Anomaly detection device, anomaly detection method, and anomaly detection program

Abstract

To provide an anomaly detection device, an anomaly detection method, and an anomaly detection program, which eliminate the need for design of a quantity of characteristic of a waveform to be handled, and which have high precision corresponding to waveforms of a large number of parameters.SOLUTION: The present invention includes: an acquisition unit that acquires a measurement waveform that is time-series data relating to a plurality of parameters of a target machine; a waveform generation unit that, by inputting the measurement waveform, generates a generated waveform that is the time-series data relating to the plurality of parameters; a deviation degree calculation unit that calculates a deviation degree by comparing the measurement waveform with the generated waveform; and a determination unit that, when the deviation degree exceeds a predetermined threshold value, determines that the measurement waveform is anomalous; wherein the waveform generation unit includes a trained learning model learned based on the measurement waveform at normal time in the target machine.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an anomaly detection device, an anomaly detection method, and an anomaly detection program. [Background technology]

[0002] Conventionally, anomaly detection technology has been proposed that predicts failures based on waveform data from sensors installed in production equipment. In general, anomaly detection does not collect enough abnormal data, and it is difficult to detect unknown anomalies. Therefore, a machine learning method that learns only normal data and detects anomalies has been proposed as a conventional technology.

[0003] Generally, production equipment may operate under multiple operating conditions even in a normal state. Therefore, Patent Document 1 proposes a method for detecting anomalies that takes multiple operating conditions into consideration. Meanwhile, Patent Document 2 proposes a method for determining anomalies by inputting a measured waveform into a model that has learned only normal data, outputting a generated waveform, and comparing the measured waveform with the generated waveform. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-51698 [Patent Document 2] Patent Publication No. 2021-9441 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the technology of Patent Document 1, when all inference results of each learning model prepared for each operating condition exceed all thresholds, it is judged as an anomaly, so there is a problem with detection accuracy. In addition, although it is a method for detecting an anomaly by extracting features of a generated waveform, it is not easy to design effective features in a state where abnormal data is insufficient. In contrast, the technology of Patent Document 2 does not require feature design because it is a comparison of waveforms. However, the input to the autoencoder proposed in Patent Document 2 is only a univariate explanatory variable, and multivariate cases in which many parameters are used are not assumed. Furthermore, in both Patent Documents 1 and 2, only the degree of anomaly of a univariate is notified as an index of an anomaly, so there is a problem that the explanatory power of the anomaly judgment is poor.

[0006] The present invention has been made to solve the above problems, and aims to provide an anomaly detection device, an anomaly detection method, and an anomaly detection program that do not require designing features of the waveforms to be handled and are highly explainable when dealing with waveforms of a large number of parameters. [Means for solving the problem]

[0007] The present invention comprises an acquisition unit that acquires a measured waveform, which is time series data regarding multiple parameters of a target machine, a waveform generation unit that generates a generated waveform, which is time series data regarding the multiple parameters, by inputting the measured waveform, a deviation calculation unit that calculates a deviation by comparing the measured waveform with the generated waveform, and a judgment unit that judges the measured waveform to be abnormal if the deviation exceeds a predetermined threshold value, and the waveform generation unit has a trained learning model that has been learned based on the measured waveform when the target machine is operating normally.

[0008] In the above-mentioned anomaly detection device, the learning model may be learned based on the measured waveforms under normal conditions acquired under multiple operating conditions of the target machine.

[0009] In the above anomaly detection device, the learning model can be generated by unsupervised learning.

[0010] In the above anomaly detection device, the learning model can be configured by a neural network.

[0011] In the above anomaly detection device, the neural network includes an encoding unit and a generating unit, the encoding unit receives the measured waveform as an input and outputs a latent variable, the generating unit receives the latent variable as an input and outputs the generated waveform, and the encoding unit and the generator can be configured by a convolutional neural network.

[0012] In the above anomaly detection device, the learning model can be constructed, for example, by an Efficient GAN.

[0013] In the above anomaly detection device, the deviation degree may be a single scalar value.

[0014] In the above anomaly detection device, the deviation calculation unit can be configured to calculate an individual deviation of a time series for each of the parameters.

[0015] The above anomaly detection device may further include a display unit that displays at least one of the measured waveform, the generated waveform, the deviation degree, and an individual deviation degree of the time series for each parameter.

[0016] The abnormality determination method of the present invention includes the steps of acquiring a measured waveform, which is time-series data for a plurality of parameters of a target machine, generating a generated waveform, which is time-series data for the plurality of parameters, by inputting the measured waveform, calculating a deviation by comparing the measured waveform with the generated waveform, and determining that the measured waveform is abnormal if the deviation exceeds a specified threshold value, wherein the step of generating the generated waveform is performed by a trained learning model that is learned based on the measured waveform of the target machine when it is normal.

[0017] The abnormality judgment program of the present invention causes a computer to execute the steps of acquiring a measured waveform, which is time-series data for multiple parameters of a target machine, generating a generated waveform, which is time-series data for the multiple parameters, by inputting the measured waveform, calculating a deviation by comparing the measured waveform with the generated waveform, and determining that the measured waveform is abnormal if the deviation exceeds a specified threshold value, wherein the step of generating the generated waveform is performed by a trained learning model that has been learned based on the measured waveform of the target machine when it is normal. Effect of the Invention

[0018] According to the present invention, it is possible to provide an anomaly detection device, an anomaly detection method, and an anomaly detection program that do not require the design of features of the waveform to be handled and are highly interpretable for waveforms of a large number of parameters. [Brief description of the drawings]

[0019] [Figure 1] 1 is a configuration example of an embodiment of a system including an anomaly detection device of the present invention. [Diagram 2] FIG. 2 is a block diagram showing a hardware configuration of the anomaly detection device. [Diagram 3] FIG. 2 is a block diagram showing the hardware configuration of the learning device. [Figure 4] FIG. 2 is a block diagram showing the software configuration of the learning device. [Diagram 5] 13 is a graph showing an overlay of measured and generated waveforms under different operating conditions. [Figure 6] FIG. 1 is a diagram illustrating the correspondence between an encoding unit, a generating unit, and a CNN. [Figure 7] FIG. 2 is a block diagram showing a software configuration of the anomaly detection device. [Figure 8] 13 is a graph showing an overlapping of measured waveforms and generated waveforms under normal conditions for three parameters, and a graph showing deviation time series data for each parameter. [Figure 9]13 is a graph showing an overlap of measured waveforms and generated waveforms during an abnormality for three parameters, and a graph showing deviation time series data for each parameter. [Figure 10] FIG. 13 is a diagram showing rankings of deviation degrees. [Figure 11] FIG. 13 is a diagram illustrating a determination of normality or abnormality based on the relationship between the deviation and a threshold value. [Figure 12] 4 is a flowchart showing a learning process of the learning device. [Figure 13] 4 is a flowchart showing an anomaly detection method in the anomaly detection device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] An embodiment of an anomaly detection device according to the present invention will be described below with reference to the drawings. As an example, an anomaly detection device that detects anomalies in a ball screw mechanism of production equipment will be described below. FIG. 1 shows an example of the configuration of a system including this anomaly detection device, which includes an anomaly detection device 1, a learning device 2 that generates a learning model used in this anomaly detection device 1, and a transport device 3 that includes a ball screw mechanism. Each component will be described in detail below.

[0021] <1. Hardware configuration> <1-1.Transportation device> 1, the transport device 3 includes a ball screw 31, a nut 32 screwed onto the ball screw 31, and a stage 33 for a workpiece W provided on the nut 32. A servo motor 34 is connected to the ball screw 31, and the ball screw 31 is rotated by the servo motor 34. A servo driver 35 that controls the servo motor 34 is also connected to the servo motor 34. In this embodiment, the rotation speed, output torque, and stage position of the servo motor 34, which are controlled by the servo driver 35, are used as parameters for abnormality detection, and abnormalities are judged using the time-series waveforms of these parameters.

[0022] <1-2. Anomaly detection device> Next, an example of the hardware configuration of the anomaly detection device 1 according to this embodiment will be described. Fig. 2 shows an example of the hardware configuration of the anomaly detection device according to this embodiment.

[0023] The abnormality detection device 1 is a computer in which a control unit 11, a storage unit 12, an external interface 13, and a communication interface 14 are electrically connected, and can be configured, for example, by a PLC. In Fig. 2, the external interface 13 and the communication interface 14 are written as "external I / F" and "communication I / F". This also applies to various devices described later.

[0024] The control unit 11 includes a CPU, RAM, ROM, etc., and is configured to execute various information processing based on programs and data. The storage unit 12 is configured, for example, as an auxiliary storage device such as an HDD or SDD, and stores an anomaly detection program 121, learning result data 122, and various data for anomaly determination and for driving the device. The anomaly detection program 121 is a program for causing the anomaly detection device 1 to determine whether or not there is an anomaly in the measured waveform. The control unit 11 is configured to execute the processing of each step described below by interpreting and executing this anomaly detection program 121.

[0025] The external interface 13 is an interface for connecting to an external device, and is configured appropriately according to the external device to be connected. In this embodiment, the external interface 13 is connected to a display (display unit) 4. This display 4 is used, for example, to display the results of anomaly determination. The display 4 is not particularly limited, and a known liquid crystal display, touch panel display, or the like can be used. In addition, an input device, an output device, and the like can be connected to the external interface 13 as appropriate.

[0026] The communication interface 14 is, for example, a wired LAN (Local Area Network) module, a wireless LAN module, or the like, and is an interface for performing wired or wireless communication. That is, the communication interface 14 is an example of a communication unit configured to communicate with other devices. In this embodiment, the communication interface 14 is connected to the learning device 2 and the transport device 3 described above via a network.

[0027] Regarding the specific hardware configuration of the anomaly detection device 1, components can be omitted, replaced, or added as appropriate depending on the embodiment. For example, the control unit 11 may include multiple processors. The control unit 11 may also be configured by an FPGA. The storage unit 12 may be configured by a RAM and a ROM included in the control unit 11. The anomaly detection device 1 can use a PLC as well as an information processing device designed specifically for the service to be provided. Furthermore, the anomaly detection device 1 may be a general-purpose desktop PC, tablet PC, or the like depending on the target for which anomaly judgment is performed. The above points are the same for the learning device 2 described later.

[0028] <1-3. Learning device> Next, an example of the hardware configuration of the learning device 2 according to this embodiment will be described. Fig. 3 shows an example of the hardware configuration of the learning device according to this embodiment.

[0029] As shown in FIG. 3, similar to the anomaly detection device 1, the learning device 2 is a computer in which a control unit 21, a storage unit 22, an external I / F 23, and a communication I / F 24 are electrically connected.

[0030] The configurations of the control unit 21 and the storage unit 22 are the same as those of the anomaly detection device. The storage unit stores a learning program 221, a learning model 222, a first learning result data 223, a second learning result data 224, and various data for driving the learning device. The learning program 221 is a program for learning a learning model 222 for outputting a generated waveform from a measured waveform. The control unit 21 is configured to interpret and execute this learning program 221, thereby executing the processing of each step described below.

[0031] The other hardware configuration is the same as that of the anomaly detection device 1, so it is omitted here.

[0032] <2. Software configuration> <2-1. Learning device>

[0033] The learning device 2 learns the learning model 222 using measured waveforms, which are time-series data relating to the rotation speed and output torque of the servo motor 34 and the stage position transmitted from the servo driver 35 of the transport device 3. In particular, this learning model 222 is learned using measured waveforms in normal operation when the transport device 3 is operating normally. The software configuration of the learning device 2 will be described below with reference to Fig. 4. Fig. 4 is a block diagram showing the software configuration of the learning device.

[0034] <2-1-1. Learning model> As shown in Figure 4, when the control unit 21 of the learning device 2 expands the learning program 221 stored in the memory unit 22 into the RAM, the control unit 21 interprets and executes the learning program 221 using the CPU, and functions as a computer equipped with an acquisition unit 211, an encoding unit 212, a generation unit 213, a discrimination unit 214, and a storage processing unit 215.

[0035] As described above, the acquiring unit 211 acquires a measurement waveform, which is time-series data relating to the rotation speed and output torque of the servo motor 34, and the position of the stage. This measurement waveform is transmitted after preprocessing by the anomaly detection device 1, as described later. However, the measurement waveform may be transmitted directly from the servo driver 35, and preprocessing may be performed by the acquiring unit 211, as described later.

[0036] Furthermore, the measurement waveforms acquired by the acquisition unit 211 are not only those obtained under the same operating conditions, but also those obtained when the conveying device 3 operates under different operating conditions. For example, a measurement waveform when a light workpiece is conveyed (operating condition 1) and a measurement waveform when a heavy workpiece is conveyed (operating condition 2) are input. Examples of measurement waveforms under each operating condition are shown in FIG. 5. In this example, it can be seen that when a heavy workpiece W is conveyed, the speed decreases and the torque increases. Note that there are various other different operating conditions in addition to those mentioned above, such as when different processed parts are used (for example, when the stage is replaced due to a change in setup), and when the conveying device operates under different environmental temperatures (for example, when the temperature changes between morning and afternoon).

[0037] Next, the encoding unit 212, the generating unit 213, and the discriminating unit 214 will be described. The encoding unit 212, the generating unit 213, and the discriminating unit 214 configure a learning model using an Efficient Generative Adversarial Network (Efficient GAN). The encoding unit 212 and the generating unit 213 are configured, for example, by a convolutional neural network (CNN). This will be described later. The discriminating unit 214 is also configured by a similar CNN. However, the configuration of the discriminating unit 214 may be different from that of the encoding unit 212 and the generating unit 213.

[0038] When learning is performed, a first training step for training the discrimination unit 214 and a second training step for training the encoding unit 212 and the generation unit 213 are alternately performed. In the first training step, the discrimination unit 214 is trained to discriminate whether a generated waveform input to the discrimination unit 214 is a generated waveform generated by the generation unit 213 or a measured waveform transmitted from the acquisition unit 211. That is, the discrimination unit 214 is trained to discriminate whether the input waveform originates from the acquisition unit 211 or the generation unit 213. In addition, in the second training step, the encoding unit 212 and the generation unit 213 are trained to generate a waveform that causes the discrimination unit 214 to make an erroneous discrimination. Here, originating from the acquisition unit 211 is expressed as "true", and originating from the generation unit 213 is expressed as "false". However, the method of expressing each origin is not limited to such an example, and may be appropriately selected according to the embodiment.

[0039] Specifically, in the first training step, a measured waveform is input to the encoding unit 212. The encoding unit 212 outputs a latent variable z. When this latent variable z is input to the generating unit 213, the generating unit 213 outputs a generated waveform that is approximate to the measured waveform input to the encoding unit 212.

[0040] This generated waveform is input to discrimination unit 214, where arithmetic processing is executed. As a result, an output value corresponding to the result of discriminating whether the generated waveform originates from generation unit 213 or acquisition unit 211 is output. In this case, since the generated waveform input to discrimination unit 214 is the generated waveform generated by generation unit 213, discrimination unit 214's correct judgment is "false". Then, the error between the output value and this correct answer is calculated. In this way, the error between the output value and this correct answer is calculated for each generated waveform.

[0041] Similarly, the measured waveform is input to discrimination unit 214, which executes calculation processing. As a result, discrimination unit 214 outputs an output value corresponding to the result of discriminating whether the input waveform originates from generation unit 213 or acquisition unit 211. In this case, since the input waveform is a measured waveform, discrimination unit 214 correctly discriminates as "true." In this way, the error between the output value and this correct answer is calculated for each waveform.

[0042] Then, the parameter values ​​of the CNN of the discrimination unit 214 are adjusted so that the sum of the calculated errors becomes smaller. The adjustment of the parameter values ​​of the discrimination unit 214 is repeated by the above series of processes until the sum of the errors between the output value and the correct answer of true or false becomes equal to or less than the threshold value. Thereby, in the first training step, the discrimination unit 214 is trained by the backpropagation method or the like so as to discriminate whether the waveform is a generated waveform generated by the generation unit 213 or a measured waveform.

[0043] On the other hand, in the second training step, the measured waveform is input to the encoding unit 212. The encoding unit 212 outputs a latent variable z. When this latent variable z is input to the generating unit 213, the generating unit 213 outputs a generated waveform that is approximate to the measured waveform input to the encoding unit 212.

[0044] Next, the generated generated waveform is input to discrimination unit 214, and the discrimination unit 214 executes calculation processing. As a result, an output value corresponding to the result of discriminating whether the input generated waveform originates from generation unit 213 or acquisition unit 211 is output. In this training of generation unit 213, the correct answer is to make the discrimination result by discrimination unit 214 erroneous. In other words, it is sufficient that the generated waveform generated by generation unit 213 is similar to the measured waveform so that it can be mistaken for a measured waveform, and the correct answer is that the obtained output value corresponds to the "true". For the combination of the measured waveform and the generated waveform, the error between the output value obtained by a series of processes and this correct answer (i.e., the "true") is calculated.

[0045] Then, the parameter values ​​of the CNN of the encoding unit 212 and the generating unit 213 are adjusted so that the sum of the calculated errors becomes smaller. The adjustment of the parameter values ​​of the encoding unit 212 and the generating unit 213 is repeated by the above series of processes until the sum of the errors between the output value obtained by the series of processes and the "true" for each combination of the measured waveform and the generated waveform becomes equal to or less than a threshold value. As a result, in the second training step, the generating unit 213 (and the encoding unit 212) is trained by backpropagation or the like so as to generate a generated waveform that will cause an erroneous discrimination by the discriminator 214.

[0046] In this way, by alternately performing the first training step and the second training step, the accuracy of the discrimination unit 214 and the generation unit 213 (and the encoding unit 212) is alternately improved. This allows the generation unit 213 to acquire the ability to appropriately generate a generated waveform that is almost the same as a measured waveform.

[0047] After this learning is completed, the storage processing unit 215 generates first learning result data 223 and second learning result data 224 indicating the configurations of the constructed encoding unit 212 and generating unit 213, and the discrimination unit 214 (e.g., the number of layers of the neural network, the number of neurons in each layer, the connection relationships between the neurons, and the transfer function of each neuron) and calculation parameters (e.g., the weight of the connection between each neuron, the threshold value of each neuron), respectively. Then, the storage processing unit 215 stores the generated first and second learning result data 223, 224 in the memory unit 22.

[0048] In this way, when a measured waveform is input to the learned encoding section 212 and generating section 213, a generated waveform that is close to the measured waveform can be generated.

[0049] <2-1-2.CNN> Next, an example of the CNN constituting the encoding unit 212 and the generating unit 213 will be described with reference to Fig. 6. This CNN is constituted by a one-dimensional CNN, specifically, the encoding unit 212 is constituted by a one-dimensional CNN, and the generating unit 213 is constituted by a one-dimensional transposed CNN. Fig. 6 is a diagram showing the correspondence between the encoding unit and the generating unit and the CNN. The measurement waveform input to the encoding unit 212 is generated by the pre-processing unit 112 of the anomaly detection device 1.

[0050] This point will be explained first. The pre-processing unit 112 performs pre-processing on the collected measurement waveform. As the pre-processing, for example, the following three processes can be performed. First, in the transport device 3, the waveform during the period when the stage 33 moves from the start point to the end point through the section (frame) is cut out and used for judgment. In addition, the value of the measurement waveform is normalized to be a numerical value between 0 and 1. Furthermore, a linear conversion is performed so that the frame length of the time series becomes 64.

[0051] As shown in Fig. 6, in this CNN, data represented by (C, T) converted to a frame length of 64 for the above-mentioned three parameters is input to the encoding unit 212. Here, C is the number of dimensions of the variable, and T is the frame length of the data. The data input to the encoding unit 212 is (3, 64) because the number of parameters is 3 and the frame length is 64. In other words, 3*64=192 pieces of data are input. Therefore, as shown in Fig. 6, the data for each layer of the CNN is represented by (C, T).

[0052] In addition, in this CNN, the encoding unit 212 performs calculations to reduce the data frame length of the data in four layers to (32,32), (64,16), and (128,8), and generates 10 data corresponding to the latent variable z in the fully connected layer. That is, this latent variable z reduces the dimension of the measured waveform and shows the characteristics of the measured waveform in low dimensions. By expressing the measured waveform in low dimensions in this way, learning can be easily performed. After that, the data corresponding to this latent variable z is input to the generation unit 213, and calculations are performed to expand the reduced data frame length of the data in five layers to (512,4), (256,8), and (128,16), and finally generates data of (3,64) corresponding to the measured waveform, that is, a generated waveform.

[0053] Note that this CNN is just an example, and a normal neural network, etc. can also be used.

[0054] <2-2. Anomaly detection device> Next, a software configuration of the anomaly detection device 1 will be described. Fig. 7 is a block diagram showing the software configuration of the anomaly detection device. As shown in Fig. 7, when the control unit 11 of the anomaly detection device 1 loads the anomaly detection program 121 stored in the storage unit 12 into the RAM, the control unit 11 interprets and executes the anomaly detection program 121 by the CPU, and functions as a computer including a collection unit 111, a preprocessing unit 112, a waveform generation unit 113, a deviation calculation unit 114, a determination unit 115, and an output unit 116.

[0055] The collecting unit 111 collects measurement waveforms, which are time-series data relating to the rotation speed, output torque, and stage position of the servo motor 34 transmitted from the servo driver 35 of the transport device 3. The collected measurement waveforms are stored in the storage unit 12 or the RAM.

[0056] The preprocessing unit 112 performs preprocessing on the collected measurement waveform. This processing is as described above. The preprocessed measurement waveform is transmitted to the learning device 2 and used for the above-mentioned learning.

[0057] As described above, the waveform generating unit 113 is composed of the trained encoding unit 212 and generating unit 213 to which the first learning result data 223 generated by the learning device 2 has been introduced.

[0058] The deviation calculation unit 114 compares the measured waveform with the generated waveform and calculates the deviation. As the deviation, an individual deviation and a total deviation are calculated. The calculation of the individual deviation is performed as follows. First, for each of the above three parameters, deviation time series data is calculated by calculating the difference between the measured waveform and the generated waveform shown in the time series data in a time series. For example, the difference is calculated for each of 64 data, which is the frame length, and deviation time series data consisting of the data of the 64 differences is calculated. Next, the 64 data are summed up to calculate the individual deviation. That is, the deviation for each of the three parameters can be expressed.

[0059] FIG. 8 is a graph showing the measured waveforms and generated waveforms superimposed on each other in normal conditions for three parameters, and the deviation time series data for each parameter. Meanwhile, FIG. 9 is a graph showing the measured waveforms and generated waveforms superimposed on each other in abnormal conditions for three parameters, and the deviation time series data for each parameter. In the deviation time series data in normal conditions shown in FIG. 8, the deviation is approximately 0 for each parameter in any time period. Note that in the graph in FIG. 8 where the measured waveforms and generated waveforms are superimposed, the two are almost the same, so they appear as a single line. Meanwhile, in the deviation time series data in abnormal conditions shown in FIG. 9, it can be seen that the deviation changes. For example, it can be seen that the deviation time series data for torque changes significantly at the start of conveyance (circled area).

[0060] Also, based on the individual deviations, a ranking of deviations can be generated as shown in Fig. 10. On the other hand, the overall deviation is a scalar value obtained by summing up the individual deviations. In this way, the individual deviations and the overall deviations are calculated for each combination of the measured waveform and the generated waveform.

[0061] The judgment unit 115 judges that the overall deviation exceeds a specified threshold value as an abnormality, and judges that the overall deviation is normal if the overall deviation is equal to or less than the threshold value. The waveform generation unit 113 includes a learning model learned based on a normal measured waveform, so that when a normal measured waveform is input, a generated waveform that is close to the measured waveform is generated. On the other hand, when an abnormal measured waveform is input, a generated waveform different from the normal generated waveform is generated, so the deviation becomes large. Therefore, when the overall deviation exceeds a predetermined threshold value, it is judged to be abnormal. FIG. 11 is an example. It can be seen that when a normal measured waveform is input, an overall deviation lower than the threshold value is calculated, whereas when an abnormal measured waveform is input, an overall deviation higher than the threshold value is calculated.

[0062] The output unit 116 outputs at least one of the measured waveform, the generated waveform, the deviation time-series data, the individual deviation, and the overall deviation, and displays it on the display 4. In addition, when an abnormality occurs, the output unit 116 can perform output such as displaying an alert on the display 4 or generating a sound.

[0063] <3. Learning process in the learning device> Next, the learning process of the learning device configured as above will be described with reference to the flowchart of FIG.

[0064] As shown in FIG. 12, first, the conveying device 3 is operated (step S11), and measurement waveforms in normal times are collected for the above-mentioned three parameters (step S12). Next, preprocessing is performed on the measurement waveform (step S13). Next, the preprocessed measurement waveform is input to the learning model 222, and the above-mentioned first and second learning result data 223, 224 are generated (step S14). Following this, a threshold is set (step S15). The setting of the threshold is not particularly limited, but for example, a normal distribution of the above-mentioned overall deviation degree is calculated, and the threshold can be set to 3σ or more, for example, 3σ, 4σ, etc. Note that the threshold can be set automatically in the learning device 2, or can be set in the anomaly detection device 1. Alternatively, the user can arbitrarily set the threshold. Thereafter, the first learning result data 223 is transmitted to the anomaly detection device 1 and incorporated into the waveform generation unit 113 (step S16). The first learning result data 223 can be created periodically and incorporated each time into the waveform generating unit 113. In this way, the learning process in the learning device 2 is completed.

[0065] <4. Anomaly detection method using an anomaly detection device> Next, an anomaly detection method in the anomaly detection device configured as above will be described with reference to the flowchart of FIG.

[0066] As shown in FIG. 13, first, while the transport device 3 is in operation, the collecting unit 111 collects measured waveforms for the above-mentioned three parameters (step S21). Next, the preprocessing unit 112 performs preprocessing on the measured waveform (step S22). Next, the preprocessed measured waveform is input to the waveform generating unit 113 incorporating the first learning result data 223, and a generated waveform is generated (step S23). Following this, the deviation calculating unit 114 compares the calculated generated waveform with the measured waveform and calculates the overall deviation (step S24). Finally, if the determining unit 115 determines that the overall deviation is less than the threshold (YES in step S25), the transport device is considered to be normal and the operation of the transport device is continued. On the other hand, if the overall deviation is equal to or greater than the threshold (NO in step S25), it is determined that an abnormality has occurred, and an abnormality alert is displayed on the display 4 (step S26).

[0067] <5.Features> As described above, according to this embodiment, the waveform generator 113 generates a generated waveform, which is time series data on a plurality of parameters, from a measured waveform under normal conditions, which is time series data on the plurality of parameters measured by the transport device 3, and the measured waveform is compared with the generated waveform to calculate a deviation, and if this exceeds a predetermined threshold, it is determined that an abnormality has occurred. Therefore, the following effects can be obtained.

[0068] (1) It is possible to detect abnormalities with high accuracy corresponding to multiple operating conditions. (2) Since the deviation time series data and individual deviations can be calculated, it is possible to check, for example, at what time the deviation becomes large for each parameter. This improves the interpretability of anomaly determination. (3) Since the measured waveform that has been subjected to simple preprocessing is input and the generated waveform is output, there is no need to design the waveform features.

[0069] <6. Variations> Although the embodiment of the present invention has been described above, the above description is merely an example of the present invention in every respect, and it goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. For example, the following modifications are possible. In the following, the same reference numerals are used for the same components as in the above embodiment, and the description of the same points as in the above embodiment is omitted as appropriate. In addition, the following modifications can be combined as appropriate.

[0070] <6-1> In the above embodiment, the learning model is based on GAN, but is not limited thereto, and may be based on neural networks such as autoencoder, GANomaly, Skip GANomaly, etc. In other words, any method may be used as long as it can learn multivariate and multidimensional data such as time-series data related to multiple parameters as described above.

[0071] <6-2> The method of calculating the deviation for performing anomaly determination in deviation calculation section 114 is not particularly limited as long as it is possible to calculate the difference between the measured waveform and the generated waveform as the deviation, which can be evaluated by determination section 115.

[0072] <6-3> In the above embodiment, the anomaly detection device 1 and the learning device 2 are configured as separate computers, but they may be configured as one computer. That is, a learning function may be added to the anomaly detection device 1.

[0073] <6-4> The method of pre-processing the measured waveform is not particularly limited and can be changed as appropriate depending on the learning model used. Alternatively, learning can be performed or a generated waveform can be generated without pre-processing.

[0074] <6-5> The abnormality detection device 1 of the above embodiment judges an abnormality in a conveyance device having a ball screw mechanism, but is not limited to this and can be applied to judge abnormalities in various target machines. [Explanation of symbols]

[0075] 1. Anomaly detection device 11 Acquisition Department 13 Waveform generator 14 Deviation calculation section 15 Judgment section 2 Learning device 212 Encoding section 213 Generation part 214 Discrimination part 222 Learning Model 3. Transport equipment (target machine) 4 Display

Claims

1. an acquisition unit that acquires a measurement waveform, which is a set of time series data relating to each of a plurality of parameters of a target machine; a waveform generating unit that receives the measured waveform and generates a generated waveform, which is a set of time-series data relating to each of the plurality of parameters; a deviation calculation unit that calculates a deviation by comparing the measured waveform with the generated waveform; a determination unit that determines that the measured waveform is abnormal when the deviation exceeds a predetermined threshold value; Equipped with An anomaly detection device, wherein the waveform generation unit is equipped with a trained learning model that has been learned based on the measured waveform of the target machine under normal conditions.

2. The anomaly detection device according to claim 1 , wherein the learning model is learned based on the measured waveforms under normal conditions acquired under a plurality of operating conditions of the target machine.

3. The anomaly detection device according to claim 1 , wherein the learning model is generated by unsupervised learning.

4. The anomaly detection device according to claim 3 , wherein the learning model is configured by a neural network.

5. The neural network includes an encoding unit and a generating unit, The encoding unit receives the measured waveform as an input and outputs a latent variable, The generating unit receives the latent variables as an input and the generated waveform as an output, The anomaly detection device according to claim 4 , wherein the encoding unit and the generating unit are configured with a convolutional neural network.

6. The anomaly detection device according to claim 4 , wherein the learning model is configured using an Efficient GAN.

7. The anomaly detection device according to claim 1 , wherein the deviation is a single scalar value.

8. The anomaly detection device according to claim 1 , wherein the deviation calculation unit is configured to calculate an individual deviation of a time series for each of the parameters.

9. The anomaly detection device according to claim 1 , further comprising a display unit that displays at least one of the measured waveform, the generated waveform, the deviation, and an individual deviation of the time series for each parameter.

10. acquiring a measurement waveform which is a set of time series data relating to each of a plurality of parameters of a target machine; generating a generated waveform, which is a set of time-series data relating to each of the plurality of parameters, by inputting the measured waveform; calculating a deviation by comparing the measured waveform with the generated waveform; determining that the measured waveform is abnormal when the deviation exceeds a specified threshold value; Equipped with An anomaly detection method, in which the step of generating the generated waveform is performed by a trained learning model that has been learned based on the measured waveform of the target machine under normal conditions.

11. On the computer, acquiring a measurement waveform which is a set of time series data relating to each of a plurality of parameters of a target machine; generating a generated waveform, which is a set of time-series data relating to each of the plurality of parameters, by inputting the measured waveform; calculating a deviation by comparing the measured waveform with the generated waveform; determining that the measured waveform is abnormal when the deviation exceeds a specified threshold value; Run the command, An anomaly detection program, wherein the step of generating the generated waveform is performed by a trained learning model that has been learned based on the measured waveform of the target machine under normal conditions.

Images