Rolling bearing replacement time prediction method and rolling bearing replacement time prediction device

Abstract

The invention provides a rolling bearing replacement time prediction method and a rolling bearing replacement time prediction device which can predict replacement time with higher precision. In the present invention, a frequency spectrum of vibration data indicating vibration of a rolling bearing is obtained for each sampling period and stored in a storage unit (6) in association with the sampling period, and a specific peak value that does not occur during normal is detected from the frequency spectrum within a first frequency range including a theoretical frequency at which a peak value occurs during an abnormality. When a specific peak value is detected, the present invention detects, as a re-detected specific peak value, a specific peak value within a second frequency range, which includes the specific peak value and is narrower than the first frequency range, for one or more frequency spectrums stored in the storage unit (6) before the sampling period. The replacement time of the rolling bearing is predicted on the basis of the specific peak value and the re-detection specific peak value.

Description

TECHNICAL FIELD

[0001] The present application relates to a rolling bearing replacement timing prediction method and a rolling bearing replacement timing prediction device for predicting a replacement timing of a rolling bearing. BACKGROUND

[0002] A rolling bearing is a device for supporting a load by interposing a rolling element such as a ball, a roller, or the like between two members (a shaft and a track wheel), and is equipped in a device for various uses with a rotating body. The rolling bearing can be obstructed from smoothly rolling due to an abnormality such as wear (abrasion, damage), fatigue due to deformation, and fusion due to pressure, and a failure of the aforementioned device can occur. Therefore, it is desirable to predict a remaining life of the rolling bearing to predict a replacement timing, and a rolling bearing abnormality detection device for predicting a remaining life of a rolling bearing is disclosed in Patent Document 1.

[0003] The rolling bearing abnormality detection device disclosed in Patent Document 1 is provided with a vibration sensor that detects a vibration frequency of a rolling bearing, a band pass filter that passes an output signal of the vibration sensor with a band pass characteristic selected in accordance with a contact frequency among an inner ring, an outer ring, and a ball calculated based on a geometric dimension of a bearing that is a measurement target, a conversion device that frequency-converts a time axis signal from the filter, and a calculation device that calculates a frequency spectrum intensity corresponding to the contact frequency based on an output signal of the device, and calculates a deterioration index of each of the inner ring, the outer ring, and the ball from a ratio of the calculated value to a reference frequency spectrum intensity, and stores an increase in the calculated frequency spectrum intensity over time, and predicts a remaining life of the rolling bearing from a speed of the increase.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT DOCUMENTS

[0006] Patent Document 1: Japanese Patent Application Laid-Open No. Hei 2-140451 SUMMARY

[0007] PROBLEMS TO BE SOLVED BY THE INVENTION

[0008] The abnormality detection device of the rolling bearing disclosed in the aforementioned Patent Document 1 calculates the contact frequency among the inner ring, the outer ring, and the balls based on the geometric dimensions, that is, the theoretical frequency that brings a peak on the frequency spectrum at the time of abnormality occurrence, and performs the abnormality detection and the remaining life prediction based on the calculated spectrum intensity. However, in reality, it is not necessarily the case that a peak is generated at the theoretical frequency at the time of abnormality occurrence, and there are cases where a peak is generated at a frequency deviated from the aforementioned theoretical frequency. In such a case, even if the spectrum intensity is calculated at the aforementioned theoretical frequency, the peak at the time of abnormality occurrence cannot be detected, and it can be impossible to predict the remaining life. Furthermore, in the aforementioned Patent Document 1, there is no description of the relationship between the increase speed of the spectrum intensity and the remaining life, and the manner of estimating the remaining life based on the increase speed of the spectrum intensity is not clear. Therefore, there is room for improvement in the prediction of the remaining life, that is, the prediction of the replacement timing, disclosed in the aforementioned Patent Document 1.

[0009] The present application has been made in view of the above-described circumstances, and an object thereof is to provide a rolling bearing replacement timing prediction method and a rolling bearing replacement timing prediction device that can more accurately predict the replacement timing of a rolling bearing.

[0010] Means for solving the technical problem

[0011] The present inventors have conducted various studies, and as a result, have found that the above object can be achieved by the following invention. That is, a rolling bearing replacement timing prediction method according to a technical solution of the invention includes: a spectrum processing step of acquiring vibration data representing vibrations generated in a rolling bearing at a predetermined acquisition interval and for a predetermined sampling period, and storing a frequency spectrum of the vibration data for the sampling period in a storage unit in correspondence with the sampling period; a peak detection step of detecting a specific peak that does not occur when the rolling bearing is normal from the frequency spectrum in a predetermined first frequency range that includes a theoretical frequency that brings about a peak on the frequency spectrum when an anomaly occurs; a peak re-detection step of, in a case where a specific peak is detected by the peak detection step, detecting a specific peak that does not occur when the rolling bearing is normal as a re-detected specific peak in a predetermined second frequency range that is narrower than the first frequency range and that includes a frequency of the specific peak detected by the peak detection step, from one or more frequency spectra stored in the storage unit before the sampling period in which the specific peak is detected; and a replacement timing prediction step of predicting a replacement timing of the rolling bearing based on the sampling period in which the specific peak is detected by the peak detection step and an amplitude of the detected specific peak, and the sampling period in which the re-detected specific peak is detected by the peak re-detection step and an amplitude of the detected re-detected specific peak. Preferably, in the above rolling bearing replacement timing prediction method, the replacement timing prediction step finds a fitting curve that best fits a central point of the sampling period in which the specific peak is detected by the peak detection step and the amplitude of the detected specific peak, and a central point of the sampling period in which the re-detected specific peak is detected by the peak re-detection step and the amplitude of the detected re-detected specific peak, in a coordinate space having time and amplitude as coordinate axes, finds a point at which the found fitting curve intersects with a predetermined determination amplitude value used to determine the replacement timing, as the replacement timing. Preferably, in the above rolling bearing replacement timing prediction method, the first frequency range is set so that the theoretical frequency is a frequency in the middle thereof.

[0012] Generally, in the case where some kind of failure that hinders smooth rolling of the rolling bearing occurs in a normal state, after the amplitude of the peak value that indicates the development of the aforementioned failure becomes large, the aforementioned peak value can be determined, and the frequency of the peak value can be specified. On the other hand, it is presumed that the peak value that is not specified occurs in the vicinity of the specified frequency. The rolling bearing replacement timing prediction method described above can more reliably detect the specified peak value because the specified peak value is detected in the first frequency range that includes the aforementioned theoretical frequency, and can more correctly detect the specified peak value that is caused by the occurrence of the aforementioned failure by excluding the confusing peak value that is included in the first frequency range but is not caused by the occurrence of the aforementioned failure from the detection target, and by re-detecting the specified peak value in the second frequency range that is narrower than the aforementioned first frequency range.

[0013] On the other hand, the time until the rolling bearing needs to be replaced, i.e., the remaining life ((the end point of the sampling period in the frequency spectrum (vibration data) in which the specified peak value is detected) + (the remaining life at that time) = (replacement timing)) depends on the development mode (tendency, trend over time) of the aforementioned failure. If the progress of the aforementioned failure develops gently, the remaining life thereof becomes long, and on the other hand, if the progress of the aforementioned failure develops rapidly, the remaining life thereof becomes short. The rolling bearing replacement timing prediction method described above can more accurately predict the replacement timing because it predicts the replacement timing based on the amplitude of the specified peak value detected by the aforementioned peak value detection process and the amplitude of the re-detected specified peak value detected by the aforementioned peak value re-detection process, in addition to detecting the specified peak value and re-detecting the specified peak value before it, and because it predicts the replacement timing taking into account the development mode of the aforementioned failure. Furthermore, the rolling bearing replacement timing prediction method described above can more accurately predict the replacement timing in the case where the replacement timing is predicted using a plurality of re-detected specified peak values, compared to the case where the replacement timing is predicted using one re-detected specified peak value.

[0014] In another aspect, in the above-described rolling bearing replacement timing prediction device, the peak re-detection process, in the case of detecting the re-detection specific peak in the past direction for the plurality of frequency spectra, with respect to each of the plurality of frequency spectra, takes each of the peak detection process and the specific peak detected by the peak detection process as each of an individual first peak re-detection process and a first re-detection specific peak, and detects the re-detection specific peak in the second frequency range including the re-detection specific peak detected by the first peak re-detection process for that frequency spectrum. Preferably, in the above-described rolling bearing replacement timing prediction method, the second frequency range is set so that the frequency of the re-detection specific peak becomes a central frequency. Preferably, in the above-described rolling bearing replacement timing prediction method, the peak re-detection process detects the re-detection specific peak in the past direction for the plurality of frequency spectra in succession. Preferably, in the above-described rolling bearing replacement timing prediction method, the peak re-detection process detects the re-detection specific peak in the past direction for the plurality of frequency spectra non-continuously.

[0015] Such a rolling bearing replacement timing prediction method can more reliably detect the re-detection specific peak even in the case where the frequency of the re-detection specific peak changes over time, since the re-detection specific peak is detected in the second frequency range including the re-detection specific peak detected by the first peak re-detection process.

[0016] In another aspect, in the above-described rolling bearing replacement timing prediction method, the peak detection process further detects the specific peak in the second frequency range including the specific peak detected by the first peak detection process for the sampling period in which the frequency spectrum stored in the storage section in the sampling period later than the sampling period in which the specific peak is detected.

[0017] Such a rolling bearing replacement timing prediction method can more reliably detect the specific peak even in the case where the frequency of the specific peak changes over time after detection, since the specific peak is detected in the second frequency range including the specific peak detected by the first peak detection process.

[0018] In another aspect, in the above-described rolling bearing replacement timing prediction method, the length of the sampling period is set based on the second frequency range; the narrower the second frequency range, the longer the sampling period is set to be.

[0019] Such a rolling bearing replacement timing prediction method can detect peaks from frequency spectrum data with a suitable frequency resolution corresponding to the range of the second frequency range.

[0020] The rolling bearing replacement timing prediction device according to the present application is a rolling bearing replacement timing prediction device that predicts a replacement timing of a rolling bearing, and includes: a vibration detection sensor that acquires vibration data representing vibration generated by the rolling bearing; a storage unit; and a control processing unit that performs a frequency spectrum processing, a peak value detection processing, a peak value re-detection processing, and a replacement timing prediction processing, wherein the frequency spectrum processing is processing of acquiring, at a predetermined acquisition interval, frequency spectra of the vibration data in a predetermined sampling period, for the vibration data of the sampling period, storing the acquired frequency spectra in the storage unit in association with the sampling period, and the peak value detection processing is processing of detecting, from the frequency spectra, a specific peak value that does not occur when the rolling bearing is normal, in a predetermined first frequency range that includes a theoretical frequency that brings about a peak on the frequency spectrum when an abnormality occurs.

[0021] The rolling bearing replacement timing prediction device can predict the replacement timing of the rolling bearing with higher accuracy.

[0022] Effects of the Invention

[0023] The rolling bearing replacement timing prediction method and the rolling bearing replacement timing prediction device of the present application can more reliably detect the specific peak value because the specific peak value is detected in the first frequency range including the theoretical frequency, and can more correctly detect the specific peak value because the specific peak value is detected again in the second frequency range narrower than the first frequency range, and the easily confusing peak value included in the first frequency range but not caused by the occurrence of the failure that hinders smooth rotation of the rolling bearing is excluded from the detection object. The rolling bearing replacement timing prediction method and the rolling bearing replacement timing prediction device of the present application can more accurately predict the replacement timing because the replacement timing is predicted based on the sampling period and the amplitude of the detected specific peak value in the case where the specific peak value is detected, and the sampling period and the amplitude of the detected again specific peak value in the case where the again detected specific peak value is detected, in addition to detecting the specific peak value and the again detected specific peak value before the specific peak value, and the replacement timing is predicted taking into account the development mode of the failure. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is a block diagram showing the structure of the rolling bearing replacement timing prediction device of the embodiment.

[0025] Figure 2 is a graph for explaining a predetermined length of time in vibration data.

[0026] Figure 3 is a graph for explaining a mechanical device provided with a rolling bearing.

[0027] Figure 4 is a graph for explaining the first and second frequency ranges.

[0028] Figure 5 is a graph for explaining the mode of prediction of the replacement timing.

[0029] Figure 6 is a flowchart showing the operation of the rolling bearing replacement timing prediction device with respect to the detection of the specific peak value before the specific peak value is detected.

[0030] Figure 7 is a flowchart showing the operation of the rolling bearing replacement timing prediction device with respect to the detection of the specific peak value after the specific peak value is detected.

[0031] Figure 8 is a block diagram showing the structure of the rolling bearing replacement timing prediction device of the modification of the embodiment. DETAILED DESCRIPTION

[0032] One or more embodiments of the present application will be described below with reference to the accompanying drawings. However, the scope of the application is not limited to the disclosed embodiments. In addition, structures indicated by the same reference numerals in the respective drawings represent the same structures, and the description thereof will be appropriately omitted. In the present specification, the reference numerals without the subscript are used to indicate the structures in the general case, and the reference numerals with the subscript are used to indicate the structures in the individual case.

[0033] Figure 1 is a block diagram showing the structure of a rolling bearing replacement timing prediction device according to an embodiment. Figure 2 is a graph for explaining a predetermined time length in vibration data. Figure 3 is a graph for explaining a mechanical device equipped with a rolling bearing. Figure 4 is a graph for explaining a first and a second frequency range. Figure 4 A indicates a first frequency range, Figure 4 B indicates a second frequency range. Figure 4 A and Figure 4 Each of the horizontal axes in the graphs of A and B is frequency [Hz], and each of the vertical axes is amplitude. Figure 5 is a graph for explaining a manner of predicting a replacement timing. Figure 5 The horizontal axis of is time (elapsed time), and the vertical axis thereof is amplitude.

[0034] The rolling bearing replacement timing prediction device 1000 according to an embodiment, for example, as shown in Figure 1 is provided with a vibration detection sensor 1, a control processing section 2, an input section 3, an output section 4, an interface section (IF section) 5, and a storage section 6.

[0035] The vibration detection sensor 1 is connected to the control processing section 2, and is a device that acquires vibration data indicating vibration generated by a rolling bearing in accordance with the control of the control processing section 2. In the present embodiment, a plurality of vibration data acquired at a predetermined acquisition interval for a predetermined sampling period by a vibration detection sensor that detects vibration generated at a rolling bearing is stored in the storage section 6.

[0036] As described later, the vibration data is converted from time space to frequency space by means of high-speed Fourier transform. In this case, the frequency resolution of the frequency spectrum depends on the number of data points used at the time of high-speed Fourier transform. The more the number of data points described above, the higher the frequency resolution, and vibration data for the predetermined sampling period described above corresponding thereto is required. Here, in a case where the sampling interval is set to SP, the number of data points is set to Nfft, and the length of the predetermined sampling period described above is set to TW, TW = SP x Nfft is obtained. The length TW of the predetermined sampling period described above is set based on the second frequency range FW2 described later in the present embodiment, and the narrower the second frequency range described above, the longer the period for which the sampling period described above is set. In order to detect an effective peak in the second frequency range ±△fw2, for example, as shown inFigure 2 As shown, the second frequency range ±△fw2 (= 2 x△fw2) needs to be divided into four or more. Therefore, the frequency resolution becomes 2 x△fw2 / 4 =△fw2 / 2 [Hz] or more, and thus the length TW of the aforementioned predetermined sampling period becomes 2 / △fw2 [sec] or more. For example, in a case where△fw2 is set to 0.02 [Hz] and the sampling interval SP is set to 0.2 [ms] (= 0.0002 [sec]), the length TW of the aforementioned predetermined sampling period becomes TW = 2 / 0.02 = 100 [sec] or more, and the number of data points Nfft becomes Nfft = 100 / 0.0002 = 500,000 [pieces] or more. High-speed Fourier transform generally handles powers of two, so the minimum number of 500,000 or more becomes 2^19 = 524,288 [pieces], and the length TW of the aforementioned predetermined sampling period becomes TW = 0.0002 x 524,288 = 104.8576 [sec].

[0037] The aforementioned vibration detection sensor 1 is one or more and is configured in an apparatus such as a machine or the like that has a rolling bearing as a monitoring target. The aforementioned machine is an example of an apparatus that has a rolling bearing, and can be any device as long as it has a rolling bearing. For example, the aforementioned machine M is Figure 3 As shown in the reduction gear M, a first to a third rolling bearing BE-1 to BE-3, a first and a second rotating shaft AX-1, AX-2, a first and a second gear GA-1, GA-2, and a box (housing) that houses these first to third rolling bearings BE-1 to BE-3, first and second rotating shafts AX-1, AX-2, and first and second gears GA-1, GA-2 are generally provided. The first rotating shaft AX-1 is fixed to the first gear GA-1, which is a rotating shaft of the first gear GA-1, and is supported by the first rolling bearing BE-1. The second rotating shaft AX-2 is fixed to the second gear GA-2, which is a rotating shaft of the second gear GA-2, and is supported by the second and third rolling bearings BE-2, BE-3. The first gear GA-1 engages with the second gear GA-2, and for example, rotational force caused by rotation of the first rotating shaft AX-1 is transmitted to the second rotating shaft AX-2 via the first and second gears GA-1, GA-2, and the second rotating shaft AX-2 rotates.

[0038] The vibration detection sensor 1 has three first to third vibration detection sensors 1-1 to 1-3 for such a structure of the speed reducer M. The first to third vibration detection sensors 1-1 to 1-3 are respectively arranged on each outer circumference of the first to third rolling bearings BE-1 to BE-3. Note that the vibration detection sensor 1 (1-1 to 1-3) is not limited to be arranged on the rolling bearings BE, and can be arranged on the aforementioned housing, for example. In any case, the first to third vibration detection sensors 1-1 to 1-3 are arranged at positions where vibrations caused by the rolling bearings BE propagate. Such first to third vibration detection sensors 1-1 to 1-3 are, for example, acceleration sensors, AE (Acoustic Emission) sensors, or the like, and appropriate sensors are used depending on the frequency of vibrations generated in the monitored object. The first to third vibration detection sensors 1-1 to 1-3 output detection results to the control processing portion 2 in this embodiment.

[0039] In this embodiment, for example, the control processing portion 2 instructs the first to third vibration detection sensors 1-1 to 1-3 to acquire the aforementioned vibration data if it becomes the start point of the predetermined sampling period, and in accordance with the instruction, the first to third vibration detection sensors 1-1 to 1-3 detect vibrations at each sampling point corresponding to the predetermined sampling interval, and output vibration data of the length TW of the aforementioned predetermined sampling period to the control processing portion 2. The control processing portion 2 stores and saves the aforementioned vibration data in association with the aforementioned sampling period (for example, the start time of the sampling period) to the storage portion 6.

[0040] The input portion 3 is connected to the control processing portion 2, and is, for example, a device that inputs various commands such as a command to instruct the start of the replacement timing prediction, and various data such as the name of the mechanical equipment of the monitored object required to cause the rolling bearing replacement timing prediction device 1000 to operate, and is, for example, a plurality of input switches, a keyboard, a mouse, or the like to which a predetermined function is assigned. The output portion 4 is connected to the control processing portion 2, and is a device that outputs the commands, data, and replacement timing, and the like input from the input portion 3 in accordance with the control of the control processing portion 2, and is, for example, a display device such as a CRT display, a liquid crystal display, and an organic EL display, a printing device such as a printer, or the like.

[0041] In addition, the input section 3 and the output section 4 can also constitute a so-called touch panel. In the case of constituting the touch panel, the input section 3 is a position input device that detects an operation position and inputs, for example, in a resistance film method, an electrostatic capacity method, or the like, and the output section 4 is a display device. In the touch panel, the aforementioned position input device is provided on a display surface of the aforementioned display device, and the aforementioned display device displays candidates of one or a plurality of input contents that can be input, and if a user touches a display position at which an input content that the user wants to input is displayed, the position is detected by the aforementioned position input device, and a display content at the detected position is input to the rolling bearing replacement timing prediction device 1000 as an operation input content of the user. In such a touch panel, since the user can easily intuitively understand an input operation, the rolling bearing replacement timing prediction device 1000 that is easy to operate for the user is provided.

[0042] The IF section 5 is connected to the control processing section 2, and is a circuit that performs input and output of data between the external device in accordance with control of the control processing section 2, and is, for example, an interface circuit for an RS-232C of a serial communication method, an interface circuit using a Bluetooth (registered trademark) specification, an interface circuit that performs infrared communication in an IrDA (Infrared Data Association) specification, and an interface circuit using a USB (Universal Serial Bus) specification, or the like. In addition, the IF section 5 is a circuit that performs communication with the external device, and can also be, for example, a data communication card, a communication interface circuit that complies with an IEEE 802.11 specification, or the like.

[0043] The storage section 6 is connected to the control processing section 2 and is a circuit that stores various predetermined programs and various predetermined data in accordance with the control of the control processing section 2. Among the aforementioned various predetermined programs, for example, there are included a control processing program, among which there are included a control program, a spectrum processing program, a peak value detection program, a peak value re-detection program, and a replacement timing prediction program. The aforementioned control program is a program that controls each of the sections 1, 3 to 6 of the rolling bearing replacement timing prediction device 1000 in accordance with the function of each section. The aforementioned spectrum processing program is a program that, from the mutually different plurality of vibration data in each of mutually different plurality of sampling periods respectively acquired by the vibration detection sensor 1, calculates the frequency spectrum of the vibration data in each sampling period, respectively, for each of the aforementioned plurality of sampling periods, stores the aforementioned calculated frequency spectrum in the storage section 6 in correspondence with the sampling period. The aforementioned peak value detection program is a program that detects a specific peak value that does not occur at the time of normality of the rolling bearing in a predetermined first frequency range that includes a theoretical frequency that brings about a peak on the frequency spectrum at the time of occurrence of an anomaly. The aforementioned peak value re-detection program is a program that, in the case where a specific peak value is detected by the aforementioned peak value detection program, detects, as a re-detected specific peak value, a specific peak value that does not occur at the time of normality of the aforementioned rolling bearing in a predetermined second frequency range that is narrower than the aforementioned first frequency range and that includes the frequency of the aforementioned specific peak value detected by the aforementioned peak value detection program, for one or more frequency spectra stored in the storage section 6 in advance of the sampling period in which the aforementioned specific peak value is detected. The aforementioned replacement timing prediction program is a program that predicts the replacement timing of the aforementioned rolling bearing on the basis of the sampling period at the time when a specific peak value is detected by the aforementioned peak value detection program and the amplitude of the aforementioned detected specific peak value, and the sampling period at the time when a re-detected specific peak value is detected by the aforementioned peak value re-detection program and the amplitude of the aforementioned detected re-detected specific peak value. Among the aforementioned various predetermined data, for example, there are included data required for execution of each of the aforementioned programs, such as the frequency spectrum, the theoretical frequency, the first frequency range, the second frequency range, the specific peak value, the re-detected specific peak value, and the predicted replacement timing, in correspondence with each sampling period. Such a storage section 6, for example, has a ROM (Read Only Memory) that is a non-volatile storage element, an EEPROM (Electrically Erasable Programmable Read Only Memory) that is a non-volatile storage element that can be rewritten, or the like. Furthermore, the storage section 6 includes a RAM (Random Access Memory) that is a work memory of the aforementioned control processing section 2, or the like, that stores data and the like generated in the execution of the aforementioned predetermined programs. In addition, the storage section 6 can have a hard disk device that can store a relatively large capacity, a solid state drive (SSD), or the like.

[0044] The control processing section 2 is a circuit for predicting the replacement timing of the rolling bearing BE by controlling each of the sections 1, 3 to 6 of the rolling bearing replacement timing prediction device 1000 in accordance with the functions of the sections. The control processing section 2 is configured by, for example, a CPU (Central Processing Unit) and a peripheral circuit thereof. By causing the control processing section 2 to execute the aforementioned control processing program, the control section 21, the spectrum processing section 22, the peak detection section 3, the peak re-detection section 24, and the timing prediction section 25 are functionally configured.

[0045] The control section 21 controls each of the sections 1, 3 to 6 of the rolling bearing replacement timing prediction device 1000 in accordance with the functions of the sections, and is responsible for the control of the entire rolling bearing replacement timing prediction device 1000.

[0046] The spectrum processing section 22 performs spectrum processing of acquiring vibration data in each of a plurality of sampling periods different from each other by the vibration detection sensors 1, calculating a frequency spectrum of the vibration data in each of the sampling periods, and storing the calculated frequency spectrum in association with the sampling period in the storage section 6. More specifically, in the spectrum processing, at a start point of a sampling period that becomes a predetermined acquisition interval, vibration data is acquired by the first to third vibration detection sensors 1-1 to 1-3 for a predetermined length of the sampling period, and is stored in the storage section 6 as vibration data for this sampling period. Next, the spectrum acquisition section 22 calculates a frequency spectrum of the vibration data by converting time-space vibration data acquired by the first vibration detection sensor 1-1 into frequency-space vibration data by, for example, a fast Fourier transform (FFT), and stores the calculated frequency spectrum in association with the sampling period and the first vibration detection sensor 1-1 (for example, an identifier (sensor ID) of the first vibration detection sensor 1-1) in the storage section 6. Next, similarly, the spectrum acquisition section 22 calculates a frequency spectrum of the vibration data by converting time-space vibration data acquired by the second vibration detection sensor 1-2 into frequency-space vibration data by the FFT, and stores the calculated frequency spectrum in association with the sampling period and the second vibration detection sensor 1-2 (for example, an identifier (sensor ID) of the second vibration detection sensor 1-2) in the storage section 6. Also similarly, the spectrum acquisition section 22 calculates a frequency spectrum of the vibration data by converting time-space vibration data acquired by the third vibration detection sensor 1-3 into frequency-space vibration data by the FFT, and stores the calculated frequency spectrum in association with the sampling period and the third vibration detection sensor 1-3 (for example, an identifier (sensor ID) of the third vibration detection sensor 1-3) in the storage section 6. Such processing is repeated at a start point of a sampling period that becomes a predetermined acquisition interval. The predetermined acquisition interval is appropriately set in advance, for example, in accordance with an object to be detected for an anomaly. The life of a rolling bearing can be predicted, for example, in accordance with a load and a rotational speed, and the like, so the predetermined acquisition interval is appropriately set in accordance with the life of the rolling bearing. For example, in a case where the monitoring object is a rolling bearing that supports a shaft that is relatively heavily used, for example, a shaft that is rotated at a high speed by frequent operation, the predetermined acquisition interval is set to a relatively short period of time, for example, one hour, one day, or the like, or in a case where the monitoring object is a rolling bearing that supports a shaft that is relatively gently used, for example, a shaft that is rotated relatively gently, the predetermined acquisition interval is set to a relatively long period of time, for example, one month, half a year, or the like. The start point of the sampling period is indicated by, for example, a continuous number from the time when the vibration data is first acquired, a time of the start point, or the like.

[0047] The peak detection section 23 performs a peak detection process of detecting a specific peak that does not occur at the time of normality of the rolling bearing from the frequency spectrum in a predetermined first frequency range that includes a theoretical frequency that brings a peak on the frequency spectrum at the time of abnormality, when storing the aforementioned frequency spectrum to the aforementioned storage section 6. The peak detection section 23 stores the sampling period, the amplitude, and the frequency of the detected specific peak to the storage section 6 in the case where the specific peak is detected.

[0048] The aforementioned theoretical frequency ft that brings a peak on the frequency spectrum at the time of abnormality is known, and differs depending on the site of occurrence of damage (bearing damage) of the rolling bearing, and is, for example, as in Table 1 below. The aforementioned site of bearing damage is, for example, the inner ring, the outer ring, the rolling element, and the retainer. Here, fti is the theoretical frequency in the case where bearing damage has occurred in the inner ring, fto is the theoretical frequency in the case where bearing damage has occurred in the outer ring, ftb is the theoretical frequency in the case where bearing damage has occurred in the rolling element, and ftm is the theoretical frequency in the case where bearing damage has occurred in the retainer. d is the diameter of the rolling element, D is the pitch diameter of the rolling element, Z is the number of rolling elements, and a is the contact angle.

[0049] [Table 1]

[0050]

[0051]

[0052] The frequency of the peak that is generated on the frequency spectrum due to some failure of the rolling bearing in the normal state (for example, a rolling bearing that is not used, etc.) that hinders smooth rolling does not actually coincide with the theoretical frequency because of the dimensional tolerance of the rolling bearing, deformation by load, etc. Therefore, in order to detect the unclear peak that occurs on the aforementioned frequency spectrum, a predetermined first frequency range that includes the aforementioned theoretical frequency ft is appropriately set in advance. For example, as shown in A, the first frequency range FW1 is set so that the theoretical frequency ft becomes the frequency at the center thereof (ft - Δfw1 ≤ FW1 ≤ ft + Δfw1, Δfw1 is, for example, about 1 to 5 [%] of ft). Figure 4

[0053] In the detection of the specific peak (peak that occurs due to the aforementioned failure) that does not occur at the time of normality of the aforementioned rolling bearing, for example, a peak is detected in the first frequency range FW1, and in the case where a peak (high frequency) also exists at a frequency that is an integral multiple (for example, a frequency that is twice, a frequency that is three times, etc.) of the frequency of the aforementioned detected peak, the aforementioned detected peak is set as the specific peak, and it is determined that the aforementioned specific peak is detected, and in the case where a peak does not exist at a frequency that is an integral multiple of the frequency of the aforementioned detected peak, it is determined that the aforementioned detected peak is not the specific peak, and it is determined that the aforementioned specific peak is not detected. For example, in the case where a peak is detected at a frequency that is an integral multiple of the frequency of the detected peak, the peak is detected at a frequency that is an integral multiple of the frequency of the detected peak, and the peak is detected at a frequency that is an integral multiple of the frequency of the detected peak.​Figure 4 In A, in the case where the peak PK1 and the peak PK2 are detected in the first frequency range FW1, and there is no peak at an integral multiple of the frequency fpl with respect to the frequency of the peak PK1, which is characteristic of the damage vibration of the bearing, and on the other hand, in the case where there is a peak at an integral multiple of the frequency fp2 with respect to the frequency of the peak PK2, it is determined that the peak PK1 is not the specific peak, and on the other hand, the peak PK2 is determined to be the specific peak, and it is determined that the specific peak is detected. In the present embodiment, since the plurality of first to third vibration detection sensors 1-1 to 1-3 are used, in the case where the specific peak is determined in the common frequency in two or more frequency spectra among the first to third vibration detection sensors 1-1 to 1-3, it is finally determined that the specific peak is detected.

[0054] The peak re-detection unit 24, in the case where the specific peak is detected by the aforementioned peak detection unit 23, performs, for one or more frequency spectra stored in the aforementioned storage unit 6 earlier than the sampling period in which the aforementioned specific peak is detected, a peak re-detection process of detecting, as a re-detection specific peak, a specific peak that does not occur at normal times within a predetermined second frequency range FW2 that is narrower than the aforementioned first frequency range FW1 and that includes the frequency of the aforementioned specific peak detected by the aforementioned peak detection unit 23. The peak re-detection unit 24 stores the sampling period, the amplitude, and the frequency of the detected re-detection specific peak to the storage unit 6.

[0055] Since it is for detecting the development mode (tendency, trend over time) of the aforementioned failure, the peak re-detection unit 24 can detect the aforementioned specific peak as a re-detection specific peak for one frequency spectrum stored in the storage unit 6 earlier than the sampling period of the frequency spectrum in which the aforementioned specific peak is detected by the aforementioned peak detection unit 23, or can detect the aforementioned specific peak as a re-detection specific peak for some of the frequency spectra (a plurality of frequency spectra less than all) stored in the storage unit 6 earlier than the aforementioned sampling period, or can detect the aforementioned specific peak as a re-detection specific peak for all of the frequency spectra stored in the storage unit 6 earlier than the aforementioned sampling period. In the case where the aforementioned specific peak is detected as a re-detection specific peak for a plurality of frequency spectra, the peak re-detection unit 24 can, for example, detect the aforementioned re-detection specific peak for the aforementioned plurality of frequency spectra in the past direction continuously in order, or, for example, the peak re-detection unit 24 can detect the aforementioned re-detection specific peak for the aforementioned plurality of frequency spectra in the past direction discontinuously. In the present embodiment, the peak re-detection unit 24 continuously takes out in order in the past direction a predetermined number of frequency spectra stored in the storage unit 6 in correspondence with a predetermined number of sampling periods earlier than the aforementioned sampling period, respectively, and detects the aforementioned re-detection specific peak continuously in order in the past direction for the aforementioned predetermined number of frequency spectra taken out.

[0056] Generally, the peak value can be determined after the amplitude of the peak value becomes large with development of some obstacle that hinders smooth rolling of the rolling bearing in a normal state. Therefore, the frequency of the specific peak value is not clear until the specific peak value is detected by the peak value detection section 23, so it is generally necessary to set the first frequency range FW1 relatively wide. On the other hand, it is presumed that the specific peak value before the specific peak value is generated in the vicinity of the frequency of the specific peak value. Therefore, a predetermined second frequency range narrower than the first frequency range FW1 is set to include the frequency of the specific peak value appropriately in advance. For example, as shown in FIG. 2B, the second frequency range FW2 is set so that the frequency fp2 of the specific peak value PK2 becomes the center frequency (fp2 - Δfw2 ≤ FW2 ≤ fp2 + Δfw2, Δfw2 is about 0.1 to 0.5 [%] of the theoretical frequency ft, for example). Figure 4 B.

[0057] In the detection of the re-detection specific peak value, a peak value that is the maximum amplitude in the second frequency range FW2 and in which a peak value also exists in an integral multiple portion of the frequency is set as the re-detection specific peak value. For example, in FIG. 2B, a peak value PK2' that is the maximum amplitude in the second frequency range FW2 and in which a peak value also exists in an integral multiple portion of the frequency is set as the re-detection specific peak value. Figure 4 B.

[0058] The second frequency range FW2 can be the same in the case where the re-detection specific peak value is detected from a plurality of frequency spectra, but in the present embodiment, the peak value re-detection section 24, in the case where the re-detection specific peak value is detected in the past direction for the plurality of frequency spectra, respectively, regards the peak value detection process and each specific peak value detected in the peak value detection process as each first peak value re-detection process and a first re-detection specific peak value for the plurality of frequency spectra, and detects the re-detection specific peak value in the second frequency range FW2 including the re-detection specific peak value detected by the first peak value re-detection process for the frequency spectrum. Even if a deviation occurs in the frequency of the re-detection specific peak value due to wear or the like, for example, the re-detection specific peak value can be detected.

[0059] Furthermore, after the initial detection of a specific peak, the peak detection unit 23 can detect the specific peak again during a sampling period later than the sampling period in which the specific peak was detected, within the second frequency range FW2 that includes the initially detected specific peak PK2, for the frequency spectrum stored in the storage unit 6 during that sampling period. However, in this embodiment, the peak detection unit 23 detects the specific peak again during a sampling period later than the sampling period in which the specific peak was detected, within the second frequency range FW2 that includes the specific peak detected by the previous peak detection process during that sampling period, for the frequency spectrum stored in the storage unit 6 during that sampling period. Even if the frequency of the specific peak deviates due to factors such as wear, the specific peak can still be detected.

[0060] The replacement period prediction unit 25 performs replacement period prediction processing: based on the sampling period when a specific peak is detected by the peak detection unit 23 and the amplitude of the aforementioned specific peak, and the sampling period when a specific peak is re-detected by the peak re-detection unit 24 and the amplitude of the aforementioned re-detected specific peak, the replacement period of the rolling bearing is predicted.

[0061] When the peak re-detection unit 24 detects one re-detected specific peak using one frequency spectrum, the period prediction unit 25 predicts the replacement period based on the specific peak detected by the peak detection unit 23 and the one re-detected specific peak detected by the peak re-detection unit 24. When the peak re-detection unit 24 detects multiple re-detected specific peaks using multiple frequency spectra, the period prediction unit 25 predicts the replacement period based on the specific peak detected by the peak detection unit 23 and the multiple re-detected specific peaks detected by the peak re-detection unit 24. During the sampling period after the peak detection unit 23 initially detects the specific peak, the specific peak detected during the sampling period, and one or more re-detected specific peaks, the period prediction unit 25 predicts the replacement period based on the initially detected specific peak, the specific peak detected during the sampling period, and one or more re-detected specific peaks. The period prediction unit 25 outputs the predicted replacement period to the output unit 4.

[0062] For example, in Figure 5 In the example shown, at time point TD0 during the sampling period (e.g., the representative time point of the sampling period is the central time point TD0), if a specific peak indicated by ● is detected by the peak detection unit 23, the peak re-detection unit 24 detects the peak one time point TD0 that is earlier than time point TD0. －1 The frequency spectrum is used to detect and re-detect a specific peak within the second frequency range FW2, which contains a specific peak ●, for example, detecting the time point TD represented by ×. －1 The specific peak value is re-detected. Additionally, four frequency spectra are assumed to be retrieved and detected by tracing back in the past direction. Next, the peak re-detection unit 24 compares the peak value with the time point TD.－1 The first one in time TD －2 The frequency spectrum, including time point TD －1 A specific peak value was detected within the second frequency range FW2 of the re-detection specific peak value ×, for example, the time point TD represented by × was detected. －2 The specific peak value is then re-detected. Next, the peak re-detection unit 24 compares the peak value with the time point TD. －2 The first one in time TD －3 The frequency spectrum, including time point TD －2 The specific peak value is detected again within the second frequency range FW2. Figure 5 In the example shown, from time point TD －3 No peak was detected in the frequency spectrum. Next, the peak re-detection unit 24 compared the peak at time point TD. －3 The first one in time TD －4 The frequency spectrum, including time point TD －3 The specific peak value is detected again within the second frequency range FW2. Figure 5 In the example shown, from time point TD －4 No peak was detected in the frequency spectrum. Then, the period prediction unit 25, based on the specific peak value of time point TD0, time point TD －1 Re-detection of specific peak values ​​× time point TD －2 Re-detection of specific peak values ​​× predicts replacement periods.

[0063] For example, the period prediction unit 25 calculates, in a coordinate space with time and amplitude as coordinate axes, a fitting curve of the center time point and amplitude of the sampling period for which the specific peak value is best fitted and the center time point and amplitude of the sampling period for which the specific peak value is re-detected, and calculates the time point at which the aforementioned fitting curve intersects with a preset determination amplitude value used to determine the replacement period, and uses this as the aforementioned replacement period. Figure 5 In the example shown, the period prediction unit 25 calculates the time point TD0 and its amplitude corresponding to the aforementioned specific peak value ●, and the time point TD －1 The re-detection of a specific peak × the aforementioned time point TD －1 Its amplitude and time point TD －2 The specific peak value × is re-detected at the aforementioned time point TD-2 and the best fitting curve α of its amplitude. The time point ESα at which the obtained fitting curve α intersects with the preset judgment amplitude value Th used to determine the replacement period is obtained, and this is taken as the aforementioned replacement period ESα.

[0064] Time TD after time point TD0 +1 The peak detection unit 23 measures TD at that time point. +1The frequency spectrum of time point TD0 is detected within the second frequency range FW2, which contains the specific peak value of time point TD0. +1 Specific peak values, such as the time point TD represented by ×, are detected. +1 The specific peak value. Then, the period prediction unit 25, based on the specific peak value of time point TD0, time point TD... +1 Specific peak value × and TD at each time point －1 TD －2 Each device re-detects a specific peak value × to predict the replacement period.

[0065] Alternatively, for example, if a specific peak indicated by ● is detected by the peak detection unit 23 at time point TD0, the peak re-detection unit 24 will detect a peak one time point TD preceding time point TD0. － The frequency spectrum of 1 is used to detect and re-detect a specific peak within a second frequency range FW2 that contains a specific peak ●, for example, detecting the time point TD represented by ○. －1 The specific peak value is then re-detected. Next, the peak re-detection unit 24 compares the peak value with the time point TD. －1 The first one in time TD －2 The frequency spectrum, including time point TD －1 The re-detection of a specific peak value ○ within the second frequency range FW2, for example, detecting the time point TD represented by ○. －2 The specific peak value is then re-detected. Next, the peak re-detection unit 24 compares the peak value with the time point TD. －2 The first one in time TD －3 The frequency spectrum, including time point TD －2 The re-detection of a specific peak value ○ within the second frequency range FW2, for example, detecting the time point TD represented by ○. －3 The specific peak value is then re-detected. Next, the peak re-detection unit 24 compares the peak value with the time point TD. －3 The first one in time TD －4 The frequency spectrum, including time point TD －3 The re-detection of a specific peak value ○ within the second frequency range FW2, for example, detecting the time point TD represented by ○. －4 The specific peak value is then re-detected. Then, the period prediction unit 25, based on the specific peak value at time point TD0, and time point TD... －1 Re-detection of specific peak values ​​× and TD at each time point －1 TD －2 TD －3 TD －4 Each unit re-detects a specific peak value × and predicts the replacement period. For example, similarly to the above, the period prediction unit 25 calculates the time point TD0 and its amplitude corresponding to the aforementioned specific peak value ●, as well as the TD at each time point. －1 TD－2 , TD －3 , TD －4 each of the re-detected specific peaks －1 , TD －2 , TD －3 , TD －4 and their respective amplitudes, the fitting curve β that best fits the amplitudes is found, and the point ESβ at which the found fitting curve β crosses the determination amplitude value Th is taken as the aforementioned replacement period ESβ.

[0066] the point TD +1 , the frequency spectrum of the point TD +1 , the specific peak of the point TD +1 , for example, the specific peak of the point TD +1 is detected. Then, the period prediction section 25 predicts the replacement period based on the specific peak ● of the point TD +1 , the specific peak ○ of the point TD －1 , TD －2 , TD －3 , TD －4 each of the re-detected specific peaks ○.

[0067] the point TD +2 , the frequency spectrum of the point TD +2 , the specific peak of the point TD +1 , for example, the specific peak of the point TD +2 is detected. Then, the period prediction section 25 predicts the replacement period based on the specific peak ● of the point TD +2 , the specific peak ○ of each of the points TD +1 , TD +2 , and the specific peak ○ of each of the points TD －1 , TD －2 , TD －3 , TD －4 each of the re-detected specific peaks ○.

[0068] Also, at the point TD +3 , the frequency spectrum of the point TD +3 , the specific peak of the point TD +2 , for example, the specific peak of the point TD +3 is detected. Then, the period prediction section 25 predicts the replacement period based on the specific peak ● of the point TD +3Specific peak values. Then, the period prediction unit 25 predicts the replacement period based on the specific peak value ● at the time point TD0 and each specific peak value ○ at each time point TD +1 、TD +2 、TD +3 and each re - measurement peak value ○ at each time point TD －1 、TD －2 、TD －3 、TD －4 respectively.

[0069] In Figure 5 the example shown, when represented by the specific peak value ● and the re - measurement specific peak value ×, the progress of the aforementioned failure develops relatively rapidly, and the remaining life from the time point TD0 is relatively short. The replacement period ESα reflecting the development mode (change trend over time) of this failure is predicted. On the other hand, when represented by the specific peak value ● and the re - measurement specific peak value ○, the progress of the aforementioned failure develops relatively gently, and the remaining life from the time point TD0 is relatively long. The replacement period ESβ reflecting the development mode of this failure is predicted (ESα < ESβ). Thus, even if the time point TD0 when the specific peak value ● is detected is the same, if the development mode of the aforementioned failure is different, the remaining life is different, and thus the replacement period is also different. The rolling bearing replacement period prediction device 1000 of the embodiment can predict the replacement period considering such a development mode of the failure.

[0070] These control processing unit 2, input unit 3, display unit 4, IF unit 5, and storage unit 6 can be constituted by, for example, a desktop computer, a laptop computer, etc.

[0071] Next, the operation of this embodiment will be described. Figure 6 is a flowchart showing the operation of the aforementioned rolling bearing replacement period prediction device before detecting the specific peak value. Figure 7 is a flowchart showing the operation of the aforementioned rolling bearing replacement period prediction device after detecting the aforementioned specific peak value.

[0072] When the power of the rolling bearing replacement period prediction device 1000 with such a structure is turned on, it executes the initialization of each necessary part and starts its operation. In the control processing unit 2, through the execution of this control processing program, the control unit 21, spectrum processing unit 22, peak detection unit 23, peak re - detection unit 24, and period prediction unit 25 are functionally constituted.

[0073] If the aforementioned operation starts and it becomes the start time point of the sampling period, then at Figure 6In the present embodiment, first, the rolling bearing replacement timing prediction device 1000 acquires, from the storage section 6 by the spectrum processing section 22 of the control processing section 2, vibration data of the length TW of the sampling period acquired by the first to third vibration detection sensors 1-1 to 1-3, calculates the frequency spectrum of the vibration data in the present sampling period (Sll), and stores the calculated frequency spectrum in correspondence with the present sampling period to the storage section 6 (S12).

[0074] Next, the rolling bearing replacement timing prediction device 1000 detects, from the frequency spectrum in the present sampling period by the peak detection section 23 of the control processing section 2, a specific peak in a predetermined first frequency range including the theoretical frequency (S13). In a case where the result of the detection does not detect the specific peak (NO), the rolling bearing replacement timing prediction device 1000 ends the present processing in the present sampling period. On the other hand, in a case where the result of the aforementioned determination detects the specific peak (YES), the rolling bearing replacement timing prediction device 1000 executes the processing S14 next.

[0075] In the processing S14, the rolling bearing replacement timing prediction device 1000 stores the aforementioned detected specific peak (the sampling period, the amplitude, and the frequency thereof) to the storage section 6 by the peak detection section 23 of the control processing section 2.

[0076] Next, the rolling bearing replacement timing prediction device 1000 detects the specific peak as a re-detected specific peak by the peak re-detection section 24 of the control processing section 2, and stores the detected re-detected specific peak (the sampling period, the amplitude, and the frequency thereof) to the storage section 6 (S15).

[0077] Next, the rolling bearing replacement timing prediction device 1000 predicts the replacement timing of the rolling bearing BE based on the specific peak and the re-detected specific peak by the timing prediction section 25 of the control processing section 2 (S16).

[0078] Then, the rolling bearing replacement timing prediction device 1000 outputs the predicted replacement timing to the output section 4 by the timing prediction section 25 (S17), and ends the present processing in the present sampling period. In addition, the aforementioned replacement timing can be output to an external device via the IF section 5 as necessary.

[0079] On the other hand, if it becomes a sampling period after the aforementioned specific peak is detected, in the processing S21, the rolling bearing replacement timing prediction device 1000 acquires, from the storage section 6 by the spectrum processing section 22 of the control processing section 2, vibration data of the length TW of the sampling period acquired by the first to third vibration detection sensors 1-1 to 1-3, and calculates the frequency spectrum of the vibration data in the present sampling period (S21). Figure 7 In the present embodiment, first, the rolling bearing replacement timing prediction device 1000 acquires, from the storage section 6 by the spectrum processing section 22 of the control processing section 2, vibration data of the length TW of the sampling period acquired by the first to third vibration detection sensors 1-1 to 1-3, calculates the frequency spectrum of the vibration data in the present sampling period (Sll), and stores the calculated frequency spectrum in correspondence with the present sampling period to the storage section 6 (S12).

[0080] Next, the rolling bearing replacement timing prediction device 1000 stores the detected specific peak value (sampling period and amplitude) to the storage section 6 (S23).

[0081] Next, the rolling bearing replacement timing prediction device 1000 predicts the replacement timing of the rolling bearing BE based on the detected specific peak values and the re-detected specific peak value by the timing prediction section 25 of the control processing section 2 (S24).

[0082] Then, the rolling bearing replacement timing prediction device 1000 outputs the predicted replacement timing to the output section 4 by the timing prediction section 25 as in the above-described process S17 (S25), and ends the present process in the present sampling period.

[0083] As described above, the rolling bearing replacement timing prediction device 1000 and the rolling bearing replacement timing prediction method installed therein of the embodiment can more reliably detect the specific peak value because the specific peak value is detected in the first frequency range including the theoretical frequency, and can more correctly detect the specific peak value caused by the occurrence of the above-described failure by excluding the confusing peak value included in the first frequency range but not caused by the occurrence of the above-described failure from the detection object, and re-detecting the specific peak value by narrowing to the second frequency range narrower than the above-described first frequency range.

[0084] The time until the rolling bearing needs to be replaced, i.e., the remaining life ((end point of the sampling period in the frequency spectrum (vibration data) in which the specific peak value is detected) + (the remaining life at this time) = (replacement timing)) depends on the development mode (tendency, trend over time) of the above-described failure. The above-described rolling bearing replacement timing prediction device 1000 and the rolling bearing replacement timing prediction method can more accurately predict the replacement timing because the replacement timing is predicted based on the sampling period in which the specific peak value is detected and the amplitude of the above-described detected specific peak value, and the sampling period in which the re-detected specific peak value is detected and the amplitude of the above-described detected re-detected specific peak value, in addition to detecting the specific peak value and the re-detected specific peak value before it, and predict the replacement timing taking into account the development mode of the above-described failure. In addition, the above-described rolling bearing replacement timing prediction device 1000 and the rolling bearing replacement timing prediction method can more accurately predict the replacement timing in the case where the replacement timing is predicted using a plurality of re-detected specific peak values than in the case where the replacement timing is predicted using one re-detected specific peak value.

[0085] The rolling bearing replacement timing prediction device 1000 and the rolling bearing replacement timing prediction method described above can more reliably detect the re-detection specific peak value in the case where the frequency of the re-detection specific peak value changes over time, because the re-detection specific peak value is detected in the second frequency range including the re-detection specific peak value detected by the first peak value re-detection processing.

[0086] The rolling bearing replacement timing prediction device 1000 and the rolling bearing replacement timing prediction method described above can more reliably detect the specific peak value in the case where the frequency of the specific peak value changes over time, because the specific peak value is detected in the second frequency range including the specific peak value detected by the first peak value detection processing. The rolling bearing replacement timing prediction device 1000 and the rolling bearing replacement timing prediction method described above can predict the replacement timing with higher accuracy, because the replacement timing is predicted using the specific peak value in the sampling period later than the sampling period in which the specific peak value is detected.

[0087] In addition, in the embodiment described above, the data recorder 7 can be provided between the vibration detection sensor 1 and the control processing section 2, and the spectrum processing section 22 can be provided in the data recorder 7.

[0088] Figure 8 is a block diagram showing the structure of a rolling bearing replacement timing prediction device of a modification example of the embodiment. The rolling bearing replacement timing prediction device 1000a of the modification example has, for example, as shown in Figure 8 the vibration detection sensor 1, the data recorder 7, the control processing section 2, the input section 3, the output section 4, the IF section 5, and the storage section 6.

[0089] The vibration detection sensor 1, the control processing section 2, the input section 3, the output section 4, the IF section 5, and the storage section 6 in the rolling bearing replacement timing prediction device 1000a are the same as those in the rolling bearing replacement timing prediction device 1000 except that a part of the functions of the control processing section 2 and a part of the functions of the storage section 6 are moved to the data recorder 7, so the description thereof is omitted. The control processing section 2 has, in terms of functions, the control section 21, the peak value detection section 23, the peak value re-detection section 24, and the timing prediction section 25, and the functions of the spectrum processing are moved to the data recorder 7. The functions of storing the vibration data and the frequency spectrum thereof are moved to the data recorder 7.

[0090] The data logger 7 is connected to the vibration detection sensor 1 and the control processing section 2, for example, is configured with a computer, and has a data logger control processing section 71 and a data logger storage section 72. The data logger control processing section 71 is configured with a CPU and its peripheral circuit, and has a spectrum processing section 22 that functions in the same way as described above, except that it stores in the data logger storage section 72 instead of the storage section 6. The data logger storage section 72 has a ROM, an EEPROM, a RAM, a hard disk device, and the like, and stores the aforementioned vibration data and frequency spectrum. The data logger 7 outputs the frequency spectrum of the aforementioned vibration data to the control processing section 2 as required (requested) by the control processing section 2.

[0091] In order to embody the present application, the above-described Figure 1 The present application has been described adequately and sufficiently by the embodiments above with reference to the drawings, but it should be recognized by those skilled in the art that the above-described embodiments can be easily changed and / or modified. Thus, as long as the changed or modified embodiments by those skilled in the art are not beyond the scope of the claims recited in the claims, the changed or modified embodiments are interpreted as being included in the scope of the claims.

[0092] Explanation of Reference Numerals

[0093] BE (BE-1 to BE-3) Rolling bearing (first to third rolling bearings)

[0094] FW1 First frequency range FW2 Second frequency range

[0095] 1000 Rolling bearing replacement timing prediction device

[0096] 1 (1-1 to 1-3) Vibration detection sensor (first to third vibration detection sensors) 2 Control processing section

[0097] 6 Storage section

[0098] 7 Data logger

[0099] 21 Control section

[0100] 22 Spectrum processing section

[0101] 23 Peak detection section

[0102] 24 Peak re-detection section

[0103] 25 Timing prediction section

Claims

1. A method for predicting the replacement period of rolling bearings, characterized in that, have: The spectrum processing step acquires vibration data representing the vibration generated in the rolling bearing at predetermined acquisition intervals and sampling periods. For the aforementioned vibration data, the frequency spectrum of the vibration data in the sampling period is calculated, and the calculated frequency spectrum is established in correspondence with the sampling period and stored in the storage unit. The peak detection process involves detecting specific peaks that would not occur when the aforementioned rolling bearing is functioning normally, within a predetermined first frequency range that includes the theoretical frequencies that would cause peaks in the frequency spectrum when an anomaly occurs. In the peak re-detection process, if a specific peak is detected by the aforementioned peak detection process, for one or more frequency spectra stored in the aforementioned storage unit earlier than the sampling period when the aforementioned specific peak was detected, within a predetermined second frequency range that includes the frequency of the aforementioned specific peak detected by the aforementioned peak detection process and is narrower than the aforementioned first frequency range, a specific peak that would not appear when the aforementioned rolling bearing is functioning normally is detected as a re-detected specific peak; and The replacement period prediction process is a process that predicts the replacement period of the rolling bearing based on the sampling period and amplitude of the specific peak detected by the peak detection process, and the sampling period and amplitude of the specific peak detected by the peak redetection process.

2. The method for predicting the replacement period of rolling bearings as described in claim 1, characterized in that, In the case of detecting the aforementioned re-detected specific peak in the past direction for the aforementioned plurality of frequency spectra, for each of the aforementioned plurality of frequency spectra, the aforementioned peak detection process and the specific peak detected by the aforementioned peak detection process are each regarded as the respective initial peak re-detection process and the initial re-detected specific peak. For the frequency spectrum, the aforementioned re-detected specific peak is detected in the aforementioned second frequency range including the re-detected specific peak detected by the previous aforementioned peak re-detection process.

3. The method for predicting the replacement period of rolling bearings as described in claim 1, characterized in that, The aforementioned peak detection process also detects the specific peak in the aforementioned second frequency range that includes the specific peak detected by the previous peak detection process during the aforementioned second frequency range, for the frequency spectrum stored in the aforementioned storage unit during the sampling period that is later than the sampling period in which the aforementioned specific peak was detected.

4. The method for predicting the replacement period of rolling bearings as described in claim 1, characterized in that, The length of the aforementioned sampling period is set based on the aforementioned second frequency range; The narrower the aforementioned second frequency range, the longer the aforementioned sampling period is set.

5. A rolling bearing replacement period prediction device, characterized in that, have: Vibration detection sensors acquire vibration data representing the vibrations generated by rolling bearings; Storage department; and The control and processing unit performs spectrum processing, peak detection processing, peak re-detection processing, and replacement period prediction processing. The aforementioned spectrum processing is a process of obtaining the aforementioned vibration data during a predetermined sampling period at predetermined acquisition intervals, calculating the frequency spectrum of the vibration data during the aforementioned sampling period, establishing a correspondence between the calculated frequency spectrum and the sampling period, and storing it in a storage unit. The aforementioned peak detection process is a process of detecting specific peaks that would not occur when the aforementioned rolling bearing is normal, within a predetermined first frequency range that includes the theoretical frequencies that would cause peaks in the frequency spectrum when an anomaly occurs. The aforementioned peak re-detection process involves detecting a specific peak that would not occur when the rolling bearing is functioning normally, within a predetermined second frequency range that includes the frequency of the specific peak detected by the aforementioned peak detection process and is narrower than the aforementioned first frequency range, for one or more frequency spectra stored in the aforementioned storage unit earlier than the sampling period when the aforementioned specific peak was detected. This is a process for re-detecting the specific peak. The aforementioned replacement period prediction process is based on the sampling period and amplitude of the aforementioned specific peak detected by the aforementioned peak detection process, and the sampling period and amplitude of the aforementioned specific peak detected by the aforementioned peak re-detection process.

Images