WO2018016098A1 - Dispositif de contrôle d'installation et procédé de contrôle d'installation - Google Patents

Dispositif de contrôle d'installation et procédé de contrôle d'installation Download PDF

Info

Publication number
WO2018016098A1
WO2018016098A1 PCT/JP2016/084959 JP2016084959W WO2018016098A1 WO 2018016098 A1 WO2018016098 A1 WO 2018016098A1 JP 2016084959 W JP2016084959 W JP 2016084959W WO 2018016098 A1 WO2018016098 A1 WO 2018016098A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
period
alarm
detection data
learning
Prior art date
Application number
PCT/JP2016/084959
Other languages
English (en)
Japanese (ja)
Inventor
玄太 吉村
北上　眞二
米山　純一
剛維 木村
Original Assignee
三菱電機株式会社
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 三菱電機株式会社 filed Critical 三菱電機株式会社
Publication of WO2018016098A1 publication Critical patent/WO2018016098A1/fr

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

Definitions

  • the present invention relates to a facility monitoring apparatus and a facility monitoring method, and more particularly, to a facility monitoring apparatus and a facility monitoring method for monitoring the state of a facility based on information collected from the facility.
  • a learning model and equipment that accumulates the detection data of sensors installed in equipment such as power plants, extracts the detection data when the equipment is normal from the accumulated detection data, and models the detection data when the equipment is normal Compared with the detection data of the sensor detected at the time of operation, if these differences exceed a predetermined threshold, the equipment is not in a normal state, that is, there is a possibility that the equipment is abnormal
  • An invariant analysis technique (SIAT) for determination is known (see Non-Patent Document 1).
  • the detection data when the equipment is normal is collected from the accumulated sensor detection data. For this reason, it is important to improve the accuracy of collecting detection data when the equipment is normal. In other words, it is important to exclude the detection data at the time of equipment abnormality from the accumulated data with high accuracy.
  • Patent Documents 1 and 2 disclose that the warning is restored from the time when a warning due to an abnormality, failure, or the like regarding the accumulated data occurs. It is described that the detection data of the sensor in the period until it is removed.
  • Patent Document 3 accurately describes from the first warning to the recovery of the last warning even if the occurrence of a plurality of warnings related to facilities and the recovery of those warnings are mixed in a time-series order. It is described that the detection data of the sensor during this period is not adopted.
  • Kei Fukushima 5 others, “Plant Failure Prediction Monitoring System Using Invariant Analysis Technology (SIAT)”, NEC Technique, NEC Corporation, November 2014, Vol. 67 No. 1. Special feature on public solutions that support social safety and security, p. 119-122
  • Equipment alarm occurrence (abnormality occurrence) and alarm recovery (abnormality recovery) and sensor detection data are closely related.
  • alarm generation and alarm recovery and sensor detection data changes instantaneously in response to alarm generation and alarm recovery almost simultaneously with facility alarm generation and alarm recovery. In this case, it is possible to determine the abnormality of the facility and the recovery from the abnormality by the change of the detection data of the sensor.
  • the sensor detection data detects an alarm sign (an abnormality sign) from before the equipment alarm occurs, that is, When a slight abnormality occurs in the facility, and the abnormality progresses and a facility alarm is generated, the detection data is gradually changed from before the alarm is generated.
  • the sensor detection data may not be stable after the facility alarm is restored, and it may take a predetermined period of time to return to normal detection data.
  • the detection data of the sensor at the time of equipment abnormality is removed from the accumulated detection data of the sensor by the technique described in Patent Literature 1-3. be able to.
  • the detection data indicating the abnormality or the detection data that is not stable is included in the data for generating the learning model. As a result, the accuracy of the learning model is lowered, and as a result, the accuracy of equipment abnormality determination is also lowered.
  • an object of the present invention is to improve the accuracy of a learning model by excluding abnormality detection data from learning data for generating a learning model or adding normal detection data to learning data.
  • the facility monitoring device of the present invention is a facility monitoring device that monitors the state of the facility based on detection data of a sensor provided in the facility, the detection data, information on occurrence of an alarm due to an abnormality in the facility, and restoration of the alarm
  • a data storage unit for storing information in time series, a first data period from alarm occurrence to alarm recovery in the detection data by associating the detection data with the occurrence information and the recovery information in time series; And a second data period from alarm recovery to alarm occurrence in the detected data, respectively, and a predetermined period of time from the boundary between the first data period and the second data period to the second data period.
  • a learning model generation unit that generates a learning model indicating the detection data of the sensor when in operation, the learning model, and the detection data of the sensor detected during operation of the facility And an equipment state determination unit for determining the state.
  • the learning data generation unit may exclude the third data period from the second data period when the detection data in the third data period is abnormal data.
  • the learning data generation unit adds the fourth data period to the second data period when the detection data in the fourth data period is normal data.
  • the facility state determination unit compares the generated learning model with the detection data of the sensor detected during operation of the facility, and a difference between the learning model and the detection data is predetermined.
  • a singular point extraction unit is provided that extracts a singular point to be used for determining the equipment abnormality when a threshold value is exceeded.
  • the facility monitoring method of the present invention is a facility monitoring method for monitoring the state of the facility based on detection data of a sensor provided in the facility, wherein the detection data, alarm occurrence information due to facility abnormality, and the alarm
  • a data period and a fourth data period of a predetermined period from the boundary into the first data period are set, and the third data period is excluded from the second data period.
  • the learning data is generated, or the fourth data period is added to the second data period to generate the learning data, and the sensor when the equipment is normally operated based on the learning data
  • a learning model indicating the detected data is generated, and the learning state is compared with the detection data of the sensor detected during operation of the equipment, thereby determining the equipment state.
  • the accuracy of the learning model can be improved by excluding the abnormality detection data from the learning data for generating the learning model or adding the normal detection data to the learning data.
  • the determination accuracy can be improved.
  • FIG. 1 It is a block block diagram of the system containing an equipment monitoring apparatus. It is a hardware block diagram of the computer of an equipment monitoring apparatus. It is a flowchart which shows the process which sets the learning model in Embodiment 1 of this invention. It is a characteristic view which shows the relationship of history data, trend data, a learning model, and a singular point extraction in Embodiment 1 of this invention, (A) is a characteristic view which shows the relationship between history data, trend data, and a learning model, ( B) shows a characteristic diagram of the generated learning model, and (C) shows a characteristic diagram for extracting a singular point.
  • FIG. 1 It is a block block diagram of the system containing an equipment monitoring apparatus. It is a hardware block diagram of the computer of an equipment monitoring apparatus. It is a flowchart which shows the process which sets the learning model in Embodiment 1 of this invention. It is a characteristic view which shows the relationship of history data, trend data, a learning model, and a singular point extraction in Embodiment 1
  • FIG. 2 is a schematic diagram of each margin when calculating an evaluation value of a learning model in Embodiment 1 of the present invention
  • A is a schematic diagram in a state where the margin is not adjusted
  • B is an initial margin
  • C shows a schematic diagram in a state where the next margin is reduced
  • D shows a schematic diagram in a state where the next margin is further reduced.
  • It is a flowchart which shows the singular point extraction process using the learning model in Embodiment 1 of this invention.
  • It is a characteristic view of the history data, trend data, and learning model which show the modification of Embodiment 1 of this invention.
  • It is a characteristic view of history data, trend data, and a learning model in Embodiment 2 of the present invention.
  • It is a characteristic view of the history data, trend data, and learning model which show the modification of Embodiment 2 of this invention.
  • FIG. 1 is a block configuration diagram showing a schematic configuration of a facility monitoring system 1 including a facility monitoring device 10 according to the first embodiment.
  • FIG. 4 is a characteristic diagram showing the relationship between history data, trend data, a learning model, and singularity extraction in the first embodiment.
  • the facility monitoring system 1 includes a building 2 and a facility monitoring apparatus 10 that is connected to the building 2 via a network and remotely monitors the building 2.
  • the building 2 is provided with an air conditioner 3 as equipment.
  • the air conditioner 3 is provided with an air supply sensor SA that detects the temperature of the air that is sent from the air conditioner 3 into the room, and a return air sensor RA that detects the temperature of the air that is drawn into the air conditioner 3 from the room.
  • SA air supply sensor
  • RA return air sensor
  • the facility monitoring apparatus 10 includes a data storage unit 11 that stores various data related to the building 2.
  • the data accumulating unit 11 supplies the air supply sensor SA, the return air sensor RA, and the ratio of these sensors SA and RA, that is, the detection data of the air supply sensor SA / return air sensor RA (hereinafter, these data are referred to as trend data TD).
  • Trend data storage unit 11A that accumulates in time series, generation information that indicates that an alarm has occurred due to an abnormality in air conditioner 3, and recovery information that indicates that the alarm has been recovered (hereinafter referred to as history data HD) in time series
  • history data HD recovery information that indicates that the alarm has been recovered
  • a history data storage unit 11B for storing.
  • the facility monitoring apparatus 10 uses the trend data TD and the history data HD to generate learning data SD11, SD12, SD13, and so on, and the learning data SD11, SD12, SD13, and so on.
  • a learning model generation unit 13 that generates a learning model SM and an equipment state determination unit 14 that determines the state of the air conditioner 3 using the learning model SM.
  • the learning data generation unit 12 and the generation of the learning model SM by the learning model generation unit 13 will be described later.
  • the equipment state determination unit 14 compares the detection data of the air supply sensor SA / return air sensor RA during operation of the air conditioner 3 and the learning model SM, and determines when the air conditioner 3 has operated differently from the normal state.
  • a singular point extraction unit 14A that extracts the singular point P (see FIG. 4) is provided.
  • the singular point extraction unit 14A has a threshold value L (see FIG. 4) for determining that the air conditioner 3 has performed an operation different from the normal state.
  • the singular point P indicates when the air conditioner 3 is abnormal or when there is a possibility that the air conditioner 3 may be abnormal.
  • the singular point P indicates that the air conditioner 3 is operating differently from the normal state, and the extracted information of the singular point P is used as maintenance inspection information of the air conditioner 3.
  • the building 2 is provided with a number of sensors in addition to the supply air sensor SA and the return air sensor RA of the air conditioner 3, but in the first embodiment, as an example, the supply air sensor SA / return air sensor.
  • the detection data of RA will be described.
  • FIG. 2 is a hardware configuration diagram of a computer forming the equipment monitoring apparatus 10.
  • the computer forming the facility monitoring apparatus 10 can be realized by a hardware configuration of a general-purpose personal computer (PC) that has existed in the past. That is, the computer is provided as a CPU 21, ROM 22, RAM 23, HDD controller 25 connected with a hard disk drive (HDD) 24, a mouse 26 and a keyboard 27 provided as input means, and a display device as shown in FIG.
  • An input / output controller 29 for connecting each display 28 and a network controller 30 provided as a communication means are connected to an internal bus 31.
  • the program for executing the equipment monitoring method by the equipment monitoring apparatus 10 can be provided not only by communication means but also by storing in a computer-readable recording medium such as a CD-ROM or DVD-ROM.
  • the program provided from the communication means or the recording medium is installed in the computer, and various processes are realized by the CPU of the computer sequentially executing the program.
  • step S101 of the flowchart shown in FIG. 3 a first data period from alarm generation to alarm recovery (hereinafter referred to as alarm generation period TA1) based on the alarm generation information and recovery information stored in the history data storage unit 11B. , TA2, TA3, etc And second data periods (hereinafter referred to as non-alarm periods TB1, TB2, TB3%) From alarm recovery to alarm generation are determined, and the process proceeds to step S102.
  • an alarm is generated or alarmed at the boundary between the first data period (alarm generation periods TA1, TA2, TA3...) And the second data period (non-alarm periods TB1, TB2, TB3). There will be a recovery.
  • step S102 first, based on the trend data TD, the characteristics of the detection data of the air supply sensor SA / return air sensor RA are determined.
  • the characteristics of the detection data of the air supply sensor SA / return air sensor RA are determined in advance by the sensor type and the like.
  • the characteristic of the detection data of the air supply sensor SA / return air sensor RA in the first embodiment is that the trend data TD of the air supply sensor SA / return air sensor RA before the alarm occurrence of the air conditioner 3 is an alarm sign (an abnormality sign). ).
  • the trend data TD gradually changes from before the alarm occurs (from the time of occurrence of the abnormality). . That is, the trend data TD includes data indicating an abnormality sign before the alarm is generated. Since the characteristic of such detection data can be grasped in advance, the following control is performed based on this characteristic.
  • the trend data TD includes data indicating an abnormality sign before the occurrence of an alarm.
  • step S102 as shown in FIG. 4A, the boundary between the alarm generation periods TA1, TA2, TA3... And the non-alarm periods TB1, TB2, TB3.
  • Margins ⁇ 11, ⁇ 12, ⁇ 13... As third data periods are set in the period.
  • the margins ⁇ 11, ⁇ 12, ⁇ 13,... Correspond to a period including detection data indicating an abnormality sign before the occurrence of an alarm.
  • the margins ⁇ 11, ⁇ 12, ⁇ 13... Are excluded from the non-alarm periods TB1, TB2, TB3...
  • the learning data SD11, SD12, SD13 are obtained by excluding data indicating signs of abnormality before the alarm is generated.
  • step S103 a learning model SM is generated based on the learning data SD11, SD12, SD13... Generated in step S102. Since the generation method of the learning model SM based on the learning data SD11, SD12, SD13,... Is known, detailed description thereof is omitted. By modeling the learning data SD11, SD12, SD13..., A model for predicting the detection data of the supply air sensor SA / return air sensor RA detected when the air conditioner 3 is in a normal state is generated. Then, the process proceeds to step S104.
  • step S104 the evaluation value of the generated learning model SM is calculated, and the process proceeds to step S105.
  • step S105 the margins ⁇ 11, ⁇ 12, ⁇ 13,. That is, if the periods of the margins ⁇ 11, ⁇ 12, ⁇ 13,... Are lengthened, detection data indicating an abnormality sign before the occurrence of an alarm can be reliably excluded, but normal detection data is also excluded. On the other hand, if the periods of the margins ⁇ 11, ⁇ 12, ⁇ 13,... Are shortened, it becomes impossible to exclude detection data indicating an abnormality sign before the occurrence of an alarm, and detection data indicating an abnormality sign is included. there is a possibility. Therefore, the margins ⁇ 11, ⁇ 12, ⁇ 13...
  • the margins ⁇ 11, ⁇ 12, ⁇ 13,... are optimized.
  • An optimization algorithm called metaheuristics is used.
  • the margins ⁇ 11, ⁇ 12, ⁇ 13,... are optimized using the optimization algorithm of the hill-climbing method of the neighborhood search method.
  • the evaluation data is detection data of the supply air sensor SA / return air sensor RA that the operator of the equipment monitoring apparatus 10 determines to be normal data, and serves as an index of the evaluation value.
  • the evaluation value indicates a scale that the learning model SM approximates to the evaluation data. The higher the evaluation value, the closer the learning model SM and the evaluation data are. That is, the learning model SM having a high evaluation value is approximated to normal detection data.
  • step S105 it is determined whether the learning model SM is optimal. As a method for determining the optimum learning model SM, in the first embodiment, steps S102 to S104 are repeated according to the number of margins, and the learning model SM that provides the highest evaluation value is selected.
  • steps S102 to S104 are repeated, and then only margin ⁇ 11 is set to a smaller value as shown in FIG. 5B.
  • learning data SD11, SD12, SD13,... are generated to generate a learning model SM.
  • the learning model SM and the evaluation data are compared to calculate an evaluation value of the learning model SM.
  • step S102 to step S104 are repeated, and as shown in FIG. 5C, only the margin ⁇ 12 is set to a smaller value, and learning data SD11, SD12, SD13,. Generate.
  • the learning model SM and the evaluation data are compared to calculate an evaluation value of the learning model SM.
  • FIG. 5D only the margin ⁇ 13 is set to a smaller value, a learning model SM is created, and an evaluation value is calculated.
  • Step S102 to Step S104 are repeated according to the number of margins, and as a result, the learning model SM that provides the highest evaluation value among the obtained evaluation values is set as the optimum learning model SM. That is, the learning model SM closest to the evaluation data is set.
  • the evaluation value becomes the evaluation value 80, indicating that this learning model SM is the learning model SM closest to the evaluation data compared to other learning models SM.
  • step S106 the learning model SM when the margin ⁇ 12 is set to a small value, here, the learning model SM with the evaluation value 80 is set as the learning model SM used during the operation of the air conditioner 3, and learning is performed.
  • the setting of the model SM is finished.
  • the learning model SM thus set is shown in FIG.
  • the target evaluation value of the learning model SM may be set in advance, and steps S102 to S105 may be repeated until the calculated evaluation value approaches the target evaluation value.
  • the optimization algorithm includes neighborhood search methods such as annealing method and tabu search, genetic algorithm, evolution strategy, evolutionary programming, genetic programming, and ant colony optimization. Evolutionary computation methods such as particle swarm optimization can also be applied. Also, algorithms such as simulated, evolution, artificial immune system, neural network, etc. can be applied.
  • step S110 of the flowchart shown in FIG. 6 the difference between the learning model SM set in step S106 of the flowchart of FIG. 3 and the detection data of the supply air sensor SA / return air sensor RA detected during operation of the air conditioner 3 is calculated. Calculate and go to step S111. A characteristic TM of the difference between the two is shown in FIG.
  • step S111 it is determined whether or not the difference between the two exceeds a predetermined threshold value L.
  • the threshold value L is used to determine that the learning model SM and the detection data of the supply air sensor SA / return air sensor RA have greatly deviated. Further, in the characteristic TM, a point where the difference between the two exceeds the threshold value L is extracted as a singular point P.
  • the singular point P is a point where the difference between the learning model SM and the detection data of the supply air sensor SA / return air sensor RA exceeds the threshold L, and is not necessarily directly related to the occurrence of an abnormality in the air conditioner 3. Absent. That is, the singular point P may be extracted even if the alarm of the air conditioner 3 does not occur. For this reason, when the singular point P is extracted, this is notified to the operator of the equipment monitoring apparatus 10.
  • step S111 when the singular point P is not extracted (No), the process returns to step S110, the detection data of the supply air sensor SA / return air sensor RA is acquired again, and steps S110 and 111 are repeated.
  • the process proceeds to step S112, and the information regarding the extracted singular point P is used as the maintenance check information of the air conditioner 3. Thereafter, the detection data of the air supply sensor SA / return air sensor RA is acquired again, and steps S110 and S111 are repeated.
  • the detection data SD11, SD12, SD13,... are generated excluding the detection data indicating the sign of abnormality of the air conditioner 3 before the alarm is generated.
  • a learning model SM can be generated based on normal detection data. As a result, the accuracy of the learning model SM can be improved, and the accuracy of the state determination of the air conditioner 3 using the learning model SM can also be improved.
  • the detection data of the supply air sensor SA / return air sensor RA has a characteristic that requires a predetermined period to return to the normal detection state after the alarm is restored, that is, the detection data of the air supply sensor SA / return air sensor RA.
  • margins ⁇ 21, ⁇ 22, ⁇ 23... are set at boundaries between the alarm generation periods TA1, TA2, TA3... And the non-alarm periods TB1, TB2, TB3.
  • the margins ⁇ 21, ⁇ 22, ⁇ 23,... Correspond to a period including detection data in an unstable state after alarm recovery.
  • the margins ⁇ 21, ⁇ 22, ⁇ 23... Are excluded from the non-alarm periods TB1, TB2, TB3...
  • learning model SM is produced
  • the optimization of the margins ⁇ 21, ⁇ 22, ⁇ 23... And the setting of the optimal learning model SM are the same as in the first embodiment.
  • the accuracy of the learning model SM can be improved by excluding the detection data in an unstable state after the alarm recovery.
  • the detection data of both the detection data indicating the abnormality sign before the occurrence of the alarm and the detection data in an unstable state after the alarm recovery described above may be excluded, or one of the detection data May be excluded.
  • Embodiment 2 will be described with reference to FIG.
  • the detection data of the supply air sensor SA / return air sensor RA is the data after the alarm is generated.
  • the detection data of the supply air sensor SA / return air sensor RA is the data after the alarm is generated.
  • margins ⁇ 31, ⁇ 32, ⁇ 33,... At predetermined periods after the boundaries between the alarm generation periods TA1, TA2, TA3... And the non-alarm periods TB1, TB2, TB3.
  • the margins ⁇ 31, ⁇ 32, ⁇ 33,... Correspond to a period of detection data that has hardly changed from the detection data before the occurrence of the alarm. In other words, it can be considered that the detection data is substantially the same as the normal detection data before the alarm is generated.
  • the learning model SM having the highest evaluation value is set as the learning model SM used when the air conditioner 3 is operated, and the air supply sensor SA / detected when the air conditioner 3 is operated using the learning model SM thus set.
  • the state of the detection data of the return air sensor RA is determined.
  • the detection data state is determined by extracting the singular point P.
  • the detection data SD31, SD32, SD33,... are added to the learning data SD31, SD32, SD33. ⁇ ⁇ Data amount can be increased.
  • the data amount of the learning data SD31, SD32, SD33... Increases an accurate learning model SM can be generated, and as a result, the accuracy of the state determination of the air conditioner 3 can be improved.
  • the amount of learning data is small, the amount of learning data can be increased, which is effective.
  • the detection data of the supply air sensor SA / return air sensor RA has detected normal detection data before the alarm is restored, for example, the detection data is in a substantially normal state at the same time as the equipment abnormality is resolved. After that, when the alarm is recovered, margins ⁇ 41, ⁇ 42, ⁇ 43,... As the fourth data period are set before the alarm recovery.
  • the learning data SD41, SD42,... Is acquired by taking in the learning data SD41, SD42, SD43,. , SD43... Can be increased, and an accurate learning model SM can be obtained.
  • both detection data after the alarm is generated and before the alarm is recovered may be acquired, or any one of the detection data may be acquired.
  • the equipment monitoring apparatus can improve the accuracy of the learning model by excluding the abnormality detection data from the learning data for generating the learning model, or by adding normal detection data to the learning data. Moreover, the determination accuracy of equipment abnormality can be improved, and it is suitable for use in equipment monitoring devices that monitor the state of equipment based on information collected from equipment.
  • 1 equipment monitoring system 2 buildings, 3 air conditioners, 10 equipment monitoring equipment, 11 data storage section, 11A trend data storage section, 11B history data storage section, 12 learning data generation section, 13 learning model generation section, 14 equipment status determination Part, 14A singularity extraction part, RA return air sensor, SA air supply sensor, SD11, SD12, SD13, SD21, SD22, SD23, SD31, SD32, SD33, SD41, SD42, SD43 learning data, SM learning model, TA alarm Occurrence period, TB non-alarm period, ⁇ 11, ⁇ 12, ⁇ 13, ⁇ 21, ⁇ 22, ⁇ 23, ⁇ 31, ⁇ 32, ⁇ 33, ⁇ 41, ⁇ 42, ⁇ 43 margin.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

Selon l'invention, une période de génération d'alerte (TA1, TA2, TA3…) et une période de non-alerte (TB1, TB2, TB3…) dans des données de tendances (TD) sont respectivement déterminées, une marge (σ11, σ12, σ13…) dirigée dans la période de non-alerte (TB) depuis une limite entre la période de génération d'alerte (TA) et la période de non-alerte (TB) est définie, des données d'apprentissage (SD11, SD12, SD13…) sont générées en excluant la marge (σ11, σ12, σ13…) de la période de non-alerte (TB) et un modèle d'apprentissage (SM) est généré sur la base des données d'apprentissage (SD11, SD12, SD13…).
PCT/JP2016/084959 2016-07-20 2016-11-25 Dispositif de contrôle d'installation et procédé de contrôle d'installation WO2018016098A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2016142424A JP6632941B2 (ja) 2016-07-20 2016-07-20 設備監視装置及び設備監視方法
JP2016-142424 2016-07-20

Publications (1)

Publication Number Publication Date
WO2018016098A1 true WO2018016098A1 (fr) 2018-01-25

Family

ID=60993278

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/084959 WO2018016098A1 (fr) 2016-07-20 2016-11-25 Dispositif de contrôle d'installation et procédé de contrôle d'installation

Country Status (2)

Country Link
JP (1) JP6632941B2 (fr)
WO (1) WO2018016098A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2020027342A (ja) * 2018-08-09 2020-02-20 富士電機株式会社 情報処理装置、監視装置、及び情報処理方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7038629B2 (ja) * 2018-08-31 2022-03-18 三菱電機ビルテクノサービス株式会社 機器状態監視装置及びプログラム
JP7517810B2 (ja) * 2019-11-19 2024-07-17 旭化成株式会社 診断装置、診断方法及び診断プログラム
WO2021193396A1 (fr) * 2020-03-27 2021-09-30 住友重機械工業株式会社 Dispositif et système de gestion de pelle, dispositif de support de pelle et pelle
JP7016195B1 (ja) 2021-09-06 2022-02-18 Wota株式会社 プログラム、方法、情報処理装置、システム

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012073289A1 (fr) * 2010-12-02 2012-06-07 株式会社日立製作所 Dispositif de diagnostic d'installation et procédé de diagnostic d'installation
JP2015197850A (ja) * 2014-04-02 2015-11-09 三菱電機ビルテクノサービス株式会社 設備監視装置及びプログラム

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012073289A1 (fr) * 2010-12-02 2012-06-07 株式会社日立製作所 Dispositif de diagnostic d'installation et procédé de diagnostic d'installation
JP2015197850A (ja) * 2014-04-02 2015-11-09 三菱電機ビルテクノサービス株式会社 設備監視装置及びプログラム

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2020027342A (ja) * 2018-08-09 2020-02-20 富士電機株式会社 情報処理装置、監視装置、及び情報処理方法

Also Published As

Publication number Publication date
JP2018013914A (ja) 2018-01-25
JP6632941B2 (ja) 2020-01-22

Similar Documents

Publication Publication Date Title
WO2018016098A1 (fr) Dispositif de contrôle d'installation et procédé de contrôle d'installation
JP7069269B2 (ja) デジタル・ツイン・シミュレーション・データを利用した時系列データに基づく、大規模な産業用監視システム向けの半教師あり深層異常検出のための方法およびシステム
CN107636619B (zh) 信息处理装置、信息处理系统、信息处理方法及记录介质
JP6896432B2 (ja) 故障予知方法、故障予知装置および故障予知プログラム
EP2905665B1 (fr) Appareil de traitement d'informations, procédé de diagnostic et programme
JP6141235B2 (ja) 時系列データにおける異常を検出する方法
WO2013011745A1 (fr) Procédé de surveillance d'état d'installation et dispositif pour celui-ci
JP6354755B2 (ja) システム分析装置、システム分析方法、及びシステム分析プログラム
JP5364530B2 (ja) 設備状態監視方法、監視システム及び監視プログラム
EP3847586A1 (fr) Procédé mis en oeuvre par ordinateur, produit-programme d'ordinateur et système de détection d'anomalie et/ou de maintenance prédictive
CN107480440A (zh) 一种基于两阶段随机退化建模的剩余寿命预测方法
WO2019133316A1 (fr) Détection d'anomalies basée sur la reconstruction
US11494252B2 (en) System and method for detecting anomalies in cyber-physical system with determined characteristics
US10047757B2 (en) Predicting a surge event in a compressor of a turbomachine
US20140188777A1 (en) Methods and systems for identifying a precursor to a failure of a component in a physical system
Harrou et al. Improved detection of incipient anomalies via multivariate memory monitoring charts: Application to an air flow heating system
EP3674946B1 (fr) Système et procédé de détection des anomalies dans un système cyber-physique ayant des caractéristiques déterminées
KR101960755B1 (ko) 미취득 전력 데이터 생성 방법 및 장치
KR101997580B1 (ko) 상관도를 고려한 데이터 분류 방법 및 이 방법을 수행하기 위한 프로그램이 저장된 컴퓨터 판독가능한 저장매체
JP6875199B2 (ja) 機器診断システム
JP2016177676A (ja) 診断装置、診断方法、診断システムおよび診断プログラム
JP6896380B2 (ja) 故障予兆判定方法、故障予兆判定装置および故障予兆判定プログラム
JP5948998B2 (ja) 異常診断装置
EP3324259A2 (fr) Appareil et procédé de récupération de signaux de défaillance
JP6710913B2 (ja) 情報提供装置、情報提供方法、及びプログラム

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: 16909573

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16909573

Country of ref document: EP

Kind code of ref document: A1