System, method, and program

Abstract

To provide a system, a method, and a program that are able to extract a normal waveform for generating a unit space in an MT method.SOLUTION: The present invention is a system for detecting an anomaly in a target object by using an MT method, including: a reception unit configured to receive a waveform of a digital signal output from the target object; a reception unit configured to receive an input of a threshold for extracting a division waveform as dividing conditions for extracting from the waveform the division waveform serving as a candidate for a normal waveform for generating a unit space in the MT method, an input of a minimum time taken for the threshold to be exceeded, an input of an extraction start point before the threshold is exceeded, and an input of an extraction time that is a time taken to extract the waveform from the extraction start point; and an extraction unit configured to extract the division waveform as a candidate for the normal waveform from the waveform, based on the division condition.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a system, a method, and a program.

Background Art

[0002] A technique for extracting a waveform necessary for extracting a feature amount from a radio wave signal based on the waveform of the radio wave signal to be monitored has been disclosed.

Summary of the Invention

Problems to be Solved by the Invention

[0003] However, in the prior art, a waveform for determining an abnormality is extracted, and for example, it is difficult to extract a normal waveform for generating a unit space used in the MT method.

[0004] The present invention has been made in view of the above, and an object thereof is to provide a system, a method, and a program capable of extracting a normal waveform for generating a unit space in the MT method.

Means for Solving the Problems

[0005] In order to solve the above-described problems and achieve the object, the present invention is a system for detecting an abnormality of an object using the MT method, including a receiving unit that receives a waveform of a digital signal output from the object, and a normal waveform for generating a unit space in the MT method. A receiving unit that receives an input of a threshold value for extracting the divided waveform, a minimum time until the threshold value is exceeded, an extraction start point before the time when the threshold value is exceeded, and an extraction time that is a time for extracting a waveform from the extraction start point, as a division condition for extracting a divided waveform that is a candidate for the above; and an extraction unit that extracts the divided waveform that is a candidate for the normal waveform from the waveform based on the division condition.

Effects of the Invention

[0006] According to the present invention, there is an effect that a normal waveform for generating a unit space in a method (MT method) for detecting an abnormality can be extracted.

Brief Description of the Drawings

[0007]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Embodiments for Carrying Out the Invention

[0008] Hereinafter, embodiments of a system, a method, and a program will be described in detail with reference to the accompanying drawings.

[0009] FIG. 1 is a diagram showing an example of the configuration of a predictive maintenance system according to the present embodiment. As shown in FIG. 1, the predictive maintenance system according to the present embodiment includes an electronic blackboard 2, an inkjet printer 3, a smartphone 4, a PC (Personal Computer) 5, an omnidirectional imaging device 6, a video conferencing terminal 7, a projector 8, an MFP (Multi-Function Peripheral) 9, and the like. The electronic blackboard 2, the inkjet printer 3, the smartphone 4, the PC 5, the omnidirectional imaging device 6, the video conferencing terminal 7, the projector 8, the MFP 9, and the like are communicably connected to each other via a communication network 100. In the present embodiment, the inkjet printer 3, the MFP 9, and the like may be an example of production equipment.

[0010] FIG. 2 is a hardware configuration diagram of a PC (or server) according to the present embodiment. Here, the hardware configuration of the server 5 will be described. As shown in FIG. 2, the server 5 is constructed by a computer and includes, as shown in FIG. 2, a CPU (Central Processing Unit) 501, a ROM (Read Only Memory) 502, a RAM (Random Access Memory) 503, an HD 504, an HDD (Hard Disk Drive) controller 505, a display 506, an external device connection I / F (Interface) 508, a network I / F 509, a data bus 510, a keyboard 511, a pointing device 512, a DVD-RW (Digital Versatile Disk Rewritable) drive 514, and a media I / F 516.

[0011] Among these, the CPU 501 controls the operation of the entire server 5. The ROM 502 stores programs used for driving the CPU 501 such as the IPL. The RAM 503 is used as a work area for the CPU 501. The HD 504 stores various data such as programs. The HDD controller 505 controls the reading or writing of various data to and from the HD 504 according to the control of the CPU 501. The display 506 displays various information such as a cursor, menu, window, characters, or images. The external device connection I / F 508 is an interface for connecting various external devices. External devices in this case are, for example, a USB (Universal Serial Bus) memory, a printer, etc. The network I / F 509 is an interface for performing data communication using the communication network 100. The data bus (bus line) 510 is an address bus, a data bus, etc. for electrically connecting each component such as the CPU 501 shown in FIG. 2.

[0012] Also, the keyboard 511 is a type of input means having a plurality of keys for inputting characters, numerical values, various instructions, etc. The pointing device 512 is a type of input means for selecting and executing various instructions, selecting a processing target, moving a cursor, etc. The DVD-RW drive 514 controls the reading or writing of various data to and from the DVD-RW 513 as an example of a removable recording medium. Note that it may be not limited to DVD-RW, but may also be DVD-R, etc. The media I / F 516 controls the reading or writing (storage) of data to and from the recording medium 515 such as a flash memory.

[0013] Incidentally, when the product type of the production equipment (manufacturing apparatus) such as the inkjet printer 3 and the MFP 9 changes, the operation of the production equipment changes accordingly, and thus different waveforms appear even if the production equipment is normal. Here, the waveform may be the waveform of a digital signal output from the production equipment (an example of the object) targeted for predictive maintenance. For example, the waveform may be the vibration waveform of the production equipment (vibration waveform), or the current waveform (or voltage waveform) output from the motor that operates the production equipment. When the production equipment is performing the same operation, the same waveform appears if the production equipment is normal, provided there are no disturbances such as the opening and closing of the door or vibrations picked up by people walking.

[0014] As an example of a change in the operation of the production equipment, when the mounting position varies depending on the substrate mounted on the chip mounter, there is an example where the movement of the arm that moves the nozzle changes. Also, as an example of a change in the operation of the production equipment, in the drying process, when switching the model according to the production plan for a plurality of types of products, the product is heated by a heater, but since the heating time varies depending on the product, there is an example where the operation time of the production equipment itself varies.

[0015] And in order to determine whether the production equipment is normal or abnormal, the MT method may be used. In the present embodiment, the MT method will be used as a method for detecting abnormalities. For example, the MT method uses the normal waveform of the current waveform of the motor of the production equipment or the vibration waveform of the production equipment as the unit space (reference), and determines whether the production equipment is normal or abnormal based on whether it deviates from that reference. Here, the MT method mainly analyzes whether the production equipment is normal or abnormal based on the Mahalanobis distance. Also, here, the normal population is called the unit space and is used as a criterion for determining whether the production equipment is normal or abnormal. And the MT method quantitatively determines whether the production equipment is normal or abnormal based on the degree of deviation of the sample to be judged (for example, the waveform for judging whether it is normal or abnormal) from the unit space.

[0016] If the facilities in a factory or the like suddenly break down, there will be problems such as a decrease in the operating rate and a delay in the production plan. Although the factory staff conducts regular inspections manually, there are also problems such as differences in inspection accuracy depending on the person, facilities located at high places, and difficulties in inspecting facilities in dangerous places such as those handling chemicals. Also, although parts are regularly replaced to prevent equipment failures, it is not known whether the replacement cycle is appropriate. Equipment failure prediction maintenance systems that monitor the vibration and voltage waveforms of equipment by attaching sensors to the equipment to monitor the deterioration of the equipment are already known.

[0017] For equipment such as fans and pumps that are operating stably continuously, waveforms in a constant operating state can be obtained no matter when the measurement is taken. However, most manufacturing equipment operates intermittently, and regular measurements will pick up a lot of data at inappropriate timings including when it is stopped. To avoid this, by setting a trigger which is the measurement start condition, it becomes possible to not start the measurement in the stopped state and wait until the measurement target generates the set acceleration. Also, for measurement targets where disturbances occur at startup, fine settings such as setting a certain acceleration or more as a disturbance to be ignored as a trigger condition, or waiting for a certain time after exceeding the set threshold value and then starting the measurement are possible. With these advanced trigger functions, it is possible to collect waveforms under certain operating conditions even for equipment operating in complex vibration modes, and technologies to improve the accuracy of detecting failure signs have also been developed. Also, it can flexibly respond to additional developments such as obtaining a measurement start signal by communication from an external PLC or the like.

[0018] However, since the waveforms measured differ for each type of workpiece being measured, it is difficult to easily find abnormal states. For each type of workpiece being measured, the normal state varies, making it difficult to distinguish it from the variation in the abnormal state.

[0019] Therefore, in the predictive maintenance system according to the present embodiment, by extracting similar (close) waveforms from the waveforms of various operations of an object such as a motor of production equipment, even for equipment that is not in a steady operation, a normal waveform for generating a unit space can be defined, and an abnormal state of the object can be detected.

[0020] FIG. 3 is a diagram showing an example of the functional configuration of a server included in the predictive maintenance system according to the present embodiment. The server 5 according to the present embodiment is an example of a system that detects an abnormality of an object using the MT method. As shown in FIG. 3, the server 5 includes a reception unit 301, a reception section 302, an extraction unit 303, a display control unit 304, a calculation unit 305, and a unit space generation unit 306.

[0021] The reception unit 301 is an example of a reception unit that receives waveforms of a plurality of patterns of digital signals output from a motor (an example of an object) of production equipment that is a target of predictive maintenance.

[0022] The reception section 302 receives, as a division condition for extracting a divided waveform that is a candidate for a normal waveform (an example of a unit space) for generating a unit space in the MT method, a threshold value for extracting the divided waveform, a minimum time until the threshold value is exceeded, an extraction start point before the time when the threshold value is exceeded, and an extraction time for extracting a waveform from the extraction start point. Here, the unit space is used in a method for detecting an abnormality such as the MT method. The MT method determines whether an object is normal or abnormal based on the Mahalanobis distance from the unit space. Further, the reception section 302 receives an operation for changing the division condition displayed on the display 506 by the display control unit 304 described later. Further, the reception section 302 receives settings of the MT method including determination of feature amounts, generation of unit spaces, deletion of highly correlated feature amounts, and confirmation of the validity of test data.

[0023] The extraction unit 303 is an example of an extraction unit that extracts divided waveforms that are candidates for normal waveforms from the waveforms received by the reception unit 301 based on the division conditions received by the reception unit 302. As a result, similar (close) waveforms appear at the start of the operation of the object or at other locations with a large load. By extracting similar (close) waveforms from various operation waveforms, it is possible to define the normal state even for objects that are not in a steady operation. Therefore, it is possible to extract normal waveforms for generating a unit space in a method (for example, the MT method) for detecting abnormalities in an object. Further, when the extraction unit 303 receives a change in the division conditions from the reception unit 302, the extraction unit 303 may extract the divided waveforms based on the changed division conditions.

[0024] When a plurality of divided waveforms are extracted, the calculation unit 305 extracts an extraction waveform from the plurality of divided waveforms based on preset extraction conditions. Specifically, the calculation unit 305 is an example of a calculation unit that calculates the score of each of the plurality of divided waveforms and selects an extraction waveform from among the plurality of divided waveforms based on the score. Here, the score may be the difference between the maximum value and the minimum value of the divided waveform, or the average of a predetermined number of upper or lower values of the maximum value or the longest value of the divided waveform. Further, the calculation unit 305 may change a predetermined number based on the calculation result of the score. Further, the calculation unit 305 may change the score to 0 when the difference between the maximum value and the minimum value of the divided waveform exceeds the upper limit value. Further, the calculation unit 305 may calculate the standard deviation of any number of upper scores and change the score to 0 when the standard deviation is greater than or equal to a predetermined value.

[0025] The display control unit 304 is an example of a display control unit that displays the divided waveforms extracted by the extraction unit 303 on the display 506 (an example of a display unit). The display control unit 304 may display the divided waveforms, the division conditions, and the number of divisions of the divided waveforms divided from the received waveforms on the display 506. Further, the display control unit 304 may display a list box (an example of a list) indicating information about the divided waveforms, the divided waveforms, the extraction waveforms, and the selected waveforms that are the divided waveforms selected from the list box on the display 506. Further, the display control unit 304 may display the divided waveforms extracted based on the changed division conditions on the display 506.

[0026] The unit space generation unit 306 is an example of a unit space generation unit that generates a unit space based on the divided waveforms extracted by the extraction unit 303. The unit space generation unit 306 applies the conditions for extracting similar (close) waveforms set by the calculation unit 305 to a plurality of measured waveforms of about 1000. It determines the feature quantities to be used in the MT method for those waveforms and generates a unit space. Here, whether the unit space is an effective unit space in which abnormal test data can be determined as abnormal by the MT method may include determination of feature quantities, generation of unit spaces, deletion of highly correlated feature quantities, confirmation of the validity of test data, deletion of waveforms that cannot be extracted, deletion of waveforms that clearly deviate, etc. Further, the unit space generation unit 306 issues an alert when the abnormality score of the waveform calculated by the MT method is equal to or higher than a threshold value.

[0027] FIG. 4 is a diagram for explaining an example of the current waveform of a motor during the operation of a manufacturing apparatus that operates intermittently. Specifically, the manufacturing apparatus (an example of an object) may be a chip mounter. The waveform of the digital signal output from the manufacturing apparatus may be a current waveform measured from the trigger signal when the stage on which the substrate of the chip mounter is placed rises to the working position. The three signals (current waveforms) shown in FIG. 4 are examples of waveforms measured from the trigger signals of different types of substrates among the substrates of the chip mounter.

[0028] Also, in FIG. 4, the vertical axis represents current and the horizontal axis represents time. Also, in FIG. 4, 0 on the horizontal axis is the time when the current serving as the trigger signal is detected. Also, in FIG. 4, the waveforms indicated by reference numerals 401 to 403 are similar waveforms among the three current waveforms. Since the waveform of the current waveform is different for each substrate, if the entire current waveform is used as training data, there is a risk that an abnormality in the manufacturing apparatus cannot be detected.

[0029] Since the MT method uses the normal waveform (the waveform when the production equipment is normal) as the unit space (reference), when generating the unit space with the movement of the entire different current waveform as the normal waveform as shown in Fig. 4, there may be cases where the degree of abnormality does not appear for abnormal current waveforms (the waveforms when the production equipment is abnormal). Therefore, in this embodiment, the predictive maintenance system can detect abnormal waveforms by extracting current waveforms with similar shapes (for example, the current waveforms indicated by reference numerals 401 to 403) from different current waveforms by the extraction unit 303 and generating the unit space.

[0030] Specifically, the extraction unit 303 determines characteristic and approximate current waveforms (divided waveforms) from a plurality of patterns of current waveforms (an example of the waveforms received by the reception unit 301).

[0031] Fig. 5 is a diagram for explaining an example of setting division conditions so that the divided waveforms to be extracted can be extracted. Here, an example of setting division conditions in a server 5 (such as a waveform analysis device) having the following four division conditions (1) to (4) will be described. In this embodiment, the reception unit 302 can display the UI shown in Fig. 5 on the display 506 or the like to set the four division conditions (1) to (4).

[0032] (1) The number of samples near the amplitude of the waveform being 0 (the minimum time until exceeding the threshold value, around 500 ms in the example shown in Fig. 5), in other words, the minimum time during which the waveform with an amplitude within the threshold value continues. That is, the minimum time is the time when it is determined that the waveform is not present. (2) The threshold value for determining the reference position of the divided waveform (extraction target waveform) to be extracted (the amplitude value of the divided waveform to be extracted (for example, the amplitude value in the + direction)) (the threshold value when it is determined that the waveform is not present, and after the time of (1) has elapsed while the waveform is within this threshold value, the position where the waveform exceeds the threshold value is used as the reference position) (3) The extraction start position for starting the extraction of the divided waveform to be extracted (the time before the reference position (the position where the threshold value is exceeded in the extraction target waveform)) The value of the extraction start position is a certain period from the reference position for determining the time before the reference position. (4) Number of samples to be extracted (extraction time for extracting the divided waveform from the extraction start position)

[0033] The procedure for setting the division conditions in the method for extracting the divided waveform to be extracted may be the procedures shown in the following (1) to (3). (1) Based on the state where no waveform appears during the intermittent operation (almost no waveform appears, only a little noise appears, or the equipment is stopped). (2) The determination of the state where no waveform appears is defined by an arbitrary threshold value and a minimum time.

[0034] Here, as the threshold value, a value with a margin (indicated by reference numeral 404) is set for the value of the vertical axis (sensor output value, amplitude of the waveform) that is determined to have almost no waveform appearing (or it may be input as "0+-").

[0035] Also, here, the minimum time may be the time (indicated by reference numeral 405) for determining that no waveform appears. For example, the minimum time may be input as "the number of samples near 0". If the minimum time is set to be as long as possible, the division conditions become strict and waveforms that should be extracted may not be extracted. If the minimum time is set too short, the division conditions become loose and waveforms that should not be extracted may be extracted.

[0036] The point exceeding the threshold value from the state determined that no waveform appears under the above conditions is taken as the reference position (the reference position indicated by reference numeral 406). The reference position, which is the point exceeding the threshold value, is also called the threshold or threshold position. Then, by inputting and specifying how many samples before (for example, "-50" before) the extraction start time (extraction start position) is before the reference position, since the reference position is about "530", the extraction start time is set to about "480". Further, the time for which extraction is to be performed from the extraction start time (for example, "480 to 700" or "480 to 800" can be set longer) is input as the number of samples to be extracted.

[0037] (3) When a plurality of divided waveforms are extracted by the procedures of (1) and (2) above, the extraction waveform is selected from among the plurality of divided waveforms.

[0038] FIG. 6 is a diagram for explaining an example of a process of selecting a waveform to be extracted in the predictive maintenance system according to the present embodiment. In the present embodiment, as shown in FIG. 6, the display control unit 304 displays the divided waveform and the number of divisions on the display 506 or the like. Since the division condition becomes stricter as the number of samples near 0 of the voltage (V), which is an example of the amplitude of the waveform, is larger, the user can adjust the voltage (V) so that the number of divisions becomes smaller.

[0039] Also, in the present embodiment, as shown in FIG. 6, the display control unit 304 displays on the display 506 or the like a list box showing the extracted divided waveforms, a divided waveform that is a candidate for a normal waveform, an extracted waveform extracted from a plurality of divided waveforms based on a preset extraction condition, and a selected waveform that is a divided waveform selected from the list box. In the list box, the threshold position, score, range average, and standard deviation are listed for each of the extracted divided waveforms. The user can check the threshold value and the like and confirm whether the divided waveform to be extracted is displayed in the list box. If it is not displayed, the user reviews the division condition. When the division condition is reviewed and any of the division conditions is changed, the extraction unit 303 extracts the divided waveform based on the changed division condition, and the display control unit 304 changes the display of the list box, the divided waveform, the extracted waveform, and the selected waveform on the display 506 according to the changed division condition.

[0040] When a plurality of divided waveforms are extracted under the division condition set above, a score for selecting the divided waveform to be extracted may be determined, and the divided waveform with the highest score may be selected as the extracted waveform. Here, examples of the score may be the following (1) to (3). (1) The difference between the maximum value and the minimum value of the extracted divided waveform is used as the score. (2) Since the maximum value (or minimum value) of the extracted divided waveform may suddenly take a high value due to external disturbance or the like, the average of a predetermined number of upper (or lower) values of the maximum value (or minimum value) of the extracted divided waveform is used as the score. (3) If the score of the segmented waveform to be extracted is not the highest value, adjust the score so that the score becomes the highest using the following method. Change the predetermined number of upper (or lower) values used as scores among the maximum or minimum values of the extracted segmented waveform. Determine the upper limit value of the difference to be adopted as the score among the differences between the maximum and minimum values of the extracted segmented waveform. Then, set the score of the difference exceeding the upper limit value to 0. Calculate the standard deviation of any number of upper scores, and change the scores when the standard deviation is greater than or equal to a predetermined value to 0. Then, to determine whether the desired extracted waveform has been extracted, check the screens shown in FIGS. 6 and 7.

[0041] FIGS. 7 and 8 are diagrams for explaining an example of a setting method for using the MT method in the predictive maintenance system according to the present embodiment. In the present embodiment, the predictive maintenance system determines whether an object such as production equipment is normal or abnormal by the MT method. Settings for determining whether an object is abnormal or normalized by the MT method include determination of feature quantities, generation of unit spaces, deletion of highly correlated feature quantities, and confirmation of the validity of test data if available. FIGS. 7 and 8 are screens for setting unit spaces that are displayed after extracting the segmented waveform on the screen of FIG. 6. On this screen, the unit space is set based on the waveform extracted in FIG. 6. The reception unit 302 receives the settings input on this screen.

[0042] For example, as shown in FIG. 7, the predictive maintenance system may display on the display 506 a screen on which feature quantities such as the maximum value (MAX) and minimum value (MIN) of the entire waveform and the extracted waveform, the difference (RANGE) between the maximum value and the minimum value, the average (AVE) of the entire waveform and the extracted waveform, and the standard deviation (DEV) of the entire waveform and the extracted waveform can be set as settings for determining whether an object is abnormal or normalized by the MT method. At this time, since the average of the entire waveform becomes 0, it is not included in the feature quantities. Also, the maximum and minimum values of the entire waveform and the extracted waveform, and the difference between the maximum value and the minimum value have a strong correlation, and if a plurality of them are included, the unit space may not be set properly.

[0043] Also, for example, as shown in FIG. 8, the predictive maintenance system may display on the display 506 a screen for setting feature quantities (quantity of existence, quantity of change, existence, small existence, no large, no small, threshold position, etc.) to be included as settings for determining whether an object is abnormal or normal by the MT method.

[0044] FIG. 9 is a diagram for explaining an example of the storage process of a unit space in the predictive maintenance system according to the present embodiment. In the MT method, generally, it is said that four or more deviate from the set. Therefore, a threshold value is set based on four, but the threshold value may be individually adjusted with data on malfunctions of actual equipment, etc. Examples of the setting of the threshold value may be a Caution threshold value for sending an alert email when it appears several times at an arbitrary time, a Danger threshold value for sending an alert email when it appears several times at an arbitrary time, etc. In the present embodiment, as shown in FIG. 9, when a unit space is determined, the respective threshold values of Caution and Danger, Saveflag (whether to save the raw data of OK itself), etc. are specified, and when the save button is pressed, the unit space is saved in the DB and the save completion is lit.

[0045] As described above, according to the predictive maintenance system according to the present embodiment, since similar (close) waveforms appear at the start of the operation of the object or at other locations with a large load, by extracting similar (close) waveforms from various operation waveforms, it is possible to define a normal state even for non-steady operation objects, and thus it is possible to extract a normal waveform for generating a unit space in the MT method.

[0046] Each function of the embodiments described above can be realized by one or more processing circuits. Here, the "processing circuit" in this specification refers to a processor programmed to execute each function by software, such as a processor implemented by an electronic circuit, an ASIC (Application Specific Integrated Circuit) designed to execute each function described above, a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), and devices such as conventional circuit modules.

[0047] The device group described in the embodiments merely shows one of a plurality of computing environments for implementing the embodiments disclosed in this specification. In one embodiment, the server 5 includes a plurality of computing devices such as a server cluster. The plurality of computing devices are configured to communicate with each other via any type of communication link including a network or a shared memory, and implement the processing disclosed in this specification.

[0048] Note that the object such as production equipment is not limited to an image forming apparatus as long as it is a device having a communication function. The object may be, for example, an output device such as a PJ (Projector), an IWB (Interactive White Board), a digital signage, a HUD (Head Up Display) device, an industrial machine, an imaging device, a sound collecting device, a medical device, a network home appliance, a connected car, a notebook PC (Personal Computer), a mobile phone, a smartphone, a tablet terminal, a game machine, a PDA (Personal Digital Assistant), a digital camera, a wearable PC, or a desktop PC.

[0049] The correspondence table (table) in the specification may be generated by the learning effect of machine learning. Also, by classifying keywords and accounting items that may be included in the description of transaction details using machine learning, it may not be necessary to use the correspondence table. Here, machine learning is a technology for enabling a computer to acquire learning capabilities like those of a human. It refers to a technology in which a computer autonomously generates algorithms necessary for judgments such as data identification from pre-loaded learning data and applies this to new data to make predictions. The learning method for machine learning may be any one of supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, and deep learning. Furthermore, a learning method combining these learning methods may also be used, and the learning method for machine learning is not limited.

[0050] The program executed by the server 5 of the present embodiment is provided by being recorded on a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, a DVD (Digital Versatile Disk) in an installable format or an executable format file.

[0051] Also, the program executed by the server 5 of the present embodiment may be configured to be stored on a computer connected to a network such as the Internet and downloaded via the network for provision. Also, the program executed by the server 5 of the present embodiment may be configured to be provided or distributed via a network such as the Internet.

[0052] Alternatively, the program of the present embodiment may be configured to be provided by being pre - incorporated into a ROM502 or the like. The program executed on the server 5 of the present embodiment has a module configuration including the above - described respective parts (reception unit 301, reception section 302, extraction unit 303, display control unit 304, calculation unit 305, and unit space generation unit 306). As actual hardware, an example of a processor such as a CPU501 reads the program from the above - described storage medium and executes it, so that the above - described respective parts are loaded onto the main storage device, and the reception unit 301, reception section 302, extraction unit 303, display control unit 304, calculation unit 305, and unit space generation unit 306 are generated on the main storage device.

[0053] Aspects of the present invention are as follows, for example. <1> A system for detecting an abnormality of an object using the MT method, a reception unit that receives a waveform of a digital signal output from the object; a reception section that receives, as split conditions for extracting a split waveform that is a candidate for a normal waveform for generating a unit space in the MT method, an input of a threshold value for extracting the split waveform, a minimum time until the threshold value is exceeded, an extraction start point before the time point when the threshold value is exceeded, and an extraction time for extracting a waveform from the extraction start point, from the waveform; an extraction unit that extracts the split waveform that is a candidate for the normal waveform from the waveform based on the split conditions; A system comprising: <2> The system according to <1>, further comprising a display control unit that displays the split waveform extracted by the extraction unit on a display unit. <3> When a plurality of the split waveforms are extracted, further comprising a calculation unit that calculates scores of each of the plurality of split waveforms and selects an extraction waveform from among the plurality of split waveforms based on the scores, The system according to <2>, wherein the display control unit displays the extraction waveform on the display unit. <4> The score is a difference between a maximum value and a minimum value of the split waveform, or an average of a predetermined number of upper or lower values of the maximum value or the minimum value of the split waveform. The calculation unit changes the predetermined number based on the calculation result of the score, changes the score to 0 when the difference exceeds the upper limit value, calculates the standard deviation of any number of the top scores, and changes the score when the standard deviation is greater than or equal to a predetermined value. The system according to <3>. <5> The display control unit displays the divided waveform, the division condition, and the number of divisions of the divided waveform divided from the waveform on the display unit. The system according to any one of <2> to <4>. <6> The reception unit receives an operation for changing the division condition displayed on the display unit. When the reception unit receives the change of the division condition, the extraction unit extracts the divided waveform based on the changed division condition. The display control unit displays the divided waveform extracted by the extraction unit on the display unit based on the changed division condition. The system according to <5>. <7> The display control unit displays a list indicating information about the divided waveform, the divided waveform, the extracted waveform, and a selected waveform that is the divided waveform selected from the list on the display unit. The system according to <3>. <8> The system according to any one of <1> to <7>, further comprising a unit space generation unit that generates the unit space based on the divided waveform extracted by the extraction unit. <9> The reception unit receives the setting of the MT method, including determination of feature amounts, generation of unit spaces, deletion of highly correlated feature amounts, and confirmation of the validity of test data. The system according to any one of <1> to <8>. <10> The unit space generation unit issues an alert when the abnormality score of the waveform calculated by the MT method is greater than or equal to a threshold value. The system according to <8>. <11> A method executed by a system for detecting an abnormality of an object using the MT method, receiving a waveform of a digital signal output from the object; As a splitting condition for extracting a split waveform that is a candidate for a normal waveform for generating a unit space in the MT method from the waveform, a step of receiving an input of a threshold value for extracting the split waveform, a minimum time until the threshold value is exceeded, an extraction start point before the time when the threshold value is exceeded, and an extraction time that is the time for extracting the waveform from the extraction start point; a step of extracting the split waveform that is a candidate for the normal waveform from the waveform based on the splitting condition; A method comprising: <12> A computer, a receiving unit that receives a waveform of a digital signal output from an object to be detected for abnormality using the MT method; a reception unit that receives an input of a threshold value for extracting the split waveform, a minimum time until the threshold value is exceeded, an extraction start point before the time when the threshold value is exceeded, and an extraction time that is the time for extracting the waveform from the extraction start point, as a splitting condition for extracting a split waveform that is a candidate for a normal waveform for generating a unit space in the MT method from the waveform; an extraction unit that extracts the split waveform that is a candidate for the normal waveform from the waveform based on the splitting condition; A program for causing the computer to function as such.

Explanation of Signs

[0054] 5 Server 301 Receiving unit 302 Reception unit 303 Extraction unit 304 Display control unit 305 Calculation unit 306 Unit space generation unit 501 CPU 502 ROM 503 RAM 504 HD 506 Display

Prior Art Documents

Patent Documents

[0055]

Patent Document 1

Claims

1. A system for detecting an abnormality of an object using the MT method, comprising: a receiving unit that receives a waveform of a digital signal output from the object; a receiving unit that receives, as input for a division condition for extracting, from the waveform, a division waveform that is a candidate for a normal waveform for generating a unit space in the MT method, a threshold value for extracting the division waveform, a minimum time until the threshold value is exceeded, an extraction start point before the time when the threshold value is exceeded, and an extraction time for extracting a waveform from the extraction start point; an extraction unit that extracts, from the waveform, the division waveform that is a candidate for the normal waveform based on the division condition; A system comprising:

2. The system according to claim 1, further comprising a display control unit that displays the division waveform extracted by the extraction unit on a display unit.

3. When a plurality of the division waveforms are extracted, further comprising a calculation unit that calculates scores of the respective plurality of the division waveforms and selects an extraction waveform from among the plurality of the division waveforms based on the scores, The system according to claim 2, wherein the display control unit displays the extraction waveform on the display unit.

4. The score is a difference between a maximum value and a minimum value of the division waveform, or an average of a predetermined number of upper or lower values of the maximum value or the minimum value of the division waveform, The calculation unit changes the predetermined number based on a calculation result of the score, changes the score to 0 when the difference exceeds an upper limit value, calculates a standard deviation of any number of upper scores, and changes the score to 0 when the standard deviation is equal to or greater than a predetermined value. The system according to claim 3.

5. The system according to any one of claims 2 to 4, wherein the display control unit displays the division waveform, the division condition, and the number of divisions of the division waveform divided from the waveform on the display unit.

6. The receiving unit receives an operation for changing the division condition displayed on the display unit, When the receiving unit receives the change of the division condition, the extraction unit extracts the division waveform based on the changed division condition, The system according to claim 5, wherein the display control unit displays the division waveform extracted by the extraction unit based on the changed division condition on the display unit.

7. The system according to claim 3, wherein the display control unit displays, on the display unit, a list indicating information regarding the divided waveform, the divided waveform, the extracted waveform, and a selected waveform that is the divided waveform selected from the list.

8. The system according to claim 1 or 2, further comprising a unit space generation unit that generates the unit space based on the divided waveform extracted by the extraction unit.

9. The system according to claim 8, wherein the reception unit receives the setting of the MT method, including determination of feature amounts, generation of a unit space, deletion of highly correlated feature amounts, and confirmation of the validity of test data.

10. The system according to claim 8, wherein the unit space generation unit issues an alert when an abnormality score of the waveform calculated by the MT method is equal to or greater than a threshold value.

11. A method executed in a system for detecting an abnormality of an object using the MT method, the method comprising: receiving a waveform of a digital signal output from the object; receiving, as an input for extraction conditions for extracting, from the waveform, a divided waveform that is a candidate for a normal waveform for generating a unit space in the MT method, the threshold for extracting the divided waveform, the minimum time until the threshold is exceeded, the extraction start point before the time when the threshold is exceeded, and the extraction time for extracting the waveform from the extraction start point; extracting, from the waveform, the divided waveform that is a candidate for the normal waveform based on the extraction conditions; The method including the above steps.

12. A program for causing a computer to function as: a reception unit that receives a waveform of a digital signal output from an object for detecting an abnormality using the MT method; a reception unit that receives, as an input for extraction conditions for extracting, from the waveform, a divided waveform that is a candidate for a normal waveform for generating a unit space in the MT method, the threshold for extracting the divided waveform, the minimum time until the threshold is exceeded, the extraction start point before the time when the threshold is exceeded, and the extraction time for extracting the waveform from the extraction start point; an extraction unit that extracts, from the waveform, the divided waveform that is a candidate for the normal waveform based on the extraction conditions. ​

Images