WO2022215232A1 - 異常診断装置および異常診断方法 - Google Patents
異常診断装置および異常診断方法 Download PDFInfo
- Publication number
- WO2022215232A1 WO2022215232A1 PCT/JP2021/014928 JP2021014928W WO2022215232A1 WO 2022215232 A1 WO2022215232 A1 WO 2022215232A1 JP 2021014928 W JP2021014928 W JP 2021014928W WO 2022215232 A1 WO2022215232 A1 WO 2022215232A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- feature
- abnormality
- abnormality diagnosis
- diagnosis
- unit
- Prior art date
Links
- 230000005856 abnormality Effects 0.000 title claims abstract description 139
- 238000003745 diagnosis Methods 0.000 title claims abstract description 133
- 238000000034 method Methods 0.000 title claims description 24
- 238000009826 distribution Methods 0.000 claims abstract description 75
- 238000004364 calculation method Methods 0.000 claims abstract description 31
- 239000013598 vector Substances 0.000 claims description 35
- 239000011159 matrix material Substances 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 238000005259 measurement Methods 0.000 description 23
- 239000003507 refrigerant Substances 0.000 description 15
- 238000001514 detection method Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 11
- 238000000513 principal component analysis Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 239000007788 liquid Substances 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000005494 condensation Effects 0.000 description 1
- 238000009833 condensation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000007274 generation of a signal involved in cell-cell signaling Effects 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
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
- 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
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B49/00—Arrangement or mounting of control or safety devices
- F25B49/005—Arrangement or mounting of control or safety devices of safety devices
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P29/00—Arrangements for regulating or controlling electric motors, appropriate for both AC and DC motors
- H02P29/02—Providing protection against overload without automatic interruption of supply
- H02P29/024—Detecting a fault condition, e.g. short circuit, locked rotor, open circuit or loss of load
-
- 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
- G05B2219/20—Pc systems
- G05B2219/26—Pc applications
- G05B2219/2638—Airconditioning
Definitions
- This application relates to an abnormality diagnosis device and an abnormality diagnosis method.
- the present application discloses a technology for solving the above problems, and aims to obtain an abnormality diagnosis device and an abnormality diagnosis method that can appropriately diagnose the presence or absence of an abnormality in a diagnosis target during driving.
- An abnormality diagnosis device disclosed in the present application is an abnormality diagnosis device that performs abnormality diagnosis of a diagnosis target based on a first state quantity and a second state quantity respectively indicating states of the diagnosis target during operation, wherein the first either one or both of the time-series data and frequency-series data of the state quantity and the data of the second state quantity, and either the time-series data or the frequency-series data of the first state quantity
- a feature amount calculation unit that calculates a plurality of feature amounts from one or both of them, a driving mode determination unit that determines a driving mode to be diagnosed based on the second state amount, and a plurality of feature amount values as components
- a feature quantity distribution generating unit that generates a feature quantity distribution in the feature space based on the feature vector, and a reference distribution or reference region that is generated in the feature space based on the feature quantity distribution obtained from the diagnosis subject in the normal state.
- a reference generation unit a storage unit that stores a reference corresponding to an operation mode when the reference is generated, a feature amount distribution obtained during abnormality diagnosis, and a reference corresponding to the operation mode at the time of abnormality diagnosis are compared. and a judgment unit for judging the presence or absence of an abnormality to be diagnosed.
- the abnormality diagnosis method disclosed in the present application is an abnormality diagnosis method for diagnosing an abnormality of a diagnosis target based on a first state quantity and a second state quantity respectively indicating the state of the diagnosis target during driving, acquiring either one or both of the time-series data and the frequency-series data of the first state quantity and the data of the second state quantity; a step of calculating a plurality of feature quantities from one or both of them; a step of determining a driving mode to be diagnosed based on the second state quantity; and a feature vector having values of the plurality of feature quantities as components , a step of generating a feature quantity distribution in the feature space; a step of generating a reference distribution or a reference region in the feature space based on the feature quantity distribution obtained from a diagnostic subject in a normal state; A step of storing a reference corresponding to an operation mode, and comparing a feature value distribution obtained at the time of abnormality diagnosis with the reference corresponding to the operation mode at the time of abnormality diagnosis to determine whether or not there
- the abnormality diagnosis device and the abnormality diagnosis method disclosed in the present application it is possible to appropriately diagnose the presence or absence of abnormality in the diagnosis target during driving.
- FIG. 1 is a diagram showing an outline of an air conditioner according to Embodiment 1;
- FIG. 1 is a schematic configuration diagram showing an abnormality diagnosis device according to Embodiment 1;
- FIG. 1 is a block diagram showing an abnormality diagnosis device according to Embodiment 1;
- FIG. 2 is a diagram illustrating a Clark transform unit and a Park transform unit according to Embodiment 1;
- FIG. FIG. 4 is a flowchart showing the operation of an abnormality diagnosis unit according to Embodiment 1;
- FIG. 4 is a flowchart showing reference region generation according to the first embodiment;
- FIG. 4 is a flowchart showing abnormality diagnosis according to Embodiment 1; It is a figure which shows the principal component analysis result of a feature-value.
- FIG. 4 is a diagram showing reference regions for different operating modes;
- FIG. 4 is a diagram showing an example of operation modes classified by drive frequency and load torque;
- 1 is a diagram showing an example of a hardware configuration of an abnormality diagnosis
- Embodiment 1 will be described with reference to FIGS. 1 to 11.
- FIG. 1 is a diagram showing an overview of an air conditioner according to Embodiment 1
- FIG. 2 is a schematic configuration diagram showing an abnormality diagnosis device according to Embodiment 1
- FIG. 3 is an abnormality diagnosis device according to Embodiment 1. 2 is a block diagram showing .
- the air conditioner 1000 includes a compressor 71, a condenser 72, an expansion valve 73, and an evaporator 74, and refrigerant circulates through the compressor 71, the condenser 72, the expansion valve 73, and the evaporator 74 in this order. and operates as a refrigeration cycle device.
- the compressor 71 compresses gaseous refrigerant (gas refrigerant) and discharges it to the condenser 72 .
- the condenser 72 condenses the gas refrigerant discharged from the compressor 71 to generate liquid refrigerant (liquid refrigerant). At this time, the refrigerant passing through the condenser 72 releases condensation heat to the surrounding air.
- the condenser 72 discharges the generated liquid refrigerant to the expansion valve 73 .
- the expansion valve 73 decompresses and expands the liquid refrigerant discharged from the condenser 72 by controlling the degree of opening thereof by a control device (not shown).
- the expanded liquid refrigerant is sent to the evaporator 74 .
- the evaporator 74 evaporates the expanded liquid refrigerant to generate gas refrigerant. At this time, the refrigerant passing through the evaporator 74 absorbs heat of evaporation from the surrounding air.
- the evaporator 74 discharges the generated gas refrigerant to the compressor 71 .
- Compressor 71 is connected to drive device 81 as shown in FIG. An angle sensor 712 is provided inside. Also, the compressor 71 is connected to a compression mechanism (not shown). The electric motor 711 is driven by the driving device 81 to rotate the rotor, and the rotation of the rotor operates the compression mechanism to compress the gas refrigerant.
- the abnormality diagnosis device 100 of Embodiment 1 also has a control function for the driving device 81, and includes a driving device control section 110, an abnormality diagnosis section 120, and a storage section .
- the drive device control unit 110 acquires a current detection value from a current detection unit 82 that detects a current flowing between the compressor 71 and the drive device 81, and controls the drive device 81 by feedback control.
- the abnormality diagnosis unit 120 detects the current value detected by the current detection unit 82, that is, the data of the first state quantity, and the load torque and driving frequency that indicate the operation mode (details of which will be described later) of the compressor 71, that is, the second
- Storage unit 130 stores the results of operations performed by drive device control unit 110 and abnormality diagnosis unit 120 as necessary.
- the current detection unit 82 is configured to have N measurement points by a plurality of (N) current sensors.
- the drive device 81 includes an inverter 811 and a converter 812 .
- the converter 812 is supplied with an alternating current from an alternating current power supply (not shown), converts the alternating current into a direct current, and supplies the direct current to the inverter 811 .
- the frequency of the alternating current supplied from the alternating current power supply to the converter 812 is predetermined, for example, 50 Hz or 60 Hz.
- the inverter 811 has an inverter main circuit including a plurality of switching elements, and is controlled by the drive device control section 110 of the abnormality diagnosis device 100 . Illustrations of the plurality of switching elements and the inverter main circuit are omitted.
- the driving device control unit 110 transmits a PWM (Pulse Width Modulation) signal to the inverter 811 to switch on and off the switching elements of the inverter main circuit, thereby driving the three-phase (U-phase) motor 711 of the compressor 71. , V phase, and W phase).
- PWM Pulse Width Modulation
- a current detection unit 82 provided between the electric motor 711 and the driving device 81 detects the U-phase current Iu and the V-phase current Iv among the three-phase currents output to the electric motor 711, and detects the U-phase current Iu. and the current value of the V-phase current Iv. Note that these current values may also be indicated as Iu and Iv. Further, since the current value of the W-phase current Iw can be calculated from the current value of the U-phase current Iu and the current value of the V-phase current Iv, the current detection unit 82 outputs the current values of at least two of the three phases. do it.
- the driving device control unit 110 outputs a PWM signal to the inverter 811 to perform vector control.
- Park transform is performed on the two-phase current values using the Clarke transformation unit 112 that performs Clarke transformation on the phase currents and obtains the two-phase current values, and the rotation angle ⁇ obtained from the angle sensor 712 .
- a voltage command value calculator 114 that calculates the voltage command value for the inverter 811; and an output voltage vector calculator that calculates an output voltage vector from the voltage command value calculated by the voltage command value calculator 114.
- 115 and a PWM signal generator 116 that generates a PWM signal to be output to the inverter 811 .
- Phase current calculation unit 111 calculates the current value of W-phase current Iw from the current value of U-phase current Iu and the current value of V-phase current Iv obtained from current detection unit 82, and calculates the current value of each phase (Iu, Iv , Iw) to the current value Clark transform unit 512 . It should be noted that the currents (Iu, Iv, Iw) of each phase change with the rotation angle ⁇ of the rotor of the electric motor 711 . In Embodiment 1, the rotation angle ⁇ is described as a value measured by the angle sensor 712, but the angle sensor 712 is not an essential component. The angle sensor 712 can be omitted if the rotation angle ⁇ is calculated by another method.
- the rotation angle ⁇ can be calculated from the currents (Iu, Iv, Iw) of each phase and the voltage command value to the inverter 811, as in position sensorless control. There is a way. Since this calculation method is known, detailed description is omitted.
- the Clark transform unit 112 and the Park transform unit 113 will be explained using FIG.
- the current values of the phase currents (Iu, Iv, Iw) output by the phase current calculator 111 are input to the Clarke transform unit 112 .
- the Clarke transformer 112 transforms the three-phase currents Iu, Iv, and Iw into two-phase ( ⁇ -phase) currents (I ⁇ , I ⁇ ) and outputs the currents to the Park transformer 113 .
- the Park transform unit 113 acquires the current values of the ⁇ -phase currents (I ⁇ , I ⁇ ) from the Clarke transform unit 112, acquires the rotation angle ⁇ from the angle sensor 712, and obtains the ⁇ -phase currents (I ⁇ , I ⁇ ) of the stationary coordinate system. I ⁇ ) is converted into dq-axis currents (Id, Iq) corresponding to the coordinates of the rotating coordinate system (dq coordinate system). Park transforming unit 113 outputs the dq-axis currents (Id, Iq) to voltage command value computing unit 114 .
- the d-axis current Id is an exciting current component and causes the electric motor 711 to generate a rotating magnetic field.
- the q-axis current Iq is a torque current component and produces torque of the electric motor 711 .
- the dq-axis currents (Id, Iq) correspond to values obtained by measuring two-phase currents (I ⁇ , I ⁇ ) rotating at a rotation angle ⁇ in a stationary coordinate system in a rotating coordinate system that follows the rotation. No change in angle ⁇ appears.
- the voltage command value calculation unit 114 calculates a command value for the output voltage of the inverter 811 so that the electric motor 711 outputs desired torque and rotational speed.
- the output voltage vector calculation unit 115 calculates an output voltage vector based on the output voltage command value calculated by the voltage command value calculation unit 114 .
- PWM signal generation section 116 generates a PWM signal for controlling inverter 811 based on the output voltage vector calculated by output voltage vector calculation section 115 .
- the abnormality diagnosis unit 120 has two functions: generation of a reference region in prior learning and abnormality diagnosis in actual operation.
- the abnormality diagnosis unit 120 includes a data acquisition unit 121 that acquires necessary data in each of the prior learning and the actual operation, an operation mode determination unit 122 that determines the operation mode in each of the prior learning and the actual operation, and a prior In each of the learning and the actual operation, the feature quantity calculation unit 123 that calculates the feature quantity used in the principal component analysis, and in the pre-learning, the dimension of the feature vector is reduced and the reference matrix that is projected onto the two-dimensional plane is calculated.
- Embodiment 1 abnormality diagnosis is performed using the U-phase current Iu.
- the wear of the sliding portion which accounts for most of the abnormalities in the compressor 71, occurs, the permeance changes due to the vibration of the gap between the rotor and the stator of the electric motor 711. This is because it is effective to use the current value.
- the vibration of the gap is similarly generated when the bearing of the electric motor 711 wears, it is effective to use the current value for abnormality diagnosis of the electric motor 711 as well.
- the current detector 82 can be realized by installing a sensor on the power cable between the driving device 81 and the compressor 71, an additional sensor for abnormality diagnosis is not required. Therefore, it can be said that the method using the current value is desirable also in terms of cost.
- the current used for abnormality diagnosis is not limited to the U-phase current Iu, but may be currents of other phases of the three-phase currents, namely Iv and Iw, or any of the ⁇ -phase currents and any of the dq-axis currents. may If ⁇ -phase currents are used, Clarke transforms are required, and if dq-axis currents are used, Clarke and Park transforms are required, but in any case no additional sensors are required.
- abnormality diagnosis unit 120 After generating a reference region by prior learning (step ST100), abnormality diagnosis section 120 performs abnormality diagnosis in actual operation (step ST200). The abnormality diagnosis is periodically performed while the air conditioner 1000 is in operation. Generation of reference regions is also performed as needed.
- FIG. 6 is a flowchart showing reference area generation according to the first embodiment.
- Generation of the reference region in pre-learning must be performed in a normal state (a state in which there is no abnormality).
- the data acquisition unit 121 acquires the time-series data of the current value of the U-phase current Iu from the N measurement points of the current detection unit 82, and also acquires the current load torque and drive frequency. (step ST101).
- Data acquisition unit 121 outputs the current value of U-phase current Iu to feature amount calculation unit 123 , and outputs the current load torque and drive frequency to operation mode determination unit 122 .
- the driving mode determination unit 122 determines the current driving mode based on the current load torque and drive frequency acquired by the data acquisition unit 121 (step ST102).
- FIG. 10 shows examples of operation modes classified by drive frequency and load torque. As shown in FIG. 10, among the classified operation modes, the operation mode with the lowest drive frequency is designated as mode (1, 1), (1, 2), and so on, and the operation mode with the lowest load torque is selected in order. Modes (1, 1), (2, 1), .
- the load torque data is acquired by, for example, measuring the rotation of the electric motor 711, acquiring the rotation speed data by the data acquisition unit 121, and calculating the slip from the ratio between the rotation speed and the drive frequency. There is a way to calculate In this case, it is necessary to add a rotation speed measuring device.
- the load torque of the electric motor 711 is determined by the outside air temperature, target temperature, and driving frequency of each device. By estimating the load torque from these pieces of information, it becomes possible to obtain the value of the load torque without the need for an additional sensor, and an increase in cost due to the addition of the sensor can be prevented.
- Each operation mode can be defined in advance, but it is also possible to define each operation mode in advance learning. Specifically, when the data of the load torque and drive frequency are acquired in the pre-learning, the operation modes are defined in order by the acquired load torque and drive frequency. Since the load torque and the drive frequency are continuous values, an error margin is set for the target value for each operation mode. It is conceivable that the margin of error is up to the midpoint between adjacent operation modes. For example, in mode (1, 1) of FIG. 10, the target values are a drive frequency of 35 Hz and a load torque of 2 Nm. If it is within the range, it is assumed to be included in mode (1, 1). In addition, if the classification of the operation mode by load torque is 0.5 Nm, consider 0.0 to less than 0.25 Nm, 0.25 Nm to less than 0.75 Nm, and 0.75 Nm to less than 1.25 Nm as each classification. can be done.
- the feature quantity calculation unit 123 calculates a feature quantity using the current value of the U-phase current Iu (step ST103).
- the feature amount calculated in the pre-learning is the normal feature amount. Note that the feature amount is calculated using all the data acquired in one measurement.
- the feature quantity of the U-phase current Iu is a parameter that indicates the current waveform of the U-phase current Iu. That is, specific examples of the feature amount include moments such as the mean, variance, kurtosis, and skewness of the current value of the U-phase current Iu, the maximum value, the minimum value and their mean and variance, and the current value Powers, square roots, logarithmic means and variances, etc. are possible.
- I is the current value of the U-phase current Iu.
- Ii is time-series data obtained at predetermined time intervals.
- a superscript bar (-) represents the average value of N current values. Also, ⁇ represents standard deviation. The same applies to the following formulas.
- the subscripts p and L respectively represent the maximum and minimum values of the current value Ii within a predetermined time at a certain measurement point. That is, pt5 is obtained by dividing the average value of the maximum values at each measurement point by the standard deviation of the maximum values at each measurement point. Also, pt6 is obtained by dividing the average value of the minimum values at each measurement point by the standard deviation of the minimum values at each measurement point.
- Nk represents the number of Ii larger than (mean value of Ii + standard deviation of Ii) at a certain measurement time
- Nh represents the number of Ii smaller than (mean value of Ii - standard deviation of Ii) at a certain measurement time
- pt11 and pt12 indicate the degree of influence of outliers at a certain measurement time.
- pt1 to pt12 among pt1 to pt12, pt5 and pt6 are feature amounts that can be obtained once within a predetermined period of time, and the other feature amounts are time-series data.
- the number of feature amounts is 12 in Embodiment 1, the number of feature amounts is not limited to this. For example, it is conceivable to use some of the above 12 feature quantities. It is also possible to combine feature quantities related to current waveforms other than the above twelve.
- the 12 calculated feature values indicate the normal state.
- a distribution indicating the normal state is generated from the characteristic amount of the normal state, and based on this distribution, a criterion used for determination of normality/abnormality is generated.
- a feature space is defined as a space that has dimensions equal to the number of feature values and the value of each feature value indicates a position coordinate
- a feature vector having the value of each feature value as a component is: Indicates position in feature space.
- a principal component analysis technique is used to reduce the dimensions of feature vectors and then generate feature quantity distributions.
- Principal component analysis is a method of reducing the dimension of a feature vector by selecting multidimensional axes with large variance as the first and second principal components. After dimensionality reduction, a feature quantity distribution is generated on a plane having the first principal component and the second principal component on the vertical axis and the horizontal axis, respectively.
- the dimension after reduction may be three or more.
- the reference matrix calculation unit 124 calculates a reference matrix for projecting a feature vector having s (s ⁇ 12) feature amounts used for abnormality diagnosis as components onto a two-dimensional plane (step ST104).
- this two-dimensional plane is a plane having the first principal component and the second principal component on the vertical axis and the horizontal axis, respectively. Since the s feature quantities, which are the components of the feature vector in Embodiment 1, are obtained as time-series data (except for pt5 and pt6), the feature vectors having the s feature quantities as components are also obtained as time-series data. be able to.
- the corresponding components should be constant at pt5 and pt6.
- the calculated feature amount is stored in the storage unit 130 each time.
- the reference matrix is a matrix of s rows and 2 columns. By applying this reference matrix to the feature vector of the s-dimensional vector, the first principal component and the second principal component are projected onto a two-dimensional plane having vertical and horizontal axes.
- Which of the 12 types of feature amount is used is determined by whether the value of the feature amount changes depending on whether the compressor 71 and the electric motor 711 are normal or abnormal. By preliminarily excluding feature quantities that are not affected by normal/abnormality from the objects of calculation, and subjecting only feature quantities whose values change depending on whether it is normal or abnormal, the amount of calculation and the amount of storage can be reduced.
- the number of measurements of the current value of the U-phase current Iu can be set to 100, for example, for the calculation of the reference matrix, but is not limited to this.
- a large amount of calculation time is required to calculate the number of feature values required for principal component analysis. For this reason, it is desirable to leave a certain period of time between measurements of the U-phase current Iu and to calculate the feature quantity during the interval. About 30 minutes can be considered as an example of the measurement interval.
- the feature quantity distribution generation unit 125 applies the reference matrix to the feature vector to project it onto a two-dimensional plane to generate a feature quantity distribution (step ST105).
- a feature quantity distribution By sequentially projecting feature vectors, which are time-series data, onto a two-dimensional plane, distributions as shown in FIGS. 8 and 9 are generated.
- the distribution generated here is stored in the storage unit 130 as a reference distribution representing the normal distribution.
- the reference region generation unit 126 generates a reference region based on the reference distribution generated in step ST105 (step ST106).
- a reference region is generated such that a certain percentage of the points of the reference distribution are contained within the region.
- This ratio is arbitrary, but if it is too large, the reference area will also be large, and there is a possibility that normal will be judged as abnormal in abnormality diagnosis. For example, if this ratio is set to 100% and a reference region is generated to include all points of the reference distribution, the reference region is generated to include so-called "outliers.” In this case, even the data that should be excluded is treated as normal data, and there is a possibility that normal data is judged to be abnormal.
- the reference area is generated as, for example, a circular or elliptical shape, but it may be a polygonal reference area as long as an appropriate area is generated according to the shape of the reference distribution.
- the reference area is used as the reference for abnormality diagnosis, but it is also possible to use the reference distribution as the reference.
- the storage unit 130 stores the reference matrix calculated by the reference matrix calculation unit 124 and the reference region generated by the reference region generation unit 126 in association with the current driving mode determined in step ST102 (step ST107). If the driving modes are different, different reference areas S11 and S12 are generated as in the example shown in FIG. 9, so the reference areas need to be stored in association with the current driving mode. Similarly, the reference matrix must be stored in association with the current driving mode.
- the operation mode is determined to prevent misdiagnosis.
- the degree of opening of expansion valve 73 is adjusted, for example, based on the pressure on the low-pressure side and the temperature at the outlet of the pipe of evaporator 74 so that the degree of superheat at the outlet of pipe of evaporator 74 approaches a target value. (superheat control), load torque fluctuates.
- the driving device control section 110 of the abnormality diagnosis device 100 controls the driving frequency of the compressor 71 via the driving device 81 . In this way, the compressor 71, which requires continuous operation, is operated in a state where the load torque and drive frequency are constantly fluctuating based on the difference between the air temperature and the target temperature.
- FIG. 9 shows the reference distribution and reference area when the drive frequency is 35 Hz and the reference distribution and reference area when the drive frequency is 55 Hz. It can be seen from FIG. 9 that the reference area S11 when the driving frequency is 35 Hz and the reference area S12 when the driving frequency is 55 Hz are different. If the reference region is different, the diagnostic result of normality/abnormality also differs.
- the time required for one measurement is assumed to be approximately 4 s to 60 s.
- the duration of a specific operation mode is known in advance, it is conceivable to use the duration as the measurement time.
- the storage unit 130 also stores each feature value and reference distribution during pre-learning. However, if all of these data are stored, the capacity becomes enormous, so they are deleted from the storage unit 130 when they become unnecessary. is desirable.
- the current value of the U-phase current Iu is acquired as continuous time-series data. become necessary.
- Such a sampling frequency is about 10 to 500 times the maximum frequency (200 Hz) for the U-phase current Iu whose frequency is about 20 Hz to 200 Hz during continuous operation when performing an abnormality diagnosis of the compressor 71, that is, About 2 kHz to 100 kHz is required.
- the number of data is a large value of 8000 (2 kHz, 4 S) to 6 million.
- the time-series data of the current value of the U-phase current Iu becomes unnecessary after each feature amount calculation.
- the number of feature data for calculating the variance is also large, it is desirable to delete it when it becomes unnecessary.
- the data of each feature amount (feature vector) becomes unnecessary after the reference distribution is generated.
- the reference distribution becomes unnecessary after the reference area is generated. Therefore, the data of the reference distribution may be deleted if unnecessary. It is also conceivable to store data limited to the operation mode having a long duration during pre-learning.
- time-series data of current values are used for abnormality diagnosis, but frequency-series data may be used for abnormality diagnosis, and both time-series data and frequency-series data may be used in combination. is also conceivable.
- frequency series data it is desirable to set the number of data to a power of two.
- FIG. 7 is a flowchart showing abnormality diagnosis according to the first embodiment.
- the data acquisition unit 121 acquires the time-series data of the current value of the U-phase current Iu from the N measurement points of the current detection unit 82 for the compressor 71 in actual operation, and also acquires the current load torque. and drive frequency are obtained (step ST201).
- Data acquisition unit 121 outputs the current value of U-phase current Iu to feature amount calculation unit 123 , and outputs the current load torque and drive frequency to operation mode determination unit 122 .
- the driving mode determining section 122 determines the current driving mode based on the current load torque and drive frequency acquired by the data acquiring section 121 (step ST202).
- the classification of the operation mode is as described in the explanation of step ST102, but in actual operation, there is a possibility that an abnormality will occur in the load torque and the drive frequency, which are the basis for determining the operation mode. Therefore, instead of the measured value of the load torque, the command value of the load torque is acquired from the drive device control unit 110, and the operation mode is determined based on the acquired command value. If the command value of the load torque cannot be acquired, it is conceivable to determine the operation mode based on the current value (not limited to the U-phase current). This is because, in general, fluctuations in the current value due to changes in the load torque are greater than fluctuations due to abnormalities within the same operating mode. For example, between rated torque and no load, there is a difference in current value of about 1.5 times.
- the feature amount calculation unit 123 calculates feature amounts (step ST203).
- the calculation of the feature amount is as described in the explanation of step ST103, so the explanation is omitted.
- the feature quantity distribution generation unit 125 generates a feature quantity distribution on a two-dimensional plane from the feature quantity obtained by the calculation in step ST203 (step ST204).
- the feature quantity distribution generation unit 125 first determines which feature quantity is to be used for abnormality diagnosis among the feature quantities obtained by the calculation in step ST203. Assume that r (r ⁇ 12) feature quantities are used here.
- the feature quantity distribution generation unit 125 uses an r-dimensional vector having r feature quantities as components as a feature vector, and projects the r-dimensional feature vector corresponding to the current driving mode onto a two-dimensional plane.
- a two-column reference matrix is read from the storage unit 130 .
- the feature quantity distribution generation unit 125 applies the reference matrix to the r-dimensional feature vector, and projects the first principal component and the second principal component onto a two-dimensional plane having vertical and horizontal axes, thereby generating a feature quantity. Generate a distribution. Thereby, for example, a distribution as shown in FIG. 8 can be obtained.
- the determination unit 127 performs abnormality diagnosis by comparing the feature quantity distribution and the reference region. That is, the determination unit 127 counts the number of points outside the reference area S1 among the points forming the feature amount distribution generated in step ST204 (step ST205), and determines the number of points in the feature amount distribution outside the reference area S1. is equal to or greater than a predetermined threshold value (for example, 50%) (step ST206). If it is equal to or greater than the predetermined threshold, determination section 127 determines that there is an abnormality (step ST207). If it is equal to or greater than the predetermined threshold, determination section 127 determines that there is no abnormality (step ST208). The determination unit 127 outputs the diagnosis result to the diagnosis result output unit 128 .
- a predetermined threshold value for example, 50%
- an abnormality diagnosis method there is also a method such as comparing the distance between groups. Any method can be used as long as the method can recognize the difference between the feature quantity distribution and the reference region S1. Also in the method of using the ratio of points in the feature quantity distribution outside the reference region S1 as the diagnostic criteria, it is also conceivable to set the threshold in stages. For example, if 90% or more, replacement is required, if 70% or more, replacement is recommended, if 50% or more, replacement is considered, and so on.
- a region S2 in FIG. 8 shows the distribution of feature quantities when the compressor 71 has an abnormality such as wear of the sliding portion. In the example of FIG. 8, since 70% of the points are outside the reference area S1, it is determined to be abnormal when the predetermined threshold is 50%.
- the distribution of the feature amount during an anomaly also differs depending on the type of anomaly. Therefore, it is also possible to identify whether the origin of the abnormality is in the electric motor 711 or in another part of the compressor 71 from the position of the distribution of the feature amount.
- the diagnosis result output unit 128 outputs the diagnosis result sent from the determination unit 127 to the outside (step ST209).
- the output destination of the diagnostic result is not particularly limited, and may be a display device such as a monitor, an audio output such as an alarm, or an external storage device.
- Embodiment 1 In addition to the principal component analysis adopted in Embodiment 1, there are methods for diagnosing anomalies that handle a plurality of feature values, such as the Taguchi method and methods using the Mahalanobis distance. If there is, the type does not matter.
- FIG. 11 is a diagram illustrating an example of a hardware configuration of an abnormality diagnosis device according to Embodiment 1.
- the abnormality diagnosis device 100 is mainly composed of a processor 91 and a memory 92 and an auxiliary storage device 93 as main storage devices.
- the processor 91 includes, for example, a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), a DSP (Digital Signal Processor), and an FPGA (Field Programmable Gate Array).
- the memory 92 is composed of a volatile storage device such as a random access memory
- the auxiliary storage device 93 is composed of a non-volatile storage device such as a flash memory or a hard disk.
- a predetermined program to be executed by the processor 91 is stored in the auxiliary storage device 93, and the processor 91 appropriately reads and executes this program to perform various arithmetic processing.
- the predetermined program is temporarily stored in the memory 92 from the auxiliary storage device 93 , and the processor 91 reads the program from the memory 92 .
- Arithmetic processing by each functional unit shown in FIG. 4 is realized by the processor 91 executing a predetermined program as described above.
- the results of arithmetic processing by the processor 91 are temporarily stored in the memory 92 and then stored in the auxiliary storage device 93 depending on the purpose of the executed arithmetic processing.
- the abnormality diagnosis device 100 outputs an input circuit 94 that receives various inputs from the outside such as a current value from the current detection unit 82, a PWM signal to the inverter 811, and a diagnosis result from the abnormality diagnosis unit 120. It has an output circuit 95 for
- the abnormality diagnosis device 100 is configured to include both the driving device control section 110 and the abnormality diagnosis section 120. However, they are separate devices, and a device that performs only control of the driving device and a device that performs only abnormality diagnosis. may be divided into devices that
- Embodiment 1 it is possible to appropriately diagnose the presence or absence of an abnormality in the diagnosis target during operation. More specifically, a feature amount calculation unit that calculates a plurality of feature amounts from time-series data of the current value of the current that drives the motor, and an operation that determines the operation mode of the compressor based on the load torque and the drive frequency.
- a mode determination unit a feature amount distribution generation unit that generates a feature amount distribution from a plurality of feature amount values, a reference area generation unit that generates a reference area based on the feature amount distribution during normal operation, and an operation mode and a determination unit that compares the feature quantity distribution obtained at the time of abnormality diagnosis with the reference region corresponding to the operation mode at the time of abnormality diagnosis to determine whether there is an abnormality in the compressor and the electric motor. and
- the operation mode determination unit determines the operation mode at that time, and when generating the reference region, the reference region is stored in association with the operation mode, and the abnormality is detected.
- the reference region corresponding to the operation mode at that time is used as the reference for abnormality diagnosis.
- the feature quantity distribution is generated after reducing the dimension of the feature vector, the amount of calculation and storage can be reduced.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Thermal Sciences (AREA)
- Artificial Intelligence (AREA)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Control Of Ac Motors In General (AREA)
- Air Conditioning Control Device (AREA)
Abstract
Description
実施の形態1を図1から図11に基づいて説明する。実施の形態1では、「診断対象」として、常時運転する空調機の圧縮機、およびこの電動機に備えられた電動機を例に説明する。図1は、実施の形態1に係る空調機の概要を示す図、図2は、実施の形態1における異常診断装置を示す概要構成図であり、図3は、実施の形態1における異常診断装置を示すブロック図である。空調機1000は、圧縮機71と、凝縮器72と、膨張弁73と、蒸発器74とを備え、圧縮機71、凝縮器72、膨張弁73、および蒸発器74の順に冷媒が循環することで、冷凍サイクル装置として動作する。圧縮機71は、気体状の冷媒(ガス冷媒)を圧縮し、凝縮器72に吐出する。凝縮器72は、圧縮機71から吐出されたガス冷媒を凝縮して液体状の冷媒(液冷媒)を生成する。この際、凝縮器72を通過する冷媒は、周囲の空気に凝縮熱を放出する。凝縮器72は、生成した液冷媒を膨張弁73に吐出する。膨張弁73は、図示しない制御装置により開度が制御されることにより、凝縮器72から吐出された液冷媒を減圧し、膨張させる。膨張した液冷媒は蒸発器74に送られる。蒸発器74は、膨張した液冷媒を蒸発させてガス冷媒を生成する。この際、蒸発器74を通過する冷媒は、周囲の空気から蒸発熱を吸収する。蒸発器74は、生成したガス冷媒を圧縮機71に吐出する。
従って、例示されていない無数の変形例が、本願に開示される技術の範囲内において想定される。例えば、少なくとも1つの構成要素を変形する場合、追加する場合または省略する場合が含まれるものとする。
Claims (14)
- 運転中の診断対象の状態をそれぞれ示す第1の状態量および第2の状態量に基づいて前記診断対象の異常診断を行う異常診断装置であって、
前記第1の状態量の時系列データおよび周波数系列データのいずれか一方もしくは両方、および前記第2の状態量のデータを取得するデータ取得部と、
前記第1の状態量の時系列データおよび周波数系列データのいずれか一方もしくは両方から複数の特徴量を計算する特徴量計算部と、
前記第2の状態量に基づいて、前記診断対象の運転モードを判定する運転モード判定部と、
前記複数の特徴量の値を成分に持つ特徴ベクトルに基づいて、特徴空間に特徴量分布を生成する特徴量分布生成部と、
正常状態の前記診断対象から得た特徴量分布に基づいて、前記特徴空間に基準分布または基準領域を基準として生成する基準生成部と、
前記基準を生成したときの前記運転モードに対応させて、前記基準を記憶する記憶部と、
異常診断時に得られる前記特徴量分布と、前記異常診断時の前記運転モードに対応する前記基準とを比較して、前記診断対象の異常の有無を判定する判定部と
を備えたことを特徴とする異常診断装置。 - 前記特徴ベクトルの次元を削減する基準行列を演算する基準行列演算部をさらに備え、前記特徴量分布生成部は、次元が削減された前記特徴ベクトルに基づいて、前記特徴量分布を生成する請求項1に記載の異常診断装置。
- 前記基準を生成するときに、同じ前記運転モードが継続する時間を測定し、測定された時間を前記記憶部に記憶させて、前記異常診断時には、前記測定された時間の範囲内で前記第1の状態量を取得する請求項1または2に記載の異常診断装置。
- 前記基準は前記基準領域であって、前記判定部は、異常診断時の前記特徴量分布を構成する点のうち、前記基準領域の外にある点の割合が予め定められた閾値以上である場合に、前記診断対象に異常があると判定する請求項1から3のいずれか1項に記載の異常診断装置。
- 前記閾値は、互いに大きさが異なる複数の閾値であり、前記判定部は、前記割合と前記複数の閾値とを比較して、異常の程度を段階的に判定する請求項4に記載の異常診断装置。
- 前記判定部は、異常診断時の前記特徴量分布に基づいて、前記診断対象に生じている異常の由来を特定する請求項1から5のいずれか1項に記載の異常診断装置。
- 前記特徴ベクトルは、前記特徴量のうち、前記診断対象が正常か異常かにより値が変化する特徴量の値のみを成分に持つ請求項1から6のいずれか1項に記載の異常診断装置。
- 前記判定部による診断結果を出力する診断結果出力部をさらに備える請求項1から7のいずれか1項に記載の異常診断装置。
- 前記診断対象は、空調機の圧縮機および前記圧縮機に備えられた電動機を含む請求項1から8のいずれか1項に記載の異常診断装置。
- 前記第2の状態量は、前記電動機の負荷トルク、回転速度、および駆動周波数のうちの少なくとも1つを含む請求項9に記載の異常診断装置。
- 前記負荷トルクは、前記空調機の設定気温および外気温から推定された推定値である請求項10に記載の異常診断装置。
- 前記第1の状態量は、前記電動機を駆動する電流の電流値を含む請求項9から11のいずれか1項に記載の異常診断装置。
- 前記電流値は、三相電流のうちのいずれかの相電流の電流値、および回転座標系のいずれかの軸の電流の電流値のうちの、少なくとも1つを含む請求項12に記載の異常診断装置。
- 運転中の診断対象の状態をそれぞれ示す第1の状態量および第2の状態量に基づいて前記診断対象の異常診断を行う異常診断方法であって、
前記第1の状態量の時系列データおよび周波数系列データのいずれか一方もしくは両方、および前記第2の状態量のデータを取得するステップと、
前記第1の状態量の時系列データおよび周波数系列データのいずれか一方もしくは両方から複数の特徴量を計算するステップと、
前記第2の状態量に基づいて、前記診断対象の運転モードを判定するステップと、
前記複数の特徴量の値を成分に持つ特徴ベクトルに基づいて、特徴空間に特徴量分布を生成するステップと、
正常状態の前記診断対象から得た特徴量分布に基づいて、前記特徴空間に基準分布または基準領域を基準として生成するステップと、
前記基準を生成したときの前記運転モードに対応させて、前記基準を記憶するステップと、
異常診断時に得られる前記特徴量分布と、前記異常診断時の前記運転モードに対応する前記基準とを比較して、前記診断対象の異常の有無を判定するステップと
を備えたことを特徴とする異常診断方法。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/014928 WO2022215232A1 (ja) | 2021-04-08 | 2021-04-08 | 異常診断装置および異常診断方法 |
CN202180096614.2A CN117136494A (zh) | 2021-04-08 | 2021-04-08 | 异常诊断装置以及异常诊断方法 |
US18/285,386 US20240120870A1 (en) | 2021-04-08 | 2021-04-08 | Abnormality diagnosis device and abnormality diagnosis method |
JP2023512609A JPWO2022215232A1 (ja) | 2021-04-08 | 2021-04-08 | |
EP21936036.9A EP4322396A4 (en) | 2021-04-08 | 2021-04-08 | ANOMALY DIAGNOSIS DEVICE AND ANOMALY DIAGNOSIS METHOD |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/014928 WO2022215232A1 (ja) | 2021-04-08 | 2021-04-08 | 異常診断装置および異常診断方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022215232A1 true WO2022215232A1 (ja) | 2022-10-13 |
Family
ID=83545318
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/014928 WO2022215232A1 (ja) | 2021-04-08 | 2021-04-08 | 異常診断装置および異常診断方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US20240120870A1 (ja) |
EP (1) | EP4322396A4 (ja) |
JP (1) | JPWO2022215232A1 (ja) |
CN (1) | CN117136494A (ja) |
WO (1) | WO2022215232A1 (ja) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005266965A (ja) * | 2004-03-16 | 2005-09-29 | Toshiba Corp | データ監視装置およびデータ監視方法 |
JP2006161677A (ja) * | 2004-12-07 | 2006-06-22 | Mitsubishi Electric Corp | 圧縮機検査装置 |
JP2006250818A (ja) * | 2005-03-11 | 2006-09-21 | Toenec Corp | 電気設備診断システム |
WO2020202567A1 (ja) * | 2019-04-05 | 2020-10-08 | 株式会社Ihi原動機 | 振動音響解析方法及び装置と機器異常部位推定方法及び装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5740459B2 (ja) * | 2009-08-28 | 2015-06-24 | 株式会社日立製作所 | 設備状態監視方法 |
JP6880992B2 (ja) * | 2017-04-27 | 2021-06-02 | 富士通株式会社 | 評価方法、評価装置および評価プログラム |
WO2020079843A1 (ja) * | 2018-10-19 | 2020-04-23 | 三菱電機株式会社 | エレベーターのドア装置異常診断システム |
JP6837612B1 (ja) * | 2020-05-28 | 2021-03-03 | 三菱電機株式会社 | 機器状態監視装置および機器状態監視方法 |
-
2021
- 2021-04-08 JP JP2023512609A patent/JPWO2022215232A1/ja active Pending
- 2021-04-08 CN CN202180096614.2A patent/CN117136494A/zh active Pending
- 2021-04-08 WO PCT/JP2021/014928 patent/WO2022215232A1/ja active Application Filing
- 2021-04-08 US US18/285,386 patent/US20240120870A1/en active Pending
- 2021-04-08 EP EP21936036.9A patent/EP4322396A4/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005266965A (ja) * | 2004-03-16 | 2005-09-29 | Toshiba Corp | データ監視装置およびデータ監視方法 |
JP2006161677A (ja) * | 2004-12-07 | 2006-06-22 | Mitsubishi Electric Corp | 圧縮機検査装置 |
JP2006250818A (ja) * | 2005-03-11 | 2006-09-21 | Toenec Corp | 電気設備診断システム |
WO2020202567A1 (ja) * | 2019-04-05 | 2020-10-08 | 株式会社Ihi原動機 | 振動音響解析方法及び装置と機器異常部位推定方法及び装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP4322396A4 * |
Also Published As
Publication number | Publication date |
---|---|
EP4322396A4 (en) | 2024-05-08 |
JPWO2022215232A1 (ja) | 2022-10-13 |
CN117136494A (zh) | 2023-11-28 |
EP4322396A1 (en) | 2024-02-14 |
US20240120870A1 (en) | 2024-04-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2312159B1 (en) | Drive unit for compressor and refrigerator | |
EP2198160B1 (en) | System and method for evaluating parameters for a refrigeration system with a variable speed compressor | |
JP6173530B1 (ja) | 空調機の故障徴候検出装置 | |
CN1098512C (zh) | 冷却器压缩机电机过热的预测方法 | |
JP6765320B2 (ja) | 交流電動機の制御装置 | |
US20070101735A1 (en) | Heat pump apparatus using expander | |
CN108267649B (zh) | 压缩机相序检测方法和装置及其启动控制方法和设备 | |
CN115136487A (zh) | 电动机的诊断装置 | |
WO2022215232A1 (ja) | 異常診断装置および異常診断方法 | |
CN117482443B (zh) | 一种消防设备健康监测方法及系统 | |
WO2006085475A1 (ja) | 冷凍サイクル装置 | |
WO2023176873A1 (ja) | 状態判定装置、モータ駆動システム、冷凍システム、ファンシステム、状態判定方法 | |
US20230057924A1 (en) | Improved anomaly detection for rotating machines | |
JP2924445B2 (ja) | 空気調和機用圧縮機の駆動装置 | |
JP4670825B2 (ja) | 圧縮機内部状態推定装置及び空気調和装置 | |
US11626825B2 (en) | Rotary machine control device, refrigerant compression device, refrigeration cycle apparatus, and air conditioner | |
KR101936261B1 (ko) | 전동기 제어 장치 및 그 동작 방법 | |
JP7442742B2 (ja) | 電動機付設備の故障徴候検出装置および電動機付設備の故障徴候検出方法 | |
WO2020075262A1 (ja) | 故障前兆検出装置 | |
JP2004147488A (ja) | モータ駆動制御装置 | |
KR20120079375A (ko) | 냉동기 압축기 구동용 모터 제어 장치, 이의 제어 방법, 및 이를 포함하는 냉동기 | |
WO2021149203A1 (ja) | モータ制御装置および空気調和装置 | |
JPWO2022215232A5 (ja) | ||
JP2020133643A (ja) | 異常検出装置および異常検出方法 | |
CN118117928A (zh) | 马达驱动装置、电气常数测定方法以及冷冻设备 |
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: 21936036 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023512609 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18285386 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2021936036 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2021936036 Country of ref document: EP Effective date: 20231108 |