WO2017163569A1 - データ処理装置、データ処理システム、データ処理方法及びプログラム - Google Patents
データ処理装置、データ処理システム、データ処理方法及びプログラム Download PDFInfo
- Publication number
- WO2017163569A1 WO2017163569A1 PCT/JP2017/001877 JP2017001877W WO2017163569A1 WO 2017163569 A1 WO2017163569 A1 WO 2017163569A1 JP 2017001877 W JP2017001877 W JP 2017001877W WO 2017163569 A1 WO2017163569 A1 WO 2017163569A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- waveform
- operation state
- target device
- distance
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
- G05B23/024—Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D27/00—Simultaneous control of variables covered by two or more of main groups G05D1/00 - G05D25/00
- G05D27/02—Simultaneous control of variables covered by two or more of main groups G05D1/00 - G05D25/00 characterised by the use of electric means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/33—Director till display
- G05B2219/33034—Online learning, training
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/33—Director till display
- G05B2219/33038—Real time online learning, training, dynamic network
-
- 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
Definitions
- the present invention relates to a data processing device, a data processing system, a data processing method, and a program.
- Patent Document 1 discloses an apparatus for diagnosing the presence / absence of an abnormality sign of a mechanical facility based on multidimensional sensor data measured by a plurality of sensors.
- Patent Documents 2 and 3 disclose methods for determining the type and operating status of an electrical device based on a current waveform.
- Patent Documents 1 to 3 it is necessary to previously generate measurement data (teacher data) representing a predetermined state (eg, a state where there is no abnormality, a predetermined operation state, etc.).
- This measurement data may vary depending on the operating state of the target device and the external environment. It takes a lot of time and effort to generate various teacher data corresponding to various operating states and various external environments.
- An object of the present invention is to provide a new technique for generating teacher data for estimating the state of an electrical device.
- Waveform data acquisition means for acquiring waveform data of current consumption and / or voltage of the target device; Feature amount extraction means for extracting a waveform feature amount from the waveform data;
- Environmental data acquisition means for acquiring environmental data indicating the environment of the target device when the waveform data is acquired;
- Operation state data acquisition means for acquiring operation state data indicating an operation state of the target device when the waveform data is acquired;
- a distance calculating means for calculating a distance between the member including the waveform feature, the environment data, and the operation state data, and a plurality of reference members; Grouping means for grouping the members based on a distance from each of the plurality of reference members;
- Registration means for registering a group satisfying a predetermined condition as teacher data;
- a data processing apparatus is provided.
- the terminal device Waveform data acquisition means for acquiring waveform data of current consumption and / or voltage of the target device; Feature amount extraction means for extracting a waveform feature amount from the waveform data;
- Environmental data acquisition means for acquiring environmental data indicating the environment of the target device when the waveform data is acquired;
- Operation state data acquisition means for acquiring operation state data indicating an operation state of the target device when the waveform data is acquired;
- Transmitting / receiving means for transmitting the waveform feature quantity, the environmental data and the operation state data to the server;
- Have The server Transmitting / receiving means for receiving the waveform feature value, the environmental data, and the operation state data from each of the plurality of terminal devices;
- a distance calculating means for calculating a distance between the member including the waveform feature, the environment data, and the operation state data, and a plurality of reference members;
- Grouping means for grouping the members based on a distance from each of the plurality of reference members;
- Registration means for registering
- Waveform data acquisition means for acquiring waveform data of current consumption and / or voltage of the target device; Feature amount extraction means for extracting a waveform feature amount from the waveform data;
- Environmental data acquisition means for acquiring environmental data indicating the environment of the target device when the waveform data is acquired;
- Operation state data acquisition means for acquiring operation state data indicating an operation state of the target device when the waveform data is acquired;
- a distance calculating means for calculating a distance between each of the plurality of reference members and a member including the waveform feature, the environment data, and the operation state data;
- Grouping means for grouping the members based on a distance from each of the plurality of reference members;
- Registration means for registering a group satisfying a predetermined condition as teacher data;
- a program is provided that functions as:
- Each unit included in the apparatus of the present embodiment is stored in a CPU (Central Processing Unit), a memory, a program loaded into the memory, a storage unit such as a hard disk storing the program (from the stage of shipping the apparatus in advance).
- a storage unit such as a hard disk storing the program (from the stage of shipping the apparatus in advance).
- storage media such as CDs (Compact Discs) and programs downloaded from servers on the Internet can also be stored.) Realized by any combination of hardware and software, centering on the network connection interface Is done. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus.
- FIG. 1 is a block diagram illustrating the hardware configuration of the apparatus according to the present embodiment.
- the apparatus includes a processor 1A, a memory 2A, an input / output interface 3A, a peripheral circuit 4A, and a bus 5A.
- the peripheral circuit includes various modules.
- the bus 5A is a data transmission path through which the processor 1A, the memory 2A, the peripheral circuit 4A, and the input / output interface 3A transmit / receive data to / from each other.
- the processor 1A is an arithmetic processing device such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit).
- the memory 2A is a memory such as a RAM (Random Access Memory) or a ROM (Read Only Memory).
- the input / output interface 3A includes an interface for acquiring information from an external device, an external server, an external sensor, and the like.
- the processor 1A issues a command to each module and performs a calculation based on the calculation result.
- the data processing apparatus 10 provides a function of collecting teacher data used for estimating the state of the target device. Details will be described below.
- FIG. 2 shows an example of a functional block diagram of the data processing apparatus 10 of the present embodiment.
- the data processing apparatus 10 includes a waveform data acquisition unit 11, a feature amount extraction unit 12, an environment data acquisition unit 13, an operation state data acquisition unit 14, a grouping unit 15, a registration unit 16, A distance calculation unit 19.
- the waveform data acquisition unit 11 acquires waveform data of current consumption and / or voltage of the target device.
- the target device may be a household electrical appliance used in a general household, an office device used in an office, a factory device used in a factory, or the like. There may be.
- a measurement sensor installed at a predetermined position continuously measures waveform data of current consumption and / or voltage of one corresponding target device.
- the waveform data acquisition unit 11 continuously acquires the waveform data measured by the measurement sensor.
- the waveform data is associated with date / time information of the measurement timing by any means.
- the waveform data measured by the measurement sensor may be stored in the waveform data DB (database) in association with the date / time information.
- the waveform data acquisition part 11 may acquire waveform data from waveform data DB.
- the waveform data acquisition unit 11 may acquire waveform data without using the waveform data DB.
- the feature amount extraction unit 12 extracts a predetermined waveform feature amount from the waveform data acquired by the waveform data acquisition unit 11.
- the feature amount extraction unit 12 may divide the waveform data into processing unit data for each predetermined time, and extract the waveform feature amount for each processing unit data.
- Waveform features are the frequency intensity and phase of current consumption (harmonic component), phase, current consumption change, average value, peak value, effective value, crest factor, waveform rate, current change convergence time, energization time, peak , The time difference between the peak position of the voltage and the peak position of the current consumption, the power factor, and the like are conceivable, but are not limited thereto.
- the feature amount extraction unit 12 extracts one or more types of predetermined waveform feature amounts.
- the environment data acquisition unit 13 acquires environment data indicating the environment of the target device when the waveform data is acquired.
- the environmental data acquisition unit 13 can acquire at least one of temperature, humidity, and weather as environmental data.
- the environmental data is associated with the observed date / time information by any means.
- the environmental data acquisition unit 13 may acquire the temperature and / or humidity measured by a measurement sensor installed around the target device, for example.
- the environmental data acquisition unit 13 continuously acquires the temperature and / or humidity measured by the measurement sensor.
- the data is associated with date / time information of the measurement timing by any means.
- the temperature and / or humidity measured by the measurement sensor may be stored in the temperature and / or humidity DB in association with the date / time information.
- the environmental data acquisition part 13 may acquire the said data from temperature and / or humidity DB.
- the environment data acquisition unit 13 may acquire the data without using the temperature and / or humidity DB.
- the environmental data acquisition unit 13 may acquire, for example, weather information on the installation location of the target device that is disclosed on the network from a predetermined server.
- the environment data acquisition unit 13 holds information indicating the installation location of the target device in advance.
- the server requests and obtains weather information corresponding to the installation location.
- the operator may input weather information to the environment data acquisition unit 13.
- the environment data acquisition unit 13 obtains the latest actual weather information (concept different from the weather forecast, weather information at that timing actually observed at each timing) announced every predetermined time as the above-mentioned predetermined weather information. You may acquire from a server or an operator for every time. In addition, the environment data acquisition unit 13 may acquire a weather forecast announced in advance from a server or an operator, and acquire weather information when waveform data is acquired from the weather forecast.
- the operation state data acquisition unit 14 acquires operation state data indicating the operation state of the target device when the waveform data is acquired.
- the operation state data acquisition unit 14 can acquire operation state data from the target device and / or from a measurement sensor attached to the target device.
- the operation state data is associated with date / time information when each operation state is observed by any means.
- the operation state data acquisition unit 14 continuously acquires operation state data from the target device and / or measurement sensor.
- the operation state data is associated with date / time information of the measurement timing by any means.
- the operation state data measured by the target device and / or the measurement sensor may be stored in the operation state data DB in association with the date / time information. Then, the operation state data acquisition unit 14 may acquire operation state data from the operation state DB. In addition, the operation state data acquisition unit 14 may acquire the operation state data without using the operation state DB.
- the operation state data will be described.
- the contents (shape, etc.) of the waveform data of the current consumption and / or voltage may be different from each other.
- an example of operation state data will be described for each type of target device.
- the operation status data includes data indicating the operating status (standby, operating, etc.), data indicating the operation mode (heating, cooling, dehumidification, etc.), set temperature, ambient temperature of the outdoor unit, The number of rotations of the outdoor unit fan, the temperature in the room, the number of rotations of the indoor unit fan, the operating state of the compressor (maximum XX%, etc.) can be considered.
- the operation status data may be the operating status (standby, operating, etc.), operating mode (normal mode, eco mode, etc.), set temperature, fan speed, etc.
- the operation status data includes operating status (freezing, defrosting, etc.), set temperature, outdoor unit ambient temperature, compressor operating status (maximum XX%, etc.) Conceivable.
- the operation status data includes operating status (refrigeration, defrosting, etc.), set temperature, outdoor unit ambient temperature, outdoor unit fan speed, indoor temperature, indoor unit The rotation speed of the fan, the operating state of the compressor (maximum XX%, etc.), etc. are conceivable.
- the operation status data may be the operating status (standby, heating, etc.), power setting, etc.
- the operation state data may be in the operating state (standby, watching, recording, etc.).
- the operation state data may be the operation state (refrigeration, defrosting, etc.), the internal temperature, the ambient temperature, the compressor operation state (maximum XX%, etc.), and the like.
- the operating status data may be the operating status (standby, operating, etc.).
- the operation state data may be the operating state (during washing, rinsing, dehydrating, waiting, etc.).
- the operating state data may be the operating state (standby, lighting, etc.), dimming state (XX%), etc.
- the operating status data includes the operating status (standby, warming, cold, etc.), operating mode (warming, cold, etc.), compressor operating status (maximum XX%) Etc.), internal temperature, ambient temperature, etc. can be considered.
- the operation status data may be the operating status (standby, copying, etc.).
- the operation state data may be the operating state (standby, depositing, withdrawal, etc.).
- the operating state data may be the operating state (lighted, extinguished, etc.).
- the operating state data may be the operating state (standby, heating, etc.).
- the operating state data may be the operating state (standby, washing, drying, etc.).
- the operation status data may be the operating status (standby, extraction, boiling, cleaning, etc.).
- the operation status data may be the operating status (standby, input, deposit, withdrawal, printing, etc.).
- the above-described waveform data acquisition unit 11, feature amount extraction unit 12, environment data acquisition unit 13 and operation state data acquisition unit 14 obtain values other than the distance in the data shown in FIG.
- the data shown in FIG. 3 includes information (processing unit data ID (identifier)) for identifying each processing unit data, which is a unit for extracting the waveform feature value from the waveform data, and the waveform feature value extracted from each processing unit data.
- processing unit data ID identifier
- the external environment data at the timing when each processing unit data is measured, the operation state data at the timing when each processing unit data is measured, and the distance calculated by the distance calculation unit 19 described below are associated with each other. ing.
- the data shown in FIG. 3 is updated as needed (new data is added).
- “External environment data at the timing when the processing unit data is measured” means the same timing as the date and time (target timing) when the processing unit data is measured, the timing closest to the target timing, and the closest after the target timing It may be external environment data measured at the timing or the closest timing before the target timing.
- processing unit data waveform data of processing unit
- multiple values corresponding to various external environment data are measured within the time width. obtain.
- the statistical values (average value, maximum value, minimum value, mode value, intermediate value, etc.) of the plurality of values may be external environment data at the timing when the processing unit data is measured.
- operation state data at the timing when the processing unit data is measured means the same timing as the date and time (target timing) when the processing unit data is measured, the timing closest to the target timing, and the most recent after the target timing. It means the operation state data measured at the closest timing or the closest timing before the target timing.
- the processing unit data waveform data of the processing unit
- a plurality of values corresponding to various operation state data are measured within the time width. obtain.
- the statistical values average value, maximum value, minimum value, mode value, intermediate value, etc.
- the plurality of values may be used as the operation state data at the timing when the processing unit data is measured.
- the distance calculation unit 19 includes the waveform feature amount extracted by the feature amount extraction unit 12, the environment data acquired by the environment data acquisition unit 13, and the operation state data acquired by the operation state data acquisition unit 14. The distance between the member and each of the plurality of reference members is calculated. The waveform feature amount, environment data, and operation state data corresponding to one processing unit data ID shown in FIG. 3 become one member.
- Each reference member includes waveform features, environmental data, and operating state data.
- each of the teacher data registered at that time is set as each reference member.
- FIG. 4 schematically shows an example of registered teacher data.
- information for identifying each of a plurality of teacher data, a waveform feature amount, external environment data, and operation state data are associated with each other.
- the information is updated by editing by the operator and registration of new teacher data by the registration unit 16 described below.
- an operator registers two or more teacher data.
- the teacher data registered in this way becomes a reference member.
- the number of reference members increases accordingly.
- the distance calculation unit 19 calculates the distance (similarity) between the newly acquired member and each of the teacher data (reference member) registered at that time.
- the calculation method of the distance is a design matter. For example, the distance may be calculated by machine learning.
- various waveform feature quantities, various environment data, and various operation state data are normalized according to predetermined rules, and multidimensional coordinates are obtained by arranging them in a predetermined order. May be. Then, the distance between the coordinate values may be calculated.
- the value in the distance column in FIG. 3 is determined by the processing by the distance calculation unit 19.
- N is an integer of 2 or more
- N distances are calculated corresponding to each member and registered as N-dimensional data in the distance column.
- the grouping unit 15 groups the members based on the distance from each of the plurality of reference members. Specifically, the grouping unit 15 groups items having similar distance (N-dimensional data) values.
- the grouping method is a design matter, but an example will be described below.
- the grouping unit 15 issues a group ID and makes the first member belong to the first group. Then, the distance (N-dimensional data) of the first member is registered as the representative distance of the first group. Thereby, the information shown in FIG. 5 is updated.
- the grouping unit 15 determines the distance of the second member (N-dimensional data) and the representative distance (N-dimensional data) of the first group registered at that time.
- the similarity is calculated.
- the grouping unit 15 treats N-dimensional data as N-dimensional coordinates, and calculates the distance between the coordinates as the similarity.
- the grouping unit 15 determines whether the similarity is equal to or less than a predetermined value (design item). When the value is equal to or less than the predetermined value (design item), the grouping unit 15 causes the second member to belong to the first group. Then, based on the distance (N-dimensional data) of the first member and the distance (N-dimensional data) of the second member, the representative distance (N-dimensional data) of the first group is calculated and updated. For example, the average value of the distance of the first member (N-dimensional data) and the distance of the second member (N-dimensional data) is used as the representative distance (N-dimensional data) of the first group. Thereby, the information shown in FIG. 5 is updated.
- the grouping unit 15 issues a new group ID and makes the second member belong to the second group. Then, the distance (N-dimensional data) of the second member is registered as the representative distance of the second group. Thereby, the information shown in FIG. 5 is updated.
- the grouping unit 15 determines the new member distance (N-dimensional data) and the representative distance ( N-dimensional data) may be calculated, and those having similarities below a predetermined value may be grouped.
- the registration unit 16 registers a group satisfying the predetermined condition as teacher data indicating a predetermined state of the target device.
- the predetermined condition include “the number of members is a predetermined number or more”.
- the registration unit 16 When new teacher data is registered by the registration unit 16, the information shown in FIG. 4 is updated. In addition, when new teacher data is registered by the registration unit 16, the number of reference members increases. Along with this, the dimension of the N-dimensional data indicating the distance between the reference member and each member also increases.
- the distance calculation unit 19 can update the value in the distance column shown in FIG. 3 when new teacher data is registered by the registration unit 16. For example, it is assumed that the number of reference members has increased from 3 to 4 due to the registration of new teacher data by the registration unit 16. In this case, the distance calculation unit 19 obtains the three-dimensional data (distance) indicating the distance between each member and each of the three reference members up to that point, between each member and each of the four new reference members. Updates to four-dimensional data (distance) indicating the distance.
- Each teacher data includes a waveform feature amount, environment data, and operation state data.
- wave teacher data at the time of activation, light load, and heavy load may be registered.
- the waveform data acquisition unit 11 acquires current consumption and / or voltage waveform data of the target device. Further, the environment data acquisition unit 13 acquires environment data indicating the environment of the target device when the waveform data is acquired. Further, the operation state data acquisition unit 14 acquires operation state data indicating the operation state of the target device when the waveform data is acquired.
- the feature quantity extraction unit 12 processes the processing unit data in the waveform data acquired in S10, and extracts a predetermined waveform feature quantity.
- the distance calculation unit 19 is registered at that time with the member including the waveform feature amount of the waveform data extracted in S11, the environmental data and the operation state data when the waveform data acquired in S10 is acquired.
- the distance to each teacher data (reference member) is calculated and associated with the member (see FIG. 3).
- the grouping unit 15 groups members based on the distance from each of the plurality of teacher data (reference members) (grouping process). Thereby, the information of FIG. 5 is updated.
- the registration unit 16 determines whether there is a group that satisfies a predetermined condition for registering as teacher data (whether it has newly occurred) (S14). ).
- the predetermined condition here is, for example, that the number of members is a predetermined number or more.
- the registration unit 16 newly registers the group as teacher data indicating the predetermined state of the target device (S15). Thereafter, the distance calculation unit 19 updates the value in the distance column shown in FIG. 3 (S16). Specifically, the value in the distance column is updated to information indicating the distance between each member and each of a plurality of teacher data (reference members) including newly registered teacher data.
- the grouping unit 15 groups the members based on the updated distance (grouping process). Thereby, the information of FIG. 5 is updated. Then, it progresses to S14 and repeats the same process.
- S17 it is determined whether or not to end the teacher data generation process. If it is determined not to end (No in S17), the process returns to S10 and is repeated. On the other hand, if it is determined to end (Yes in S17), the process ends. For example, “no new teacher data is registered for a predetermined time or more”, “the number of registered teacher data has reached a predetermined number”, “the execution time of teacher data generation processing has reached a predetermined time”, “operator If a condition such as “there was a process end instruction input from” is satisfied, the data processing apparatus 10 determines to end the process.
- teacher data indicating a predetermined state of the target device while operating the target device. For this reason, for example, teacher data can be generated while the target device is actually used in daily life.
- teacher data can be generated while being actually used in daily life, it is possible to reduce the labor of operating the target device only to generate teacher data. In this case, since it is not necessary to operate the target device wastefully, it is expected to save electricity costs.
- the waveform feature quantity of the group satisfying a predetermined condition can be registered as teacher data. Therefore, it is possible to register various teacher data corresponding to various external environments and various operation states.
- the waveform feature quantities of groups having a predetermined number of members or more is not registered as teacher data.
- the waveform feature amount that appears temporarily is not registered as teacher data due to an unexpected operation of the target device.
- registration of unnecessary teacher data can be reduced, and the inconvenience that the number of teacher data is unnecessarily large can be reduced.
- a plurality of members can be grouped based on the distance to each of a plurality of teacher data (reference members) registered at that time. For this reason, it is not necessary to prepare a lot of teacher data to be prepared first, and at least two teacher data are sufficient. In addition, there is no special condition for the teacher data to be prepared first.
- the teacher data can be collected in an environment that can be easily prepared, for example, temperatures of 10 ° C. and 20 ° C., which is one of the environmental data. Time and effort to prepare can be greatly reduced.
- the data processing apparatus 10 provides a function for estimating the state of the target device using teacher data generated by the functions described in the first embodiment. To do. Details will be described below.
- FIG. 7 shows an example of a functional block diagram of the data processing apparatus 10 of the present embodiment.
- the data processing apparatus 10 includes a waveform data acquisition unit 11, a feature amount extraction unit 12, an environment data acquisition unit 13, an operation state data acquisition unit 14, a grouping unit 15, and a registration unit 16.
- the configurations of the waveform data acquisition unit 11, the feature amount extraction unit 12, the environment data acquisition unit 13, the operation state data acquisition unit 14, the grouping unit 15, the registration unit 16, and the distance calculation unit 19 are the same as those in the first embodiment. is there.
- the teacher data storage unit 17 stores the teacher data registered by the registration unit 16 (see FIG. 4).
- the estimation unit 18 estimates the state of the target device using the teacher data stored in the teacher data storage unit 17. Specifically, it is estimated whether the state of the target device is abnormal or normal.
- the estimation unit 18 generates an estimation model by machine learning using the teacher data (normal waveform feature amount), and the feature amount extraction unit 12, the environment data acquisition unit 13, and the operation state data are added to the generated estimation model.
- An estimation result (normal or above) can be obtained by inputting a new member value acquired by the acquisition unit 14.
- the estimation model for example, a multiple regression analysis, a neural network, a hidden Markov model, or the like can be used.
- the estimation process by the estimation unit 18 may be performed in parallel with the teacher data generation process described with reference to FIG. 6 in the first embodiment, or may be performed after the teacher data collection process is completed. . In any case, the estimation accuracy by the estimation unit 18 increases as the number of teacher data stored in the teacher data storage unit 17 increases.
- the same operational effects as those of the first embodiment can be realized.
- the state (presence / absence of abnormality) of the target device can be estimated.
- various teacher data corresponding to various external environments and various operation states are generated and registered. By performing estimation processing based on such various teacher data, it is possible to improve estimation accuracy.
- various teacher data are collected by processing linked with a server (for example, a cloud server) and a plurality of terminal devices.
- FIG. 8 shows an example of a functional block diagram of the server 40 and the terminal device 30.
- each of the plurality of terminal devices 30 includes a waveform data acquisition unit 11, a feature amount extraction unit 12, an environment data acquisition unit 13, an operation state data acquisition unit 14, a teacher data storage unit 17-2, an estimation unit 18, and A transmission / reception unit 21 is included.
- the teacher data storage unit 17-2 and the estimation unit 18 may not be provided.
- the configurations of the waveform data acquisition unit 11, the feature amount extraction unit 12, the environment data acquisition unit 13, the operation state data acquisition unit 14, and the estimation unit 18 are the same as those in the first and second embodiments.
- the transmission / reception unit 21 transmits / receives data to / from another device (for example, the server 40).
- the teacher data storage unit 17-2 stores a plurality of teacher data.
- the server 40 includes a grouping unit 15, a registration unit 16, a teacher data storage unit 17-1, a distance calculation unit 19, and a transmission / reception unit 20.
- the configurations of the grouping unit 15, the registration unit 16, and the distance calculation unit 19 are the same as those in the first and second embodiments.
- the transmission / reception unit 20 transmits / receives data to / from other devices (for example, a plurality of terminal devices 30).
- the teacher data storage unit 17-1 stores a plurality of teacher data.
- a plurality of terminal devices 30 are installed corresponding to each of a plurality of target devices.
- the plurality of terminal devices 30 may be installed at geographically distant locations.
- the plurality of terminal devices 30 acquire waveform feature amounts, environment data, and operation state data corresponding to each of the plurality of target devices, and transmit them to the server 40.
- the terminal device 30 may transmit the waveform feature amount, the environmental data, and the operation state data to the server 40 in association with the identification information (model number or the like) of the corresponding target device.
- the server 40 groups the waveform feature amounts, environment data, and operation state data acquired from the plurality of terminal devices 30. Then, a group satisfying a predetermined condition is registered as teacher data.
- the server 40 can transmit the teacher data registered in the teacher data storage unit 17-1 to each of the plurality of terminal devices 30.
- Each terminal device 30 registers the teacher data received from the server 40 in the teacher data storage unit 17-2.
- the estimation unit 18 estimates the state of the target device using the teacher data stored in the teacher data storage unit 17-2.
- data (waveform feature values, environmental data, and operation state data) can be collected by a plurality of terminal devices 30, so that a large amount of data can be collected in a short time.
- teacher data can be generated efficiently.
- various teacher data can be generated that incorporates environmental factors (differences in temperature and humidity, and differences in temperature between day and night) due to geographical location. it can.
- Waveform data acquisition means for acquiring waveform data of current consumption and / or voltage of the target device; Feature amount extraction means for extracting a waveform feature amount from the waveform data;
- Environmental data acquisition means for acquiring environmental data indicating the environment of the target device when the waveform data is acquired;
- Operation state data acquisition means for acquiring operation state data indicating an operation state of the target device when the waveform data is acquired;
- a distance calculating means for calculating a distance between the member including the waveform feature, the environment data, and the operation state data, and a plurality of reference members; Grouping means for grouping the members based on a distance from each of the plurality of reference members;
- Registration means for registering a group satisfying a predetermined condition as teacher data;
- a data processing apparatus for registering a group satisfying a predetermined condition as teacher data;
- the distance calculation means is a data processing device that calculates the distance using the registered teacher data as the reference member.
- the registration means is a data processing apparatus for registering the group having the predetermined number of members or more as the teacher data. 4).
- the environmental data acquisition means is a data processing device that acquires at least one of temperature, humidity, and weather as the environmental data. 5.
- the operation state data acquisition unit is a data processing device that acquires the operation state data from the target device and / or from a sensor attached to the target device. 6).
- a data processing apparatus further comprising an estimation unit that estimates the state of the target device using the teacher data. 7).
- the terminal device Waveform data acquisition means for acquiring waveform data of current consumption and / or voltage of the target device; Feature amount extraction means for extracting a waveform feature amount from the waveform data;
- Environmental data acquisition means for acquiring environmental data indicating the environment of the target device when the waveform data is acquired;
- Operation state data acquisition means for acquiring operation state data indicating an operation state of the target device when the waveform data is acquired; Transmitting / receiving means for transmitting the waveform feature quantity, the environmental data and the operation state data to the server;
- Have The server Transmitting / receiving means for receiving the waveform feature value, the environmental data, and the operation state data from each of the plurality of terminal devices;
- a distance calculating means for calculating a distance between the member including the waveform feature, the environment data, and the operation state data, and a plurality of reference members;
- the data processing system in which the transmission / reception unit of the server transmits the teacher data registered by the registration unit to each of the plurality of terminal devices.
- Computer A waveform data acquisition step for acquiring waveform data of current consumption and / or voltage of the target device; A feature amount extraction step of extracting a waveform feature amount from the waveform data; An environmental data acquisition step of acquiring environmental data indicating an environment of the target device when the waveform data is acquired; An operation state data acquisition step of acquiring operation state data indicating an operation state of the target device when the waveform data is acquired; A distance calculating step of calculating a distance between each of the plurality of reference members and a member including the waveform feature, the environment data, and the operation state data; A grouping step of grouping the members based on a distance to each of the plurality of reference members; A registration process for registering a group satisfying a predetermined condition as teacher data; Data processing method to execute.
- 9-2 The data processing method according to 9, In the distance calculation step, a data processing method for calculating the distance by using the registered teacher data as the reference member. 9-3. In the data processing method according to 9 or 9-2, A data processing method for registering, as the teacher data, the group in which the number of members is a predetermined number or more in the registration step. 9-4. In the data processing method according to any one of 9 to 9-3, In the environmental data acquisition step, at least one of temperature, humidity, and weather is acquired as the environmental data. 9-5. In the data processing method according to any one of 9 to 9-4, In the operation state data acquisition step, a data processing method for acquiring the operation state data from the target device and / or from a sensor attached to the target device. 9-6.
- a data processing method in which the computer further executes an estimation step of estimating a state of the target device using the teacher data includes Computer Waveform data acquisition means for acquiring waveform data of current consumption and / or voltage of the target device; Feature amount extraction means for extracting a waveform feature amount from the waveform data; Environmental data acquisition means for acquiring environmental data indicating the environment of the target device when the waveform data is acquired; Operation state data acquisition means for acquiring operation state data indicating an operation state of the target device when the waveform data is acquired; A distance calculating means for calculating a distance between each of the plurality of reference members and a member including the waveform feature, the environment data, and the operation state data; Grouping means for grouping the members based on a distance from each of the plurality of reference members; Registration means for registering a group satisfying a predetermined condition as teacher data; Program to function as.
- the distance calculation means calculates the distance using the registered teacher data as the reference member. 10-3.
- the registration means is a program for registering the group in which the number of members is a predetermined number or more as the teacher data. 10-4.
- the environmental data acquisition means is a program for acquiring at least one of temperature, humidity and weather as the environmental data. 10-5.
- the operating state data acquisition means is a program for acquiring the operating state data from the target device and / or from a sensor attached to the target device. 10-6.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
Description
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段と、
前記波形データから波形特徴量を抽出する特徴量抽出手段と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段と、
所定条件を満たすグループを教師データとして登録する登録手段と、
を有するデータ処理装置が提供される。
複数の端末装置とサーバとを有し、
前記端末装置は、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段と、
前記波形データから波形特徴量を抽出する特徴量抽出手段と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを前記サーバに送信する送受信手段と、
を有し、
前記サーバは、
複数の前記端末装置各々から前記波形特徴量、前記環境データ及び前記動作状態データを受信する送受信手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段と、
所定条件を満たすグループを教師データとして登録する登録手段と、
を有するデータ処理システムが提供される。
コンピュータが、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得工程と、
前記波形データから波形特徴量を抽出する特徴量抽出工程と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得工程と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得工程と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出工程と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化工程と、
所定条件を満たすグループを教師データとして登録する登録工程と、
を実行するデータ処理方法が提供される。
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段、
前記波形データから波形特徴量を抽出する特徴量抽出手段、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段、
所定条件を満たすグループを教師データとして登録する登録手段、
として機能させるプログラムが提供される。
本実施形態のデータ処理装置10は、対象機器の状態を推定するために用いられる教師データを収集する機能を提供する。以下、詳細に説明する。
本実施形態のデータ処理装置10は、第1の実施形態で説明した機能に加えて、第1の実施形態で説明した機能で生成した教師データを用い、対象機器の状態を推定する機能を提供する。以下、詳細に説明する。
本実施形態では、サーバ(例:クラウドサーバ)と、複数の端末装置との連動した処理により、多様な教師データを収集する。
1. 対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段と、
前記波形データから波形特徴量を抽出する特徴量抽出手段と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段と、
所定条件を満たすグループを教師データとして登録する登録手段と、
を有するデータ処理装置。
2. 1に記載のデータ処理装置において、
前記距離算出手段は、登録されている前記教師データを前記参照メンバーとし、前記距離を算出するデータ処理装置。
3. 1又は2に記載のデータ処理装置において、
前記登録手段は、前記メンバーの数が所定数以上となった前記グループを、前記教師データとして登録するデータ処理装置。
4. 1から3のいずれかに記載のデータ処理装置において、
前記環境データ取得手段は、前記環境データとして、温度、湿度及び天候の中の少なくとも1つを取得するデータ処理装置。
5. 1から4のいずれかに記載のデータ処理装置において、
前記動作状態データ取得手段は、前記対象機器から、及び/又は、前記対象機器に取り付けられたセンサーから、前記動作状態データを取得するデータ処理装置。
6. 1から5のいずれかに記載のデータ処理装置において、
前記教師データを用いて、前記対象機器の状態を推定する推定手段をさらに有するデータ処理装置。
7. 複数の端末装置とサーバとを有し、
前記端末装置は、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段と、
前記波形データから波形特徴量を抽出する特徴量抽出手段と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを前記サーバに送信する送受信手段と、
を有し、
前記サーバは、
複数の前記端末装置各々から前記波形特徴量、前記環境データ及び前記動作状態データを受信する送受信手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段と、
所定条件を満たすグループを教師データとして登録する登録手段と、
を有するデータ処理システム。
8. 7に記載のデータ処理システムにおいて、
前記サーバの前記送受信手段は、前記登録手段により登録された前記教師データを、複数の前記端末装置各々に送信するデータ処理システム。
9. コンピュータが、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得工程と、
前記波形データから波形特徴量を抽出する特徴量抽出工程と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得工程と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得工程と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出工程と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化工程と、
所定条件を満たすグループを教師データとして登録する登録工程と、
を実行するデータ処理方法。
9-2. 9に記載のデータ処理方法において、
前記距離算出工程では、登録されている前記教師データを前記参照メンバーとし、前記距離を算出するデータ処理方法。
9-3. 9又は9-2に記載のデータ処理方法において、
前記登録工程では、前記メンバーの数が所定数以上となった前記グループを、前記教師データとして登録するデータ処理方法。
9-4. 9から9-3のいずれかに記載のデータ処理方法において、
前記環境データ取得工程では、前記環境データとして、温度、湿度及び天候の中の少なくとも1つを取得するデータ処理方法。
9-5. 9から9-4のいずれかに記載のデータ処理方法において、
前記動作状態データ取得工程では、前記対象機器から、及び/又は、前記対象機器に取り付けられたセンサーから、前記動作状態データを取得するデータ処理方法。
9-6. 9から9-5のいずれかに記載のデータ処理方法において、
前記コンピュータが、前記教師データを用いて、前記対象機器の状態を推定する推定工程をさらに実行するデータ処理方法。
10. コンピュータを、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段、
前記波形データから波形特徴量を抽出する特徴量抽出手段、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段、
所定条件を満たすグループを教師データとして登録する登録手段、
として機能させるプログラム。
10-2. 10に記載のプログラムにおいて、
前記距離算出手段は、登録されている前記教師データを前記参照メンバーとし、前記距離を算出するプログラム。
10-3. 10又は10-2に記載のプログラムにおいて、
前記登録手段は、前記メンバーの数が所定数以上となった前記グループを、前記教師データとして登録するプログラム。
10-4. 10から10-3のいずれかに記載のプログラムにおいて、
前記環境データ取得手段は、前記環境データとして、温度、湿度及び天候の中の少なくとも1つを取得するプログラム。
10-5. 10から10-4のいずれかに記載のプログラムにおいて、
前記動作状態データ取得手段は、前記対象機器から、及び/又は、前記対象機器に取り付けられたセンサーから、前記動作状態データを取得するプログラム。
10-6. 10から10-5のいずれかに記載のプログラムにおいて、
前記教師データを用いて、前記対象機器の状態を推定する推定手段をさらに有するプログラム。
Claims (10)
- 対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段と、
前記波形データから波形特徴量を抽出する特徴量抽出手段と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段と、
所定条件を満たすグループを教師データとして登録する登録手段と、
を有するデータ処理装置。 - 請求項1に記載のデータ処理装置において、
前記距離算出手段は、登録されている前記教師データを前記参照メンバーとし、前記距離を算出するデータ処理装置。 - 請求項1又は2に記載のデータ処理装置において、
前記登録手段は、前記メンバーの数が所定数以上となった前記グループを、前記教師データとして登録するデータ処理装置。 - 請求項1から3のいずれか1項に記載のデータ処理装置において、
前記環境データ取得手段は、前記環境データとして、温度、湿度及び天候の中の少なくとも1つを取得するデータ処理装置。 - 請求項1から4のいずれか1項に記載のデータ処理装置において、
前記動作状態データ取得手段は、前記対象機器から、及び/又は、前記対象機器に取り付けられたセンサーから、前記動作状態データを取得するデータ処理装置。 - 請求項1から5のいずれか1項に記載のデータ処理装置において、
前記教師データを用いて、前記対象機器の状態を推定する推定手段をさらに有するデータ処理装置。 - 複数の端末装置とサーバとを有し、
前記端末装置は、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段と、
前記波形データから波形特徴量を抽出する特徴量抽出手段と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを前記サーバに送信する送受信手段と、
を有し、
前記サーバは、
複数の前記端末装置各々から前記波形特徴量、前記環境データ及び前記動作状態データを受信する送受信手段と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段と、
所定条件を満たすグループを教師データとして登録する登録手段と、
を有するデータ処理システム。 - 請求項7に記載のデータ処理システムにおいて、
前記サーバの前記送受信手段は、前記登録手段により登録された前記教師データを、複数の前記端末装置各々に送信するデータ処理システム。 - コンピュータが、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得工程と、
前記波形データから波形特徴量を抽出する特徴量抽出工程と、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得工程と、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得工程と、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出工程と、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化工程と、
所定条件を満たすグループを教師データとして登録する登録工程と、
を実行するデータ処理方法。 - コンピュータを、
対象機器の消費電流及び/又は電圧の波形データを取得する波形データ取得手段、
前記波形データから波形特徴量を抽出する特徴量抽出手段、
前記波形データが取得された時の前記対象機器の環境を示す環境データを取得する環境データ取得手段、
前記波形データが取得された時の前記対象機器の動作状態を示す動作状態データを取得する動作状態データ取得手段、
前記波形特徴量、前記環境データ及び前記動作状態データを含むメンバーと、複数の参照メンバー各々との距離を算出する距離算出手段、
複数の前記参照メンバー各々との距離に基づき、前記メンバーをグループ化するグループ化手段、
所定条件を満たすグループを教師データとして登録する登録手段、
として機能させるプログラム。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/085,665 US10831184B2 (en) | 2016-03-23 | 2017-01-20 | Data processing apparatus, data processing system, data processing method, and non-transitory storage medium |
CN201780019327.5A CN108885448A (zh) | 2016-03-23 | 2017-01-20 | 数据处理装置、数据处理***、数据处理方法和程序 |
EP17769616.8A EP3435185A4 (en) | 2016-03-23 | 2017-01-20 | DATA PROCESSING DEVICE, DATA PROCESSING SYSTEM, DATA PROCESSING METHOD, AND PROGRAM |
CA3018660A CA3018660A1 (en) | 2016-03-23 | 2017-01-20 | Data processing apparatus, data processing system, data processing method, and storage medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016-058004 | 2016-03-23 | ||
JP2016058004A JP6686593B2 (ja) | 2016-03-23 | 2016-03-23 | データ処理装置、データ処理システム、データ処理方法及びプログラム |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017163569A1 true WO2017163569A1 (ja) | 2017-09-28 |
Family
ID=59899928
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2017/001877 WO2017163569A1 (ja) | 2016-03-23 | 2017-01-20 | データ処理装置、データ処理システム、データ処理方法及びプログラム |
Country Status (7)
Country | Link |
---|---|
US (1) | US10831184B2 (ja) |
EP (1) | EP3435185A4 (ja) |
JP (1) | JP6686593B2 (ja) |
CN (1) | CN108885448A (ja) |
CA (1) | CA3018660A1 (ja) |
TW (1) | TW201738677A (ja) |
WO (1) | WO2017163569A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019107360A1 (ja) * | 2017-11-28 | 2019-06-06 | 株式会社安川電機 | 制御システム、工場システム、学習システム、推定用モデルの生成方法及びアクチュエータの状態推定方法 |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018146747A1 (ja) * | 2017-02-08 | 2018-08-16 | 三菱電機株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
JP6810675B2 (ja) | 2017-11-16 | 2021-01-06 | 東京エレクトロンデバイス株式会社 | 情報処理装置及びプログラム |
WO2020194716A1 (ja) * | 2019-03-28 | 2020-10-01 | 三菱電機株式会社 | 信号選択装置、学習装置、信号選択方法及びプログラム |
TWI687696B (zh) * | 2019-06-26 | 2020-03-11 | 國立中興大學 | 回饋型隱藏式馬可夫模型辨識器的建立方法與基於此辨識器的辨識系統的建立方法 |
EP4094433A4 (en) | 2020-01-22 | 2024-02-21 | Nodar Inc. | NON-RIGID STEREOSCOPIC VISION CAMERA SYSTEM |
US11427193B2 (en) * | 2020-01-22 | 2022-08-30 | Nodar Inc. | Methods and systems for providing depth maps with confidence estimates |
US11577748B1 (en) | 2021-10-08 | 2023-02-14 | Nodar Inc. | Real-time perception system for small objects at long range for autonomous vehicles |
TWI783826B (zh) * | 2021-12-14 | 2022-11-11 | 國立中山大學 | 電力系統狀態的分析方法 |
US11782145B1 (en) | 2022-06-14 | 2023-10-10 | Nodar Inc. | 3D vision system with automatically calibrated stereo vision sensors and LiDAR sensor |
CN116302897B (zh) * | 2023-05-15 | 2023-10-17 | 合肥联宝信息技术有限公司 | 一种数据集的建立方法、装置、电子设备及存储介质 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010191556A (ja) * | 2009-02-17 | 2010-09-02 | Hitachi Ltd | 異常検知方法及び異常検知システム |
JP2014167667A (ja) * | 2013-02-28 | 2014-09-11 | Mitsubishi Electric Corp | 設備点検順位設定装置 |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW200919210A (en) | 2007-07-18 | 2009-05-01 | Steven Kays | Adaptive electronic design |
CN101802561B (zh) | 2007-09-11 | 2012-03-21 | 日本电气株式会社 | 数据记录器、数据保存方法和程序 |
JP5048625B2 (ja) | 2008-10-09 | 2012-10-17 | 株式会社日立製作所 | 異常検知方法及びシステム |
JP5392533B2 (ja) | 2008-10-10 | 2014-01-22 | ソニー株式会社 | 固体撮像素子、光学装置、信号処理装置及び信号処理システム |
BRPI0923437B1 (pt) | 2008-12-15 | 2021-03-09 | Accenture Global Services Limited | método de determinar um tipo de falha em uma rede de energia de múltiplas fases |
TW201037637A (en) | 2009-04-03 | 2010-10-16 | Lin Wen Yuan | Method of monitoring an object using ID recognization |
US8156055B2 (en) | 2009-05-04 | 2012-04-10 | ThinkEco, Inc. | System and method for utility usage, monitoring and management |
US9014685B2 (en) | 2009-06-12 | 2015-04-21 | Microsoft Technology Licensing, Llc | Mobile device which automatically determines operating mode |
JP2011155712A (ja) | 2010-01-25 | 2011-08-11 | Sony Corp | 電子機器、電力管理装置、及び機器特定方法 |
JP4832609B1 (ja) | 2011-06-22 | 2011-12-07 | 株式会社日立エンジニアリング・アンド・サービス | 異常予兆診断装置および異常予兆診断方法 |
JP5639562B2 (ja) | 2011-09-30 | 2014-12-10 | 株式会社東芝 | サービス実行装置、サービス実行方法およびサービス実行プログラム |
JP2013201230A (ja) | 2012-03-23 | 2013-10-03 | Seiko Instruments Inc | ホールセンサ |
JP5991942B2 (ja) | 2013-04-11 | 2016-09-14 | 日本電信電話株式会社 | 電源電圧歪みによる電流の特徴への影響除去法 |
US11002773B2 (en) * | 2013-07-17 | 2021-05-11 | Nec Corporation | Monitoring apparatus, monitoring method, and storage medium |
CN104699685B (zh) * | 2013-12-04 | 2018-02-09 | 富士通株式会社 | 模型更新装置及方法、数据处理装置及方法、程序 |
JP6561980B2 (ja) * | 2014-03-24 | 2019-08-21 | 日本電気株式会社 | 監視装置、監視システム、監視方法及びプログラム |
WO2015151789A1 (ja) * | 2014-03-31 | 2015-10-08 | 日本電気株式会社 | 教師データ生成装置、電気機器監視システム、教師データ生成方法及びプログラム |
-
2016
- 2016-03-23 JP JP2016058004A patent/JP6686593B2/ja active Active
-
2017
- 2017-01-20 CN CN201780019327.5A patent/CN108885448A/zh active Pending
- 2017-01-20 EP EP17769616.8A patent/EP3435185A4/en not_active Withdrawn
- 2017-01-20 WO PCT/JP2017/001877 patent/WO2017163569A1/ja active Application Filing
- 2017-01-20 US US16/085,665 patent/US10831184B2/en active Active
- 2017-01-20 CA CA3018660A patent/CA3018660A1/en not_active Abandoned
- 2017-02-02 TW TW106103438A patent/TW201738677A/zh unknown
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010191556A (ja) * | 2009-02-17 | 2010-09-02 | Hitachi Ltd | 異常検知方法及び異常検知システム |
JP2014167667A (ja) * | 2013-02-28 | 2014-09-11 | Mitsubishi Electric Corp | 設備点検順位設定装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3435185A4 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019107360A1 (ja) * | 2017-11-28 | 2019-06-06 | 株式会社安川電機 | 制御システム、工場システム、学習システム、推定用モデルの生成方法及びアクチュエータの状態推定方法 |
CN111433691A (zh) * | 2017-11-28 | 2020-07-17 | 株式会社安川电机 | 控制***、工厂***、学习***、估计用模型的生成方法和致动器的状态估计方法 |
JPWO2019107360A1 (ja) * | 2017-11-28 | 2020-12-03 | 株式会社安川電機 | 制御システム、工場システム、学習システム、推定用モデルの生成方法及びアクチュエータの状態推定方法 |
CN111433691B (zh) * | 2017-11-28 | 2024-03-08 | 株式会社安川电机 | 控制***、工厂***、学习***、估计用模型的生成方法和致动器的状态估计方法 |
US11947322B2 (en) | 2017-11-28 | 2024-04-02 | Kabushiki Kaisha Yaskawa Denki | Factory system for machine learning of an actuator |
Also Published As
Publication number | Publication date |
---|---|
EP3435185A4 (en) | 2019-05-01 |
CN108885448A (zh) | 2018-11-23 |
EP3435185A1 (en) | 2019-01-30 |
JP2017174045A (ja) | 2017-09-28 |
CA3018660A1 (en) | 2017-09-28 |
US20190087186A1 (en) | 2019-03-21 |
US10831184B2 (en) | 2020-11-10 |
JP6686593B2 (ja) | 2020-04-22 |
TW201738677A (zh) | 2017-11-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017163569A1 (ja) | データ処理装置、データ処理システム、データ処理方法及びプログラム | |
US9625518B2 (en) | Multi-node electrical power monitoring, analysis, and related services | |
US6996508B1 (en) | System and method for remote retrofit identification of energy consumption systems and components | |
US11002773B2 (en) | Monitoring apparatus, monitoring method, and storage medium | |
JP6583265B2 (ja) | 監視装置、監視システム、監視方法及びプログラム | |
CN102708419A (zh) | 数据中心效率分析及优化的方法和*** | |
JP6958599B2 (ja) | 監視装置、監視システム、監視方法及びプログラム | |
US9857414B1 (en) | Monitoring and fault detection of electrical appliances for ambient intelligence | |
US20110288793A1 (en) | Event identification | |
JP2003316922A (ja) | エネルギー情報分析方法、エネルギー情報分析装置、エネルギー情報分析システム | |
CN107078543A (zh) | 用于远程电负载管理的方法和装置 | |
JP2010218077A (ja) | 省エネ支援装置、省エネ支援システム | |
JP2019212131A (ja) | 予測装置、電気機器、管理システム、予測方法、及び制御プログラム | |
US9785902B1 (en) | Computer-implemented engineering review of energy consumption by equipment | |
CN112101665A (zh) | 故障检测预警方法、装置、存储介质及电子设备 | |
CN103616863A (zh) | 家用空调和冰箱的风险预估方法 | |
US11125790B2 (en) | Method for operating a power consumption metering system and power consumption metering system | |
CN110736221A (zh) | 空调的控制方法、装置和*** | |
US11508020B2 (en) | Method for operating a power consumption metering system and power consumption metering system | |
JP2006178886A (ja) | 店舗管理システム | |
CN114811891B (zh) | 用于空调新风管防凝露的方法、装置及空调 | |
JP2012048409A (ja) | 電力管理システムおよび管理サーバ | |
WO2015194248A1 (ja) | 情報提供装置、情報提供方法、及び、プログラム | |
US20240242136A1 (en) | Data Substitution System and Data Substitution Method | |
JP6656093B2 (ja) | 情報提供システム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 3018660 Country of ref document: CA |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2017769616 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2017769616 Country of ref document: EP Effective date: 20181023 |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17769616 Country of ref document: EP Kind code of ref document: A1 |