WO2022054782A1 - 状態判定装置及び状態判定方法 - Google Patents
状態判定装置及び状態判定方法 Download PDFInfo
- Publication number
- WO2022054782A1 WO2022054782A1 PCT/JP2021/032790 JP2021032790W WO2022054782A1 WO 2022054782 A1 WO2022054782 A1 WO 2022054782A1 JP 2021032790 W JP2021032790 W JP 2021032790W WO 2022054782 A1 WO2022054782 A1 WO 2022054782A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- state
- learning
- industrial machine
- statistical
- data
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims description 51
- 230000006870 function Effects 0.000 claims abstract description 56
- 238000003860 storage Methods 0.000 claims abstract description 34
- 238000006243 chemical reaction Methods 0.000 claims abstract description 23
- 230000005856 abnormality Effects 0.000 claims description 45
- 238000010801 machine learning Methods 0.000 claims description 41
- 238000001746 injection moulding Methods 0.000 description 59
- 230000008569 process Effects 0.000 description 37
- 238000002347 injection Methods 0.000 description 33
- 239000007924 injection Substances 0.000 description 33
- 238000012545 processing Methods 0.000 description 26
- 238000004519 manufacturing process Methods 0.000 description 18
- 238000013075 data extraction Methods 0.000 description 17
- 238000000465 moulding Methods 0.000 description 13
- 230000002950 deficient Effects 0.000 description 11
- 230000007547 defect Effects 0.000 description 10
- 239000011347 resin Substances 0.000 description 10
- 229920005989 resin Polymers 0.000 description 10
- 238000012549 training Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 238000013500 data storage Methods 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 6
- 238000005303 weighing Methods 0.000 description 6
- 230000001133 acceleration Effects 0.000 description 4
- 238000002845 discoloration Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000001816 cooling Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 239000011800 void material Substances 0.000 description 2
- KNMAVSAGTYIFJF-UHFFFAOYSA-N 1-[2-[(2-hydroxy-3-phenoxypropyl)amino]ethylamino]-3-phenoxypropan-2-ol;dihydrochloride Chemical compound Cl.Cl.C=1C=CC=CC=1OCC(O)CNCCNCC(O)COC1=CC=CC=C1 KNMAVSAGTYIFJF-UHFFFAOYSA-N 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 235000019219 chocolate Nutrition 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 239000007787 solid Substances 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
-
- 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
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0267—Fault communication, e.g. human machine interface [HMI]
- G05B23/027—Alarm generation, e.g. communication protocol; Forms of alarm
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
- B29C45/17—Component parts, details or accessories; Auxiliary operations
- B29C45/76—Measuring, controlling or regulating
- B29C45/768—Detecting defective moulding conditions
Definitions
- the present invention relates to a state determination device and a state determination method for industrial machines.
- Maintenance of industrial machines such as injection molding machines is performed regularly or when an abnormality occurs.
- the maintenance person determines whether or not there is an abnormality in the operating state of the industrial machine by using a physical quantity indicating the operating state of the industrial machine recorded at the time of operation of the industrial machine. Then, perform maintenance work such as replacement of parts with abnormalities.
- Patent Document 1 shows that an abnormality is determined by supervised machine learning such as a load of a driving unit and a resin pressure.
- machine learning such as a load of a driving unit and a resin pressure.
- Patent Document 2 regarding the abnormality degree estimation value derived by machine learning, a correction coefficient associated with the injection molding model or equipment is used for the abnormality degree estimation value calculated from one learning model. It has been shown to derive a corrected anomaly correction value.
- Patent Document 3 discloses that a plurality of learning models according to conditions related to injection operation such as operating conditions and environmental conditions are prepared in advance. What is shown in Patent Document 3 is to select one learning model from a plurality of learning models based on the conditions and processing capacity of the injection operation when calculating the evaluation value for the state of the injection operation. , The judgment accuracy of machine learning is improved. Further, Patent Document 4 shows that a plurality of learning models are prepared, the learning data classified according to the classification conditions are associated with the learning model, and one learning model is selected from the plurality of learning models. Has been done.
- Japanese Unexamined Patent Publication No. 2017-20632 Japanese Unexamined Patent Publication No. 2020-404718 Japanese Unexamined Patent Publication No. 2019-067138 Japanese Unexamined Patent Publication No. 2020-066178
- the state determination device with respect to the degree of abnormality estimated by machine learning, the time-series physical quantities (current, speed, etc.) acquired from the control device that controls the industrial machine are used as data indicating the state related to the industrial machine. Then, this state determination device calculates a plurality of estimated values (abnormality degrees) using a plurality of various learning models. Next, the statistical functions associated with the model and equipment of the industrial machine and the learning model are applied to the calculated estimated values for each of the plurality of learning models, and the statistic for evaluating the degree of abnormality of the industrial machine is calculated. .. Since the characteristics of a plurality of learning models are taken into consideration in the calculated statistic, it is possible to comprehensively determine the degree of abnormality in which a wide variety of characteristics are reflected by the statistic.
- the degree of abnormality can be comprehensively determined based on the statistic calculated by applying the statistical function associated with the model and incidental equipment to the value. For example, two learning models, one for large machines and the other for small machines, are prepared in advance, and learning is performed when determining the degree of abnormality of a medium-sized machine different from the model used to create the learning model.
- the two estimates calculated by the model were derived by applying weights corresponding to the size of the machine (eg, 70% for a learning model with a large machine size and 30% for a learning model with a small machine size).
- weights corresponding to the size of the machine eg, 70% for a learning model with a large machine size and 30% for a learning model with a small machine size.
- one aspect of the present invention is a state determination device for determining a state related to an industrial machine, and has learned the correlation between the data related to a predetermined physical quantity acquired from the industrial machine and the state related to the industrial machine.
- a learning model storage unit that stores a plurality of learning models, designation of a plurality of the learning models used when determining a state related to the industrial machine, and numerical conversion of estimation results by the designated learning model to obtain statistics.
- a statistical condition storage unit that stores statistical conditions including at least a statistical function for calculating the above, a data acquisition unit that acquires data related to a predetermined physical quantity as data indicating a state of the industrial machine, and the data acquisition unit.
- the estimation unit that estimates the state related to the industrial machine using the plurality of learning models stored in the learning model storage unit, and the statistical function with reference to the statistical condition storage unit. It is a state determination device including a numerical conversion unit that numerically converts the estimation results for each of a plurality of learning models by the estimation unit using the acquired statistical function and calculates a statistic.
- another aspect of the present invention is a state determination method for determining a state related to an industrial machine, and learns the correlation between the data related to a predetermined physical quantity acquired from the industrial machine and the state related to the industrial machine.
- the designation of the plurality of learning models used when determining the state related to the industrial machine and the estimation result by the designated learning model are numerically converted to calculate the statistic.
- Statistical conditions including at least the statistical function to be performed are stored in advance, and based on the step of acquiring the data related to a predetermined physical quantity as the data indicating the state of the industrial machine and the data acquired in the acquired step.
- the step of estimating the state related to the industrial machine using the plurality of learning models stored in advance, and the statistical function included in the statistical condition stored in advance are acquired, and the acquired statistical function is obtained. It is a state determination method that executes a step of numerically converting the estimation results for each of a plurality of learning models by the estimation step and calculating a statistic.
- statistically processed statistics are calculated based on estimated values obtained from a plurality of learning models, so that it is possible to make a comprehensive judgment on the state of industrial machines.
- FIG. 1 is a schematic hardware configuration diagram showing a main part of a state determination device according to an embodiment of the present invention.
- the state determination device 1 according to the present embodiment can be implemented as a control device that controls an industrial machine based on, for example, a control program. Further, the state determination device 1 according to the present embodiment is a personal computer attached to a control device that controls an industrial machine based on a control program, a personal computer connected to the control device via a wired / wireless network, and a cell computer. , Can be mounted on the fog computer 6 and the cloud server 7.
- the state determination device 1 is mounted on a personal computer connected to the control device 3 via the network 9 .
- the industrial machine for which the state determination device of the present invention determines the state include an injection molding machine, a machine tool, a mining machine, a woodworking machine, an agricultural machine, and a construction machine.
- an injection molding machine will be described as an example of an industrial machine.
- the CPU 11 included in the state determination device 1 is a processor that controls the state determination device 1 as a whole.
- the CPU 11 reads the system program stored in the ROM 12 via the bus 22 and controls the entire state determination device 1 according to the system program. Temporary calculation data, display data, various data input from the outside, and the like are temporarily stored in the RAM 13.
- the non-volatile memory 14 is composed of, for example, a memory backed up by a battery (not shown), an SSD (Solid State Drive), or the like, and the storage state is maintained even when the power of the state determination device 1 is turned off.
- the non-volatile memory 14 stores data read from the external device 72 via the interface 15, data input via the input device 71, data acquired from the injection molding machine 4 via the network 9, and the like.
- the stored data includes, for example, the motor current, voltage, torque, position, speed, acceleration, and inside of the mold of the drive unit detected by various sensors 5 attached to the injection molding machine 4 controlled by the control device 3.
- Data related to physical quantities such as pressure, injection cylinder temperature, resin flow rate, flow velocity, vibration and sound may be included.
- the data stored in the non-volatile memory 14 may be expanded in the RAM 13 at the time of execution / use. Further, various system programs such as a known analysis program are written in the ROM 12 in advance.
- the interface 15 is an interface for connecting the CPU 11 of the state determination device 1 and an external device 72 such as an external storage device.
- an external device 72 such as an external storage device.
- a system program, a program related to the operation of the injection molding machine 4, parameters, and the like can be read.
- the data or the like created / edited on the state determination device 1 side can be stored in an external storage medium such as a CF card or a USB memory (not shown) via the external device 72.
- the interface 20 is an interface for connecting the CPU 11 of the state determination device 1 and the wired or wireless network 9.
- the network 9 communicates using technologies such as serial communication such as RS-485, Ethernet (registered trademark) communication, optical communication, wireless LAN, Wi-Fi (registered trademark), and Bluetooth (registered trademark). It's okay to have it.
- a control device 3 for controlling the injection molding machine 4, a fog computer 6, a cloud server 7, and the like are connected to the network 9, and these devices exchange data with each other with the state determination device 1. ..
- each data read into the memory, data obtained as a result of executing a program, etc., data output from the machine learning device 2 described later, and the like are output and displayed via the interface 17. Will be done.
- the input device 71 composed of a keyboard, a pointing device, and the like passes commands, data, and the like based on operations by the operator to the CPU 11 via the interface 18.
- the interface 21 is an interface for connecting the CPU 11 and the machine learning device 2.
- the machine learning device 2 stores a processor 201 that controls the entire machine learning device 2, a ROM 202 that stores a system program, a RAM 203 that temporarily stores each process related to machine learning, a learning model, and the like.
- the non-volatile memory 204 used for the above is provided.
- the machine learning device 2 has data that can be acquired by the state determination device 1 via the interface 21 (for example, the motor current, voltage, torque, position of the drive unit detected by various sensors 5 attached to the injection molding machine 4). It is possible to observe speed, acceleration, mold pressure, injection cylinder temperature, resin flow rate, flow velocity, data related to physical quantities such as vibration and sound). Further, the state determination device 1 acquires the processing result output from the machine learning device 2 via the interface 21, stores and displays the acquired result, and transmits the acquired result to other devices via the network 9 or the like. And send.
- FIG. 2 is a schematic configuration diagram of the injection molding machine 4.
- the injection molding machine 4 is mainly composed of a mold clamping unit 401 and an injection unit 402.
- the mold clamping unit 401 is provided with a movable platen 416 and a fixed platen 414. Further, a movable side mold 412 is attached to each of the movable platen 416, and a fixed side mold 411 is attached to the fixed platen 414.
- the injection unit 402 includes an injection cylinder 426, a hopper 436 for storing the resin material to be supplied to the injection cylinder 426, and a nozzle 440 provided at the tip of the injection cylinder 426.
- the mold clamping unit 401 closes and molds the mold by moving the movable platen 416, and the injection unit 402 presses the nozzle 440 against the fixed side mold 411, and then the nozzle.
- the resin is injected into the mold via the 440.
- sensors 5 for detecting physical quantities are attached to each part of the injection molding machine 4, and the motor current, voltage, torque, position, speed, acceleration, mold internal pressure, temperature of the injection cylinder 426, and resin of the drive part are attached.
- the physical quantities such as the flow rate, the flow velocity, the vibration and the sound of the sensor 5 are detected by the sensor 5.
- the physical quantity detected by the sensor 5 is sent to the control device 3.
- each detected physical quantity is stored in a RAM, a non-volatile memory, or the like (not shown), and is transmitted to the state determination device 1 via the network 9 as needed.
- FIG. 3 shows as a schematic block diagram the functions included in the state determination device 1 according to the first embodiment of the present invention.
- the CPU 11 included in the state determination device 1 and the processor 201 included in the machine learning device 2 respectively execute a system program, and the state determination device 1 and the machine are provided. It is realized by controlling the operation of each part of the learning device 2.
- the state determination device 1 of the present embodiment includes a data acquisition unit 100, a data extraction unit 110, an estimation command unit 120, and a numerical conversion unit 140. Further, the machine learning device 2 includes an estimation unit 207. Further, in the RAM 13 to the non-volatile memory 14 of the state determination device 1, the acquired data storage unit 300 as an area for storing the data acquired by the data acquisition unit 100 from the control device 3 and the like, and the numerical value by the numerical conversion unit 140. A statistical condition storage unit 310 for storing statistical conditions used for conversion in advance is prepared in advance.
- a learning model storage unit 210 is prepared in advance as an area for storing the learning model 214 of the above.
- the data acquisition unit 100 executes a system program read from the ROM 12 by the CPU 11 included in the state determination device 1 shown in FIG. 1, mainly performs arithmetic processing using the RAM 13 and the non-volatile memory 14 by the CPU 11, and the interfaces 15 and 18. Alternatively, it is realized by performing the input control process according to 20.
- the data acquisition unit 100 includes the motor current, voltage, torque, position, speed, acceleration, mold internal pressure, temperature of the injection cylinder 426, and resin flow rate of the drive unit detected by the sensor 5 attached to the injection molding machine 4. , Flow velocity, acquisition of data related to physical quantities such as vibration and sound.
- the data related to the physical quantity acquired by the data acquisition unit 100 may be so-called time-series data indicating the value of the physical quantity for each predetermined cycle. Further, the data acquisition unit 100 may acquire data directly from the control device 3 that controls the injection molding machine 4 via the network 9. The data acquisition unit 100 may acquire data acquired and stored by the external device 72, the fog computer 6, the cloud server 7, and the like. The data acquisition unit 100 may acquire data related to physical quantities for each step constituting one molding cycle by the injection molding machine 4.
- FIG. 4 is a diagram illustrating a molding cycle for manufacturing one molded product. In FIG.
- the mold closing process, the mold opening process, and the protrusion process which are the processes of the shaded frame, are the operations of the mold clamping unit 401, and are the injection process, the pressure holding process, the measuring process, and the depressurizing process, which are the processes of the white frame.
- the cooling step is performed by the operation of the injection unit 402.
- the data acquisition unit 100 acquires data related to physical quantities so that each of these steps can be distinguished.
- the data related to the physical quantity acquired by the data acquisition unit 100 is stored in the acquisition data storage unit 300.
- the data extraction unit 110 is realized by executing a system program read from the ROM 12 by the CPU 11 included in the state determination device 1 shown in FIG. 1 and performing arithmetic processing mainly by the CPU 11 using the RAM 13 and the non-volatile memory 14. Will be done.
- the data extraction unit 110 extracts from the acquisition data storage unit 300 data used for processing related to machine learning such as estimation processing by the machine learning device 2 from the data related to the physical quantity acquired by the data acquisition unit 100.
- the data extraction by the data extraction unit 110 is performed based on the statistical conditions stored in the statistical condition storage unit 310.
- Statistical conditions define how to calculate the estimation result of the current state of industrial machinery.
- the statistical condition includes at least the designation of a plurality of learning models used for determining the state of the industrial machine and the statistical function for numerically converting the estimation result by the designated learning model to calculate the statistic.
- Statistical conditions may be created for each model (scale, etc.) of the industrial machine and the type of equipment attached to the industrial machine.
- Statistical functions to be included in the statistical conditions include, for example, weighted mean, arithmetic mean, weighted harmonic mean, harmonic mean, pruned mean, log mean, sum of squared mean square root, minimum, maximum, median, weighted median, and mode. It may be possible to set a predetermined statistical function in consideration of the relationship between the judgment state such as a value and each learning model.
- a weighted average in which the weights are changed according to the correlation between the model of the industrial machine and the estimation result of each learning model may be used. Further, when it is determined that even one of the estimation results of the plurality of learning models is in a predetermined state (for example, an abnormal state), it is determined that the state of the industrial machine has become an abnormal state.
- the maximum value (or minimum value) among the estimation results of the learning model of the above may be selected and used.
- the estimation results of a plurality of learning models include outliers that deviate significantly from the average value of the estimation results and the outliers are excluded, the center of the estimation results of the plurality of learning models. You may use a value or a mode value.
- FIG. 5 is a diagram illustrating a model of an injection molding machine and statistical conditions in which a screw diameter and a statistical function are associated with each other.
- the model of the injection molding machine (the screw diameter is arbitrary) is 30 tons
- the estimation results of the learning model a and the learning model b are used, and the weight for the estimation result by the learning model a is 0.30.
- a weighted average with a weight of 0.70 for the estimation result by the learning model b is calculated, and this is defined as an estimated value for determining the state of the injection molding machine.
- the estimation results of the learning models a, b, and c are used to calculate the pruning average of these estimation results, and this is used as the injection molding machine. It is defined as an estimated value to judge the state of.
- Parameters such as weights related to the estimation result by the learning model used for the calculation of the statistical function may be defined by fixed values such as weights as described above. Further, parameters such as weights may be calculated by using a function such as a trigonometric function, a hyperbolic function, or a sigmoid function that takes a predetermined value (hyperparameter) determined in advance by an experiment or the like as an argument.
- a hyperbolic function that takes a hyperparameter x as an argument when calculating these weighted averages using the estimation results of the learning models a and b.
- tanh (x) which is a kind of, be a function f (x) for calculating the weight of the weighted average.
- a weighted average is calculated in which the weight for the estimation result by the learning model a is the value calculated by the function f (x) and the weight value for the estimation result by the learning model b is 1-f (x). It is defined as an estimated value for determining the state of the injection molding machine.
- the functions f (x), g (x), and h (x) for calculating the weights related to the learning model in the example of FIG. 5 use the hyper parameter x as an argument in addition to the above-mentioned function tanh (x). It may be a trigonometric function such as sin (x) or cos (x), or a sigmoid function.
- the hyperparameter x which is an argument of the function at this time, may be set manually by the operator on the operation screen, or may be predetermined by conducting an experiment. This has the advantage that parameters such as weights can be easily adjusted in an analog manner. Further, parameters such as weights may be set directly from the user setting screen.
- the data extraction unit 110 identifies a plurality of learning models necessary for determining the state related to the industrial machine. Then, the data necessary for executing the estimation process in the specified plurality of learning models is extracted from the acquired data storage unit 300 that stores the data acquired by the data acquisition unit 100.
- the statistical conditions for each model of the injection molding machine and the screw diameter are shown, but different statistics are obtained when the operating conditions are different, such as energy-saving operation, stability-oriented operation, and productivity-oriented operation.
- the injection molding machine used to create the training model is used. It is possible to determine the abnormal state of an injection molding machine of a different type from the above.
- statistical conditions should be defined so as to use the estimation results of multiple learning models corresponding to the differences in operating conditions and environmental conditions, and the estimation results of multiple learning models corresponding to the differences in the accuracy and production efficiency of products. This makes it possible to make comprehensive judgments and general-purpose judgments that meet the needs of a wide variety of production environments and operators.
- the statistical conditions may be expressed in a table format as illustrated in FIG. 5, but may be expressed in another format such as a mathematical formula. Regardless of which expression is used, the statistical condition may be defined by associating a plurality of learning models to be used with a statistical function applied to the estimation result by the learning model. This makes it possible to comprehensively determine the state of industrial machines based on the statistics calculated by applying statistical functions to the estimation results of a plurality of learning models.
- the estimation command unit 120 executes a system program read from the ROM 12 by the CPU 11 included in the state determination device 1 shown in FIG. 1, and mainly uses the arithmetic processing by the CPU 11 using the RAM 13 and the non-volatile memory 14 and the interface 21. It is realized by performing the input / output processing that was performed.
- the estimation command unit 120 refers to the statistical condition storage unit 310 to specify a learning model for performing estimation processing. Then, the machine learning device 2 is instructed to execute the estimation process using each of the specified learning models.
- the numerical conversion unit 140 mainly performs arithmetic processing using the RAM 13 and the non-volatile memory 14 by the CPU 11 and the interface 21 by executing the system program read from the ROM 12 by the CPU 11 included in the state determination device 1 shown in FIG. It is realized by performing input / output processing using.
- the numerical conversion unit 140 calculates the statistical function by referring to the statistical condition storage unit 310 and using the values as the estimation results by the plurality of learning models acquired from the machine learning device 2. Then, the calculation result is output as an estimated value for determining the state related to the industrial machine.
- the estimated value output by the numerical conversion unit 140 may be displayed and output to the display device 70.
- the estimated value may be displayed and output as it is, or the state determination may be performed by comparing with a predetermined threshold value or the state classification determination may be performed, and the determination result may be output. Further, transmission and output may be performed to the control device 3 of the injection molding machine 4 which is the target of determining the operating state, or to a higher-level device such as the fog computer 6 or the cloud server 7 via the network 9.
- the estimation unit 207 included in the machine learning device 2 executes the system program read from the ROM 202 by the processor 201 included in the machine learning device 2 shown in FIG. 1, and mainly uses the RAM 203 and the non-volatile memory 204 by the processor 201. It is realized by performing the arithmetic processing that was performed.
- the estimation unit 207 selects a plurality of learning models 214 from the learning model storage units 210 based on the command from the estimation command unit 120, and executes the estimation process using each learning model 214. Then, a plurality of estimation results are output to the numerical conversion unit 140.
- a plurality of learning models 214 are stored in advance in the learning model storage unit 210.
- the learning model 214 stores a learning model created in advance.
- Each learning model 214 is trained in different situations and has a wide variety of different characteristics.
- the learning model used to determine the state of an injection molding machine is data related to physical quantities that differ for each molding cycle process (injection process, pressure holding process, weighing process, depressurization process, cooling process, etc.) (in the injection process, the injection speed and In the mold internal pressure and weighing process, the screw rotation speed, screw torque, cylinder internal pressure, etc.) may be acquired and used as learning data, and the learning model may be created for each process (for each operating condition).
- the learning model used to determine the condition of the injection molding machine is the type of equipment (motor, gear, etc.) that composes the injection molding machine, the type of production material, and the type of incidental equipment (mold, mold temperature controller, resin dryer). Etc.) It may be a learning model created for each configuration situation by acquiring data related to physical quantities and using it as learning data each time the configuration status is different.
- the learning model used to determine the state of the injection molding machine acquires data related to physical quantities for each different production environment (stability of power supply, seasonal factors in summer and winter) in which the injection molding machine is operated, and uses it as learning data. It may be a learning model created for each environment situation. Due to the difference in these situations, the learning model has a difference in a suitable model and a suitable environment.
- the learning models used for state determination related to industrial machines are supervised learning (multilayer perceptron, regression coupling neural network, convolutional neural network, etc.), unsupervised learning (autoencoder, k-average method, hostile generation network, etc.), and enhanced learning. It may be created for a different learning method such as (Q-learning, etc.). Further, the components (types of hyperparameters, types of optimization functions at the time of machine learning, etc.) in each algorithm may be different. Due to these differences, each learning model has a difference in the calculation load (calculation time) during the training process and the estimation process, the accuracy of the estimated value, and the robustness (stability, robustness) to the training data.
- the learning model used for state determination related to industrial machines may be stored in a compressed state and decompressed at the time of calculation. As a result, the memory can be used efficiently and the amount of memory can be reduced, which has the merit of cost reduction. Further, the learning model may be encrypted and stored. It is preferable to encrypt and store the learning model from the viewpoint of security and information confidentiality.
- the learning model a uses time-series data of the injection speed and the pressure inside the mold acquired in the injection process from a 30-ton scale injection molding machine as training data, and data indicating whether the operation at that time was normal or abnormal. It was created by performing supervised learning as label data.
- the learning model b uses time-series data of screw rotation speed, screw torque, and cylinder pressure acquired in the weighing process from a 30-ton injection molding machine as training data, and determines whether the operation at that time was normal or abnormal.
- the statistical condition storage unit 310 stores the statistical conditions exemplified in FIG. 5 in advance. In this case, the operating state when a screw having a screw diameter of 20 mm is attached to an injection molding machine having a scale of 50 tons is determined.
- the data extraction unit 110 refers to statistical conditions matching the injection molding machine to be determined, and as data used for estimation, time-series data of injection speed and mold internal pressure in the injection process, and screw rotation speed in the weighing process. , Screw torque, and time-series data of cylinder pressure are extracted as data used for estimation.
- the estimation command unit 120 causes the machine learning device 2 to perform estimation processing of the operating state of the injection molding machine using each of the learning model a and the learning model b using the data extracted by the data extraction unit 110. Command against.
- the estimation unit 207 performs estimation processing using the learning model a and the learning model b stored in the learning model storage unit 210, and outputs each abnormality degree to the numerical conversion unit 140 as the estimation result. do.
- the numerical conversion unit 140 refers to the statistical condition storage unit 310, and as a statistical function having a screw diameter of 20 mm in a 50-ton scale injection molding machine, the weight is 0.4 with respect to the estimation result of the learning model a, and the learning model b has a weight of 0.4. An operation using a weighted average function with a weight of 0.6 is performed on the estimation result.
- the numerical conversion unit 140 is 0.4 ⁇ 0.7 + 0.6 ⁇ .
- the statistic is plotted as an abnormality score on the screen displayed on the display device 70, and the warning message “Abnormality has been detected. Please check the injection cylinder.” Is output as an alarm. This is a display example. Then, the operation of the injection molding machine may be stopped or decelerated, or the drive torque of the prime mover for driving the drive unit of the injection molding machine may be limited.
- the learning model a uses time-series data of the injection speed and the internal pressure of the mold acquired in the injection process from a 30-ton scale injection molding machine as training data, and if the product molded at that time has a molding defect, the molding is performed.
- the learning model b created by performing supervised learning using (1,0,0,0,0)) as label data is Time-series data of screw rotation speed, screw torque, and cylinder pressure acquired in the weighing process from a 30-ton injection molding machine is used as learning data, and if there is a molding defect in the product molded at that time, the molding defect is found.
- the statistical condition storage unit 310 stores the statistical conditions exemplified in FIG. 5 in advance.
- a defective state of the molded product is determined when a screw having a screw diameter of 20 mm is attached to an injection molding machine having a scale of 50 tons.
- the data extraction unit 110 refers to the statistical conditions of the injection molding machine to be determined, and as data used for estimation, time-series data of the injection speed and the internal pressure in the mold in the injection process, and the screw rotation speed in the weighing process. , Screw torque, and time-series data of cylinder pressure are extracted as data used for estimation.
- the estimation command unit asks the machine learning device 2 to use the data extracted by the data extraction unit 110 to estimate the defective state of the molded product using each of the learning model a and the learning model b. Command.
- the estimation unit 207 performs estimation processing using the learning model a and the learning model b stored in the learning model storage unit 210, and as the estimation result, the vector value indicating each defective state is converted into a numerical value conversion unit. Output to 140.
- the numerical conversion unit 140 refers to the statistical condition storage unit 310, and as a statistical function having a screw diameter of 20 mm in a 50-ton scale injection molding machine, the weight is 0.4 with respect to the estimation result of the learning model a, and the learning model b has a weight of 0.4. An operation using a weighted average function with a weight of 0.6 is performed on the estimation result.
- the numerical conversion unit 140 is set to 0.4 ⁇ .
- ya + 0.6 ⁇ y b (0.16, 0.14, 0.26, 0.24, 0.20 ) is output as a statistic for determining the defective state of the molded product.
- 0.26 which is the largest value among the vector values indicating the defective state of the molded product output by the numerical conversion unit 140, is located at the third position in the vector, so that the defective state of the molded product (sink: molding). Defects in which the product is dented, warp: deformation of the molded product due to residual stress, burning: discoloration of the molded product, voids: vacancies, cracks: cracks or cracks in the molded product) are judged as "burning: discoloration of the molded product" Will be done. Further, if there is a type of molding defect whose statistics exceed the preset threshold value Th b of the defective state of the molded product, it may be determined that the molding defect has occurred and an alarm may be sounded. ..
- the state determination device 1 comprehensively describes the state related to the industrial machine by calculating statistically processed statistics based on the estimated values obtained from a plurality of learning models. It is possible to make a good judgment. In addition, it is not necessary to prepare a learning model for all models, configurations, operating conditions, periods from the start of production, operating conditions such as environmental conditions, and the like. In this case, a learning model for some typical models, configurations, etc. is prepared in advance, and the relationship between the learning model and other models, configurations, operating conditions, etc. is confirmed by experiments, etc., and statistical conditions are obtained. By creating it in advance, it is possible to perform estimation processing with a certain degree of accuracy without collecting a huge amount of learning data. This makes it possible to reduce the cost of operating the machine learning device.
- a warning indicating the state of abnormality is displayed on the display device based on the statistics calculated by statistically processing the degree of abnormality obtained as the output of multiple machine learning. Or, when the statistic exceeds a predetermined threshold, the operation of the industrial machine is stopped, or the motor that drives the movable part is decelerated so that the movable part operates in a safe state, or the drive torque of the motor. Can also be limited to a small size.
- FIG. 6 shows as a schematic block diagram the functions included in the state determination device 1 according to the second embodiment of the present invention.
- the CPU 11 included in the state determination device 1 and the processor 201 included in the machine learning device 2 respectively execute a system program, and the state determination device 1 and the machine are provided. It is realized by controlling the operation of each part of the learning device 2.
- the state determination device 1 of the present embodiment further includes a learning command unit 150 in addition to each function provided by the state determination device 1 according to the first embodiment. Further, the machine learning device 2 further includes a learning unit 206.
- the data extraction unit 110 functions in the same manner as the data extraction unit 110 according to the first embodiment when performing estimation processing.
- the data extraction unit 110 when the data extraction unit 110 is instructed by an operator or the like to proceed with learning by the machine learning device 2, the data extraction unit 110 extracts data used for processing related to machine learning such as learning processing from the acquisition data storage unit 300.
- the data extraction unit 110 extracts data required for learning processing of one or a plurality of designated learning models from the acquisition data storage unit 300 that stores the data acquired by the data acquisition unit 100.
- the learning command unit 150 executes a system program read from the ROM 12 by the CPU 11 included in the state determination device 1 shown in FIG. 1, and mainly uses the arithmetic processing by the CPU 11 using the RAM 13 and the non-volatile memory 14 and the interface 21. It is realized by performing the input / output processing that was performed.
- the learning command unit 150 instructs the machine learning device 2 to execute a learning process using the data extracted by the data extraction unit 110 for each of the designated one or a plurality of designated learning models.
- the learning unit 206 included in the machine learning device 2 executes the system program read from the ROM 202 by the processor 201 included in the machine learning device 2 shown in FIG. 1, and mainly uses the RAM 203 and the non-volatile memory 204 by the processor 201. It is realized by performing the arithmetic processing that was performed.
- the learning unit 206 selects one or a plurality of learning models 214 to be learned from the learning model storage unit 210 based on the command from the learning command unit 150, and performs learning processing using each learning model 214. Execute.
- the learning unit 206 may create a new learning model when the learning model commanded by the learning command unit 150 is not stored in the learning model storage unit 210.
- the state determination device 1 can proceed with the learning process of one or a plurality of learning models 214 based on the command from the operator. By updating the learning model when useful data can be acquired to proceed with learning, it can be expected that the accuracy of estimation in the estimation process will be further improved.
- the present invention is not limited to the examples of the above-described embodiments, and can be implemented in various embodiments by making appropriate changes.
- the injection molding machine has been described as an example, but the target of the state determination may be another industrial machine.
- the degree of abnormality of the spindle may be determined from a plurality of learning models corresponding to the cutting tool assembled on the spindle, the type and flow rate of the machining fluid for cooling the cutting tool, the work material, and the like.
- the degree of abnormality of the rotating tool may be determined from a plurality of learning models corresponding to the type of the rotating tool, the rotation speed, and the like.
- the degree of abnormality of the drive unit may be determined from a plurality of learning models corresponding to the driving force applied to the drive unit, the equipment provided in the drive unit, and the like.
- the degree of abnormality of the hydraulic cylinder may be determined from a plurality of learning models corresponding to the type of hydraulic hose connected to the hydraulic cylinder, the output of the prime mover, the operating environment, and the like.
- the embodiment when the machine learning device 2 is built in the state determination device 1 is described, but the machine learning device 2 is in a state where data can be exchanged with the state determination device 1. It may be installed outside the determination device 1.
- the machine learning device 2 may be arranged on the fog computer 6 or the cloud server 7 and configured to transmit commands and receive estimation results via the network 9. With this configuration, the machine learning device 2 can be shared by a plurality of state determination devices 1 and the installation cost can be reduced.
- the statistical conditions may be set based on the operator's designation as illustrated in FIG.
- the operator specifies a weighted average as a statistical function, and the weight of the weighted average is 0.4 (40%) for the estimation result by the learning model 1 (high-precision model), and the learning model 2 (high).
- the weight for the estimation result by the production model) is specified as 0.6 (60%).
- the weighted average is calculated using the designated weight, and the calculated weighted average is used as an estimated value for determining the state of the injection molding machine.
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)
- Human Computer Interaction (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Injection Moulding Of Plastics Or The Like (AREA)
Abstract
Description
すなわち、多種多様な生産環境やオペレータの要望に対応するため、複数の学習モデルより算出された「複数の状態判定結果(推定値)」を活用した総合的な判定、汎用的な判定を実現することが望まれている。
図1は本発明の一実施形態による状態判定装置の要部を示す概略的なハードウェア構成図である。本実施形態による状態判定装置1は、例えば制御用プログラムに基づいて産業機械を制御する制御装置として実装することができる。また、本実施形態による状態判定装置1は、制御用プログラムに基づいて産業機械を制御する制御装置に併設されたパソコンや、有線/無線のネットワークを介して制御装置と接続されたパソコン、セルコンピュータ、フォグコンピュータ6、クラウドサーバ7の上に実装することができる。本実施形態では、状態判定装置1を、ネットワーク9を介して制御装置3と接続されたパソコンの上に実装した例を示す。なお、本発明の状態判定装置がその状態判定の対象とする産業機械としては、射出成形機、工作機械、鉱山機械、木工機械、農業機械、建設機械などが例示される。以下では、産業機械の一例として、射出成形機を例に説明する。
このように、射出成形機の機種やスクリュ径など射出成形機の差異に対応した複数の学習モデルの推定結果を用いるように統計条件を定義することによって、学習モデルの作成に用いた射出成形機とは異なる種類の射出成形機の異常状態の判定を可能とする。
また、稼働状況や環境状況の差異に対応した複数の学習モデルの推定結果や、生産品の精度や生産効率の差異に対応した複数の学習モデルの推定結果を用いるように統計条件を定義することによって、多種多様な生産環境やオペレータの要望に対応した総合的な判定、汎用的な判定を可能とする。
次いで、推定指令部120は、データ抽出部110が抽出したデータを用いて、学習モデルa及び学習モデルbのそれぞれを用いた射出成形機の動作状態の推定処理を行うように機械学習装置2に対して指令する。
数値変換部140は、統計条件記憶部310を参照し、50t規模の射出成形機におけるスクリュ径が20mmである統計関数として、学習モデルaの推定結果に対して重み0.4、学習モデルbの推定結果に対して重み0.6とした加重平均関数を用いた演算を行う。例えば、学習モデルaによる異常度の推定結果が0.7、学習モデルbによる異常度の推定結果が0.5であった場合、数値変換部140は、0.4×0.7+0.6×0.5=0.58を、スクリュ径が20mmのスクリュを取り付けた50t規模の射出成形機の状態を判定するための統計量として出力する。こうして出力された統計量には学習モデルaと学習モデルbのそれぞれの推定結果が反映されているので、統計量によって射出成形機の総合的な異常度の判定が可能となる。また、予め設定されている異常度の閾値Theを統計量が超えている場合には、射出成形機の動作に異常が発生していると判定して警報を出力する。図8は、表示装置70に表示される画面に対して統計量を異常スコアとしてプロットし、警告のメッセージ「異常を検出しました。射出シリンダを点検して下さい。」を警報として出力した画面の表示例である。そして、射出成形機の運転を停止、減速したり、射出成形機の駆動部を駆動させる原動機の駆動トルクを制限したりするようにしても良い。
次いで、推定指令部は、データ抽出部110が抽出したデータを用いて、学習モデルa及び学習モデルbのそれぞれを用いた成形品の不良状態の推定処理を行うように機械学習装置2に対して指令する。
数値変換部140は、統計条件記憶部310を参照し、50t規模の射出成形機におけるスクリュ径が20mmである統計関数として、学習モデルaの推定結果に対して重み0.4、学習モデルbの推定結果に対して重み0.6とした加重平均関数を用いた演算を行う。例えば、学習モデルaによる成形品の不良状態を示すベクトル値の推定結果がya=(0.10,0.20,0.20,0.30,0.20)、学習モデルbによる成形品の不良状態のベクトル値を示す推定結果がyb=(0.20,0.10,0.30,0.20,0.20)であった場合、数値変換部140は、0.4×ya+0.6×yb=(0.16,0.14,0.26,0.24,0.20)を、成形品の不良状態を判定するための統計量として出力する。ここで、数値変換部140が出力した成形品の不良状態を示すベクトル値の中で最も大きな値である0.26はベクトル内で3番目に位置するので、成形品の不良状態(ヒケ:成形品が凹んだ不良、そり:残留応力による成形品の変形、焼け:成形品の変色、ボイド:空孔、クラック:成形品の割れやひび)としては、「焼け:成形品の変色」と判定される。また、予め設定されている成形品の不良状態の閾値Thbを統計量が超えている成形不良の種類がある場合には、成形不良が発生していると判定して警報を鳴らしても良い。
上記の実施形態では射出成形機を例に説明したが、状態判定の対象は他の産業機械であっても良い。例えば、工作機械では、主軸に組付ける切削工具、切削工具を冷却する加工液の種類や流量、ワーク材料、などに対応した複数の学習モデルより、主軸の異常度を判定しても良い。木工機械では、回転工具の種類、回転速度などに対応した複数の学習モデルより回転工具の異常度を判定しても良い。農業機械では、駆動部に掛かる駆動力、駆動部が備える機材、などに対応した複数の学習モデルより、駆動部の異常度を判定しても良い。建設機械や鉱山機械では、油圧シリンダに接続された油圧ホースの種類、原動機の出力、運転環境、などに対応した複数の学習モデルより、油圧シリンダの異常度を判定しても良い。
2 機械学習装置
3 制御装置
4 射出成形機
5 センサ
6 フォグコンピュータ
7 クラウドサーバ
9 ネットワーク
11 CPU
12 ROM
13 RAM
14 不揮発性メモリ
15,17,18,20,21 インタフェース
22 バス
70 表示装置
71 入力装置
72 外部機器
100 データ取得部
110 データ抽出部
120 推定指令部
140 数値変換部
150 学習指令部
201 プロセッサ
202 ROM
203 RAM
204 不揮発性メモリ
206 学習部
207 推定部
210 学習モデル記憶部
214 学習モデル
300 取得データ記憶部
310 統計条件記憶部
Claims (11)
- 産業機械に係る状態を判定する状態判定装置であって、
前記産業機械から取得した所定の物理量に係るデータと該産業機械に係る状態との相関性を学習した複数の学習モデルを記憶する学習モデル記憶部と、
前記産業機械に係る状態を判定する際に用いる複数の前記学習モデルの指定と、指定された前記学習モデルによる推定結果を数値変換して統計量を算出する統計関数とを少なくとも含む統計条件を記憶する統計条件記憶部と、
前記産業機械に係る状態を示すデータとしての所定の物理量に係るデータを取得するデータ取得部と、
前記データ取得部が取得したデータに基づいて、前記学習モデル記憶部に記憶された複数の前記学習モデルを用いて前記産業機械に係る状態を推定する推定部と、
前記統計条件記憶部を参照して前記統計関数を取得し、取得した前記統計関数を用いて、前記推定部による複数の学習モデル毎の推定結果を数値変換して統計量を算出する数値変換部と、
を備える状態判定装置。 - 前記統計条件は、状態を判定する対象となる前記産業機械の機種及び該産業機械に取り付けられた機材の少なくともいずれかと関連付けて作成され、
前記数値変換部は、状態を判定する対象となる前記産業機械の機種及び該産業機械に取り付けられた機材の少なくともいずれかに基づいて用いる統計関数を取得する、
請求項1に記載の状態判定装置。 - 前記統計条件は、前記産業機械の運転条件と関連付けて作成され、
前記数値変換部は、前記運転条件に基づいて用いる統計関数を取得する、
請求項2に記載の状態判定装置。 - 前記統計関数は、加重平均、算術平均、重み付き調和平均、調和平均、刈り込み平均、対数平均、二乗和平均平方根、最小値、最大値、中央値、加重中央値、最頻値のいずれか1つである、
請求項1に記載の状態判定装置。 - 前記統計関数のパラメータは、所定の固定値、または、所定の関数を用いて算出された数値である、
請求項1に記載の状態判定装置。 - 前記データ取得部が取得したデータを用いた機械学習を行い、学習モデルを生成または更新する学習部を更に備える、
請求項1に記載の状態判定装置。 - 前記統計条件記憶部に記憶された統計条件の要素をオペレータが編集可能なインタフェースを提供する、
請求項1に記載の状態判定装置。 - 前記推定部は、前記産業機械の動作状態に係る異常度を推定し、
前記数値変換部が前記統計関数を用いて前記推定部による複数の学習モデル毎の推定結果を数値変換して算出した統計量が予め定めた所定の閾値を超えた場合に警告メッセージを表示する、
請求項1に記載の状態判定装置。 - 前記推定部は、前記産業機械の動作状態に係る異常度を推定し、
前記数値変換部が前記統計関数を用いて前記推定部による複数の学習モデル毎の推定結果を数値変換して算出した統計量が予め定めた所定の閾値を超えた場合に、前記産業機械の運転を停止、減速、または前記産業機械を駆動する原動機の駆動トルクを制限する、
請求項1に記載の状態判定装置。 - 前記データ取得部が取得するデータは、有線または無線のネットワークによって接続された複数の産業機械から取得されるデータのうち少なくとも1つである、
請求項1に記載の状態判定装置。 - 産業機械に係る状態を判定する状態判定方法であって、
前記産業機械から取得した所定の物理量に係るデータと該産業機械に係る状態との相関性を学習した複数の学習モデルを予め記憶すると共に、
前記産業機械に係る状態を判定する際に用いる複数の前記学習モデルの指定と、指定された前記学習モデルによる推定結果を数値変換して統計量を算出する統計関数とを少なくとも含む統計条件を予め記憶しておき、
前記産業機械に係る状態を示すデータとしての所定の物理量に係るデータを取得するステップと、
前記取得するステップで取得したデータに基づいて、予め記憶してある複数の前記学習モデルを用いて前記産業機械に係る状態を推定するステップと、
予め記憶してある前記統計条件に含まれる前記統計関数を取得し、取得した前記統計関数を用いて、前記推定するステップによる複数の学習モデル毎の推定結果を数値変換して統計量を算出するステップと、
を実行する状態判定方法。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2022547601A JP7538234B2 (ja) | 2020-09-10 | 2021-09-07 | 状態判定装置及び状態判定方法 |
US18/024,820 US20240028020A1 (en) | 2020-09-10 | 2021-09-07 | State determination device and state determination method |
DE112021004168.1T DE112021004168T5 (de) | 2020-09-10 | 2021-09-07 | Zustandsbestimmungsvorrichtung und Zustandsbestimmungsverfahren |
CN202180054492.0A CN116034006B (zh) | 2020-09-10 | 2021-09-07 | 状态判定装置以及状态判定方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2020152256 | 2020-09-10 | ||
JP2020-152256 | 2020-09-10 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022054782A1 true WO2022054782A1 (ja) | 2022-03-17 |
Family
ID=80632134
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/032790 WO2022054782A1 (ja) | 2020-09-10 | 2021-09-07 | 状態判定装置及び状態判定方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US20240028020A1 (ja) |
JP (1) | JP7538234B2 (ja) |
CN (1) | CN116034006B (ja) |
DE (1) | DE112021004168T5 (ja) |
WO (1) | WO2022054782A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024084741A1 (ja) * | 2022-10-17 | 2024-04-25 | 株式会社日本製鋼所 | 情報処理方法、情報処理装置及びコンピュータプログラム |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018025945A (ja) * | 2016-08-09 | 2018-02-15 | 株式会社リコー | 診断装置、学習装置および診断システム |
JP2019138152A (ja) * | 2018-02-06 | 2019-08-22 | 株式会社日立製作所 | 機械制御装置 |
Family Cites Families (9)
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 |
JP2015176175A (ja) | 2014-03-13 | 2015-10-05 | 日本電気株式会社 | 情報処理装置、情報処理方法、およびプログラム |
US20160358099A1 (en) | 2015-06-04 | 2016-12-08 | The Boeing Company | Advanced analytical infrastructure for machine learning |
US10614830B2 (en) | 2016-01-14 | 2020-04-07 | National Institute Of Advanced Industrial Science And Technology | System, method, and computer program for estimation of target value |
JP6698603B2 (ja) | 2017-09-29 | 2020-05-27 | ファナック株式会社 | 数値制御システム、及び運転状態異常検知方法 |
JP6885321B2 (ja) * | 2017-12-22 | 2021-06-16 | Jfeスチール株式会社 | プロセスの状態診断方法及び状態診断装置 |
JP6928031B2 (ja) * | 2018-10-02 | 2021-09-01 | ファナック株式会社 | 制御装置及び制御システム |
JP7015261B2 (ja) * | 2019-02-13 | 2022-02-02 | ファナック株式会社 | 管理装置及び管理システム |
JP2021015573A (ja) | 2019-07-16 | 2021-02-12 | ファナック株式会社 | 異常判定装置及び異常判定システム |
-
2021
- 2021-09-07 US US18/024,820 patent/US20240028020A1/en active Pending
- 2021-09-07 WO PCT/JP2021/032790 patent/WO2022054782A1/ja active Application Filing
- 2021-09-07 CN CN202180054492.0A patent/CN116034006B/zh active Active
- 2021-09-07 JP JP2022547601A patent/JP7538234B2/ja active Active
- 2021-09-07 DE DE112021004168.1T patent/DE112021004168T5/de active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018025945A (ja) * | 2016-08-09 | 2018-02-15 | 株式会社リコー | 診断装置、学習装置および診断システム |
JP2019138152A (ja) * | 2018-02-06 | 2019-08-22 | 株式会社日立製作所 | 機械制御装置 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024084741A1 (ja) * | 2022-10-17 | 2024-04-25 | 株式会社日本製鋼所 | 情報処理方法、情報処理装置及びコンピュータプログラム |
Also Published As
Publication number | Publication date |
---|---|
CN116034006B (zh) | 2024-09-17 |
JP7538234B2 (ja) | 2024-08-21 |
CN116034006A (zh) | 2023-04-28 |
DE112021004168T5 (de) | 2023-06-29 |
US20240028020A1 (en) | 2024-01-25 |
JPWO2022054782A1 (ja) | 2022-03-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11628609B2 (en) | State determination device and method | |
JP6975188B2 (ja) | 状態判定装置及び状態判定方法 | |
JP6867358B2 (ja) | 状態判定装置及び状態判定方法 | |
JP6773738B2 (ja) | 状態判定装置及び状態判定方法 | |
JP6826086B2 (ja) | 状態判定装置及び状態判定方法 | |
JP7010861B2 (ja) | 状態判定装置及び状態判定方法 | |
WO2022075224A1 (ja) | 状態判定装置及び状態判定方法 | |
WO2022054782A1 (ja) | 状態判定装置及び状態判定方法 | |
CN110962315B (zh) | 状态判定装置以及状态判定方法 | |
JP7525631B2 (ja) | 状態判定装置及び状態判定方法 | |
JP7495513B2 (ja) | 状態判定装置及び状態判定方法 | |
JP7184997B2 (ja) | 状態判定装置及び状態判定方法 | |
JP7495514B2 (ja) | 状態判定装置及び状態判定方法 | |
WO2024057461A1 (ja) | 判定装置及び判定方法 | |
JP2021057064A (ja) | 状態判定装置及び状態判定方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21866741 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2022547601 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18024820 Country of ref document: US |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21866741 Country of ref document: EP Kind code of ref document: A1 |