JP4935157B2 - Abnormality diagnosis apparatus and abnormality diagnosis method - Google Patents

Description

  The present invention relates to an abnormality diagnosis device and an abnormality diagnosis method for a rolling device (such as an axle bearing, a gear or a wheel) used in a railway vehicle, and more particularly, whether or not there is an abnormality in the rolling device (an axle bearing, a gear or a wheel). The present invention relates to an abnormality diagnosis device and an abnormality diagnosis method for identifying a warning sign or an abnormal part thereof.

  Conventionally, after rotating parts of a railway vehicle are used for a certain period, the axle bearings and other rotating parts are regularly inspected for abnormalities such as damage and wear. This periodic inspection is performed by disassembling the mechanical device in which the rotating part is incorporated, and the operator can detect damage and wear generated in the rotating part by visual inspection. And the main defects found in the inspection are in the case of bearings, indentations caused by biting of foreign matter, peeling due to rolling fatigue, other wear, etc., in the case of gears, tooth defects and wear, etc. In the case of a wheel, there is a rolling surface defect or wear, and in any case, if irregularities or wear that are not found in a new article are found, it is replaced with a new one.

  However, in the method of disassembling the entire mechanical equipment and visually inspecting by the operator, disassembly work to remove the rotating body and sliding member from the device, and reassembly of the inspected rotating body and sliding member into the device There has been a problem that a great deal of labor is required for the work, and the maintenance cost of the apparatus is greatly increased.

  Further, when reassembling, the inspection itself may cause a defect in the rotating body or the sliding member, such as attaching a dent to the rotating body or the sliding member that was not present before the inspection. Further, since a large number of bearings are visually inspected within a limited time, there is a problem that a possibility of overlooking a defect remains. Further, the determination of the degree of this defect also varies from person to person, and parts are exchanged even if there is virtually no defect, resulting in unnecessary costs.

  Accordingly, various methods have been proposed for performing abnormality diagnosis of rotating parts in an actual operating state without disassembling a mechanical device in which the rotating parts are incorporated (see, for example, Patent Documents 1 to 3). Most commonly, as described in Patent Document 1, an accelerometer is installed in the bearing section, the vibration acceleration of the bearing section is measured, and FFT (Fast Fourier Transform) processing is performed on this signal. There is known a method of performing diagnosis by extracting a signal of vibration generation frequency components.

In addition, various methods have been proposed for detecting flat portions called flats caused by friction and wear of the rails due to wheel locks and sliding due to malfunctions of brakes on the rolling surfaces of the wheels of railway vehicles (for example, patents). References 4-6). In particular, Patent Documents 4 and 7 propose a device that detects a defect state of a railroad vehicle wheel and a track through which the train passes by a vibration sensor, a rotation measuring device, or the like.
JP 2002-22617 A JP 2003-202276 A Japanese Patent Laid-Open No. 2004-257836 Japanese National Patent Publication No. 9-500452 JP-A-4-148839 Special table 2003-535755 gazette US Pat. No. 5,433,111

  However, in the defect state detection devices described in Patent Documents 4 and 7, since the abnormality diagnosis accuracy by frequency analysis is poor, whether the abnormal vibration is caused by a flat wheel, an axle bearing, or a line or other abnormality There is a problem that it cannot be identified.

  The present invention has been made in view of the above-described circumstances, and an object of the present invention is to detect an abnormal vibration of the rolling device and to identify with high accuracy which component the abnormal vibration is due to. An object of the present invention is to provide a diagnostic device and an abnormality diagnostic method. Furthermore, abnormal vibrations of the axle bearing and the wheel are detected from the output signal of the vibration sensor that detects the vibration of the axle bearing or the wheel, and whether the abnormal vibration is caused by the wheel flat or the axle bearing is specified with high accuracy. An object of the present invention is to provide an abnormality diagnosis device and an abnormality diagnosis method for railway vehicles.

The object of the present invention is achieved by the following constitution.
( 1 ) a vibration sensor for detecting the vibration of the rolling device;
The frequency distribution of the signal from the vibration sensor is obtained, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution corresponds to the frequency from the first order to the nth order indicating the abnormality. A degree of coincidence between the two frequencies compared to each order to be scored, and a diagnosis processing unit that performs abnormality diagnosis of the rolling device based on the result,
An externally rewritable setting value storage unit that stores the number N of coincidence calculations and the reference total score P for the calculation results over N times the coincidence,
The diagnosis processing unit
By sequentially reading the values N and P from the set value storage unit, obtaining the coincidence N times, obtaining the coincidence total score, and comparing the N coincidence total score with the reference total score P An abnormality diagnosis apparatus characterized by performing abnormality diagnosis.
( 2 ) The setting storage unit stores a value K that determines a threshold value of the peak value ,
It said value K is a rewritable from the outside,
The diagnosis processing unit
The abnormality diagnosis apparatus according to (1), wherein the value K is read from the set value storage unit, and a frequency indicating a peak value equal to or higher than a threshold value determined by the value K is obtained.
( 3 ) In the abnormality diagnosis device according to ( 1 ) or ( 2 ) , the values N, P, and / or K stored in the set value storage unit differ depending on the type or part of abnormality of the rolling device. An abnormality diagnosis device characterized by that.
( 4 ) The diagnosis processing unit
When the average value of the amplitude of the signal from the vibration sensor in a predetermined period falls below a predetermined reference value, the frequency indicating the peak value included in the frequency distribution matches the frequency indicating the abnormality. The abnormality diagnosis device according to any one of ( 1 ) to ( 3 ) , wherein the comparison of whether or not is invalidated.
( 5 ) The diagnosis processing unit
The abnormality diagnosis apparatus according to any one of (1) to (4) , wherein an amplification gain of a signal from the vibration sensor is adjusted according to a rotation speed or a movement speed of the rolling device.
( 6 ) The diagnosis processing unit
A comparison is made as to whether or not the difference between the frequency indicating the peak value to be compared and the frequency indicating the abnormality is within an allowable error range, and the allowable error is preset as a function of the frequency. The abnormality diagnosis device according to any one of (1) to (5), wherein the abnormality diagnosis device is defined by an allowable error function.
(7) (1) In the abnormality diagnosis device according to any one of - (6), the abnormality diagnosis, wherein the rolling device axle bearing for a railway vehicle, either of the gear or wheel apparatus.
( 8 ) The vibration of the rolling device is detected by the vibration sensor, and the frequency indicating the peak value from the primary (fundamental wave) to the nth order included in the frequency distribution of the signal from the vibration sensor and the primary indicating the abnormality In the abnormality diagnosis method for comparing the frequencies up to the nth order for each corresponding order and scoring as a degree of coincidence between both frequencies, and performing abnormality diagnosis of the rolling device based on the result,
The number N of coincidence calculations and the reference total score P for the calculation results over N times of the coincidence are stored in externally rewritable setting value storage means,
The values N and P are sequentially read from the set value storage means, and the degree of coincidence is obtained N times in succession to obtain the degree of coincidence total score, and the total degree of coincidence score and the reference total score P at the N times are obtained. An abnormality diagnosis method characterized in that an abnormality diagnosis is performed by comparison.
( 9 ) A value K that determines the threshold value of the peak value is stored in a setting value storage means that can be rewritten from the outside,
The abnormality diagnosis method according to (8), wherein the value K is sequentially read from the set value storage unit, and a frequency indicating a peak value equal to or higher than a threshold value determined by the value K is obtained.
( 10 ) In the abnormality diagnosis method according to ( 8 ) or ( 9 ) , the values N, P and / or K stored in the set value storage means differ depending on the type or part of the abnormality of the rolling device. An abnormality diagnosis method characterized by the above.
( 11 ) When the average value of the amplitude in a predetermined period of the vibration generated by the rolling device is lower than a predetermined reference value, the frequency indicating the peak value included in the frequency distribution and the frequency distribution indicating the abnormality, The abnormality diagnosis method according to any one of (8) to (10), wherein the comparison of whether or not they coincide is invalidated.
( 12 ) The abnormality according to any one of (8) to (11) , wherein an amplification gain of a signal from the vibration sensor is adjusted according to a rotation speed or a movement speed of the rolling device. Diagnostic method.
( 13 ) A comparison is made as to whether or not the difference between the frequency indicating the peak value to be compared and the frequency indicating the abnormality is within an allowable error range, and the allowable error is preset as a function of the frequency. The abnormality diagnosis apparatus method according to any one of (8) to (12), wherein the abnormality diagnosis apparatus method is defined by an allowable error function.
(14) (8) in the abnormality diagnostic method according to any one of - (13), the abnormality diagnosis, wherein the rolling device axle bearing for a railway vehicle, either of the gear or wheel Method.

According to the abnormality diagnosis apparatus of the present invention, the following effects (I) to (V) can be obtained.
(I) The vibration of the rolling device is detected by the vibration sensor, the frequency distribution of the signal from the vibration sensor is obtained, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution; The peak values to be compared when comparing the first to n-th order frequencies indicating abnormality for each corresponding order and performing abnormality diagnosis of the rolling device based on the result. Is compared with whether or not the difference between the frequency indicating the abnormality and the frequency indicating the abnormality is within the allowable error range, and the allowable error is defined by a preset allowable error function as a function of the frequency. Therefore, the abnormality diagnosis of the rolling device can be performed with extremely high accuracy.
(II) The vibration of the rolling device is detected by the vibration sensor, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution of the signal from the vibration sensor and the first order indicating the abnormality. The frequency up to the nth order is compared for each corresponding order and scored as the degree of coincidence between both frequencies, and when the abnormality diagnosis of the rolling device is performed based on the result, the number of coincidence calculations N and the The reference total score P for the calculation result of N times of coincidence is stored in an externally rewritable set value storage means, and the values N and P are sequentially read from the set value storage means, and the coincidence degree Is performed N times in succession to obtain the total score of coincidence, and the abnormality diagnosis is performed by comparing the total score of coincidence at the N times with the reference total score P. Can be implemented with high accuracy In addition, since the value N and the value P can be set and changed from the outside, the diagnostic accuracy can be flexibly changed according to the situation.
(III) The vibration of the rolling device is detected by the vibration sensor, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution of the signal from the vibration sensor and the first order indicating the abnormality. The frequency up to the nth order is compared for each corresponding order, and when performing abnormality diagnosis of the rolling device based on the result, a value K that determines the threshold value of the peak value is externally applied. Since the value K is stored in a rewritable set value storage means, the value K is read from the set value storage unit, and a frequency indicating a peak value equal to or higher than a threshold value determined by the value K is obtained, the abnormality diagnosis of the rolling device is extremely high. Can be implemented with accuracy. In addition, since the value K can be changed from the outside, the threshold value can be flexibly changed according to the situation.
(IV) The vibration of the rolling device is detected by the vibration sensor, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution of the signal from the vibration sensor and the first order indicating the abnormality. The frequency up to the nth order is compared for each corresponding order, and when performing abnormality diagnosis of the rolling device based on the result, the amplitude of the signal from the vibration sensor in a predetermined period When the average value falls below a predetermined reference value, the comparison between the frequency indicating the peak value included in the frequency distribution and the frequency indicating the abnormality is invalidated. Device abnormality diagnosis can be performed with extremely high accuracy.
(V) The vibration of the rolling device is detected by the vibration sensor, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution of the signal from the vibration sensor and the first order indicating the abnormality. The frequency up to the nth order is compared for each corresponding order, and when the abnormality of the rolling device is diagnosed based on the result, the frequency depends on the rotational speed or moving speed of the rolling device. Since the amplification gain of the signal from the vibration sensor is adjusted, the abnormality diagnosis of the rolling device can be performed with extremely high accuracy regardless of the rotational speed or the moving speed of the rolling device.

  Hereinafter, a first embodiment and a second embodiment of an abnormality diagnosis apparatus according to the present invention will be described in detail with reference to the drawings.

  1A is a schematic plan view of a railway vehicle equipped with the abnormality diagnosis device of the present invention, FIG. 1B is a schematic side view of the railway vehicle, and FIG. 2 illustrates the positional relationship between an axle bearing and a vibration sensor. FIG. 3 is a block diagram of the first embodiment of the abnormality diagnosis apparatus according to the present invention, FIG. 4 is a flowchart showing the operation contents of the diagnosis processing unit of FIG. 3, and FIG. 5 is a threshold used for peak detection. FIG. 6 is a flowchart illustrating the contents of the abnormality determination process, FIG. 7 is an explanatory diagram about a range for obtaining an average value of the intensity of the detection signal, and FIG. 8 is an explanatory diagram about an evaluation section for abnormality diagnosis. 9 is an explanatory diagram of a lower limit reference value that enables abnormal vibration in the second embodiment of the abnormality diagnosis apparatus according to the present invention, and FIG. 10 is a setting of an amplifier gain (amplifier gain) according to the moving speed of the vehicle. Figure 11 is an amplification diagram Block diagram illustrating a specific configuration for changing stepwise as shown in FIG. 10 the gain, and FIG. 12 is a block diagram of a second embodiment of the abnormality diagnosis apparatus according to the present invention.

(First embodiment)
First, with reference to FIGS. 1-8, the abnormality diagnosis apparatus of 1st Embodiment is demonstrated.
As shown in FIG. 1, one rail vehicle 100 is supported by two front and rear chassis, and four wheels 101 are attached to each chassis. A vibration sensor 111 that detects vibration generated from the rotation support device 110 during operation is attached to the rotation support device (bearing box) 110 of each wheel 101.

  The control panel 115 of the railway vehicle 100 is equipped with two abnormality diagnosis devices 150 that take in sensor signals for four channels simultaneously (substantially simultaneously) and perform diagnosis processing. In other words, the output signals of the four vibration sensors 111 provided in each chassis are input to another abnormality diagnosis device 150 for each chassis via the signal lines 116. The abnormality diagnosis device 150 also receives a rotation speed pulse signal from a rotation speed sensor (not shown) that detects the rotation speed of the wheel 101.

  As shown in FIG. 2, the rotation support device 110 is provided with, for example, an axle bearing 130 that is a rotating component, and the axle bearing 130 is an inner ring 131 that is a rotating wheel fitted on a rotating shaft (not shown). An outer ring 132 that is a fixed ring fitted in a housing (not shown), a ball 133 that is a plurality of rolling elements disposed between the inner ring 131 and the outer ring 132, and the ball 133 is held in a freely rollable manner. A retainer (not shown). The vibration sensor 111 is held in a posture capable of detecting vibration acceleration in the direction of gravity and is fixed near the outer ring 132 of the housing. Various sensors such as an acceleration sensor, an AE (acoustic emission) sensor, an ultrasonic sensor, and a shock pulse sensor can be used as the vibration sensor 111.

As shown in FIG. 3, the abnormality diagnosis device 150 includes a sensor signal processing unit 150A and a diagnosis processing unit (MPU) 150B. The sensor signal processing unit 150A includes four amplification / filters (A
FILT) 151. The output signals of the four vibration sensors 111 are individually input to the amplifier / filter 151. Each amplifier / filter 151 has both an analog amplifier function and an anti-aliasing filter function. The four-channel analog signals amplified and filtered by these four amplifiers / filters 151 are taken into a diagnostic processing unit (MPU) 150B. The diagnostic processing unit (MPU) 150B includes a multiplexer (MUX) 152 and an AD converter (ADC) 153. The multiplexer (MUX) 152 converts the input 4-channel analog signals into signals for each channel. The signal is switched and converted into a digital signal by an AD converter (ADC) 153.

On the other hand, a rotational speed signal such as a speed proportional sine wave or rotational speed pulse signal from the rotational speed sensor is shaped by the waveform shaping circuit 155, and then the number of pulses per unit time is counted by a timer counter (not shown). The value is input to the diagnostic processing unit (MPU) 150B as a rotation speed signal. The diagnosis processing unit (MPU) 150B executes abnormality diagnosis based on the vibration waveform signal detected by the vibration sensor 111 and the rotation speed signal detected by the rotation speed sensor. The diagnosis result by the diagnosis processing unit (MPU) 150B is output to the communication line 120 (see FIG. 1) via the line driver (LD) 156. The communication line 120 is connected to an alarm device, and an appropriate alarm operation is performed when an abnormality such as a flatness of the wheel 101 occurs.
Further, the abnormality diagnosis device 150 includes a static random access memory (SRAM) 162 having a backup battery (Batt) 161.

  Further, among the abnormalities occurring in the axle bearing 130, the outer ring raceway of the stationary ring is most likely to peel off. Therefore, for the axle bearing 130, separation of the outer ring raceway of the stationary ring (hereinafter referred to as separation of the axle bearing) is a detection target.

  Abnormalities detected from the output signal of the vibration sensor 111 are peeling of the axle bearing 130 and flatness of the wheel 101. The defect frequency due to the separation of the axle bearing 130 and the defect frequency due to the flatness of the wheel 101 appear as peaks in the frequency distribution of the signal from the vibration sensor 111. Both defect frequencies can also be detected as vibration signals in a frequency band up to around 1 kHz. Further, since the frequency band differs by about 10 times between the defect frequency due to the separation of the axle bearing 130 and the defect frequency due to the flatness of the wheel 101, the data converted into a digital signal by the AD converter (ADC) 153 is processed by software. Thus, the abnormality diagnosis of both is performed separately for the axle bearing diagnosis and the wheel diagnosis.

  Further, the defect fundamental frequency due to the flatness of the wheel 101 is equal to the rotational speed (r / s) of the wheel 101. Accordingly, since the range of the rotational speed to be diagnosed is 3 to 30 r / s when the Shinkansen is assumed, the flat defect basic frequency of the wheel 101 is 3 to 30 Hz. On the other hand, the defect fundamental frequency due to the peeling of the axle bearing 130 is approximately 10 times the defect frequency of the flat of the wheel 101 and is 22 to 220 Hz. In both cases, when examining harmonics up to the fourth order, 3 to 120 Hz for the wheel 101 and 22 to 880 Hz for the axle bearing 130 are the required DFT (discrete Fourier transform) frequency analysis ranges.

  Here, the frequency resolution of the DFT at the time of diagnosis of the axle bearing 130 is sufficient at 1.0 Hz. At 0 Hz, the resolution is insufficient. Therefore, it is necessary to separately adjust the data so that the frequency resolution of the DFT in diagnosing the wheel 101 is 0.25 Hz.

  Therefore, in this embodiment, the data converted (sampled) into a digital signal by the AD converter (ADC) 153 is converted into two types of data having different sampling frequencies for the axle bearing separation analysis and the wheel flat analysis. To process.

  FIG. 4 shows an operation flow of the diagnostic processing unit (MPU) 150B in the first embodiment. The diagnostic processing unit (MPU) 150B switches the sensor signal output from the four vibration sensors 111 and sent through the amplifier / filter (AFILT) 151 to a signal for each channel by the multiplexer (MUX) 152. The AD converter (ADC) 153 converts the signal into a digital signal (ie, S101). In this embodiment, the sampling frequency in the AD converter (ADC) 153 is 8 kHz, and the sampling unit is 1.5 seconds.

  Then, a decimation process (that is, S102) is performed on the output signal of the AD converter (ADC) 153 by FIR low-pass filtering (LPF) realized by software. In the decimation process (ie, S102), in order to reduce the sampling frequency to 2 kHz, the decimation rate M is set to 4 and the number of data is reduced to ¼.

  The diagnostic processing unit (MPU) 150B uses the data subjected to the decimation processing (ie, S102) for analysis of axle bearing separation / separation (hereinafter referred to as “for bearing”) and for wheel flat analysis (hereinafter referred to as “for wheel”). And converted into two types of data having different sampling frequencies.

  The bearing data is obtained by performing the decimation process (that is, S102) and dividing the data of 1.5 second unit with the sampling frequency of 2 kHz into two and dividing the data into data intervals of 0.75 seconds (that is, , S111). Then, the absolute value processing (that is, S112) and the AC processing (that is, S113) are sequentially performed on the obtained data. Further, by adding zero (0) for about 0.25 seconds per section and interpolating the data (zero-padded interpolation), the data section length is about 1 second, and 2048 pieces of data are generated. (That is, S114). This data is subjected to FFT with a frequency resolution of about 1.0 Hz (ie, S115). Note that a Hanning window process is performed before the FFT process.

  After the FFT processing, a frequency Zfc indicating an abnormality in the separation of the bearing 130 obtained from the rotational speed and the internal dimensions of the bearing is obtained (see Table 1), and peak detection from the fundamental wave to the fourth order is performed on the obtained data. Perform (ie, S116). Then, the frequency indicating the peak value obtained from the peak detection result is compared with the value Zfc, and the degree of coincidence between them is calculated (ie, S117). Abnormality determination of the axle bearing 130 is performed based on the total degree of coincidence score obtained by repeating this process a predetermined number of times.

  On the other hand, the data for the wheel is decimated with the decimation rate M set to 4 by the filter (LFP) after the absolute value processing (ie, S121) after the decimation process (ie, S102) and the sampling frequency is 2 kHz. It is obtained by reducing the sampling frequency to 250 Hz by performing the process (ie, S122). The number of data at this time is 375, but zero padding interpolation (ie, S124) is performed to obtain data for about 4 seconds. FFT is performed on the obtained data with a frequency resolution of about 0.25 Hz (ie, S125). Note that a Hanning window process is performed before the FFT process.

  After the FFT processing, peak detection is performed on the obtained data (ie, S126). And the frequency which shows a peak value and the frequency which shows the abnormality in the flat of the wheel 101 are compared, and both coincidence degree is calculated (namely, S127). The abnormality determination of the wheel 101 is performed based on the total matching score obtained by repeating this process a predetermined number of times. The frequency indicating abnormality due to the flatness of the wheel is equal to the rotational speed (r / s) of the wheel 101. Accordingly, the frequency indicating the abnormality in the flatness of the wheel 101 is obtained by counting the number of pulses per unit time of the rotation speed pulse signal with a timer counter (not shown).

  Here, when calculating the coincidence between the frequency indicating the peak value and the frequency indicating the abnormality, an allowable error is set in order to accurately detect the coincidence, and the allowable error is defined as an allowable error function of the frequency. . That is, for example, when the allowable error is e (Hz) and the frequency indicating abnormality is Fx (Hz) (Fx is Zfc or Fr), e = a · Fx + b (Fx is Zfc or Fr). It is prescribed. However, a and b are setting parameters set in advance, Zfc (Hz) is a frequency indicating an abnormality in the separation of the axle bearing 130, and Fr (Hz) is a frequency indicating an abnormality in the flat of the wheel 101. is there.

When the above-described allowable error function is applied to this embodiment, the frequency resolution of FFT (1.0 Hz for bearings, 0.25 Hz for axles), and the revolution speed error and speed of the rolling elements of the bearings. In consideration of fluctuations, the allowable error for separation analysis of the axle bearing 130 from the fundamental wave to the fourth order is ef _{, b, i} (Hz) (where i = 1, 2, 3, 4), Ef _{, w, i} (Hz) for the flat analysis of the wheel 101 (where i = 1, 2, 3, 4), and the frequency resolution of the FFT for the bearing is δ _f1 (Hz), When the frequency resolution of the FFT for wheels is δ _f2 (Hz), the allowable error is defined by the following equation.
e _{f, b, i} (Hz) = a ₁ · i · Zfc + b ′ ₁ · δ _f1 (where i = 1, 2, 3, 4)
ef _{, w, i} (Hz) = a ₂ · i · Fr + b ′ ₂ · δ _f2 (where i = 1, 2, 3, 4)
Here, a ₁ and b ′ ₁ are setting parameters for bearings, a ₂ and b ′ ₂ are setting parameters for wheels, and b ′ ₁ and b ′ ₂ are setting parameters corresponding to the value b. Further, a ₁ and a ₂ are set in the range of 0.005 to 0.0035, more preferably 0.005 to 0.0030, and still more preferably 0.01 to 0.025. Further, b ′ ₁ and b ′ ₂ are set in the range of 0 to 7, more preferably 0 to 5, and further preferably 1 to 3.

For the axle bearing 130, the frequency f _{p, b, 1} indicating the peak value detected in the peak detection process (ie, S116) using the allowable error ± ef _{, b, 1} (Hz) obtained from the above formula. And the coincidence of the frequency Zfc indicating an abnormality in the separation of the axle 130 is determined by whether or not the following equation is satisfied.
Zfc−e _{f, b, 1} ≦ f _{p, b, 1} ≦ Zfc + ef _{, b, 1}
The frequency indicating the high-order peak value detected in the peak detection process (ie, S116) is similarly applied as in the following equation to analyze the coincidence of the high-order components up to the fourth order.
_{i · Zfc-e f, b} , i ≦ f p, b, i ≦ i · Zfc + e f, b, i
However, i = 2, 3, and 4.

As for the wheel, as in the case of the axle bearing separation analysis, the frequency f _{p, w, i} indicating the peak value detected in the peak detection process (ie, S126) and the basic frequency Fr indicating the abnormality due to the flatness of the wheel 101 are set. A match is determined by whether or not the following expression is satisfied.
i · Fr−e _{f, w, i} ≦ f _{p, w, i} ≦ i · Fr + ef _{, w, i}
However, i = 1, 2, 3, and 4.

  In addition, the detection of the frequency indicating the peak value in the peak detection process (ie, S116 and S126) is performed by setting a peak candidate at a point where the sign of the differential coefficient changes from positive to negative after flattening with a moving average. The threshold is determined for the value, that is, the peak amplitude spectrum value, and the candidates are narrowed down.

  That is, as shown in FIG. 5, the RMS value (root mean square) is used as an average value of amplitude spectrum values up to the Nyquist frequency (1/2 of the sampling frequency), and K times the threshold value is used as a threshold value. For example, if K = 3, a peak candidate that is less than three times the RMS value is discarded, and a peak candidate that is three times or more the RMS value is detected. As the average value, an absolute value average may be used instead of the RMS value. If there are still a large number of peak candidates even if the peak candidates below the threshold are rounded down, sorting is performed using the peak values to narrow the peaks to a certain number or less.

  The value K can be set by sending a command from the outside to the diagnosis processing unit (MPU) 150B of the abnormality diagnosis device 150 through a communication line. This value is individually stored for the axle and the wheel in the SRAM 162 backed up by the battery 161. The diagnostic processing unit (MPU) 150B operates by reading the set value from the SRAM 162 at the time of restart even when reset. This makes it possible to change the threshold value flexibly according to the situation at the work site.

The flow of the abnormality determination process in this embodiment will be described in detail according to the flow of FIG. Buffer replacement is sequentially performed on data generated by AD conversion of the vibration signal (that is, S201), filter processing such as a low-pass filter (LPF) (that is, S201) is performed, and the RMS value of the vibration signal is calculated. (That is, S203).
Next, the degree of coincidence is calculated by frequency peak detection (ie, S204). In this process, a frequency indicating a peak value equal to or higher than a threshold is compared with a frequency indicating an abnormality, and as illustrated in Table 2, a total of two of the fundamental wave component and the components up to the fourth order other than the fundamental wave are 3 If two amplitude spectra match, a score of 1 is given to the degree of matching. If all match, a score of 2 is given. Otherwise, it is 0. This degree of coincidence determination is performed with a preset number N of calculations.

  Further, as shown in FIG. 7, since it is inappropriate to perform abnormality diagnosis when the amplitude of the vibration signal is small, an average value (here, RMS value) of the signal in the sampling range and a predetermined minimum reference value are obtained. In comparison (ie, S205), if the average value of the vibration signals is below the minimum reference value, the coincidence determination is not performed, and the coincidence degree is invalid, that is, 0.

  Furthermore, in the present embodiment, in a predetermined evaluation section in which FFT during running is performed, the total degree of coincidence Sp over N times of continuous calculation in the evaluation section length is calculated (that is, S207), The total matching score Sp is compared with the reference total score P (ie, S208). As a result, if the coincidence total score Sp is larger than the reference score P, a vibration abnormality alarm is output (that is, S209, see FIG. 8).

As the number of calculations N is increased without changing the values of the reference point P, the number of calculations N, and the value P / N, the probability of occurrence of a false alarm for abnormality diagnosis can be reduced. However, as the reference point number P and the calculation number N are equal, the risk of overlooking the abnormality increases, and when the value N is excessively increased, the time required for abnormality determination increases.
Therefore, in this embodiment, a program having a communication command that can set and change the values P and N from the outside is installed in the abnormality diagnosis device 150, and the values P and N can be adjusted sequentially. Yes. The set / changed values P and N are stored in the SRAM 162 backed up by the battery 161. These values can be set separately for axles and wheels. For example, the reference total points and calculation times for axle outer ring raceway separation analysis are Pb and Nb, respectively, and the reference total points for wheel flat analysis. And the number of calculations are distinguished as Pw and Nw, respectively.

Accordingly, the abnormality diagnostic device 100 of the present embodiment, the frequency (f p which indicates the peak value from the primary included in the frequency distribution of the signal from the vibration sensor 111 (fundamental) to the fourth _{order, b, i,} f _{p , W, i} ) and the frequencies (Zfc, Fr) included in the frequency indicating abnormality match each corresponding order, and based on the result, the axle bearing is compared. The abnormality diagnosis of 130 and the wheel 101 is performed. At that time, the frequency showing the peak value to be compared _{(f p, b, i,} f p, w, i) a frequency indicating the abnormality (ZFC, Fr) difference between the tolerances _{(e f, b,} i, ef _{, w, i} ) are compared to determine whether they match or not, and the permissible error (ef _{, b, i} , ef _{, w, i} ) is preset as a function of frequency. Therefore, the abnormality diagnosis of the axle bearing 130 and the wheel 101 can be performed with extremely high accuracy. According to the abnormality diagnosis apparatus 100, it is possible to perform an extremely accurate abnormality diagnosis that is not easily influenced by noise from a track or the like and the magnitude of traveling speed.

  Further, the number N of coincidence calculations and the reference total score P for the calculation results over the N degrees of coincidence are stored in the SRAM 162 backed up by the battery 161, and the diagnosis processing unit (MPU) 150B is stored in the SRAM 162. The number of coincidence calculations N and the reference total score P are sequentially read out, the coincidence is continuously obtained N times, and abnormality diagnosis is performed based on the N coincidence total scores Sp. The abnormality diagnosis of the wheel 101 can be performed with extremely high accuracy. In addition, since the number N of coincidence calculations and the reference total score P can be set and changed from the outside, the diagnostic accuracy can be flexibly changed according to the situation.

Further, a value K that determines the threshold value of the frequency (fp _{, b, i} , fp _{, w, i} ) indicating the peak value is stored in the SRAM 162 backed up by the battery 161, and the diagnostic processing unit (MPU) 150B Since the value K is sequentially read from the SRAM 162 and the two frequencies are compared using a peak value that is equal to or greater than the threshold value determined by the value K, abnormality diagnosis of the axle bearing 130 and the wheel 101 can be performed with extremely high accuracy. Moreover, since the value K can be set and changed from the outside, the threshold can be flexibly changed according to the situation.

The average value of the frequencies (fp _{, b, i} , fp _{, w, i} ) indicating the peak values from the first order to the fourth order included in the frequency distribution of the signal from the vibration sensor 111 is a predetermined reference value. If the frequency is lower than the range, the frequency indicating the peak value included in the frequency distribution and the frequency indicating the abnormality are not compared, so that the abnormality diagnosis of the axle bearing 130 and the wheel 101 is performed with extremely high accuracy. Can be implemented.

  In the present embodiment, most of the digital processing is performed by software, but part or all of the digital processing may be realized by hardware such as an FPGA (Field Programmable Gate Array).

(Second Embodiment)
As a modification of the first embodiment, the second embodiment will be described with reference to the drawings. In the second embodiment of the abnormality diagnosis device according to the present invention, as shown in FIG. 9, a reference value of the RMS value of the effective vibration signal is set in accordance with the moving speed of the railway vehicle 100, and is equal to or less than the reference value. Abnormality diagnosis is invalidated with vibration, and abnormality diagnosis is similarly invalidated when the moving speed is below a reference value. However, for this purpose, since it is necessary to suppress variation in the RMS value of the vibration signal due to the moving speed, the gain of the amplifier before the AD conversion (amplifier gain) is stepwise as shown in FIG. 10 according to the moving speed. Change to Specifically, a device such as a PGA (Programmable Gain Amplifier) that can sequentially set the gain of the amplifier on the microcomputer side is mounted on the abnormality diagnosis apparatus 150 (see FIG. 11).

  Further, instead of the SRAM with battery 162 in FIG. 3, an FRAM (Ferroelectric RAM) 165 is adopted as shown in FIG. This eliminates the need for a battery for storing set values and data. FRAM is overwhelmingly easier to rewrite and faster than EEPROM (Electrically Erable and Programmable ROM) and flash ROM, and has a long life, and even if it is used in the same way as SRAM, it can secure a useful life of 10 years. Further, the battery may be provided only for the RTC 163. In addition, when the battery is shared between the RTC 163 and the SRAM 162, the battery life is extended by the amount of the SRAM 162 being eliminated.

  Therefore, according to the second embodiment, by adjusting the amplification gain of the signal from the vibration sensor 111 according to the moving speed of the vehicle, the axle bearings 130 and the wheels 101 are not affected by the magnitude of the moving speed of the vehicle. Each abnormality diagnosis can be performed with extremely high accuracy.

  The description of the specific embodiments is finished above, but the aspect of the present invention is not limited to these embodiments, and modifications, improvements, and the like can be made as appropriate. Moreover, although the axle bearing 130 and the wheel 101 were illustrated and demonstrated as a rolling device, it is applicable not only to this but the abnormality diagnosis of a gear.

(A) is a schematic plan view of the railway vehicle carrying the abnormality diagnosis device of the present invention, and (b) is a schematic side view of the railway vehicle. It is the schematic which illustrates the positional relationship of an axle shaft bearing and a vibration sensor. 1 is a block diagram of a first embodiment of an abnormality diagnosis apparatus according to the present invention. It is a flowchart which shows the operation | movement content of the diagnostic process part of FIG. It is explanatory drawing about the threshold value used in the case of peak detection. It is a flowchart which illustrates the content of the abnormality determination process. It is explanatory drawing about the range which calculates | requires the average value of the intensity | strength of a detection signal. It is explanatory drawing about the evaluation area of abnormality diagnosis. It is explanatory drawing about the lower limit reference value which validates abnormal vibration in 2nd Embodiment of the abnormality diagnosis apparatus which concerns on this invention. It is explanatory drawing about the signal amplification gain (amplifier gain) setting according to the moving speed of a vehicle. FIG. 11 is a block diagram exemplifying a specific configuration for changing the gain of an amplifier stepwise as shown in FIG. 10. It is a block diagram of 2nd Embodiment of the abnormality diagnosis apparatus which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Rail vehicle 101 Wheel 110 Rotation support apparatus 111 Vibration sensor 130 Axle bearing 150 Abnormality diagnosis apparatus 150A Sensor signal processing part 150B Diagnosis processing part 152 Multiplexer 162 SRAM

Claims (14)

A vibration sensor for detecting the vibration of the rolling device;
The frequency distribution of the signal from the vibration sensor is obtained, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution corresponds to the frequency from the first order to the nth order indicating the abnormality. A degree of coincidence between the two frequencies compared to each order to be scored, and a diagnosis processing unit that performs abnormality diagnosis of the rolling device based on the result,
An externally rewritable setting value storage unit that stores the number N of coincidence calculations and the reference total score P for the calculation results over N times the coincidence,
The diagnosis processing unit
By sequentially reading the values N and P from the set value storage unit, obtaining the coincidence N times, obtaining the coincidence total score, and comparing the N coincidence total score with the reference total score P An abnormality diagnosis apparatus characterized by performing abnormality diagnosis. The setting storage unit stores a value K that determines a threshold value of the peak value ,
It said value K is a rewritable from the outside,
The diagnosis processing unit
The abnormality diagnosis apparatus according to claim 1, wherein a value K is read from the set value storage unit and a frequency indicating a peak value equal to or higher than a threshold value determined by the value K is obtained. 3. The abnormality diagnosis device according to claim 1, wherein the values N, P and / or K stored in the set value storage unit are different depending on an abnormality type or a part of the rolling device. Abnormality diagnosis device. The diagnosis processing unit
When the average value of the amplitude of the signal from the vibration sensor in a predetermined period falls below a predetermined reference value, the frequency indicating the peak value included in the frequency distribution matches the frequency indicating the abnormality. The abnormality diagnosis device according to any one of claims 1 to 3, wherein the comparison of whether or not is invalidated. The diagnosis processing unit
5. The abnormality diagnosis apparatus according to claim 1 , wherein an amplification gain of a signal from the vibration sensor is adjusted according to a rotation speed or a movement speed of the rolling device. The diagnosis processing unit
A comparison is made as to whether or not the difference between the frequency indicating the peak value to be compared and the frequency indicating the abnormality is within an allowable error range, and the allowable error is preset as a function of the frequency. The abnormality diagnosis device according to claim 1, wherein the abnormality diagnosis device is defined by an allowable error function. In the abnormality diagnosis device according to any one of claims 1 to 6, wherein the rolling device axle bearing for a railway vehicle, the abnormality diagnosis apparatus characterized by either of the gear or wheel. The vibration of the rolling device is detected by the vibration sensor, and the frequency indicating the peak value from the first order (fundamental wave) to the nth order included in the frequency distribution of the signal from the vibration sensor and the first order to the nth order indicating the abnormality. In the abnormality diagnosis method for comparing each frequency with each corresponding order and scoring as a degree of coincidence between both frequencies, and performing abnormality diagnosis of the rolling device based on the result,
The number N of coincidence calculations and the reference total score P for the calculation results over N times of the coincidence are stored in externally rewritable setting value storage means,
The values N and P are sequentially read from the set value storage means, and the degree of coincidence is obtained N times in succession to obtain the degree of coincidence total score, and the total degree of coincidence score and the reference total score P at the N times are obtained. An abnormality diagnosis method characterized in that an abnormality diagnosis is performed by comparison. Stores the setting value storing means capable of rewriting the value K for determining the threshold value of the peak value from the outside,
The abnormality diagnosis method according to claim 8, wherein the value K is sequentially read from the set value storage unit, and a frequency indicating a peak value equal to or higher than a threshold value determined by the value K is obtained. 10. The abnormality diagnosis method according to claim 8 or 9, wherein the values N, P and / or K stored in the set value storage means differ depending on the type or part of the abnormality of the rolling device. Abnormal diagnosis method. When the average value of the amplitude in a predetermined period of the vibration generated by the rolling device is lower than a predetermined reference value, the frequency indicating the peak value included in the frequency distribution and the frequency distribution indicating the abnormality, The abnormality diagnosis method according to any one of claims 8 to 10, wherein the comparison of whether or not they coincide is invalidated. The abnormality diagnosis method according to claim 8 , wherein an amplification gain of a signal from the vibration sensor is adjusted according to a rotation speed or a movement speed of the rolling device. . A comparison is made as to whether or not the difference between the frequency indicating the peak value to be compared and the frequency indicating the abnormality is within an allowable error range, and the allowable error is preset as a function of the frequency. The abnormality diagnosis apparatus method according to any one of claims 8 to 12, wherein the abnormality diagnosis apparatus method is defined by an allowable error function. In the abnormality diagnosis method according to any one of claims 8 to 13, the abnormality diagnostic method, wherein the rolling device axle bearing for a railway vehicle, either of the gear or wheel.

Images