Analysis system, and analysis method

Abstract

Provided is an analysis system capable of improving the accuracy of predictive maintenance for equipment to be maintained that generates vibrations during operation. 【Solution means】The analysis system includes a signal acquisition unit that acquires vibration signals detected by sensors attached to the target equipment during operation, an intensity calculation unit that calculates a first plurality of signal intensities by analyzing the vibration signals, a distance calculation unit that calculates a first Mahalanobis distance of a signal space based on the first plurality of signal intensities at a first time for a unit space, a data acquisition unit that acquires state data of the target equipment, an abnormality determination unit that determines the presence or absence of an abnormality of the target equipment based on the state data when the first Mahalanobis distance is equal to or greater than a threshold value, and a setting unit that resets the unit space based on a second plurality of signal intensities calculated by analyzing the vibration signals acquired in a predetermined period near the first time when the target equipment is determined not to be abnormal.

Description

Technical Field

[0001] The present disclosure relates to an analysis system and an analysis method.

Background Art

[0002] Conventionally, as a method for inspecting abnormalities in machines, there is known a method of determining the presence or absence of an abnormality in a device by detecting a signal caused by abnormal vibration during the operation of the device.

[0003] For example, the vibration analysis system according to International Publication No. 2022 / 065103 (Patent Document 1) calculates a plurality of signal intensities by analyzing vibration signals corresponding to an object, and based on the Mahalanobis distance of a signal space composed of the plurality of signal intensities with respect to a preset unit space, it is configured to predict the timing of occurrence of an abnormality in the object.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] When the Mahalanobis distance of the signal space with respect to the unit space is large, the vibration analysis system according to Patent Document 1 outputs warning information for warning of the occurrence of an abnormality in the target device (for example, a pump). However, when vibrations (disturbances) occurring other than the pump due to equipment replacement or the like near the pump accumulate, the Mahalanobis distance between the signal space based on the vibration signal affected by the disturbance and the reference unit space becomes large. In this case, warning information is output even though there is no immediate need to maintain the pump, so it is not possible to perform highly accurate predictive maintenance.

[0006] An object of certain aspects of the present disclosure is to provide an analysis system and an analysis method capable of improving the accuracy of predictive maintenance for equipment to be maintained that generates vibrations during operation. **Means for Solving the Problems**

[0007] An analysis system according to an embodiment includes a signal acquisition unit that acquires vibration signals detected by sensors attached to a target device during operation, and calculates a plurality of first signal intensities respectively corresponding to a plurality of frequency bands by analyzing the vibration signals corresponding to the target device. An intensity calculation unit, a distance calculation unit that calculates a first Mahalanobis distance of a signal space based on the plurality of first signal intensities at a first time with respect to a preset unit space, and a data acquisition unit that acquires state data related to the operating state of the target device. When the first Mahalanobis distance is equal to or greater than a threshold value, an abnormality determination unit that determines the presence or absence of an abnormality of the target device based on the state data, and when it is determined that the target device is not abnormal, a plurality of frequency bands calculated by analyzing the vibration signals acquired in a predetermined period near the first time. A setting unit that resets the unit space based on the second plurality of signal intensities respectively corresponding to the above.

[0008] Preferably, the intensity calculation unit calculates a third plurality of signal intensities respectively corresponding to a plurality of frequency bands by analyzing vibration signals acquired after the first time. The distance calculation unit calculates a second Mahalanobis distance of the signal space based on the third plurality of signal intensities with respect to the unit space reset by the setting unit. When the second Mahalanobis distance is equal to or greater than a threshold value, the abnormality determination unit determines the presence or absence of an abnormality of the target device based on the state data.

[0009] Preferably, the target device is a pump. The state data includes at least one of the pressure of the fluid discharged by the pump, the current value for driving the pump, the frequency for driving the pump, and the temperature of the pump housing.

[0010] Preferably, when at least one of the first condition that the pressure is less than the first reference value, the second condition that the current value is greater than or equal to the second reference value, the third condition that the frequency is greater than or equal to the third reference value, and the fourth condition that the temperature is greater than or equal to the fourth reference value is satisfied, the abnormality determination unit determines that the pump is abnormal.

[0011] Preferably, the preset unit space is a first unit space composed of a plurality of signal intensities respectively corresponding to a plurality of frequency bands, which is calculated by analyzing the vibration signal corresponding to the target device in a normal state. The distance calculation unit calculates, as the first Mahalanobis distance, the Mahalanobis distance between the signal space composed of the plurality of signal intensities at the first time and the first unit space.

[0012] Preferably, the preset unit space is a second unit space composed of two-dimensional centroid data indicating the centroid positions of a plurality of signal intensities respectively corresponding to a plurality of frequency bands, which is calculated by analyzing the vibration signal corresponding to the target device in a normal state. The distance calculation unit calculates, as the first Mahalanobis distance, the Mahalanobis distance between the signal space composed of the two-dimensional centroid data at the first time and the second unit space.

[0013] Preferably, the analysis system further includes an output control unit that outputs warning information when it is determined that the target device is abnormal.

[0014] Preferably, when the first Mahalanobis distance is greater than or equal to the threshold value, the output control unit displays the first Mahalanobis distance on the display in a display mode different from that when the first Mahalanobis distance is less than the threshold value.

[0015] The analysis method according to another embodiment includes the steps of: obtaining a vibration signal detected by a sensor attached to a target device in operation; calculating a first plurality of signal intensities respectively corresponding to a plurality of frequency bands by analyzing the vibration signal corresponding to the target device; calculating a first Mahalanobis distance of a signal space based on the first plurality of signal intensities at a first time with respect to a preset unit space; obtaining state data regarding the operating state of the target device; determining the presence or absence of an abnormality in the target device based on the state data when the first Mahalanobis distance is equal to or greater than a threshold value; and resetting the unit space based on a second plurality of signal intensities respectively corresponding to a plurality of frequency bands, which are calculated by analyzing the vibration signal obtained in a predetermined period near the first time, when it is determined that the target device is not abnormal.

Advantages of the Invention

[0016] According to the present disclosure, it is possible to improve the accuracy of predictive maintenance for a maintenance target device that generates vibrations during operation.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Mode for Carrying Out the Invention

[0018] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed descriptions thereof will not be repeated.

[0019] <System Configuration> FIG. 1 is a diagram for explaining the outline of the system 1000. Referring to FIG. 1, the system 1000 is a system for determining the presence or absence of an abnormality in a target device by analyzing a vibration signal generated during the operation of a target device to be maintained such as a pump (hereinafter, also simply referred to as a "target device"), and performing predictive maintenance on the target device. Predictive maintenance is to detect any abnormality occurring in the target device to be maintained and perform operations such as parts maintenance and replacement before the state where the target device has to be stopped.

[0020] Hereinafter, it will be described assuming that the target device is a pump, but it is not limited thereto, and the system 1000 can be applied to any target device that generates vibration (or sound) during operation. For example, the system 1000 can also be applied to the determination of abnormalities in a motor, a part that vibrates due to vibration received from a vibrating body, etc.

[0021] System 1000 includes an analysis system 100, a plurality of sensors 30, a terminal device 40, a network 50, a monitoring device 60, and a plurality of pumps 70. The analysis system 100 performs vibration analysis of the pumps 70. The analysis system 100 includes an analysis device 10 and a sensor unit 20. The sensor unit 20 is electrically connected to the plurality of sensors 30. In the system 1000, two sensor units 20 are connected to the analysis device 10, but a configuration in which three or more sensor units 20 or one sensor unit 20 are connected to the analysis device 10 may also be used. Each sensor unit 20 may be electrically connected to one sensor 30. Each sensor unit 20 may be electrically connected to a plurality of sensors 30 respectively attached to the plurality of pumps 70.

[0022] The sensor 30 is attached to the pump 70 and acquires a detection signal (vibration signal) detected due to the vibration and sound of the pump. The analysis device 10 performs vibration analysis of the pump 70 based on the vibration signal input from the sensor 30 via the sensor unit 20.

[0023] The monitoring device 60 monitors state data regarding the operating state of the pump 70. The state data includes at least one of the pressure of the fluid discharged by the pump 70 (hereinafter, also referred to as "discharge pressure J1"), the current value J2 for driving the pump 70, the frequency J3 for controlling the output of the pump 70, and the temperature J4 of the housing of the pump 70. Typically, the discharge pressure J1, the current value J2, the frequency J3, and the temperature J4 are measured by various measuring devices (for example, a pressure sensor, a current sensor, a frequency meter, a temperature sensor) provided in the pump 70. The monitoring device 60 receives the state data measured by the various measuring devices. The state data is data for directly confirming an abnormality of the pump 70. Therefore, when the state data is an abnormal value, there is a high possibility that a substantial abnormality has occurred in the pump 70.

[0024] The analysis device 10 is configured to be communicable with the monitoring device 60, and acquires (receives) status data from the monitoring device 60. The analysis device 10 executes an abnormality determination of the pump 70 based on the vibration analysis result and the status data of the pump 70. Details of the abnormality determination method will be described later. The analysis device 10 is configured to be communicable with the terminal device 40 via the network 50. The analysis device 10 transmits the analysis result, the determination result, etc. to the terminal device 40.

[0025] Typically, the analysis device 10 has a structure that follows a general-purpose computer architecture, and realizes various processes described later by the processor executing a pre-installed program. The analysis device 10 is, for example, a laptop PC (Personal Computer). However, the analysis device 10 may be any device that can execute the functions and processes described below, and may be other devices (for example, a desktop PC, a tablet terminal device).

[0026] The network 50 includes various networks such as the Internet. The network 50 may adopt a wired communication method or may adopt other wireless communication methods such as a wireless LAN (local area network).

[0027] The terminal device 40 is, for example, a portable tablet terminal device. However, the terminal device 40 is not limited to this, and may be realized by a smartphone, a desktop PC (Personal Computer), etc.

[0028] In the example of FIG. 1, the analysis system 100 is configured as a separate device in which the analysis device 10 and the sensor unit 20 are separated, but may be configured as an integrated device of the analysis device 10 and the sensor unit 20. Also, the analysis device 10 and the monitoring device 60 may be configured as an integrated device. In this case, the analysis device 10 has the above functions of the monitoring device 60 and directly receives status data measured by various measuring instruments.

[0029] FIG. 2 is a block diagram showing an example of the overall configuration of the analysis system 100. Referring to FIG. 2, the analysis system 100 includes an analysis device 10 and a sensor unit 20.

[0030] The sensor 30 connected to the sensor unit 20 is a sensor capable of detecting signals of vibration and sound, and is constituted by, for example, an acceleration sensor using a piezoelectric element. Note that the sensor 30 may be any sensor capable of detecting signals of vibration and sound, and may be constituted by an acceleration sensor of another type (for example, servo type), or may be constituted by various other sensors.

[0031] When the signal obtained by the sensor 30 is a charge signal, a charge converter is provided between the sensor 30 and the analysis system 100. In this case, the charge converter converts the charge signal from the sensor 30 into a voltage signal and outputs it to the analysis system 100. Note that when the sensor 30 has a function of converting a charge signal into a voltage signal, the charge converter is unnecessary.

[0032] The sensor unit 20 converts the vibration signal acquired from the sensor 30 (or the charge converter) into a signal that can be processed by the analysis device 10. Specifically, the sensor unit 20 includes a filter 21, an amplifier 22, and an A / D converter 23.

[0033] The filter 21 is an analog filter that removes noise components from the vibration signal output from the sensor 30. The filter 21 is constituted by a low-pass filter, a high-pass filter, or the like.

[0034] The amplifier 22 amplifies the analog signal output from the filter 21 by a predetermined multiple and outputs the amplified signal to the A / D converter 23.

[0035] The A / D converter 23 converts the signal input from the amplifier 22 from an analog signal into a digital signal at a predetermined sampling frequency. The A / D converter 23 outputs the digitally converted signal to the analysis device 10.

[0036] FIG. 3 is a block diagram showing an example of the hardware configuration of the analysis device 10. Referring to FIG. 3, the analysis device 10 includes a processor 101, a memory 103, a display 105, an input device 107, a signal input interface (I / F) 109, and a communication interface (I / F) 111. These components are connected to each other so as to be capable of data communication.

[0037] The processor 101 is typically an arithmetic processing unit such as a CPU (Central Processing Unit) or an MPU (Multi Processing Unit). The processor 101 controls the operations of the respective components of the analysis device 10 by reading and executing a program stored in the memory 103. More specifically, the processor 101 realizes each function of the analysis device 10 by executing the program.

[0038] The memory 103 is realized by a RAM (Random Access Memory), a ROM (Read-Only Memory), a flash memory, a hard disk, or the like. The memory 103 stores programs and the like executed by the processor 101.

[0039] The display 105 is, for example, a liquid crystal display, an organic EL (Electro Luminescence) display, or the like. The display 105 may be integrally configured with the analysis device 10 or may be configured separately from the analysis device 10.

[0040] The input device 107 receives operation inputs to the analysis device 10. The input device 107 is realized by, for example, a keyboard, buttons, a mouse, or the like. Further, the input device 107 may be realized as a touch panel.

[0041] The signal input interface 109 mediates data transmission between the processor 101 and the sensor unit 20. The signal input interface (I / F) 109 receives the input of the vibration signal from the sensor 30 via the sensor unit 20. Specifically, the signal input interface 109 receives the input of the digital signal from the A / D converter 23.

[0042] The communication interface 111 mediates data transmission between the processor 101 and the terminal device 40, the monitoring device 60, etc. As the communication method, for example, a wireless communication method such as Bluetooth (registered trademark), wireless LAN (Local Area Network), etc. is used. Note that a wired communication method such as USB (Universal Serial Bus) may be used as the communication method.

[0043] <Abnormality determination method> The abnormality determination method according to this embodiment will be described. The abnormality determination method includes a preparation step of preparing reference data and an analysis step of analyzing the abnormal state of the pump 70.

[0044] (Preparation step) In the preparation step according to this embodiment, for example, the vibration state of the pump 70 at the initial stage of operation start is measured. Since the pump 70 at the initial stage of operation start is in a new state, the vibration state of the normal pump 70 is measured as a reference. However, instead of the pump 70, a normal pump of the same type as the pump 70 may be separately prepared, and the vibration state of the pump may be measured as a reference.

[0045] FIG. 4 is a flowchart showing an example of the preparation step. Typically, each of the following steps is realized by the processor 101 of the analysis device 10 executing a program stored in the memory 103.

[0046] Referring to FIG. 4, the processor 101 acquires the vibration signal output from the sensor 30 via the sensor unit 20 (step S10). Specifically, the processor 101 acquires the vibration signal corresponding to the normal pump 70 from the sensor 30 (the vibration signal indicating the vibration state of the pump 70).

[0047] The processor 101 performs octave analysis on the vibration signals accumulated for a predetermined time (for example, several tens to several hundreds of milliseconds) (step S12). In this embodiment, 1 / 3 octave analysis is used. Therefore, each vibration signal is separated into 48 bands, for example, from 0.4 Hz to 20 kHz, by a 1 / 3 band-pass filter, and the signal intensity (vibration intensity) is averaged for each band (that is, frequency band). In the following description, the signal intensity averaged in the frequency band is also simply referred to as the "signal intensity of the frequency band".

[0048] For each frequency band, the processor 101 stores, in the memory 103, the signal intensity of the corresponding frequency band of the normal pump 70 as reference data R1 (step S14).

[0049] FIG. 5 is a diagram showing an example of a data set of the signal intensities of each frequency band. Referring to FIG. 5, the data set 310 includes the signal intensities L of each frequency band f1 to fn (where n is a natural number) for each time T1 to Tm (where m is a natural number). When each vibration signal is separated into 48 bands, n = 48. For example, the data set 310 includes the signal intensities L1_1 to L1_n of each frequency band f1 to fn at time T1, and includes the signal intensities Lm_1 to Lm_n of each frequency band f1 to fn at time Tm. The reference data R1 includes, for example, the signal intensities L of each frequency band f1 to fn for each time T1 to Tx. In this case, the period from time T1 to Tx corresponds to the initial operation period.

[0050] Again, referring to FIG. 4, the processor 101 sets the unit space U1 in the Mahalanobis-Taguchi method (MT method) using the reference data R1 (step S16). Specifically, the processor 101 sets the data of the signal intensity L in each frequency band f1 to fn at each time T1 to Tx as the unit space U1. The unit space U1 is used in the first abnormality determination step described later.

[0051] The processor 101 calculates two-dimensional centroid data indicating the centroid position of the signal intensity in each frequency band (step S18). Specifically, the processor 101 generates the data set shown in FIG. 6 from the data set 310.

[0052] FIG. 6 is a diagram showing an example of the data set of the centroid position. Referring to FIG. 6, the data set 320 includes the centroid position Gx of the frequency band and the centroid position Gy of the signal intensity for the times T1 to Tm. For each time T1 to Tm, two-dimensional centroid data (centroid positions Gx, Gy) is generated. The centroid position Gx is represented by the following equation (1), and the centroid position Gy is represented by the following equation (2).

[0053]

Equation

[0054]

Equation

[0055] f i represents the i-th frequency band, and L irepresents the signal intensity of the i-th frequency band, and S represents the sum of the signal intensities of all frequency bands. According to equations (1) and (2), a data set 320 composed of two-dimensional data (center of gravity positions Gx, Gy) is generated from a data set 310 composed of n-dimensional (multi-dimensional) data (signal intensities L of each frequency band f1~fn). In the data set 320, for example, the center of gravity positions Gx, Gy at time T1 are represented by the center of gravity positions Gf_1, GL_1 respectively, and the center of gravity positions Gx, Gy at time Tm are represented by the center of gravity positions Gf_m, GL_m respectively.

[0056] Referring to FIG. 4 again, the processor 101 sets the unit space U2 in the MT method using the center of gravity data during the initial operation period (for example, times T1~Tx) (step S20). Specifically, the processor 101 sets the two-dimensional center of gravity data (center of gravity positions Gx, Gy) at each time T1~Tx as the unit space U2. The unit space U2 is used in the second abnormality determination step described later.

[0057] (Analysis step) The analysis step includes a first abnormality determination step and a second abnormality determination step for determining the presence or absence of an abnormality in the pump 70.

[0058] FIG. 7 is a flowchart showing an example of the analysis step. Referring to FIG. 7, the processor 101 acquires the vibration signal output from the sensor 30 via the sensor unit 20 (step S30). Specifically, the processor 101 acquires the vibration signal corresponding to the pump 70 during the normal period (for example, the operation period after the end of the initial operation period) from the sensor 30.

[0059] Processor 101 performs octave analysis on the vibration signal accumulated for a predetermined time (step S32). Processor 101 stores, in memory 103, the signal intensity of the frequency band corresponding to pump 70 for each frequency band (step S34). Specifically, the signal intensity of each frequency band in pump 70 at a certain time Ts is stored in the form of dataset 310. Here, a series of signal intensities Ls_1 to Ls_n of frequency bands f1 to fn in pump 70 at time Ts is also referred to as signal intensity data Ps. Since the period from time T1 to Tx corresponds to the initial operation period, the normal period corresponds to the period from time Tx + 1 to time Tm. Therefore, time Ts is any one of time Tx + 1 to time Tm.

[0060] Processor 101 executes a first abnormality determination step (step S40) and a second abnormality determination step (step S50) using the signal intensity data Ps and the data obtained in the preparation step. These steps may be executed in parallel or sequentially.

[0061] [First Abnormality Determination Step] FIG. 8 is a flowchart showing an example of the first abnormality determination step. In the first abnormality determination step, a dataset 310 composed of multidimensional data is used. Referring to FIG. 8, processor 101 calculates the Mahalanobis distance MD1 of signal space X1s composed of a plurality of signal intensities with respect to unit space U1 set in step S16 of FIG. 4 (step S41). Signal space X1s is composed of a plurality of signal intensities Ls_1 to Ls_n (that is, signal intensity data Ps) at time Ts.

[0062] Processor 101 determines whether the Mahalanobis distance MD1 is equal to or greater than a threshold Th1 (for example, 4) (step S42). If the Mahalanobis distance MD1 is less than the threshold Th1 (NO in step S42), processor 101 ends the first abnormality determination step. If the Mahalanobis distance MD1 is equal to or greater than the threshold Th1 (YES in step S42), processor 101 acquires state data regarding the operating state of pump 70 from monitoring device 60 (step S43).

[0063] FIG. 9 is a diagram showing an example of a dataset of state data. Referring to FIG. 9, the dataset 330 includes state data D1 to Dp (where p is a natural number) for each time t1 to tg (where g is a natural number). In the dataset 330, the state data D1 to Dp at time t1 are represented by state data D1_1 to Dp_1, respectively, and the state data D1 to Dp at time tg are represented by state data D1_g to Dp_g, respectively. For example, the state data D1, D2, D3, D4 are the discharge pressure J1, current value J2, frequency J3, and temperature J4 of the pump 70, respectively.

[0064] The processor 101 receives from the monitoring device 60 a dataset 330 including state data for a certain past period (for example, the period from time t1 to tg). Note that the processor 101 does not necessarily have to receive the state data in the form of the dataset 330. For example, the processor 101 may be configured to specify specific state data (for example, the state data D2_3 at time t3, etc.) and receive the state data.

[0065] Referring to FIG. 8 again, the processor 101 determines whether the state data is normal (step S44). For example, assume that the state data is the discharge pressure J1. In this case, the processor 101 determines that the state data (that is, the discharge pressure J1) is normal when the discharge pressure J1 is equal to or higher than the reference value E1, and determines that the state data is not normal (that is, abnormal) when the discharge pressure J1 is less than the reference value E1.

[0066] For example, when the pump 70 has no control mechanism, the discharge pressure of the pump 70 operating at rated conditions under certain conditions serves as an indicator of the pump capacity. Therefore, a decrease in the discharge pressure means a lack of pump capacity. On the other hand, when the pump 70 has a control mechanism and is controlled to discharge a constant pressure, the control value (such as current, frequency, etc.) for controlling the discharge pressure fluctuates to keep the discharge pressure constant. However, if the control value deviates from the control range due to an abnormality in the pump 70 or the like, the discharge pressure decreases. Therefore, regardless of whether the pump 70 has a control mechanism, when the discharge pressure of the pump 70 drops below the reference value E1, it is considered that an abnormality has occurred in the pump 70.

[0067] When the state data is abnormal (NO in step S44), the processor 101 determines that some abnormality has occurred in the pump 70 and outputs an abnormality alert (step S45). Typically, the abnormality alert is displayed on the display 105. Note that the abnormality alert may be configured to be output as sound via a speaker.

[0068] On the other hand, when the state data is normal (YES in step S44), the processor 101 determines that no abnormality has occurred in the pump 70 (that is, the pump 70 is in a normal state), and re - sets (updates) the unit space U1 preset in the preparation process (step S46).

[0069] FIG. 10 is a flowchart showing an example of the process of re - setting the unit space. Referring to FIG. 10, the processor 101 selects the signal strength L used for re - setting the unit space U1 (step S102). Specifically, the processor 101 selects the signal strength L obtained by analyzing the vibration signal acquired during a predetermined period Y near the time Ts.

[0070] Here, the processor 101 performs a determination process that the Mahalanobis distance MD1 of the signal space X1s at time Ts with respect to the unit space U1 set in step S16 is greater than or equal to the threshold Th1 (YES in step S43), and a determination process that the state data is normal (YES in step S44), and then executes a process of resetting the unit space U1 (step S46). The predetermined period Y is set to a fixed period, for example, from time T(s - a×n) to time T(s + a×n). Note that, for example, "a = 50".

[0071] As an example, considering the case where the vibration signal changes rapidly near time Ts, the predetermined period Y is set to a fixed period after time Ts. For example, the predetermined period Y is the period from time Ts to time T(s + 5×n) (the period of "5×n"). In this case, in step S102, the processor 101 selects the signal intensities L of each frequency band f1 to fn from time Ts to T(s + 5×n) as the signal intensity L for resetting the unit space U1.

[0072] As another example, when the vibration signal changes gradually near time Ts, the predetermined period Y may be set to a fixed period before time Ts. In this case, for example, the predetermined period Y is the period from time T(s - 5×n) to time Ts. As yet another example, the predetermined period Y may be set to the period before and after including time Ts. In this case, for example, the predetermined period Y is the period from time T(s - h) to time T(s + 5×n - h). However, "h" is a natural number that satisfies h < 5×n.

[0073] Note that the predetermined period Y can be appropriately changed by the user. Subsequently, the processor 101 stores the selected signal strength L in the memory 103 as new reference data R2 (step S104). The processor 101 reconfigures the unit space U1 using the reference data R2 (step S106). For example, the processor 101 sets the data of the signal strength L of each frequency band f1 to fn at each time Ts to T(s + 5×n) as a new unit space U1. Hereinafter, the reconfigured unit space U1 is also referred to as "unit space U1x".

[0074] Here, the reason for executing the above-described reconfiguration process of the unit space U1 will be described in detail. FIG. 11 is a diagram showing the time changes of the Mahalanobis distance and the discharge pressure. The horizontal axis of FIG. 11 indicates time (for example, time "1" to "10000"), the left vertical axis indicates the Mahalanobis distance MD1, and the right vertical axis indicates the discharge pressure as state data. FIG. 11 shows an example in which the pump 70 that was already in operation at the initial time point (that is, time "1") failed (reached the end of its life) near time "10000". Here, the case where the state data is the discharge pressure J1 will be described, but the state data may be the current value J2, the frequency J3, or the temperature J4.

[0075] Referring to FIG. 11, the graph 401 shows the Mahalanobis distance MD1 (hereinafter, also referred to as "Mahalanobis distance MD1_1") between the initial unit space U1 and the signal space at each time. The graph 402 shows the Mahalanobis distance MD1 (hereinafter, also referred to as "Mahalanobis distance MD1_2") between the reconfigured unit space U1x and the signal space at each time. The graph 403 shows the discharge pressure when the initial time point is "1".

[0076] In the example of FIG. 11, the threshold Th1 of the Mahalanobis distance MD1 is set to "6". The reference value E1 of the discharge pressure is set to "0.8". When the discharge pressure becomes 0.7 or less, it is determined that the pump 70 has failed. Assume that the initial unit space U1 (for example, the unit space U1 set in step S16 of the preparation process) is set at the initial time point.

[0077] From around the time "1400" to around the time "3000", the Mahalanobis distance MD1_1 rises sharply. Specifically, around the time "2500", the Mahalanobis distance MD1_1 becomes equal to or greater than the threshold value Th1 and indicates an outlier. At this time, the discharge pressure is maintained at or above the reference value E1 and remains within the normal range. In such a case, it is considered that for some reason the vibration state of the pump 70 has changed, and as a result, the Mahalanobis distance MD1_1 has increased, but the pump performance is normal.

[0078] Therefore, the processor 101 determines that the pump 70 is maintaining a normal state (not abnormal), and re - sets the unit space U1 using the signal intensity L in the predetermined period Y after the time "2500". The processor 101 sets, for example, the signal intensity L from the time "2500" to the time "2740" (that is, the predetermined period Y) as the re - set unit space U1x. By using the re - set unit space U1x, the processor 101 calculates the Mahalanobis distance MD1_2 based on the state after the vibration state of the pump 70 has changed.

[0079] After around the time "2900", the Mahalanobis distance MD1_2 of the signal space at each time with respect to the unit space U1x is shown. As shown in FIG. 11, since the Mahalanobis distance MD1_2 is the Mahalanobis distance based on the unit space U1x after the change in the vibration state, it is understood that it is smaller than the Mahalanobis distance MD1_1 based on the unit space U1.

[0080] As time passes, the discharge pressure of the pump 70 gradually decreases and becomes less than the reference value E1 around the time "6100", but there is no practical problem, and the pump 70 continues to operate. An increase in the Mahalanobis distance MD1_2 cannot be confirmed around the time "6100".

[0081] After around the time "8000", the Mahalanobis distance MD1_2 has been rising sharply. Specifically, around the time "9000", the Mahalanobis distance MD1_2 becomes equal to or greater than the threshold Th1, indicating an outlier. Also, around the time "9000", the discharge pressure is less than the reference value E1, suggesting insufficient pump performance. Thus, since the Mahalanobis distance MD1_2 becomes equal to or greater than the threshold Th1 and the discharge pressure is less than the reference value E1, the processor 101 determines that an abnormality has occurred in the pump 70. Referring to the graph 401 around the time "9000", although the Mahalanobis distance MD1_1 has an upward trend, its change is becoming difficult to observe. Then, around the time "10000", the pump 70 fails.

[0082] As described above, according to the abnormality determination method that determines the abnormality of the pump 70 using the Mahalanobis distance MD1_2 and the discharge pressure with respect to the reset unit space U1x, the abnormality can be determined at a timing slightly before the timing when a sudden failure of the pump 70 occurs (for example, around the time "9000").

[0083] On the other hand, for example, when adopting an abnormality determination method that uses only the Mahalanobis distance MD1_1 based on the initial unit space U1, it is conceivable to determine the abnormality of the pump 70 when the Mahalanobis distance MD1_1 becomes equal to or greater than the threshold Th1 (for example, around the time "2500"). However, at this point, there is no substantial abnormality in the pump 70, and it is too early as the timing for maintaining the pump 70, so it is not appropriate.

[0084] Also, when adopting an abnormality determination method that uses the Mahalanobis distance MD1_1 and the discharge pressure, it is conceivable to determine the abnormality of the pump 70 when the Mahalanobis distance MD1_1 is equal to or greater than the threshold Th1 and the discharge pressure is less than the reference value E1 (for example, around the time "6100"). At this point, although the deterioration of the pump 70 has progressed, the pump 70 can still operate normally. Therefore, it is considered to be slightly early as the timing for maintaining the pump 70.

[0085] In addition, when adopting an abnormality determination method that uses only the discharge pressure without using the Mahalanobis distance, when the discharge pressure becomes less than the reference value E1 (for example, around time "6100"), or when the discharge pressure becomes less than the limit value (for example, 0.7) (for example, around time "10000"), it is conceivable to determine the abnormality of the pump 70. When determining the abnormality of the pump 70 when the discharge pressure becomes less than the reference value E1, similar to the above, this time point is slightly early as the timing for maintaining the pump 70. Also, when determining the abnormality of the pump 70 when the discharge pressure becomes less than the limit value, at this time point, since there is a high possibility that the pump 70 has already failed (or is about to fail), this time point is too late as the timing for maintaining the pump 70.

[0086] Therefore, according to the abnormality determination method of the present embodiment based on the Mahalanobis distance MD1_2 and the discharge pressure, since the abnormality can be determined at a timing slightly before the timing when a failure occurs, the pump 70 can be operated for a longer time, and the pump 70 can be maintained before a sudden failure occurs. That is, it becomes possible to improve the accuracy of predictive maintenance for the pump 70.

[0087] [Second Abnormality Determination Step] FIG. 12 is a flowchart showing an example of the second abnormality determination step. Referring to FIG. 12, the processor 101 calculates two-dimensional centroid data indicating the centroid position of the signal intensity in each frequency band (step S51). Specifically, the processor 101 calculates two-dimensional centroid data (that is, centroid positions Gx, Gy) from the signal intensity data Ps stored in step S34 of FIG. 7 using equations (1) and (2).

[0088] The processor 101 calculates the Mahalanobis distance MD2 of the signal space X2s composed of two-dimensional centroid data with respect to the unit space U2 set in step S20 of FIG. 4 (step S52). The signal space X2s is composed of the centroid positions Gx, Gy (that is, Gf_s, GL_s) at time Ts.

[0089] The processor 101 determines whether the Mahalanobis distance MD2 is greater than or equal to a threshold Th2 (for example, 4) (step S53). If the Mahalanobis distance MD2 is less than the threshold Th2 (NO in step S53), the processor 101 ends the second abnormality determination process. If the Mahalanobis distance MD2 is greater than or equal to the threshold Th2 (YES in step S53), the processor 101 executes the process of step S54. Note that the processes of steps S54, S55, and S56 are the same as the processes of steps S43, S44, and S45 in FIG. 8, respectively, and thus their detailed descriptions will not be repeated.

[0090] When the state data is normal (YES in step S55), the processor 101 reconfigures (updates) the unit space U2 set in the preparation process (step S57). The reconfiguration method of the unit space U2 is basically the same as the reconfiguration method of the unit space U1 described in FIG. 10.

[0091] Specifically, the processor 101 selects the two-dimensional centroid data (centroid positions Gx, Gy) obtained by analyzing the vibration signals acquired in a predetermined period Y near the time Ts. For example, the processor 101 selects the centroid data (centroid positions Gx, Gy) at each time Ts to T(s + 5×n) as the centroid data for reconfiguring the unit space U2. The processor 101 sets the two-dimensional centroid data at each selected time Ts to T(s + 5×n) as the reconfigured unit space U2x.

[0092] According to the second abnormality determination process, similar to the first abnormality determination process described above, it is possible to improve the accuracy of predictive maintenance for the pump 70 as described in FIG. 11.

[0093] <Example of screen> FIG. 13 is a diagram showing an example of the layout of the user interface screen 500. However, the user interface screen 500 may have any layout as long as it can realize the functions described later, and may have a layout other than that shown in FIG. 13.

[0094] Referring to FIG. 13, the user interface screen 500 includes display areas 502, 504, 506, a display area 514 for measurement conditions and set values, various buttons 516, display areas 520, 530, 540, and graphs 550, 560.

[0095] In the display area 502, the identification number (unit number) of the sensor unit 20, the identification number (sensor number) of the sensor 30, the name of the measurement object (e.g., pump), etc. are displayed. In the display area 504, the Mahalanobis distance MD1 or MD2 (hereinafter also collectively referred to as "Mahalanobis distance MD") and the status of the abnormality determination result based on the discharge pressure are displayed. For example, according to the abnormality determination result, the status "normal" or "abnormal" is displayed.

[0096] In the display area 506, the value (score) of the Mahalanobis distance MD1 and the score of the Mahalanobis distance MD2 are displayed. The color of the score changes according to the calculated score of the Mahalanobis distance MD. For example, when the Mahalanobis distance MD1 is greater than or equal to the threshold Th1, the score is displayed in a color (e.g., red) that images "abnormal". Also, when the Mahalanobis distance MD1 is less than the threshold Th1, the score is displayed in a color (e.g., black) that images "normal". Note that it is sufficient that the "abnormal" score is emphasized (made more prominent) than the "normal" score.

[0097] In the display area 520, the time-series sensor data (raw data) detected by the sensor 30 is shown. In the display area 530, the analysis result (frequency spectrum) obtained by performing a fast Fourier transform (FFT) analysis on the time-series sensor data is displayed. In the display area 540, the signal intensity data obtained by performing 1 / 3 octave analysis on the time-series sensor data is displayed as a bar graph. The graph 550 shows the time-series data of the Mahalanobis distance MD1. The graph 560 shows the time-series data of the Mahalanobis distance MD2.

[0098] By visually recognizing the display areas 504, 506, etc., the user can check the status of the pump 70. For example, when the status "abnormal" is displayed, the user can start equipment maintenance such as inspection, maintenance, and repair of the pump 70.

[0099] <Functional configuration> FIG. 14 is a functional block diagram of the analysis device 10. Referring to FIG. 14, the analysis device 10 mainly includes a signal acquisition unit 202, an intensity calculation unit 204, a setting unit 206, a distance calculation unit 208, a data acquisition unit 210, an abnormality determination unit 212, and an output control unit 214. Each of these functions is realized, for example, by the processor 101 of the analysis device 10 executing a program stored in the memory 103. Note that some or all of these functions may be configured to be realized by hardware.

[0100] The signal acquisition unit 202 acquires a vibration signal detected by the sensor 30 attached to the operating pump 70. Specifically, the signal acquisition unit 202 receives the vibration signal (digital signal) detected by the sensor 30 via the sensor unit 20.

[0101] The intensity calculation unit 204 calculates a plurality of signal intensities corresponding to a plurality of frequency bands by analyzing the vibration signal acquired by the signal acquisition unit 202. Specifically, the intensity calculation unit 204 performs octave analysis (for example, 1 / 3 octave analysis) on the vibration signal corresponding to the pump 70 to calculate the signal intensity of each frequency band (for example, the signal intensity L of each frequency band f1 to fm). Note that the intensity calculation unit 204 may be configured to calculate the signal intensity of each frequency band by fast Fourier transform (FFT).

[0102] The setting unit 206 sets a unit space based on a plurality of signal strengths calculated by the strength calculation unit 204. Specifically, the strength calculation unit 204 analyzes a vibration signal corresponding to the pump 70 in a normal state to calculate a plurality of signal strengths (for example, signal strengths L of each frequency band f1 to fn at each time T1 to Tx) respectively corresponding to a plurality of frequency bands. In one aspect, the setting unit 206 sets a unit space composed of a plurality of signal strengths at each time T1 to Tx as an initial unit space U1. In another aspect, the setting unit 206 calculates two-dimensional centroid data indicating the centroid position of a plurality of signal strengths at each time T1 to Tx, and sets a unit space composed of the centroid data as an initial unit space U2.

[0103] The distance calculation unit 208 calculates the Mahalanobis distance of a signal space based on a plurality of signal strengths at a first time (for example, time Ts) with respect to the unit space using the MT method. For example, the distance calculation unit 208 calculates the Mahalanobis distance MD1 of the signal space X1s with respect to a preset initial unit space U1. Also, the distance calculation unit 208 calculates the Mahalanobis distance MD2 of the signal space X2s with respect to a preset initial unit space U2.

[0104] The data acquisition unit 210 acquires state data regarding the operating state of the pump 70. The state data includes at least one of the discharge pressure J1, current value J2, frequency J3, and temperature J4.

[0105] The abnormality determination unit 212 determines the presence or absence of an abnormality in the pump 70 based on the Mahalanobis distance MD and the state data. In one aspect, the abnormality determination unit 212 determines whether the Mahalanobis distance MD1 is greater than or equal to a threshold value Th1. When the Mahalanobis distance MD1 is greater than or equal to the threshold value Th1, the abnormality determination unit 212 determines the presence or absence of an abnormality in the pump 70 based on the state data. Specifically, when the state data is normal (for example, when the discharge pressure J1 is greater than or equal to the reference value E1), the abnormality determination unit 212 determines that the pump 70 is normal, and when the state data is abnormal (for example, when the discharge pressure J1 is less than the reference value E1), the abnormality determination unit 212 determines that the pump 70 is abnormal.

[0106] As another example, when the current value J2 is less than the reference value E2, the abnormality determination unit 212 determines that the pump 70 is normal, and when the current value J2 is greater than or equal to the reference value E2, the abnormality determination unit 212 determines that the pump 70 is abnormal. This is because the current value increases as the load on the pump 70 increases.

[0107] Also, when the frequency J3 is less than the reference value E3, the abnormality determination unit 212 determines that the pump 70 is normal, and when the frequency J3 is greater than or equal to the reference value E3, the abnormality determination unit 212 determines that the pump 70 is abnormal. This is because when the discharge pressure of the pump 70 does not reach the specified pressure for some reason, control to increase the frequency is executed.

[0108] Furthermore, when the temperature J4 is less than the reference value E4, the abnormality determination unit 212 determines that the pump 70 is normal, and when the temperature J4 is greater than or equal to the reference value E4, the abnormality determination unit 212 determines that the pump 70 is abnormal. This is because when the mechanical load increases due to wear of the parts of the pump 70 or the like, the temperature of the housing rises due to heat generated by poor lubrication.

[0109] From the above, when at least one of the first condition that the discharge pressure J1 is less than the reference value E1, the second condition that the current value J2 is greater than or equal to the reference value E2, the third condition that the frequency J3 is greater than or equal to the reference value E3, and the fourth condition that the temperature J4 is greater than or equal to the reference value E4 is satisfied, the abnormality determination unit 212 may determine that the pump 70 is abnormal.

[0110] When it is determined that the pump 70 is not abnormal, the setting unit 206 resets the unit space (for example, unit spaces U1, U2) based on a plurality of signal strengths corresponding to a plurality of frequency bands in a predetermined period Y (for example, the period from time Ts to time T(s + 5×n)) near the first time. Note that the plurality of signal strengths are calculated by the intensity calculation unit 204 by analyzing the vibration signal acquired in the predetermined period Y. The setting unit 206 sets the plurality of signal strengths in the predetermined period Y (for example, each time Ts to T(s + 5×n)) as the unit space U1x, and sets the two-dimensional centroid data of the plurality of signal strengths in the predetermined period Y as the unit space U2x.

[0111] Also, the intensity calculation unit 204 calculates a plurality of signal strengths corresponding to a plurality of frequency bands by analyzing the vibration signal acquired after the first time (for example, time Ts). The distance calculation unit 208 calculates the Mahalanobis distance MD (for example, Mahalanobis distance MD1_2) of the signal space based on the plurality of signal strengths with respect to the unit space (for example, unit spaces U1x, U2x) reset by the setting unit 206. When the Mahalanobis distance MD is equal to or greater than the threshold value, the abnormality determination unit 212 determines the presence or absence of an abnormality in the pump 70 based on the state data in the same manner as described above.

[0112] When it is determined that the pump 70 is abnormal, the output control unit 214 outputs warning information (for example, an abnormality alert) warning of the occurrence of an abnormality in the pump 70. Also, when the Mahalanobis distance MD is equal to or greater than the threshold value, the output control unit 214 displays the Mahalanobis distance MD on the display 105 in a display mode different from when the Mahalanobis distance MD is less than the threshold value. For example, when the Mahalanobis distance MD1 is equal to or greater than the threshold value Th1, the output control unit 214 displays the Mahalanobis distance MD1 in red, and when the Mahalanobis distance MD1 is less than the threshold value Th1, the output control unit 214 displays the Mahalanobis distance MD1 in black.

[0113] <Advantages> According to this embodiment, since an abnormality can be determined at a timing slightly before the timing when a failure of the pump 70 occurs, the pump 70 can be maintained at a more appropriate timing. As a result, it becomes possible to improve the accuracy of predictive maintenance.

[0114] <Other Embodiments> (1) In the above-described embodiment, it is also possible to provide a program that causes a computer to function and execute control as described in the above flowchart. Such a program can be recorded on a non-temporary computer-readable recording medium such as a flexible disk, a secondary storage device, a main memory device, and a memory card attached to the computer, and provided as a program product. Alternatively, the program can be recorded on a recording medium such as a hard disk built into the computer and the program can be provided. Also, the program can be provided by downloading via a network.

[0115] (2) The configurations exemplified as the above-described embodiments are examples of the configuration of the present invention, and it is possible to combine them with other known technologies, or to change the configuration by omitting a part, etc., without departing from the gist of the present invention. Also, in the above-described embodiments, it may be the case where the processes and configurations described in other embodiments are appropriately adopted and implemented.

[0116] The embodiments disclosed this time should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is shown not by the above description but by the claims, and it is intended that all modifications within the meaning and scope equivalent to the claims are included.

Description of Reference Numerals

[0117] 10 Analyzer, 20 Sensor Unit, 21 Filter, 22 Amplifier, 23 A / D Converter, 30 Sensor, 40 Terminal Device, 50 Network, 60 Monitoring Device, 70 Pump, 100 Analysis System, 101 Processor, 103 Memory, 105 Display, 107 Input Device, 109 Signal Input Interface, 111 Communication Interface, 202 Signal Acquisition Unit, 204 Intensity Calculation Unit, 206 Setting Unit, 208 Distance Calculation Unit, 210 Data Acquisition Unit, 212 Abnormality Determination Unit, 214 Output Control Unit, 1000 System.

Claims

1. a signal acquisition unit that acquires a vibration signal detected by a sensor attached to a target device in operation; an intensity calculation unit that calculates a first plurality of signal intensities corresponding to a plurality of frequency bands by analyzing the vibration signal corresponding to the target device; a distance calculation unit that calculates a first Mahalanobis distance of a signal space based on the first plurality of signal intensities at a first time with respect to a preset unit space; a data acquisition unit for acquiring status data relating to an operating status of the target device; an abnormality determination unit that determines whether or not there is an abnormality in the target device based on the status data when the first Mahalanobis distance is equal to or greater than a threshold; and a setting unit that, when it is determined that the target device is not abnormal, reconfigures the unit space based on a second plurality of signal intensities corresponding to the plurality of frequency bands, which are calculated by analyzing the vibration signal acquired during a predetermined period around the first time.

2. the intensity calculation unit calculates a third plurality of signal intensities respectively corresponding to the plurality of frequency bands by analyzing the vibration signal acquired after the first time; the distance calculation unit calculates a second Mahalanobis distance of a signal space based on the third plurality of signal intensities with respect to the unit space reset by the setting unit; The analysis system according to claim 1 , wherein, when the second Mahalanobis distance is equal to or greater than the threshold value, the abnormality determination unit determines whether or not an abnormality exists in the target device based on the status data.

3. the target equipment is a pump, 3. The analysis system according to claim 1, wherein the status data includes at least one of a pressure of a fluid discharged by the pump, a current value for driving the pump, a frequency for driving the pump, and a temperature of a housing of the pump.

4. 4. The analysis system of claim 3, wherein the abnormality determination unit determines that the pump is abnormal when at least one of a first condition that the pressure is less than a first reference value, a second condition that the current value is equal to or greater than a second reference value, a third condition that the frequency is equal to or greater than a third reference value, and a fourth condition that the temperature is equal to or greater than a fourth reference value is satisfied.

5. the predetermined unit space is a first unit space configured by a plurality of signal intensities respectively corresponding to the plurality of frequency bands, the signal intensities being calculated by analyzing a vibration signal corresponding to the target device in a normal state; 3. The analysis system according to claim 1, wherein the distance calculation unit calculates, as the first Mahalanobis distance, a Mahalanobis distance of a signal space constituted by the first plurality of signal intensities at the first time with respect to the first unit space.

6. the predetermined unit space is a second unit space configured with two-dimensional center-of-gravity data indicating a center-of-gravity position of a plurality of signal intensities respectively corresponding to the plurality of frequency bands, the center-of-gravity data being calculated by analyzing a vibration signal corresponding to the target device in a normal state; 3. The analysis system according to claim 1, wherein the distance calculation unit calculates, as the first Mahalanobis distance, a Mahalanobis distance of a signal space constituted by the two-dimensional center of gravity data at the first time with respect to the second unit space.

7. The analysis system according to claim 1 , further comprising an output control unit that outputs warning information when it is determined that the target device is abnormal.

8. The analysis system according to claim 7, wherein when the first Mahalanobis distance is equal to or greater than the threshold, the output control unit displays the first Mahalanobis distance on a display in a different display manner than when the first Mahalanobis distance is less than the threshold.

9. acquiring a vibration signal detected by a sensor attached to a target device in operation; calculating a first plurality of signal strengths corresponding to a plurality of frequency bands by analyzing the vibration signal corresponding to the target device; Calculating a first Mahalanobis distance in a signal space based on the first plurality of signal intensities at a first time for a preset unit space; acquiring status data relating to an operational status of the target device; a step of determining whether or not there is an abnormality in the target device based on the status data when the first Mahalanobis distance is equal to or greater than a threshold value; and if it is determined that the target device is not abnormal, reconfiguring the unit space based on a second plurality of signal intensities corresponding to the plurality of frequency bands, which are calculated by analyzing the vibration signal acquired during a predetermined period around the first time.

Images