage unit 300 based on the weather information, the grade (or grade information) of the confirmed (or determined) corresponding weather information, and the like. Matches and stores the class corresponding to the weather information.
  • the control unit 600 may collect the first information collected prior to the quality management method (QC) included in the first error analysis group for the first error analysis group to the fourth error analysis group. 1
  • the first error determination result is calculated by applying weather information.
  • the first error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 confirms (or determines / confirms) the corresponding first weather information as the fault reporting grade (or grade 3).
  • controller 600 applies the first weather information to the daily change pattern method and the surrounding area deviation method included in the second error analysis group based on the first error determination result calculated by the first error analysis group.
  • the error determination result is calculated.
  • the control unit 600 determines the first weather information based on the daily change pattern method and the surrounding area deviation method included in the second error analysis group. Are applied to calculate the second error determination result.
  • the second error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 confirms (or determines / confirms) the corresponding first weather information as the fault reporting grade (or grade 3).
  • the controller 600 applies the first weather information to the daily change pattern method and the surrounding area deviation method included in the second error analysis group. Each of them is applied to calculate a third error determination result.
  • the third error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 checks (or determines / confirms) the corresponding first weather information as the fault reporting grade (or grade 3).
  • control unit 600 may include the prediction-observation correlation coefficient method and the live / prediction- included in the third error analysis group based on the second error determination result or the third error determination result calculated by the second error analysis group.
  • An error determination result is calculated by applying the first weather information to the observation deviation method.
  • the control unit 600 includes the prediction-observation correlation coefficient method and the running / prediction-observation deviation method included in the third error analysis group.
  • the fourth error determination result is calculated by applying the first weather information to each.
  • the fourth error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 confirms (or determines / confirms) the corresponding first weather information as the disability report grade (or grade 3).
  • the controller 600 may predict-observe correlation included in the third error analysis group.
  • the fifth error determination result is calculated by applying the first weather information to the counting method and the running / prediction-observing deviation method, respectively.
  • the fifth error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 confirms (or determines / confirms) the corresponding first weather information as the disability report grade (or grade 3).
  • the controller 600 may determine the prediction-observation correlation coefficient method and the running / prediction-observation deviation method included in the third error analysis group.
  • the sixth error determination result is calculated by applying the first weather information, respectively.
  • the sixth error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 may include an observation value distribution method included in the fourth error analysis group based on any one of the fourth error determination result to the sixth error determination result calculated by the third error analysis group.
  • the first weather information is applied to the abnormal fluctuation / wave method using the FFT and the abnormality detection method based on the machine learning to calculate the error determination result.
  • the controller 600 may include an observation value distribution method included in the fourth error analysis group, an abnormal variation / wave method using an FFT, and The seventh error determination result is calculated by applying the first weather information to the machine learning-based abnormality detection method.
  • the seventh error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 confirms (or determines / confirms) the corresponding first weather information as the disability report grade (or grade 3).
  • the controller 600 checks the corresponding first weather information as a suspected grade (or grade 2).
  • the control unit 600 determines whether The eighth error determination result is calculated by applying the first meteorological information to the observation distribution method, the abnormal fluctuation / wave method using the FFT, and the machine learning based abnormal detection method included in the error analysis group.
  • the eighth error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 checks (or determines / confirms) the corresponding first weather information as the disability report grade (or grade 3).
  • the controller 600 checks the corresponding first weather information as the suspected grade (or grade 2). In addition, when the eighth error determination result is no error, the controller 600 checks the corresponding first weather information as a normal grade (or grade 1).
  • the control unit 600 includes the observation value distribution method included in the fourth error analysis group.
  • the first error information is calculated by applying the first weather information to the abnormal fluctuation / wave method using the FFT and the abnormal detection method based on the machine learning.
  • the ninth error determination result includes no error, within a reference range and above a reference range.
  • the controller 600 checks the corresponding first weather information as the suspected grade (or grade 2).
  • the controller 600 checks the corresponding first weather information as a normal grade (or grade 1).
  • the controller 600 checks the corresponding first weather information as a normal grade (or grade 1).
  • control unit 600 provides the first weather information for each of the plurality of error analysis groups so that the result of the error analysis method included in the error analysis group of the next step is different according to the determination result of the error analysis method of the previous step.
  • the final grade for the first weather information can be confirmed (S830).
  • the controller 600 based on the grade according to the purpose of using the weather information, among the plurality of grades of weather information stored in advance in the storage unit 300 at least one weather information corresponding to the grade according to the purpose of use To perform the control function.
  • the controller 600 when a grade according to the purpose of use is set to be high for sensitive analysis, the controller 600 includes 10 pieces corresponding to the normal grade, which is the highest grade among a plurality of grades of weather information stored in the storage 300 in advance. Perform analysis function using weather information.
  • the controller 600 when a grade according to the purpose of use is set to a general grade for general analysis, the controller 600 corresponds to a normal grade and a suspect grade among weather information for each grade stored in the storage 300 in advance.
  • An analysis function is performed using 15 weather information (for example, 7 weather information of a normal grade and 8 weather information of a suspect grade).
  • An apparatus for detecting an abnormality of a weather sensor may be prepared by a computer program, and codes and code segments constituting the computer program may be easily inferred by a computer programmer in the art.
  • the computer program is stored in a computer readable media, and the abnormality of the weather sensor by being read and executed by a computer or an abnormality detection device of a weather sensor, a server, or the like according to an embodiment of the present invention.
  • the sensing device can be implemented.
  • the information storage medium includes a magnetic recording medium, an optical recording medium and a carrier wave medium.
  • the computer program for implementing the abnormality detection device of the weather sensor according to the embodiment of the present invention may be stored and installed in an internal memory such as the abnormality detection device of the weather sensor or a server.
  • an external memory such as a smart card that stores and installs a computer program that implements the abnormality detection device of the weather sensor according to an embodiment of the present invention may be mounted on the abnormality detection device of the weather sensor through an interface.
  • the detection data measured by the weather sensor is combined to determine the error based on a plurality of error analysis methods classified into a plurality of error analysis groups according to error determination reliability, It is possible to evaluate the quality of collected weather information with high reliability, and provide accurate sensing data by checking error data precisely and accurately.
  • the embodiment of the present invention adds a daily variation pattern analysis, probability distribution analysis, frequency analysis, correlation analysis, and the like to the general QC method, which is an error analysis method, and determines an individual analysis result in a complex manner.
  • the general QC method which is an error analysis method, and determines an individual analysis result in a complex manner.
  • the present invention provides a high reliability of the collected weather information by complex determination of errors based on a plurality of error analysis methods classified into a plurality of error analysis groups according to the error determination reliability of the measured data measured by the weather sensor. It is possible to evaluate the quality and to accurately and accurately check the error data to provide reliable detection data, it can be widely used in the field of weather information sensor, meteorological information application, terminal.

Abstract

La présente invention concerne un dispositif de détection d'anomalies d'un capteur météorologique à l'aide d'une détermination composite, un procédé associé et un support d'enregistrement sur lequel est enregistré un programme informatique. En d'autres termes, la présente invention permet de déterminer de manière composite si une erreur s'est produite dans des données de détection mesurées par un capteur météorologique sur la base d'une pluralité de procédés d'analyse d'erreur classés dans une pluralité de groupes d'analyse d'erreur en fonction de la fiabilité de détermination d'erreur, et ainsi une évaluation de qualité hautement fiable des informations météorologiques collectées peut être possible, et des données de détection fiables peuvent être fournies par une vérification élaborée et précise des données d'erreur.
PCT/KR2017/011290 2016-12-08 2017-10-13 Dispositif de détection d'anomalies de capteur météorologique à l'aide d'une détermination composite, procédé associé et support d'enregistrement sur lequel est enregistré un programme informatique WO2018105872A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR1020160166669A KR101874085B1 (ko) 2016-12-08 2016-12-08 복합 판단을 통한 기상 센서의 이상 감지 장치, 그 방법 및 컴퓨터 프로그램이 기록된 기록매체
KR10-2016-0166669 2016-12-08

Publications (1)

Publication Number Publication Date
WO2018105872A1 true WO2018105872A1 (fr) 2018-06-14

Family

ID=62491550

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2017/011290 WO2018105872A1 (fr) 2016-12-08 2017-10-13 Dispositif de détection d'anomalies de capteur météorologique à l'aide d'une détermination composite, procédé associé et support d'enregistrement sur lequel est enregistré un programme informatique

Country Status (2)

Country Link
KR (1) KR101874085B1 (fr)
WO (1) WO2018105872A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116146282A (zh) * 2023-04-18 2023-05-23 枣庄矿业(集团)济宁七五煤业有限公司 一种防冲隐蔽工程施工的智能监管系统
WO2023125197A1 (fr) * 2021-12-27 2023-07-06 华为技术有限公司 Procédé de détection, appareil de communication, support et puce

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110941031B (zh) * 2019-10-25 2022-08-05 浩亚信息科技有限公司 一种民航自动气象站在线核查方法及存储介质
KR102501286B1 (ko) * 2020-12-15 2023-02-22 한국전자통신연구원 기체 센서의 이상 감지 장치 및 방법
KR102493171B1 (ko) 2020-12-24 2023-01-31 (주)휴빌론 센서 데이터 융합 기반 IoT 기상센서기기 이상감지 방법 및 시스템

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20100006417A (ko) * 2008-07-09 2010-01-19 (주)정한전자시스템 디지털 대기압 측정 센서 및 이의 보정 방법
KR20130050514A (ko) * 2011-11-08 2013-05-16 한국전자통신연구원 항공 교통 관제를 위한 기상 정보 자동 적용 장치 및 방법
KR20140044264A (ko) * 2013-09-11 2014-04-14 (주) 네오위드넷 통합 품질관리시스템 기반의 품질검사 방법
JP2015010927A (ja) * 2013-06-28 2015-01-19 Kddi株式会社 観測値処理装置、観測値処理方法及び観測値処理プログラム
KR101519012B1 (ko) * 2014-12-08 2015-05-18 대한민국(기상청장) 기상정보 오류 검증 시스템 및 방법

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101655776B1 (ko) 2015-03-04 2016-09-08 (주)한국해양기상기술 소형 자동기상관측 장치 및 이를 사용한 관측 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20100006417A (ko) * 2008-07-09 2010-01-19 (주)정한전자시스템 디지털 대기압 측정 센서 및 이의 보정 방법
KR20130050514A (ko) * 2011-11-08 2013-05-16 한국전자통신연구원 항공 교통 관제를 위한 기상 정보 자동 적용 장치 및 방법
JP2015010927A (ja) * 2013-06-28 2015-01-19 Kddi株式会社 観測値処理装置、観測値処理方法及び観測値処理プログラム
KR20140044264A (ko) * 2013-09-11 2014-04-14 (주) 네오위드넷 통합 품질관리시스템 기반의 품질검사 방법
KR101519012B1 (ko) * 2014-12-08 2015-05-18 대한민국(기상청장) 기상정보 오류 검증 시스템 및 방법

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023125197A1 (fr) * 2021-12-27 2023-07-06 华为技术有限公司 Procédé de détection, appareil de communication, support et puce
CN116146282A (zh) * 2023-04-18 2023-05-23 枣庄矿业(集团)济宁七五煤业有限公司 一种防冲隐蔽工程施工的智能监管系统
CN116146282B (zh) * 2023-04-18 2023-06-30 枣庄矿业(集团)济宁七五煤业有限公司 一种防冲隐蔽工程施工的智能监管系统

Also Published As

Publication number Publication date
KR20180065589A (ko) 2018-06-18
KR101874085B1 (ko) 2018-08-02

Similar Documents

Publication Publication Date Title
WO2018105872A1 (fr) Dispositif de détection d'anomalies de capteur météorologique à l'aide d'une détermination composite, procédé associé et support d'enregistrement sur lequel est enregistré un programme informatique
WO2018105871A1 (fr) Dispositif de gestion d'informations météorologiques utilisant un capteur de poussière fine, procédé associé et support d'enregistrement sur lequel est enregistré un programme informatique
WO2016003084A1 (fr) Procédé et appareil de notification d'hameçonnage par sms
CN104077253B (zh) 跨装置通讯传输系统及其方法
WO2015126224A1 (fr) Procédé de fourniture d'image de prévisualisation concernant le réglage d'affichage pour un dispositif
WO2015005537A1 (fr) Dispositif mobile, afficheur de tête et procédé de commande associé
WO2019225842A1 (fr) Procédé et appareil de communication faisant appel à une formation de faisceau dans un système de communication sans fil
WO2015072665A1 (fr) Appareil d'affichage et procédé de paramétrage d'une télécommande universelle
WO2016013810A1 (fr) Procédé et dispositif d'exploitation pour informations de catastrophes
WO2015133847A1 (fr) Procédé et appareil de détection d'entrée d'utilisateur dans un dispositif électronique
WO2015115852A1 (fr) Procédé et appareil d'ordonnancement de tâches
WO2016080561A1 (fr) Dispositif numérique et procédé de commande de ce dernier
WO2023058888A1 (fr) Appareil de diagnostic de dysfonctionnement d'équipement utilisant une image de spectrogramme sonore, et procédé associé
EP3138306A1 (fr) Dispositif électronique et procédé de fourniture de service d'appel vidéo d'urgence
CN110336865A (zh) 服务调用方法、装置、存储介质和电子设备
WO2015093902A1 (fr) Procédé et dispositif de recherche et de commande de dispositifs commandés dans un système domotique
WO2016006735A1 (fr) Dispositif électronique et procédé de recherche par un dispositif électronique d'un point d'accès dans un réseau local
CN104125328A (zh) 一种消息处理方法、装置及移动终端
WO2017022905A1 (fr) Système et procédé de contrôle environnemental
WO2012015091A1 (fr) Appareil et procédé d'affichage d'informations de service fournies dans une zone de desserte
WO2019124770A1 (fr) Appareil terminal et procédé de commande d'appareil terminal
WO2018105873A1 (fr) Dispositif de capteur de sol utilisant des informations météorologiques pour accroître la précision d'une détection d'erreur, procédé de commande de dispositif de capteur de sol et support d'enregistrement sur lequel est enregistré un programme informatique
CN106685702B (zh) 一种基于工业控制系统安全防护设备的大数据采集方法
WO2020184854A1 (fr) Appareil de reconnaissance d'empreintes digitales pour personne âgée, et procédé associé
WO2020159111A1 (fr) Dispositif électronique de prétraitement de données dans un environnement informatique périphérique, et son procédé de commande

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: 17878147

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 14/08/2019)

122 Ep: pct application non-entry in european phase

Ref document number: 17878147

Country of ref document: EP

Kind code of ref document: A1