JP2018004566A - Diagnosis device, diagnosis system, diagnosis method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a diagnosis device, diagnosis system, diagnosis method and program that can reduce a data volume of a learned model to be used for diagnosis of abnormality of a device, as keeping accuracy of diagnosing the abnormality thereof.SOLUTION: A diagnosis device comprises: a reception unit that receives context information corresponding to a current operation of a plural pieces of context information determined for each operation category of an object device, and detection information on a physical quantity including a frequency component varying in accordance with the operation of the object device from the object device; a first pre-processing unit that converts a frequency of the detection information into a preset reference frequency from the detection information received by the reception unit to obtain conversion information; and a determination unit that determines whether the operation of the object device is normal or not, using the conversion information converted by the first pre-processing unit, and a model corresponding to the context information received by the reception unit of one or more models to be determined with respect to each of one or more pieces of context information.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a diagnostic device, a diagnostic system, a diagnostic method, and a program.

  In a device having a rotating body such as a drill, for each operation state based on context information, a learned model learned from sensor data acquired in advance with an acceleration sensor or the like is constructed, and for sensor data to be determined, A technique for determining the presence or absence of an abnormality from a comparison with a model is already known.

  As a technique for determining the presence or absence of abnormality with respect to sensor data (for example, recorded sound) generated by a rotating body or the like in this way, a frequency at which frequency components due to abnormal sound can be observed based on operation mode information A technique for diagnosing whether or not an abnormality has occurred by comparison with a normal sound based on a predetermined abnormality index in consideration of turning back to a band has been proposed (see Patent Document 1).

  In addition, a technique has been proposed in which a learned model is constructed based on data prepared in advance to determine whether there is an abnormality (see Patent Document 2).

  However, the technique described in Patent Document 1 has a problem that an abnormality index must be determined in advance, and if the abnormality index cannot be defined well, the presence / absence of abnormality cannot be correctly determined.

  In the technique described in Patent Document 2, it is not necessary to construct a learned model based on data prepared in advance and to clearly define an abnormality index. However, a large number of models serving as criteria for determination are prepared in advance. For example, for the estimation of the deterioration state of the processing tool of the processing machine, a learned model must be prepared for each different rotation speed, and the capacity of the storage device that stores these many models There was a problem of squeezing.

  The present invention has been made in view of the above-described problems, and is a diagnostic device capable of reducing the amount of data of a learned model used for diagnosing an abnormality while maintaining the accuracy of diagnosing the abnormality of the device. An object of the present invention is to provide a diagnostic system, a diagnostic method, and a program.

  In order to solve the above-described problems and achieve the object, the present invention relates to context information corresponding to the current operation among a plurality of pieces of context information determined for each type of operation of the target device, and the operation of the target device. The detection information of the physical quantity including the frequency component that changes according to the reception unit that receives from the target device, and the detection information received by the reception unit, the frequency of the detection information is converted to a preset reference frequency A first preprocessing unit for obtaining conversion information, conversion information converted by the first preprocessing unit, and context information received by the receiving unit among one or more models defined for each of the one or more context information And a determination unit that determines whether or not the operation of the target device is normal using a model corresponding to the above.

  ADVANTAGE OF THE INVENTION According to this invention, the data amount of the learned model used in order to diagnose abnormality can be reduced, maintaining the precision which diagnoses abnormality of an apparatus.

FIG. 1 is a diagram illustrating a configuration example of a diagnostic system according to the first embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of the processing machine according to the first embodiment. FIG. 3 is a diagram illustrating an example of a hardware configuration of the diagnostic apparatus according to the first embodiment. FIG. 4 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the first embodiment. FIG. 5 is a flowchart illustrating an example of a diagnosis process according to the first embodiment. FIG. 6A is a graph illustrating an example of raw data of detection information. FIG. 6B is a graph illustrating an example of data obtained by converting the detection information into the reference rotation frequency. FIG. 7A is a graph illustrating an example of a frequency spectrum of detection information. FIG. 7-2 is a graph illustrating an example of a frequency spectrum of detection information. FIG. 7-3 is a graph illustrating an example of a frequency spectrum of data obtained by converting detection information into a reference rotation frequency. FIG. 8 is a flowchart illustrating an example of model generation processing according to the first embodiment. FIG. 9 is a diagram illustrating a specific example of the model generation process and the diagnosis process in the first embodiment. FIG. 10 is a diagram for explaining a specific example in which model generation processing and diagnosis processing are performed for some context information. FIG. 11 is a diagram illustrating an example in which a common model is used in other processing steps. FIG. 12 is a diagram for explaining an example in which the accumulated usage time is used as context information. FIG. 13 is a diagram illustrating an operation for generating a model when context information for which a model has not been generated is input. FIG. 14 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the second embodiment. FIG. 15 is a diagram illustrating that detection information that differs depending on the drive unit is set as a determination target. FIG. 16 is a diagram illustrating an example of a data structure of correspondence information used for determination of detection information. FIG. 17 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the third embodiment. FIG. 18 is a diagram illustrating an example of an operation for generating a model from detection information converted into a plurality of reference rotation frequencies. FIG. 19 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the fourth embodiment. FIG. 20 is a diagram illustrating an example of the relationship between the context information and the processing section. FIG. 21 is a diagram illustrating an example of a method for specifying a machining section. FIG. 22 is a graph illustrating an example of the relationship between the likelihood and the determination value r (k).

  Hereinafter, embodiments of a diagnostic apparatus, a diagnostic system, a diagnostic method, and a program according to the present invention will be described in detail with reference to the drawings. In addition, the present invention is not limited by the following embodiments, and constituent elements in the following embodiments are easily conceivable by those skilled in the art, substantially the same, and so-called equivalent ranges. Is included. Furthermore, various omissions, substitutions, changes, and combinations of the constituent elements can be made without departing from the scope of the following embodiments.

[First Embodiment]
(Configuration of diagnostic system)
FIG. 1 is a diagram illustrating a configuration example of a diagnostic system according to the first embodiment. The configuration of the diagnostic system according to the present embodiment will be described with reference to FIG.

  As shown in FIG. 1, the diagnostic system includes a processing machine 200 (an example of a target device) and a diagnostic device 100. The processing machine 200 is an example of a target device to be diagnosed by the diagnostic device 100.

  The processing machine 200 and the diagnostic apparatus 100 may be connected in any connection form. For example, the processing machine 200 and the diagnostic apparatus 100 are connected by a dedicated line, a wired network such as a wired LAN (Local Area Network), a wireless network, or the like.

  The processing machine 200 includes a numerical control unit 201, a communication control unit 202, a drive control unit 203, a drive unit 204, and a detection unit 211. The detection unit 211 may be provided in the processing machine 200 in advance, or may be attached later to the processing machine 200 that is a completed machine.

  The processing machine 200 is a machine that processes a processing target in accordance with the control of the numerical control unit 201. The processing machine 200 includes a drive unit 204 that operates under the control of the numerical control unit 201. The drive unit 204 is, for example, a motor or the like, and may be any device as long as it is used for machining and is subject to numerical control. Two or more drive units 204 may be provided.

  The numerical control unit 201 performs processing by the driving unit 204 by numerical control (NC: Numerical Control). For example, the numerical control unit 201 generates and outputs numerical control data for controlling the operation of the driving unit 204. The numerical control unit 201 also outputs context information to the communication control unit 202. Here, the context information is information determined in plural for each type of operation of the processing machine 200. The context information includes, for example, information for identifying the drive unit 204, the number of rotations of the drive unit 204, the rotation speed of the drive unit 204, the load related to the drive unit 204, the size of the drive unit 204, and the start of use of the drive unit 204. The cumulative usage time from is included.

  For example, the numerical control unit 201 transmits context information indicating the current operation to the diagnostic apparatus 100 via the communication control unit 202. The numerical control unit 201 changes the type of the driving unit 204 to be driven and the driving state (the rotational speed, the rotation speed, etc.) of the driving unit 204 according to the processing step when processing the processing target. Each time the operation control type is changed, the numerical control unit 201 sequentially transmits context information corresponding to the changed operation type to the diagnostic apparatus 100 via the communication control unit 202.

  The communication control unit 202 controls communication with an external device such as the diagnostic device 100. For example, the communication control unit 202 transmits context information corresponding to the current operation to the diagnostic apparatus 100.

  The drive control unit 203 controls the drive of the drive unit 204 based on the numerical control data obtained by the numerical control unit 201.

  The detection unit 211 detects a physical quantity that changes in accordance with the operation of the processing machine 200 and outputs detection information (sensor data). In addition, the kind of the detection part 211 and the physical quantity to detect may be what. For example, the detection unit 211 may be a microphone, an acceleration sensor, or an AE (Acoustic Emission) sensor, and acoustic data, acceleration data, vibration data, or data indicating an AE wave may be detection information. Further, the number of detection units 211 may be arbitrary. Moreover, the some detection part 211 which detects the same physical quantity may be provided, and the some detection part 211 which detects a mutually different physical quantity may be provided.

  For example, when bending of a blade used for processing, chipping of a blade, or the like occurs, sound during processing changes. For this reason, it is possible to detect an abnormality in the operation of the processing machine 200 by detecting acoustic data with the detection unit 211 (microphone) and comparing it with a model or the like showing normal sound.

  The diagnostic device 100 includes a communication control unit 101 and a determination unit 102.

  The communication control unit 101 controls communication with an external device such as the processing machine 200. For example, the communication control unit 101 receives context information and detection information from the processing machine 200. The determination unit 102 refers to the context information and the detection information to determine whether the operation of the processing machine 200 is normal. Details of the function of each unit will be described later.

(Processing machine hardware configuration)
FIG. 2 is a diagram illustrating an example of a hardware configuration of the processing machine according to the first embodiment. The hardware configuration of the processing machine 200 according to the present embodiment will be described with reference to FIG.

  As shown in FIG. 2, the processing machine 200 includes a CPU (Central Processing Unit) 51, a ROM (Read Only Memory) 52, a RAM (Random Access Memory) 53, a communication I / F (interface) 54, and a drive. The control circuit 55 and the motor 56 are connected by a bus 58.

  The CPU 51 controls the entire processing machine 200. For example, the CPU 51 executes a program stored in the ROM 52 or the like using the RAM 53 as a work area (working area), thereby controlling the operation of the entire processing machine 200 and realizing a processing function.

  The communication I / F 54 is an interface for communicating with an external device such as the diagnostic device 100. The drive control circuit 55 is a circuit that controls the drive of the motor 56. The motor 56 drives tools used for processing such as a drill, a cutter, and a table. The motor 56 corresponds to, for example, the drive unit 204 illustrated in FIG. The sensor 57 detects a physical quantity that changes according to the operation of the processing machine 200, and outputs detection information to the diagnostic apparatus 100. The sensor 57 corresponds to, for example, the detection unit 211 illustrated in FIG.

  The numerical control unit 201 and the communication control unit 202 shown in FIG. 1 may be realized by software by causing the CPU 51 shown in FIG. 2 to execute a program, that is, by hardware such as an IC (Integrated Circuit). You may implement | achieve and it may implement | achieve combining software and hardware.

(Hardware configuration of diagnostic device)
FIG. 3 is a diagram illustrating an example of a hardware configuration of the diagnostic apparatus according to the first embodiment. A hardware configuration of the diagnostic apparatus 100 according to the present embodiment will be described with reference to FIG.

  As illustrated in FIG. 3, the diagnostic device 100 includes a CPU 61, a ROM 62, a RAM 63, a communication I / F 64, an input / output I / F 65, an input device 66, a display 67, and an auxiliary storage device 68. It is configured to be connected by a bus 69.

  The CPU 61 controls the entire diagnostic device 100. For example, the CPU 61 executes a program stored in the ROM 62 or the like using the RAM 63 as a work area (working area), thereby controlling the operation of the entire diagnostic apparatus 100 and realizing a diagnostic function.

  The communication I / F 64 is an interface for communicating with an external device such as the processing machine 200.

  The input / output I / F 65 is an interface for connecting various devices (for example, the input device 66 and the display 67) and the bus 69.

  The input device 66 is an input device such as a mouse or a keyboard for inputting characters and numbers, selecting various instructions, and moving a cursor.

  The display 67 is a display device such as an LCD (Liquid Crystal Display), a plasma display, or an organic EL (Electro-Luminescence) display that displays various information such as a cursor, a menu, a window, characters or images.

  The auxiliary storage device 68 includes setting information of the diagnostic device 100, detection information received from the processing machine 200, an OS (Operating System), an application program, and various data, an HDD (Hard Disk Drive), an SSD (Solid State Drive). ) Or an EEPROM (Electrically Erasable Programmable Read-Only Memory) or the like. The auxiliary storage device 68 is included in the diagnostic device 100, but is not limited to this, and may be, for example, a storage device installed outside the diagnostic device 100, or the diagnostic device. 100 may be a storage device included in a server device capable of data communication with 100.

(Configuration of functional block of diagnostic device)
FIG. 4 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the first embodiment. With reference to FIG. 4, the functional block configuration of the diagnostic apparatus 100 according to the present embodiment will be described.

  As shown in FIG. 4, in addition to the communication control unit 101 and the determination unit 102 described above, the diagnostic device 100 includes a reception unit 103, a feature extraction unit 104, a generation unit 105, and a preprocessing unit 106 (first preprocessing). Part) and a storage unit 111.

  The storage unit 111 stores various types of information necessary for the diagnostic function of the diagnostic apparatus 100. The storage unit 111 can be realized by, for example, the RAM 63 and the auxiliary storage device 68 shown in FIG. For example, the storage unit 111 stores one or more models used for abnormality determination.

  The model is generated by learning using, for example, detection information detected by the detection unit 211 when the processing machine 200 is operating normally. The learning method and the model of the learning model may be any method. For example, models such as GMM (Gaussian mixture model) and HMM (Hidden Markov model) and corresponding model learning methods can be applied.

  In the present embodiment, a model is generated for each context information. For example, the storage unit 111 stores context information and a model corresponding to the context information in association with each other. Further, as will be described later, as a model corresponding to certain context information stored in the storage unit 111, a model generated from detection information of a reference rotation frequency that is a specific rotation frequency among the detection information detected by the detection unit 211 described above. Is remembered.

  The communication control unit 101 includes a reception unit 101a and a transmission unit 101b. The receiving unit 101a receives various types of information transmitted from an external device such as the processing machine 200. For example, the receiving unit 101a receives the context information corresponding to the current operation of the processing machine 200 and the detection information transmitted by the detection unit 211. The transmission unit 101b transmits various types of information to the external device.

  The preprocessing unit 106 performs conversion processing (preprocessing) according to a predetermined criterion before transmitting the detection information received by the receiving unit 101a to the feature extraction unit 104. There may be a plurality of standards for conversion. For example, if the detection information is vibration data during cutting by the drill of the processing machine 200, the preprocessing unit 106 determines the rotation frequency of the drill being processed obtained from the context information. Of the one or more predetermined reference rotation frequencies, the data is converted into data (hereinafter, referred to as “conversion information” in some cases) that has a reference rotation frequency with little change from the original vibration data. At this time, vibration data for model creation is prepared as learning data of a predetermined reference rotation frequency, a model is created based on the learning data, and the preprocessing unit 106 determines vibration data to be determined. Is converted to match the reference rotation frequency of the prepared learning data. The preprocessing unit 106 obtains the rotation frequency of the drill or the like being processed from the context information. However, the present invention is not limited to this, and the rotation frequency may be obtained by analyzing the detection information.

  The feature extraction unit 104 extracts feature information used for generating the model from the detection information, and extracts feature information used for determination by the determination unit 102 from the conversion information received from the preprocessing unit 106. The feature information may be any information as long as the information indicates the feature of the detection information. For example, when the detection information is acoustic data collected by a microphone, the feature extraction unit 104 may extract energy, frequency spectrum, MFCC (Mel Frequency Cepstrum Coefficient), and the like as feature information.

  The generation unit 105 creates a model for determining normal operation by learning using feature information extracted from detection information during normal operation. When the model is generated by an external device, the generation unit 105 may not be provided. When the context information for which the model is not defined and the detection information corresponding to the context information are input, the generation unit 105 uses the feature information extracted from the detection information, and uses the model corresponding to the context information. May be generated.

  The determination unit 102 uses the conversion information converted from the detection information by the preprocessing unit 106, the context information received by the reception unit 101a, and the model corresponding to the reference rotation frequency converted from the detection information. It is determined whether or not the operation of the machine 200 is normal. For example, the determination unit 102 requests the feature extraction unit 104 to extract feature information from the conversion information. The determination unit 102 calculates a likelihood indicating the likelihood that the feature information extracted from the conversion information is normal using a corresponding model. The determination unit 102 compares the likelihood with a predetermined threshold, and determines that the operation of the processing machine 200 is normal when the likelihood is equal to or greater than the threshold, for example. Moreover, the determination part 102 determines with the operation | movement of the processing machine 200 being abnormal, when likelihood is less than a threshold value.

  Note that the method for determining whether the operation of the processing machine 200 is normal is not limited to this, and any method that can determine whether the operation of the processing machine 200 is normal using the conversion information and the model. Any method may be used. For example, instead of directly comparing the likelihood value with the threshold value, a value indicating a variation in likelihood may be compared with the threshold value.

  The accepting unit 103 accepts input of context information different from the context information received by the receiving unit 101a from the processing machine 200. For example, the accumulated usage time can be obtained from the processing machine 200. In this case, for example, the processing machine 200 may have a function of resetting (initializing) the accumulated usage time when the tool is replaced.

  Note that the reception unit 103 can receive the accumulated usage time without acquiring the accumulated usage time from the processing machine 200. The accepting unit 103 accepts context information input from an operation unit such as a keyboard and a touch panel, for example. The context information received by the receiving unit 103 is not limited to the cumulative usage time, but includes, for example, information on the specification of the tool to be used (blade diameter, number of blades, material, whether the tool is coated, etc.) Information on the material to be performed (material, etc.) may be used. The receiving unit 103 may be configured to receive context information from an external device such as a server device and a PC (Personal Computer). When it is not necessary to receive context information from other than the processing machine 200, the receiving unit 103 may not be provided.

  4 (communication control unit 101, determination unit 102, reception unit 103, feature extraction unit 104, generation unit 105, preprocessing unit 106) causes CPU 61 in FIG. 3 to execute a program, that is, software. May be realized by hardware such as an IC, or may be realized by using software and hardware together.

  In addition, each functional unit (communication control unit 101, reception unit 101a, transmission unit 101b, determination unit 102, reception unit 103, feature extraction unit 104, generation unit 105, preprocessing unit 106, diagnostic device 100 illustrated in FIG. The storage unit 111) conceptually shows functions, and is not limited to such a configuration. For example, a plurality of functional units illustrated as independent functional units in FIG. 4 may be configured as one functional unit. On the other hand, the function of one functional unit in FIG. 4 may be divided into a plurality of functions and configured as a plurality of functional units.

(Diagnosis process by diagnostic device)
FIG. 5 is a flowchart illustrating an example of a diagnosis process according to the first embodiment. FIG. 6A is a graph illustrating an example of raw data of detection information. FIG. 6B is a graph illustrating an example of data obtained by converting the detection information into the reference rotation frequency. 7A and 7B are graphs illustrating an example of the frequency spectrum of the detection information. FIG. 7-3 is a graph illustrating an example of a frequency spectrum of data obtained by converting detection information into a reference rotation frequency. With reference to FIGS. 5, 6-1, 6-2, and FIGS. 7-1 to 7-3, diagnostic processing by the diagnostic device 100 according to the present embodiment will be described.

  As described above, the numerical control unit 201 of the processing machine 200 sequentially transmits context information indicating the current operation to the diagnostic apparatus 100. The receiving unit 101a receives the context information thus transmitted from the processing machine 200 (step S101). Further, the detection unit 211 of the processing machine 200 sequentially outputs detection information at the time of processing. The receiving unit 101a receives the detection information (sensor data) thus transmitted from the processing machine 200 (step S102).

  The preprocessing unit 106 performs conversion processing (preprocessing) according to a predetermined criterion before transmitting the detection information received by the receiving unit 101a to the feature extraction unit 104. That is, the pre-processing unit 106 sets the detection information as the reference rotation frequency with a small change from the original detection information among one or more reference rotation frequencies determined in advance as the rotation frequency of the drive unit 204 obtained from the context information. (Step S103).

  Here, the graph shown in FIG. 6A shows raw data of detection information received by the receiving unit 101a. Among these, FIG. 6-1 (a) shows detection information with a rotational frequency of 25 [Hz], and FIG. 6-1 (b) shows detection information with a rotational frequency of 2.8 [Hz]. . And the graph shown to FIGS. 6-2 has shown the conversion information at the time of converting the detection information shown to FIGS. 6-1 into 3 [Hz] which is an example of a reference | standard rotation frequency by the pre-processing part 106. FIG. Among these, FIG. 6-2 (a) shows the conversion information which converted the detection information whose rotation frequency is 25 [Hz] into 3 [Hz] which is a reference | standard rotation frequency, FIG. The conversion information which converted the detection information whose rotation frequency is 2.8 [Hz] into 3 [Hz] which is a reference rotation frequency is shown. That is, the conversion information shown in FIG. 6-2 (a) is up-sampled information that lowers the rotation frequency of the original detection information, and the conversion information shown in FIG. 6-2 (b) is a rotation of the original detection information. The information is down-sampled to increase the frequency, and is converted without changing the sampling rate. Moreover, since all the conversion information shown to FIGS. 6-2 is the information converted into 3 [Hz], the period of a vibration is substantially in agreement.

  FIGS. 7-1 to 7-3 respectively show detection information (FIG. 6-1 (a)) that is raw data of 25 [Hz] (detection information that is raw data of 2.8 [Hz] (FIG. 6-1). 6-1 (b)) and the power spectrum (frequency spectrum) of the conversion information (FIG. 6-2 (a)) converted from the detection information of 25 [Hz] to 3 [Hz]. The reason why the scale of the vertical axis of the power spectrum in FIG. 7-3 is large is that the amplitude is normalized. Further, a comb-like spectrum appears in which the power spectrum after conversion (FIG. 7-3) shows a periodic amplitude rather than the power spectrum before conversion of 25 [Hz] (FIG. 7-1). As shown in FIG. 7-2, the power spectrum is approaching. Therefore, the detection information of 25 [Hz] is detected as described above, rather than the determination of abnormality using the model corresponding to the reference rotation frequency of 3 [Hz] with respect to the detection information of 25 [Hz]. It is more accurate to determine abnormality using a model corresponding to the reference rotation frequency of 3 [Hz] with respect to the conversion information converted to be 3 [Hz] of the reference rotation frequency. it can.

  Further, in the above-described example, detection information of 25 [Hz] is converted into 3 [Hz] conversion information, so upsampling is performed about 8 times. However, since the original 25 [Hz] sensor data (detection information) includes vibration components that do not depend on a rotating body or the like driven by the drive unit 204, an unnatural signal can be obtained by conversion. There is. Therefore, in order not to perform up-sampling and down-sampling that greatly change the original sensor data (detection information), a plurality of reference rotation frequencies may be set, and the preprocessing unit 106 may include a plurality of reference You may make it convert into the reference | standard rotation frequency with the smallest change from the original detection information among rotation frequencies.

  For example, it is assumed that reference rotation frequencies of 5 [Hz], 10 [Hz], 20 [Hz], and 40 [Hz] are set in advance as a plurality of reference rotation frequencies. Then, the preprocessing unit 106 optimizes based on the index value f (x, y) (the ratio of the rotation frequency to the reference rotation frequency or the ratio of the reference rotation frequency to the rotation frequency) obtained by the following equation (1). Select the correct reference rotation frequency.

f (x, y) = x / y (when x <y)
f (x, y) = y / x (when x ≧ y) (1)

  In this formula (1), x is the rotation frequency of the detection information (sensor data) received by the receiving unit 101a, and y is the reference rotation frequency. Here, when the rotation frequency of the detection information is 25 [Hz], the index values f (x, y) corresponding to the four reference rotation frequencies described above are obtained as follows.

f (25,5) = 5/25 = 0.2
f (25,10) = 10/25 = 0.4
f (25,20) = 20/25 = 0.8
f (25,40) = 25/40 = 0.625

  Of the index values corresponding to the respective reference rotation frequencies calculated in this way, the reference rotation frequency corresponding to the maximum index value is selected as the rotation frequency to be converted. That is, it is determined that the conversion information converted by the reference rotation frequency thus selected has the smallest change from the original detection information. Here, since the index value 0.8 is the maximum, the preprocessing unit 106 selects the reference rotation frequency 20 [Hz], and converts the rotation frequency of the detection information to be the reference rotation frequency.

  As described above, other than the method of selecting the reference rotation frequency based on the index value according to Equation (1), for example, the reference rotation frequency closest to the rotation frequency of the detection information may be simply selected. .

  The feature extraction unit 104 extracts feature information from the conversion information converted by the preprocessing unit 106 (step S104). Whether the processing machine 200 is operating normally using the extracted feature information, the received context information, and the model corresponding to the reference rotation frequency of the conversion information obtained by converting the detection information. It is determined whether or not (step S105). The determination unit 102 outputs the determination result (step S106). Any method of outputting the determination result may be used. For example, the determination unit 102 may display the determination result on the display 67 of the diagnostic apparatus 100. Alternatively, the determination unit 102 may output the determination result to the server device and an external device such as a PC.

  In addition, the generation source of the detection information that is the target of the process of converting the rotation frequency described above may be a tool such as a drill, a grindstone, an end mill, or a milling cutter, or a rotating body such as a bearing instead of a tool that processes a member. There may be.

(Model generation processing by diagnostic equipment)
FIG. 8 is a flowchart illustrating an example of model generation processing according to the first embodiment. With reference to FIG. 8, a model generation process performed by the diagnostic apparatus 100 according to the present embodiment will be described. The model generation process is executed in advance before the diagnosis process, for example. Or you may comprise so that a model production | generation process may be performed when the context information in which the model is not defined is input. Further, when the model is generated outside as described above, the model generation process may not be executed.

  The receiving unit 101a receives the context information transmitted from the processing machine 200 (step S201). The receiving unit 101a receives the detection information (sensor data) transmitted from the processing machine 200 (step S202).

  The context information and detection information received in this way are used to generate a model. Since the model is generated for each context information and for each reference rotation frequency, the detection information needs to be associated with the corresponding context information and the reference rotation frequency. That is, in order to generate a model corresponding to the reference rotation frequency, the reception unit 101a needs to receive detection information whose rotation frequency is the reference rotation frequency. Therefore, for example, the receiving unit 101a stores the received detection information in the storage unit 111 or the like in association with the context information received at the same timing and the reference rotation frequency. Each information may be temporarily stored in the storage unit 111 or the like, confirmed as normal information, and a model may be generated using only normal information. That is, the model may be generated using detection information labeled as normal.

  The confirmation (labeling) of whether or not it is normal may be executed at an arbitrary timing after information is stored in the storage unit 111 or the like, or may be executed in real time while the processing machine 200 is operating. . The model may be generated on the assumption that the information is normal without performing labeling. If the information assumed to be normal is actually abnormal, the generated model will not be correctly determined. Whether such a situation is present can be determined based on the frequency at which it is determined that there is an abnormality, and it is possible to take measures such as deleting an erroneously generated model. In addition, a model generated from information that is abnormal may be used as a model for determining abnormality.

  The feature extraction unit 104 extracts feature information from the collected detection information (step S203). The generation unit 105 generates a model for the context information using the feature information extracted from the detection information associated with the same context information (step S204). In this case, the generation unit 105 may add label data to the generated model. For example, as label data, “0” is given to a model generated from detection information from a new tool, and “1” is given to a model generated from detection information from a tool with advanced wear. It is good.

  The generation unit 105 stores the generated model in the storage unit 111, for example, in association with the context information and the reference rotation frequency (step S205).

  As described above, since the model is generated using the detection information that is the raw data adjusted to the reference rotation frequency, it is possible to generate a model in which mixing of errors is suppressed. However, if it is difficult to receive the detection information by adjusting the rotation frequency of the detection information to the reference rotation frequency, the preprocessing unit 106 converts the rotation frequency of the received detection information into the reference rotation frequency, It may be generated. As a result, although the model includes a certain amount of error, as described above, abnormality determination is performed using the model generated with the detection information of the different rotation frequency with respect to the conversion information converted into the specific reference rotation frequency. The determination can be made with higher accuracy.

(Specific examples of model generation processing and diagnosis processing)
FIG. 9 is a diagram illustrating a specific example of the model generation process and the diagnosis process in the first embodiment. FIG. 10 is a diagram for explaining a specific example in which model generation processing and diagnosis processing are performed for some context information. FIG. 11 is a diagram illustrating an example in which a common model is used in other processing steps. FIG. 12 is a diagram for explaining an example in which the accumulated usage time is used as context information. FIG. 13 is a diagram illustrating an operation for generating a model when context information for which a model has not been generated is input. A specific example of the model generation process and the diagnosis process performed by the diagnosis apparatus 100 according to the present embodiment will be described with reference to FIGS.

  FIG. 9 shows, for example, a model generation process and a diagnosis process for a part of a process for processing a certain part. In the model generation process, a plurality of detection information (detection information 711a to 711c in FIG. 9) received together with the context information 701 is used. Note that the number of pieces of detection information is not limited to three and may be any number.

  The context information 701 indicates that the machining process includes an operation of driving four motors (motor A, motor B, motor C, and motor D). The feature extraction unit 104 extracts feature information from the received detection information. The generation unit 105 generates a model using the feature information extracted from the corresponding detection information for each context information corresponding to each motor and for each set reference rotation frequency. The generated model is stored in the storage unit 111 or the like. FIG. 9 shows an example in which a model (“motor B”) for each reference rotation frequency generated for the context information when the motor B is driven is stored in the storage unit 111. The stored model is referred to in subsequent diagnosis processing.

  In the diagnosis process, the detection information 721 is received together with the context information 701 as in the model generation process. When the context information indicates that “the motor B is being driven”, the preprocessing unit 106 selects the rotation frequency of the detection information 721 similarly indicated by the context information from one or more reference rotation frequencies by the above method. Conversion information is generated by converting the rotation frequency. Then, for example, the determination unit 102 converts the conversion information converted from the detection information received during the period in which the context information is received, and the model corresponding to the reference rotation frequency of the conversion information stored in the storage unit 111. Using “Motor B”, it is determined whether or not the operation of the processing machine 200 is normal.

  Similarly, when other context information is received, determination by the determination unit 102 is performed using conversion information converted from corresponding detection information and a model corresponding to the reference rotation frequency of the conversion information. .

  Note that it is not necessary to perform determination on all context information. For example, FIG. 10 is a diagram illustrating an example in which determination is performed on some context information.

  In the example of FIG. 10, the model is generated only when the context information indicates “motor B is driven”. The diagnosis process is executed when context information 701-2 indicating "motor B is driven" is received. As a result, the diagnosis process can be executed using only the detection information effective for determining the abnormality. For example, when using acoustic data as detection information, a section that does not need to be determined, such as a silent section, may be included in the machining process. By excluding such unnecessary sections from the determination target, it is possible to reduce erroneous determinations and to reduce calculation costs. That is, it is possible to achieve high accuracy and efficiency of the diagnostic processing.

  Further, even if the machining steps are different from each other, for example, when the same motor or the like is used, the diagnosis process may be executed by using the corresponding model in common. For example, FIG. 11 is a diagram illustrating an example in which a common model is used in other processing steps.

  The context information 901 in FIG. 11 indicates that the machining process includes an operation of driving four motors (motor X, motor Y, motor Z, and motor B). The point that the motor B is used is, for example, common to the processing steps shown in FIG. For this reason, also in the machining process of FIG. 11, the determination unit 102 can execute a diagnosis process using the same model “motor B” and the conversion information obtained by converting the detection information 921.

  FIG. 12 is a diagram illustrating an example in which the accumulated use time is used as context information. The context information 1001 indicates the accumulated usage time of each motor. In the example of FIG. 12, a model is generated for each cumulative usage time range (from 0 to June, from June to December,...), For each motor type, and for each reference rotation frequency. For example, the storage unit 111a stores a model of each motor whose accumulated usage time is 0 to June. The memory | storage part 111b memorize | stores the model of each motor whose accumulation | storage use time is 6 to December. The memory | storage part 111c memorize | stores the model of each motor whose accumulation use time is 12 months or more. The storage units 111a to 111c may be realized by physically different storage media, or may be realized by the physically same storage medium (storage unit 111). The determination unit 102 specifies a corresponding model using the context information 1001 together with the context information 701, and executes a diagnosis process using the specified model.

  FIG. 13 is a diagram illustrating an example of generating a model when context information for which a model is not defined is input. The context information 1101 in FIG. 13 indicates that the machining process includes an operation of driving four motors (motor X, motor C, motor D, and motor B). It is assumed that no model has been generated for the context information indicating the driving of the motor X.

  In this case, the feature extraction unit 104 corresponds to context information indicating that “the motor X is driven” for each of a plurality of pieces of detection information having the reference rotation frequency (in FIG. 11, detection information 1111a, 1111b, and 1111c). The feature information is extracted from the detection information of the period to be used. Using the extracted feature information, the generation unit 105 generates a model “motor X” corresponding to the context information indicating “motor X is being driven” and stores the model “motor X” in the storage unit 111. As a result, it is possible to determine whether the motor X is to be driven in the future.

  As described above, in the first embodiment, the detection information obtained from the detection unit 211 is converted into data of a predetermined reference rotation frequency to be converted into conversion information, and the reference rotation frequency is converted to the conversion information. It is assumed that an abnormality is determined using a model corresponding to. As a result, the abnormality is determined by comparing the conversion information corresponding to the same rotation frequency with the model, so that the accuracy of diagnosing the abnormality of the apparatus (for example, the processing machine 200) can be maintained or improved. Furthermore, as a model to be stored in the storage unit 111 or the like, it is not necessary to store a model corresponding to any rotation frequency, and a model corresponding to one or more reference rotation frequencies may be stored. The amount of model data stored in the unit 111 can be reduced.

  In the first embodiment, context information indicating the current operation is received from the processing machine 200, and an abnormality can be determined using a model corresponding to the received context information. Therefore, it is possible to specify the drive unit that operates with high accuracy and to diagnose the abnormality with higher accuracy.

[Second Embodiment]
In 1st Embodiment, it was determined whether it was normal using one type of detection information. The number of detection information used for determination is not limited to 1, and may be 2 or more. The diagnosis system of the second embodiment determines an abnormality of the processing machine 200 using a plurality of detection information.

(Configuration of functional block of diagnostic device)
FIG. 14 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the second embodiment. With reference to FIG. 14, the functional block configuration of the diagnostic apparatus 100-2 according to the present embodiment will be described.

  Since the configuration of the diagnosis system of the second embodiment is the same as that of FIG. 1 showing the configuration of the first embodiment, description thereof is omitted. As illustrated in FIG. 14, the diagnostic device 100-2 includes a communication control unit 101, a determination unit 102-2, a reception unit 103, a feature extraction unit 104, a generation unit 105, and a preprocessing unit 106-2 ( A first preprocessing unit) and a storage unit 111.

  In the second embodiment, the functions of the determination unit 102-2 and the preprocessing unit 106-2 are different from those in the first embodiment. Other configurations and functions are the same as those in FIG. 4 which is a block diagram of the diagnostic apparatus 100 according to the first embodiment, and thus the same reference numerals are given and description thereof is omitted here.

  The preprocessing unit 106-2 obtains conversion information by converting each of the plurality of pieces of detection information into a reference rotation frequency according to the context information. In this case, the reference rotation frequency corresponds to the context information, and which reference rotation frequency is selected from one or more reference rotation frequencies corresponding to the context information is as described in the first embodiment. is there.

  The determination unit 102-2 determines whether the operation of the processing machine 200 is normal using the conversion information converted from each of the plurality of detection information. For example, the determination unit 102-2 switches conversion information used for determination according to the context information.

(Specific example of processing according to this embodiment)
FIG. 15 is a diagram illustrating that detection information that differs depending on the drive unit is set as a determination target. FIG. 16 is a diagram illustrating an example of a data structure of correspondence information used for determination of detection information. A specific example of processing according to the present embodiment will be described with reference to FIGS. 15 and 16.

  In the example of FIG. 15, context information 1201 and a plurality of types of detection information 1221a and 1221b are received. The detection information 1221a is acceleration data, for example. The detection information 1221b is, for example, acoustic data.

  When the context information indicates that “the motor A is being driven”, the preprocessing unit 106-2 sets the rotation frequency of the detection information 1222b in the period corresponding to the context information in the detection information 1221b. Conversion information is obtained by converting to a reference rotation frequency corresponding to. Further, when the context information indicates “motor B is being driven”, the preprocessing unit 106-2 sets the rotation frequency of the detection information 1222a in the period corresponding to the context information in the detection information 1221a. Conversion information is obtained by converting the reference rotation frequency corresponding to the context information.

  When the context information indicates “motor A is driven”, the determination unit 102-2 is extracted from the conversion information converted from the detection information 1222b in the period corresponding to the context information in the detection information 1221b. The determination is executed using the feature information. At this time, the determination unit 102-2 reads out the context information and the model corresponding to the reference rotation frequency of the conversion information from the storage unit 111a-2 and uses the model. Further, when the context information indicates “motor B is being driven”, the determination unit 102-2 uses the conversion information converted from the detection information 1222a in the period corresponding to the context information in the detection information 1221a. Determination is performed using the extracted feature information. At this time, the determination unit 102-2 reads out the context information and the model corresponding to the reference rotation frequency of the conversion information from the storage unit 111b-2 and uses the model.

  The detection information corresponding to the context information may be determined using, for example, correspondence information stored in the storage unit 111. FIG. 16 is a diagram illustrating an example of a data structure of correspondence information used for determination of detection information. The correspondence information includes, for example, sensor data and context information. The preprocessing unit 106-2 and the determination unit 102-2 can determine the detection information corresponding to the context information by referring to such correspondence information.

[Third Embodiment]
In the first embodiment, when generating a model corresponding to the reference rotation frequency used for determining the abnormality of the apparatus, detection information in which the rotation frequency is adjusted to the reference rotation frequency is received. In the diagnosis system of the third embodiment, when the rotation frequency of the detection information received when generating the model does not match among the preset reference rotation frequencies, the detection information is converted to the mismatched reference rotation frequency. Then, a model corresponding to the reference rotation frequency is generated using the conversion information.

(Configuration of functional block of diagnostic device)
FIG. 17 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the third embodiment. With reference to FIG. 17, the functional block configuration of the diagnostic apparatus 100-3 according to the present embodiment will be described.

  As illustrated in FIG. 17, the diagnostic device 100-3 includes a communication control unit 101, a determination unit 102, a reception unit 103, a feature extraction unit 104, a generation unit 105, a first preprocessing unit 107, 2 preprocessing unit 107-2 and storage unit 111.

  In the third embodiment, the configurations of the first preprocessing unit 107 and the second preprocessing unit 107-2 are different from those of the first embodiment. Other configurations and functions are the same as those in FIG. 4 which is a block diagram of the diagnostic apparatus 100 according to the first embodiment, and thus the same reference numerals are given and description thereof is omitted here.

  The first preprocessing unit 107 performs the same processing as the preprocessing unit 106 of the first embodiment. That is, in the diagnosis process, the first pre-processing unit 107 is a reference in which the rotation frequency of the drive unit 204 obtained from the context information is less than one or more reference rotation frequencies determined in advance. It converts into conversion information which becomes rotation frequency. Here, a method for selecting one reference rotation frequency from one or more reference rotation frequencies is the same as the method described in the first embodiment.

  When the second pre-processing unit 107-2 generates a model, when the detection information of the set reference rotation frequency is not obtained, the detection information received by the receiving unit 101a is set as the set reference rotation. Conversion information is obtained by converting to frequency.

(Specific example of processing according to this embodiment)
FIG. 18 is a diagram illustrating an example of an operation for generating a model from detection information converted into a plurality of reference rotation frequencies. A specific example of processing according to the present embodiment will be described with reference to FIG.

  In the example of FIG. 18, for example, it is assumed that the rotation frequency of the detection information that can be received by the reception unit 101a is 25 [Hz], and the set reference rotation frequencies are 10 [Hz] and 5 [Hz]. In this case, when it is difficult to receive detection information of 10 [Hz] and 5 [Hz] by the receiving unit 101a, it is impossible to generate a model corresponding to each rotation frequency (reference rotation frequency). become. Therefore, the second pre-processing unit 107-2 receives the detection information received from the receiving unit 101a and whose rotation frequency is 25 [Hz], and the rotation frequency becomes the reference rotation frequency 10 [Hz] and 5 [Hz], respectively. To generate conversion information.

  Then, the feature extraction unit 104 extracts feature information from the conversion information converted by the second preprocessing unit 107-2. And the production | generation part 105 produces | generates the model about this context information and the reference | standard rotation frequency of the said conversion information using the feature information extracted from the conversion information matched with the same context information.

  Thereby, even in a situation where only the detection information different from the reference rotation frequency can be obtained by the receiving unit 101a, it is possible to generate a model corresponding to the reference rotation frequency.

  When the rotation frequency of the detection information received by the reception unit 101a matches the reference rotation frequency corresponding to the model to be created, the conversion process by the second preprocessing unit 107-2 is not necessary, and the feature extraction unit 104 The feature information may be extracted directly from the detection information received by the receiving unit 101a.

[Fourth Embodiment]
In the first embodiment, the diagnosis apparatus 100 that can communicate with the processing machine 200 receives context information from the processing machine 200 and performs model generation processing and diagnosis processing using the context information. In the fourth embodiment, context information can be manually input to the diagnostic apparatus when communication with the processing machine 200 is not possible due to a case where the diagnostic apparatus is not connected to the processing machine 200 or the like.

(Configuration of functional block of diagnostic device)
FIG. 19 is a diagram illustrating an example of a functional block configuration of the diagnostic apparatus according to the fourth embodiment. With reference to FIG. 19, the functional block configuration of the diagnostic apparatus 100-4 according to the present embodiment will be described.

  As illustrated in FIG. 19, the diagnostic device 100-4 includes a communication control unit 101, a determination unit 102, a reception unit 103, a feature extraction unit 104, a generation unit 105, and a preprocessing unit 106 (first preprocessing). Unit), an operation unit 108, and a storage unit 111.

  In the fourth embodiment, an operation unit 108 is added to the configuration of the first embodiment. Other configurations and functions are the same as those in FIG. 4 which is a block diagram of the diagnostic apparatus 100 according to the first embodiment, and thus the same reference numerals are given and description thereof is omitted here.

  The operation unit 108 is a functional unit for the user to input necessary information such as the current context information of the processing machine 200 (for example, the content of the driving unit 204, the rotation frequency, and the like). The operation unit 108 is realized by, for example, the input device 66 illustrated in FIG.

  With the above configuration, when context information cannot be used, such as when the device to be diagnosed and the diagnostic device are not physically or wirelessly connected via a network, or the purpose of the context information is If necessary information (for example, rotation frequency) is not included, necessary information such as the rotation frequency can be manually input to the diagnostic apparatus 100-4. Even in such a case, the model generation process In addition, diagnostic processing is possible.

  Next, modified examples applicable to the above-described embodiments will be described.

[Modification 1]
FIG. 20 is a diagram illustrating an example of the relationship between the context information and the processing section. FIG. 21 is a diagram illustrating an example of a method for specifying a machining section.

  The context information only indicates a section in which a certain driving unit 204 is driven. For example, when the actual processing section in which the tool rotates by the driving unit 204 and hits the material cannot be accurately extracted. There is. That is, the accuracy of abnormality determination may deteriorate.

  In FIG. 20, the context information 1501 indicates that the motor B is being driven. Based on only the context information 1501, for example, detection information corresponding to the waveform section 1512 is specified in the detection information. However, the section where the material is actually processed is the waveform section 1511. For example, when acoustic data is used as detection information, the waveform section 1511 corresponds to a section of acoustic data detected when a tool makes contact with a material to generate sound.

  Therefore, in the first modification, the actual processing section is specified by combining the context information and the detection information. That is, the determination unit 102 determines a period to be used for determination out of the period specified by the received context information based on the received detection information, and makes a determination using the detection information and the model of the determined period. Run.

  For example, the determination unit 102 obtains a machining section by specifying a time at which the feature of the detection information is switched. In the example of FIG. 21, the determination unit 102 switches to a time 1601 when the amplitude of the detection information is switched to exceed a predetermined threshold (for example, “a” and “−a”), and then switches to be below the threshold. Time 1602 is specified. The determination unit 102 extracts a section 1611 between the time 1601 and the time 1602. Then, the determination unit 102 performs determination using the feature information extracted from the detection information of the section 1611. Thereby, determination can be performed with higher accuracy by using detection information of a section corresponding to an actual machining section.

[Modification 2]
FIG. 22 is a graph illustrating an example of the relationship between the likelihood and the determination value r (k).

  As described above, when determining whether or not normal, the likelihood value itself may be used, or a value indicating a variation in likelihood may be used. In the second modification, an example of a value indicating a variation in likelihood will be described. As a value indicating the variation in likelihood, for example, the variance of likelihood can be used. For example, the variance of the likelihood in the section X is calculated by the following equation (2). nX represents the number of frames in the section X, k represents the index of the frame, xi represents the likelihood (frame likelihood) in the frame i, and μX represents the average frame likelihood in the section X. A frame corresponds to a unit interval for calculating likelihood.

  Furthermore, a value obtained by scoring on the basis of the variance as in Expression (2) may be used for the determination. For example, r (k) shown in the following formula (3) may be used for the determination.

  r (k) = VS (k) / VL (k) (3)

  S and L represent the types of the section X. S represents a short section (Short section), and L represents a long section (Long section). VS (k) indicates a value calculated by the expression (2) for a short interval. VL (k) indicates a value calculated by Expression (2) for a long section. For each of S and L, nS, μS, and nL, μL, which are values corresponding to nX, μX, are calculated.

  For example, as shown in FIG. 22, in the determination method for comparing the likelihood with the threshold value, if the threshold value is 0.97, it is determined normal between 200 and 300 frames. On the other hand, in the determination method that compares the determination value r (k) and a threshold value (for example, 1.0), it is possible to detect a variation in likelihood between 200 to 300 frames and determine an abnormality. .

  In each of the above-described embodiments and modifications, the rotation frequency and the reference rotation frequency of the detection information generated by the rotating body detected by the detection unit 211 are handled, but are not necessarily limited to the rotation frequency. If it is vibration data, the vibration frequency may be used, and information including a periodic signal component may be used. In this case, the rotation frequency may be simply referred to as “frequency”, and the reference rotation frequency may be referred to as “reference frequency”.

  The programs executed by the diagnostic devices of the above-described embodiments and modifications are provided by being incorporated in advance in a ROM or the like.

  The programs executed by the diagnostic devices of the above-described embodiments and modifications are installable or executable files in a CD-ROM (Compact Disc Read Only Memory), a flexible disk (FD), a CD- The recording medium may be recorded on a computer-readable recording medium such as R (Compact Disk-Recordable), DVD (Digital Versatile Disk), or the like and provided as a computer program product.

  Further, the program executed by the diagnostic device of each of the above-described embodiments and modifications may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. Good. Further, the program executed by the diagnostic devices of the above-described embodiments and modifications may be configured to be provided or distributed via a network such as the Internet.

  The program executed by the diagnostic device of each of the above-described embodiments and modifications has a module configuration including the above-described units (communication control unit, determination unit, etc.), and the actual hardware is a CPU (processor). When the program is read from the ROM and executed, the respective units are loaded onto the main storage device, and the respective units are generated on the main storage device.

51 CPU
52 ROM
53 RAM
54 Communication I / F
55 Drive control circuit 56 Motor 57 Sensor 58 Bus 61 CPU
62 ROM
63 RAM
64 Communication I / F
65 I / O I / F
66 Input Device 67 Display 68 Auxiliary Storage Device 69 Bus 100, 100-2 to 100-4 Diagnosis Device 101 Communication Control Unit 101a Receiving Unit 101b Transmitting Unit 102, 102-2 Determination Unit 103 Reception Unit 104 Feature Extraction Unit 105 Generation Unit 106 , 106-2 preprocessing unit 107 first preprocessing unit 107-2 second preprocessing unit 108 operation unit 111, 111a to 111c storage unit 111a-2, 111b-2 storage unit 200 processing machine 201 numerical control unit 202 communication control Unit 203 drive control unit 204 drive unit 211 detection unit 701, 701-2 context information 711a to 711c detection information 721 detection information 901 context information 921 detection information 1001 context information 1101 context information 1111a to 1111c detection information 1201 Text information 1221a, 1221b detection information 1222a, 1222b detection information 1501 context information 1511 and 1512 waveform segment 1601 and 1602 time period 1611

JP2015-021901A Japanese Patent No. 3100777

Claims (15)

Of the plurality of context information determined for each type of operation of the target device, the context information corresponding to the current operation, and physical quantity detection information including a frequency component that changes according to the operation of the target device, A receiver for receiving from the device;
A first preprocessing unit that obtains conversion information by converting the frequency of the detection information from the detection information received by the reception unit into a preset reference frequency;
The conversion information converted by the first preprocessing unit and a model corresponding to the context information received by the receiving unit among one or more models determined for each of the one or more context information items, A determination unit that determines whether the operation of the target device is normal;
Diagnostic device with   The said determination part is the said conversion information converted by the said 1st pre-processing part from the detection information of the period specified by the said context information received by the said reception part among the said detection information received by the said reception part. The diagnostic apparatus according to claim 1, wherein a determination is made as to whether or not the operation of the target apparatus during the period is normal using a model corresponding to the context information received by the receiving unit. The receiving unit receives a plurality of detection information respectively corresponding to different physical quantities,
The first preprocessing unit converts the detection information determined according to the context information from the plurality of detection information into the conversion information,
The determination unit uses the conversion information determined according to the context information among a plurality of the conversion information and a model corresponding to the context information received by the reception unit, and operates the target device. The diagnostic device according to claim 1, wherein whether or not is normal is determined. A plurality of the reference frequencies are set in advance,
The first preprocessing unit selects the reference frequency such that a difference between the frequency characteristic of the detection information and the frequency characteristic of the conversion information is reduced from the plurality of reference frequencies, and the detection information is used to detect the detection frequency. The diagnostic device according to any one of claims 1 to 3, wherein a frequency of information is converted into the selected reference frequency to obtain the converted information.   The first preprocessing unit obtains a ratio of the frequency to the reference frequency when the frequency of the detection information received by the reception unit is smaller than the reference frequency from a plurality of the reference frequencies, and the detection information The diagnostic apparatus according to claim 4, wherein when the frequency is higher than the reference frequency, the ratio of the reference frequency to the frequency is obtained, and the reference frequency having the largest obtained ratio is selected. A second preprocessing unit that converts the detection information received by the reception unit into one or more reference frequencies different from the frequency of the detection information;
A generation unit that generates a model corresponding to the context information corresponding to the detection information using the information converted by the second preprocessing unit;
The diagnostic device according to any one of claims 1 to 5, further comprising:   A generation unit configured to generate a model for context information for which no model is determined, using the detection information corresponding to the context information when context information for which the model is not determined is received by the reception unit; The diagnostic apparatus as described in any one of Claims 1-5 provided.   The diagnostic device according to any one of claims 1 to 7, wherein the first preprocessing unit obtains the frequency of the detection information received by the receiving unit from the context information received by the receiving unit.   The diagnostic device according to any one of claims 1 to 7, wherein the first preprocessing unit analyzes the detection information received by the reception unit to obtain a frequency of the detection information.   The diagnostic apparatus according to claim 1, further comprising an operation unit that inputs the context information when the context information is not received by the receiving unit.   The determination unit determines a period to be used for determination among periods specified by the context information based on the detection information received by the reception unit, and the detection information corresponding to the determined period, The diagnostic device according to claim 1, wherein a model is used to determine whether the operation of the target device is normal.   The determination unit obtains a likelihood of the detection information received by the reception unit with respect to the model, and compares at least one of the likelihood and a value indicating variation in the likelihood with a threshold value. The diagnostic device according to any one of claims 1 to 11, wherein whether or not the operation of the target device is normal is determined. The diagnostic device according to any one of claims 1 to 12,
A detection unit that detects the physical quantity as the detection information;
Diagnostic system with Of the plurality of context information determined for each type of operation of the target device, the context information corresponding to the current operation, and physical quantity detection information including a frequency component that changes according to the operation of the target device, A receiving step for receiving from the device;
From the received detection information, a preprocessing step of converting the frequency of the detection information to a preset reference frequency to obtain conversion information;
Whether the operation of the target device is normal using the converted conversion information and a model corresponding to the received context information among one or more models determined for each of the one or more context information A determination step for determining whether or not;
A diagnostic method comprising: Computer
Of the plurality of context information determined for each type of operation of the target device, the context information corresponding to the current operation, and physical quantity detection information including a frequency component that changes according to the operation of the target device, A receiver for receiving from the device;
A first preprocessing unit that obtains conversion information by converting the frequency of the detection information from the detection information received by the reception unit into a preset reference frequency;
The conversion information converted by the first preprocessing unit and a model corresponding to the context information received by the receiving unit among one or more models determined for each of the one or more context information items, A determination unit that determines whether the operation of the target device is normal;
Program to make it function.

Images