JP6854595B2 - Diagnostic device for diesel engine plain bearings - Google Patents

Description

The present invention relates to a diagnostic device for sliding bearings of a diesel engine. More specifically, the present invention relates to an improvement in a diagnostic analysis method for detecting signs of bearing contact (rubbing) peculiar to plain bearings of diesel engines in ships and power generation equipment.

Plain bearings are used for large-scale important rotating equipment such as turbines and blowers, and for general industrial applications such as compressors and other high-speed rotating equipment and important equipment, as well as bearings for diesel engines for ships and power generation. Is also used. Since the plain bearing is surrounded by an oil film and rotates in a non-contact manner, it is not normally damaged, but the rotating shaft and the plain bearing come into contact (rubbing) due to improper construction, imbalance of the blades, misalignment of the coupling, abnormal vibration due to oil whip, etc. ) And damage. If any abnormality occurs in the slide bearing, vibration or noise different from the usual one is generated, and if the driving is continued in such a state, the slide bearing may be damaged in some cases. In addition, if the rubbing abnormality progresses, seizure may occur and the equipment may be stopped.

As a diagnostic technique for a plain bearing to avoid such a situation, a method using shaft vibration by an eddy current type displacement sensor and bearing box vibration by a piezoelectric type acceleration sensor has been conventionally proposed. For example, if the spectrum pattern obtained by applying the fast Fourier transform (FFT) to the waveform data such as vibration and sound generated from the slide bearing and the preset spectrum pattern at the time of abnormality are compared, the comparison is made. It is possible to diagnose the plain bearing based on the result.

Further, as a method for determining the presence or absence of rubbing (contact between the rotating part and the stationary part) in a slide bearing or the like, for example, the sound generated when the rotating shaft and the bearing come into contact is detected and an acoustic signal is extracted, and this sound is obtained. It has also been proposed that the signal is subjected to envelope detection processing (envelope processing) to form post-envelope detection data, the Keptram value is compared with the reference value, and the rubbing judgment is performed from this judgment result (Patent Documents). 1). If the occurrence of the rubbing phenomenon, which is the initial stage of seizure damage of the slide bearing, can be detected at an early stage at a minor stage, the seizure damage can be prevented.

By the way, in diagnosing a slide bearing as described above, a method of attaching an acceleration sensor to the slide bearing to make a diagnosis is appropriately used.

Japanese Unexamined Patent Publication No. 8-261817

However, if you try to attach an accelerometer to the plain bearing of a diesel engine, you have to disassemble the engine to some extent, attach the accelerometer to the plain bearing, and then reassemble the engine. If this is done, not only will it take a considerable amount of time and effort, but there will also be problems such as the engine manufacturer's inability to guarantee operation.

Therefore, an object of the present invention is to provide a diagnostic device for a plain bearing of a diesel engine, which enables diagnosis without going through operations such as disassembly and assembly of the engine.

In order to solve such a problem, the present invention provides a device for diagnosing an abnormality caused by a contact phenomenon inside a slide bearing of a rotating shaft of a diesel engine.
The rotating shaft is a crankshaft which is a main shaft.
The diagnostic device is
A rotation speed detection sensor that detects the rotation speed of the rotation shaft, and
An acceleration sensor attached to the surface of the housing of the diesel engine to detect the acceleration of the rotating shaft during vibration, and an acceleration sensor.
Waveform data representing the acceleration of vibration detected by the acceleration sensor is converted into waveform data in the frequency domain by Fourier conversion, and predetermined signal processing in which the waveform data in the frequency domain and the rotation speed of the rotation axis are combined. A plurality of peak information generated at rotation frequency intervals in the waveform data of the frequency domain is extracted, a characteristic value is obtained from the peak information by a calculation device having a predetermined peak value as a characteristic value, and the characteristic value is determined. A monitoring device that monitors whether or not the threshold value has been exceeded, and determines that an abnormality has occurred in the sliding bearing when the characteristic value exceeds the threshold value.
When the monitoring device determines that an abnormality has occurred in the plain bearing, a notification device that outputs the determination result to the outside, and
With
The housing has reinforcing ribs that hold the slide bearings and propagate vibrations generated by the rotation of the rotating shaft.
On the surface of the housing, a region is formed in which the degree of distance attenuation of the vibration is relatively small due to the propagation of the vibration by the reinforcing ribs.
The acceleration sensor is a diagnostic device for a slide bearing of a diesel engine, which is attached to a portion of the surface of the housing where a reinforcing rib is provided between the surface and the slide bearing.

According to this diagnostic device that attaches the accelerometer to the surface of the diesel engine, it is not necessary to install the accelerometer inside the engine, so it is necessary to disassemble or modify the engine inevitably when installing the accelerometer inside the engine. Is gone. Therefore, it is possible to reduce the labor and cost that have been required in the past, and it does not affect the operation guarantee by the engine manufacturer.

The number of the acceleration sensors may be the same as that of the plurality of the slide bearings.

Each of the acceleration sensors may be arranged equidistant from the slide bearing to be detected.

The acceleration sensor may be attached to the surface of a reinforcing rib formed on the radial outer side of the slide bearing.

The acceleration sensor may be arranged on the side where the peripheral surface of the rotation shaft rotates from the lower side to the upper side in the vertical direction.

The acceleration sensor may be mounted on the surface of the housing of the diesel engine at a position closest to the slide bearing.

The accelerometer may be arranged in the horizontal direction of the plain bearing.

The acceleration sensor may be arranged at a position on the surface of the housing on an extension of the line connecting the slide bearing and the reinforcing rib.

The acceleration sensors may be arranged at equal intervals along the axial direction of the rotation axis.

According to the present invention, it is possible to make a diagnosis without going through operations such as disassembling and assembling the engine.

It is a schematic block diagram of the diagnostic apparatus of a plain bearing which shows one Embodiment of this invention. It is a figure which shows an example of the acceleration waveform data obtained by using a rotation speed detection sensor and an acceleration sensor. It is the waveform data (power spectrum) of the frequency domain obtained by Fourier transforming the acceleration waveform data of FIG. It is each characteristic value of (A) keptrum, (B) autocorrelation, and (C) cross-correlation obtained from the acceleration spectrum. This is an example of a template used for cross-correlation. It is a graph which shows an example of the time-dependent change of a characteristic value. It is (A) plan view and (B) front view which shows an example of the slide bearing made of the test metal used for the verification of this invention. It is a graph which shows the relationship between a gap and an acceleration O / A value. It is a graph which shows the acceleration spectrum at the rotation speed 1200 [rpm]. It is a graph which shows the zooming spectrum at the rotation speed 1200 [rpm]. It is a graph which shows the relationship between the dimensionless sign parameter and a gap at a rotation speed of 1200 [rpm]. It is a graph which shows the acceleration waveform and the acceleration spectrum at a rotation speed of 1200 [rpm], and a gap of 1/100 [mm]. It is a graph which shows the zooming spectrum at a rotation speed of 1200 [rpm], and a gap of 1/100 [mm]. It is a graph which shows (A) cepstrum, (B) autocorrelation and (C) cross-correlation at a rotation speed of 1200 [rpm] and a gap of 15/100 [mm]. It is a graph which shows (A) cepstrum, (B) autocorrelation and (C) cross-correlation at a rotation speed of 1200 [rpm] and a gap of 3/100 [mm]. It is a graph which shows (A) cepstrum, (B) autocorrelation and (C) cross-correlation at a rotation speed of 1200 [rpm] and a gap of 1/100 [mm]. It is a graph which shows (A) cepstrum, (B) autocorrelation and (C) cross-correlation at a rotation speed of 1200 [rpm] and a gap of 0/100 [mm]. A graph showing the relationship between the dimensionless symptom parameter at a rotation speed of 1200 [rpm] and the output ratio (ratio between the output level at the time of abnormality and the output level at the normal time), a) the comparison between the kefrenity level and the conventional method. , B) It is a graph which shows the comparison of each characteristic value. A graph showing the relationship between the dimensionless symptom parameter at a rotation speed of 2800 [rpm] and the output ratio (ratio between the output level at the time of abnormality and the output level at the normal time), a) Comparison between the kefrenity level and the conventional method. , B) It is a graph which shows the comparison of each characteristic value. It is a graph which shows the kefrenity trend and cepstrum at each gap. It is a graph which shows an example (rotation speed 79.6 rpm) of the acceleration time axis waveform in a normal state. It is a graph which shows an example (rotation speed 79.6 rpm) of the acceleration envelope spectrum in a normal state. It is a figure of (A) rotation period kefrenity value trend and (B) cepstrum, (C) autocorrelation and (D) cross-correlation at a predetermined time which shows an example of the cepstrum, autocorrelation and cross-correlation analysis result in a normal state. .. An example of cepstrum, autocorrelation and cross-correlation analysis results when slight rubbing occurs (rotation speed 85 rpm), (A) rotation period kefrenity value trend, (B) cepstrum at a predetermined time, (C) autocorrelation, and (D) It is a figure of cross-correlation. It is a graph which shows an example (rotation speed 85 rpm) of the acceleration time axis waveform at the time of the occurrence of slight rubbing. It is a graph which shows an example (rotation speed 85 rpm) of the acceleration envelope spectrum at the time of the occurrence of slight rubbing. It is a graph which shows the setting example of the threshold value. It is a block diagram which shows the outline of the structure of the apparatus which diagnoses a slide bearing of a diesel engine. It is a front view of the inside of the diesel engine which shows an example of the slide bearing of a diesel engine and its diagnostic apparatus. It is a side view of the inside of the diesel engine which shows an example of the slide bearing of a diesel engine and its diagnostic apparatus. It is a front view which shows the structural example of the piston, the crosshead, etc. of a diesel engine. It is a side view which shows the structural example of the piston, the crosshead, etc. of a diesel engine. It is a top view which shows the structural example of the piston, the crosshead, etc. of a diesel engine. It is a figure explaining the relationship between the rotation direction of the rotation shaft (crankshaft) of a diesel engine, and the arrangement of an acceleration sensor. It is a figure explaining the wedge film pressure when the load acting between a rotating shaft and a main metal (sliding bearing) is a static load. It is a figure explaining the drawing film pressure when the load acting between a rotating shaft and a main metal (sliding bearing) is a fluctuating load.

Hereinafter, the configuration of the present invention will be described in detail based on an example of an embodiment shown in the drawings. In the following, first, a diagnostic device for a plain bearing and a diagnostic method based on the diagnostic device will be described in detail, and then a device for diagnosing the plain bearing of a diesel engine will be described.

1 to 6 show a method of diagnosing a slide bearing and an embodiment of a diagnostic device. The slide bearing 1 is a bearing device for a rotating shaft 2 that can be applied to large-scale important rotating equipment such as a turbine and high-speed rotating equipment such as a compressor. The diagnostic device 10 for the slide bearing 1 according to the present invention includes a rotation speed detection sensor 11, an acceleration sensor 12, a monitoring device 13, and a notification device 14. Hereinafter, the configuration of the diagnostic apparatus 10 will be described first (see FIGS. 1, 2, etc.).

The rotation speed detection sensor 11 is a sensor for detecting the rotation speed of the rotation shaft 2. For example, in the present embodiment, the rotation pulse of the rotation shaft 2 is detected via the detected member 11b provided on the surface of the rotation shaft 2 and rotating together with the rotation shaft 2, for example, a reflective tape, and the detection member 11b. The rotation speed detection sensor 11 is configured by the pulse detector 11a (see FIG. 1). The data detected by the pulse detector 11a is transmitted to the monitoring device 13. Although the rotation speed of the rotation shaft 2 in the actual machine is controlled, the rotation speed is often fluctuated due to the influence of voltage fluctuation or the like. According to such a rotation speed detection sensor 11, it is possible to accurately detect the rotation speed by using the pulse even when the rotation speed of the rotation shaft 2 fluctuates from moment to moment. Alternatively, it is also possible to detect the rotation speed from the output voltage of a tachometer installed in a ship engine or the like.

The acceleration sensor 12 is a sensor for detecting the acceleration of the rotating shaft 2 at the time of vibration based on the vibration when a rubbing phenomenon or the like occurs. For example, in this embodiment, a piezoelectric type acceleration sensor having a piezo element is used. The acceleration sensor 12 is attached to the bearing box of the slide bearing 1 or the housing of the engine including the slide bearing 1 to detect the acceleration based on the vibration of the bearing box or the like (see FIG. 1). The detection data by the acceleration sensor 12 is transmitted to the monitoring device 13.

The monitoring device 13 diagnoses the slide bearing 1 based on the transmission data from the rotation speed detection sensor 11 and the transmission data from the acceleration sensor 12, and when it is determined that an abnormality has occurred, the result is the result. Is a device for transmitting to the reporting device 14. Specifically, the monitoring device 13 of the present embodiment converts the waveform data representing the acceleration of vibration detected by the acceleration sensor 12 into a power spectrum in the frequency domain by Fourier transforming, and the rotation frequency of the axis in the spectrum. A plurality of peak information generated at intervals is quantified by performing a predetermined signal processing in combination with the rotation frequency information of the axis, a characteristic value is obtained, and the obtained characteristic value exceeds a predetermined threshold value. When the characteristic value exceeds the threshold value, it is determined that the sliding bearing 1 has an abnormality. Further, an arithmetic processing unit (for example, a personal computer) is connected to the monitoring device 13 of the present embodiment.

The reporting device 14 is a device for outputting the determination result when the monitoring device 13 determines that an abnormality has occurred in the slide bearing 1 and notifying the user or related parties. The reporting device 14 may be a device that reports to the outside by, for example, blinking a light or sounding an alarm sound, or a device that reports to related parties by using the screen of the arithmetic processing unit 15. You may.

Subsequently, a diagnostic method of the slide bearing 1 using such a diagnostic device 10 and its principle and the like will be described (see FIG. 4 and the like).

By using the rotation speed detection sensor 11 and the acceleration sensor 12 of the diagnostic device 10 described above, it is possible to detect the acceleration waveform data on the horizontal axis of time as in the conventional case (see FIG. 2). When an abnormality has occurred in the plain bearing 1, unless the acceleration of the vibration generated by the abnormality is very large, for example, the change in the acceleration amplitude in the acceleration waveform data when a slight rubbing occurs is Since it is very small, it is extremely difficult to detect the presence or absence of an abnormality only from this waveform data.

Here, in the present embodiment, the acceleration waveform data is Fourier transformed and frequency-analyzed to obtain waveform data in the frequency domain (see FIG. 3). As a result, a power spectrum having a frequency on the horizontal axis can be obtained.

A plurality of peak information generated at the rotation frequency interval of the axis in the spectrum is quantified by performing a predetermined signal processing in combination with the rotation frequency information of the axis, and a characteristic value is obtained.

As a more specific detection method of the characteristic value, the logarithmic spectrum of the spectrum in the frequency domain is calculated, the logarithmic spectrum is subjected to the inverse Fourier transform (cepstrum calculation), and the kefrenity value obtained from the waveform data after the cepstrum calculation is obtained. Used (see FIG. 4 (A)). Since the kefrency has a peak at a position that is a multiple of the rotation cycle of the shaft, a plurality of peaks may be added to obtain a characteristic value.

Further, as the characteristic value, the following autocorrelation value of the acceleration spectrum can be used (see FIG. 4B). That is, when N (524288 in this embodiment) acceleration data are collected at a sampling frequency Z [Hz] (51200 [Hz] in this embodiment), the acceleration spectrum in the wave frequency domain is Z / N [Hz]. Can be decomposed into N frequency domains from to Z [Hz]. However, the generally effective frequency range is 1/2 or less of Z [Hz]. Therefore, the data string (vector) up to the length N'is represented by X in the spectrum. For this X, the autocorrelation as in the above equation 1 is calculated, and m is calculated from 0 to N'-1. When slight rubbing occurs, there are multiple peaks at the rotation speed interval of the shaft, so m of Rxx (m) (m * Z / N [Hz] in frequency conversion) corresponds to a multiple of the rotation cycle. Appears as a large peak when the value reaches. A peak other than a multiple of zero, for example, the first peak value can be used as a characteristic value.

The frequency domain of the acceleration spectrum in this process is limited as follows from the effective frequencies of 1000 [Hz] to 20000 [Hz]. A square wave (see FIG. 5) having peaks at rotation frequency intervals is prepared. In the present embodiment, a square wave having a peak width of 1/10 of the rotation period of the axis and having 10 peaks is prepared, and the value of the cross-correlation between the square wave and the equation 2 of the acceleration spectrum is the maximum. Find the value of m. Find the average data length w of the square wave. The value R [Hz] of frequency (m + w) × Z / N [Hz] is the center frequency of the acceleration spectrum having the highest correlation with the square wave. In this example, the range of ± 1000 [Hz] of this value R [Hz] is specified. In this calculation, if the minimum value of the range is less than the minimum value of the effective frequency (1000 [Hz] in this example), 1000 [Hz] to 3000 [Hz] is specified. If the upper limit of the effective range (20000 [Hz] in this example) is exceeded, it is set to 18000 [Hz] to 20000 [Hz].

However, other waveforms such as a Gaussian distribution can be used instead of the square wave used to determine the frequency. Further, the peak width and length are not limited to the above.

Further, the frequency of the acceleration spectrum used for the calculation of the autocorrelation can be specified by the users from the acceleration spectrum.

In addition, the peak value of autocorrelation, which is the above-mentioned characteristic value, may be affected by the baseline (see FIG. 4B), and the baseline portion may be subtracted, or the peak value may be the average value of the baseline. It is desirable to break it. In this example, the 0.7-0.8 period part of the fundamental frequency is used as the base and the average value is subtracted.

Further, since the autocorrelation has a peak at a position that is a multiple of the rotation cycle of the shaft, a plurality of peaks may be added to obtain a characteristic value.

（ｍは０からＮ’−１）

(M is 0 to N'-1)

Further, it is possible to use the following cross-correlation of acceleration spectra as the characteristic value (see FIG. 4C). That is, it is also possible to obtain a characteristic value by cross-correlation between an artificial spectrum (called a template waveform) Y having a predetermined peak such as a rectangle or a Gaussian distribution at the rotation cycle interval of the axis and the acceleration spectrum X. is there. The data length of Y of the template waveform can be composed of the same or shorter waveform as the acceleration spectrum X. Let M be the length of the Y waveform, and if M is shorter than the length N'of the acceleration spectrum X, add zero so that it has the same length as X, and calculate with the above equation 2. When slight rubbing occurs, since there are a plurality of peaks at the rotation speed interval of the shaft, it appears as a large peak when m of Rxy (m) is a value corresponding to a multiple of the rotation period. However, in the calculation with the formula 2 as it is, the head position of the first obtained peak may deviate from the rotation frequency of the shaft, and the subsequent peak positions may also deviate by the same amount. Therefore, in Rxy (m), it is possible to correct this deviation by finding m that maximizes Rxy while m = 0 to m becomes the rotation frequency and setting that m to m = 0. is there.

（ｍは０からＭ−１）

(M is from 0 to M-1)

At this time, in the present embodiment, the same method as in the case of the autocorrelation is adopted for the correction of the peak by the frequency and the baseline of the acceleration spectrum for calculating the cross-correlation.

In addition, the template used to obtain the cross-correlation has a width of 1/10 of the rotation frequency of the axis and is 2.5 cycles shorter than the rotation frequency of the axis with respect to the specified range of the acceleration spectrum.

However, the template is not limited to a square wave as long as it has a peak at the rotation frequency interval of the axis. Further, the width and length of the peak are not limited to this embodiment. Further, since the cross-correlation has a peak at a position that is a multiple of the rotation period of the shaft, a plurality of peaks may be added to obtain a characteristic value.

In the present embodiment, when the time-series data of the characteristic value obtained by quantifying the peak value at the rotation frequency interval of the shaft in this way is obtained, it is monitored whether or not the characteristic value exceeds a predetermined threshold value. In general, this characteristic value tends to increase remarkably when an abnormality such as rubbing occurs in the slide bearing 1 (see the alternate long and short dash line in FIG. 6). On the other hand, in the present embodiment, a predetermined threshold value is set in advance for this characteristic value, and when the characteristic value L that changes with time exceeds the threshold value, an abnormality occurs in the slide bearing 1. Judge that you are doing. As an example, in the present embodiment, when the normal average value + 3σ (standard deviation) of the kefrency trend level is exceeded, it is determined that an abnormality (contact) has occurred in the slide bearing 1, and the value is determined. It is set as the threshold value of. Needless to say, the above concept of the threshold value uses the cross-correlation of the acceleration spectrum as the characteristic value when the autocorrelation value of the acceleration spectrum is used as the characteristic value (see FIG. 4 (B)). The same applies to the case where there is (see FIG. 4C).

Here, the threshold value will be further described based on the specific example shown in FIG. 27. In FIG. 27, in the region (1) which is the normal state, the average value of the kefency trend level calculated from the actual data is 0.0343, and the standard deviation σ is 0.0071. In this case, the mean value + 3σ is 0.0556 (see the broken line representing the threshold value in FIG. 27).

Alternatively, as another example of the above, twice the average value of the kefrency trend level can be set as a predetermined threshold value. In such a case, the value may be slightly larger than the average value + 3σ, but there is no practical problem.

According to the analysis of the characteristic value obtained by quantifying the peak value at the rotation frequency interval of the axis as described above, the periodicity buried in the spectrum is detected and the fundamental frequency (or the reciprocal of the fundamental period) is determined. Can be sought. According to this, even a slight rubbing or the like can be detected with high accuracy, and its sign can be detected at an early stage. Moreover, in the present embodiment, it is possible to diagnose an abnormality by using the conventionally used acceleration sensor as it is.

The above-described embodiment is an example of a preferred embodiment of the present invention, but the present invention is not limited to this, and various modifications can be carried out without departing from the gist of the present invention. For example, in the above-described embodiment, it has been described that the plain bearing 1 according to the present invention can be applied to a large important rotating equipment, a high-speed rotating equipment such as a compressor, and the like. It goes without saying that the fields or scopes include plain bearings for rotating equipment in plants such as petrochemicals, nuclear power plants, petroleum refining, and steel. In addition to these, it can also be applied to bearings with vibration and acoustic noise due to piston movement and explosion of marine diesel engines and diesel generators, which have been difficult to detect in the past.

For example, in the case of a 4-stroke cycle used in a diesel generator, even in a normal state, a diesel engine generates vibration that is frequency-modulated once every two revolutions due to an explosion. When rubbing occurs, the vibration waveform undergoes frequency modulation with regularity of rotation frequency due to the strength of contact pressure due to the eccentricity of the spindle. When frequency modulation is applied, a sideband wave with a rotation period is generated in the vibration acceleration spectrum. That is, the signal due to the explosion and the signal due to the rubbing can be separated depending on the generation cycle of the kefrency.

Further, in the case of a two-stroke cycle used for a diesel engine of a large ship, a kefency once per rotation is generated in the cylinder head portion as in the case of rubbing, but the cross head pin bearing 26 and the crank pin bearing 27 (FIG. 29). In FIG. 30, only a part of the plurality of pistons 23, the cross head pin bearing 26, the crank pin bearing 27, etc. is shown.) Structural effect of absorbing the vibration of the explosion. It is considered that this is due to the influence of the distance attenuation due to the large size, but it was confirmed that the bearing part was not affected by the explosion vibration under normal conditions. That is, as with the diesel generator, it is possible to detect a rubbing abnormality by paying attention to the kefrenity of the rotation cycle.

Subsequently, the diagnostic device 10 for diagnosing the slide bearing (hereinafter, also referred to as the main metal) 1 of the diesel engine 20 will be described below. Reference numeral 23 in FIGS. 31 to 33 indicates a piston of the diesel engine 20, reference numeral 24 indicates a crosshead, and reference numeral 25 indicates a metal cap.

The diagnostic device 10 detects and diagnoses the acceleration of the rotating shaft 2 during vibration by an acceleration sensor 12 externally attached to the surface of the diesel engine 20 (see FIG. 28 and the like). The number of acceleration sensors 12 is not particularly limited, but in the present embodiment, when there are a plurality of main metals 1 of the diesel engine 20, the same number of acceleration sensors 12 as the main metal 1 is used for diagnosis. If it is a general diesel engine, the number of cylinders + 1 main metal 1 is provided. Of course, when the main metal 1 to be diagnosed is a part of all the main metals 1, the diagnosis can be made by using the acceleration sensor 12 which is smaller than the number of all the main metals 1.

The rotary shaft 2 is specifically a crankshaft which is the main shaft of the diesel engine 20 (see FIGS. 29 and 30), but other shafts such as a camshaft may also correspond to the rotary shaft 2.

The installation position of the acceleration sensor 12 is not particularly limited as long as it is on the surface of the diesel engine 20 and can be diagnosed. For example, the acceleration sensor 12 is installed on the surface of a solid member formed on the radial outer side of the main metal 1. If so, it is preferable in that the acceleration of vibration transmitted from the solid member can be detected more accurately. The solid member referred to here refers to various members such as a crankcase of the diesel engine 20, an oil pan, a reinforcing rib 22, and a part of a housing capable of transmitting vibration.

Further, if the acceleration sensor 12 is attached to the surface of the diesel engine 20 at a position closest to the main metal 1, vibration can be detected more accurately at the position closest to the main metal 1 to be diagnosed. ..

As an example, in the present embodiment, each acceleration sensor 12 is attached to the reinforcing rib 22 formed corresponding to the position of each main metal 1 (see FIGS. 29, 30, and 33). The reinforcing rib 22 having a structure having a certain degree of rigidity to ensure strength is also a member that easily transmits vibration. In the present embodiment, the position is on the outer wall 21 of the diesel engine 20 and is on an extension of the line connecting the main metal 1 and the reinforcing rib 22 configured corresponding to the main metal 1. By attaching the acceleration sensor 12, it is less likely to be affected by the signal emitted by the main metal 1 located next to it in the axial direction (see FIG. 33 and the like).

Further, the acceleration sensor 12 in the present embodiment is arranged in the horizontal direction of the main metal 1 so that all the main metals 1 are located at the same height (see FIGS. 29, 31, and 32). Further, each of the acceleration sensors 12 is arranged equidistant from the main metal 1 to be detected (see FIGS. 29 and 30). As a result, by making the vibration propagation distance between each acceleration sensor 12 and each main metal 1 equal, the degree of distance attenuation of the signal (vibration) becomes equal, and the signal from each acceleration sensor 12 can be evaluated equally.

Further, the acceleration sensors 12 in the present embodiment are arranged at equal intervals along the axial direction of the rotating shaft (crankshaft) 2, thereby corresponding to the main metal 1 arranged at equal intervals in the axial direction. (See FIGS. 30 and 32). However, this is only an example, and it goes without saying that the accelerometers 12 do not need to be arranged at equal intervals unless the main metal 1s are arranged at equal intervals.

When arranging the acceleration sensor 12 as described above, it can be arranged on either the right side or the left side of the rotating shaft 2, but more preferably, the peripheral surface of the rotating shaft 2 is arranged from the lower side to the upper side in the vertical direction. It is arranged on the side that rotates toward it (see FIG. 31). For example, if the rotation direction of the rotation shaft 2 is counterclockwise, it is preferable to arrange the acceleration sensor 12 on the right side (see FIG. 34). This will be described in detail as follows.

When the load acting between the rotating shaft 2 and the main metal 1 is a static load, a constant "wedge film pressure" (see FIG. 35) acts, while when the load is a fluctuating load, the axis center. Not only does the "wedge film pressure" change with the swing angular velocity of the shaft, but also the "throttle film pressure" (see FIG. 36) is generated with the velocity in the radial direction of the axis. In this regard, since not only the static load but also the fluctuating load due to the combustion of the cylinder portion greatly acts on the main metal 1 of the diesel engine 20, the peripheral surface of the rotating shaft 2 rotates from the lower side to the upper side in the vertical direction. On the side, the oil film may not be sufficient and there may be a range where metal contact can occur ("Lubrication of internal combustion engine", supervised by Toshio Sakurai, Koshobo, Section 1.1 by Tsuneo Someya, "Fig. 1 and 4 Journal bearing oil film pressure generation" 2 forms ”,“ Fig. 1.24 Example of calculation of axial locus in bearing of 6-cylinder diesel engine 20 ”, etc.). Therefore, it is preferable to arrange the acceleration sensor 12 on the side where the metal contact can occur to form a measurement point.

As described above, according to the diagnostic device 10 in which the acceleration sensor 12 is externally attached to the surface (outer wall 21 or the like) of the diesel engine 20, it is naturally unnecessary to attach the acceleration sensor 12 inside the engine. Therefore, it is possible to save the trouble of disassembling the engine or modifying the engine, which must be performed when the acceleration sensor is installed inside the engine.

An experimental device was manufactured, and an experiment was conducted to verify the above-mentioned diagnostic method for the slide bearing 1. Hereinafter, examples will be described.

In the experimental device, the plain bearing 1 made of the test metal (white metal as a bearing alloy) was used to rotate the rotary shaft 2 at several speeds while the main shaft of the rotary shaft 2 was pivotally supported (the rotary shaft 2 was rotated at several speeds). (See FIG. 7). Although not shown in particular detail, in this embodiment, the vicinity of both ends of the rotating shaft 2 is pivotally supported by support bearings (rolling bearings), and the rotating shaft 2 is rotated by using a motor.

The outer diameter of the main shaft of the rotating shaft 2 was set to 100 [mm]. Further, when one of the gaps formed between the rotating shaft 2 and the inner circumference of the slide bearing (metal casing) 1 is represented as A and the other as B (see FIG. 7 (A)). , The total gap (sum of A and B) was set to 30/100 [mm] (hence, when the gap in the gap A and B is 15/100 [mm], the rotating shaft 2 is located at the center of the slide bearing 1. To position). Further, the slide bearing 1 is horizontally moved in a direction perpendicular to the central axis of the rotating shaft 2 by a moving mechanism using a push bolt (a mechanism that pushes and pushes the tip of the bolt against the moving target) to change the gap B. .. Using such an experimental device, an experiment was conducted with the rotation speed of the rotating shaft 2 set to 1200 [rpm], 1800 [rpm], and 2800 [rpm]. As a result, when rubbing occurs, the acceleration value increases, but in the case of slight rubbing, the difference in acceleration O / A value is only about 0.01 G, so it is difficult to evaluate at the acceleration level. It was considered (see FIG. 8). The unit of acceleration O / A value, G, is a unit of vibration acceleration, and 1G = 9800 mm / s ² = 9.8 m / s ² .

Next, when the rotation speed of the rotating shaft 2 is 1200 [rpm], the gap is 15/100, the gap is 3/100, the gap is 1/100 (a slight rubbing state), and the gap is 0/100. Acceleration spectra were detected for each of the cases (rubbing state) (see FIGS. 9A to 9D). Further, a part of each acceleration spectrum was enlarged to obtain a zooming spectrum (see FIGS. 10A to 10D).

From the results of each of these spectra and the like, the relationship between the gap and various dimensionless symptom parameters indicating the shape change of the acceleration waveform when the rotation speed is 1200 [rpm] was obtained (see FIG. 11). From this result, it was confirmed that the kurtosis, crest factor, and skewness increased due to the occurrence of rubbing, but these changes were small in the case of slight rubbing. Skewness β ₁ , kurtosis β ₂ , crest factor CF (Crest Factor), waveform factor SF (Shaped Factor), fluctuation rate C.V (above, dimensionless sign parameter), and these dimensionless sign parameters Each of the related standard deviations s and k-th order moment μ _k can be obtained by the following mathematical formulas.

[Number 7]
Crest rate CF = maximum value / effective value [Equation 8]
Waveform rate SF = effective value / average value

Here, the present inventors also examined each waveform of the acceleration waveform and the acceleration spectrum when the rotation speed is 1200 [rpm] and the gap is 1/100 [mm] (a slight rubbing state) (see FIG. 12). From this result, it was confirmed that the rise level of the acceleration spectrum was slight in the case of slight rubbing.

Furthermore, the present inventor also examined the waveform of the zooming spectrum at a rotation speed of 1200 [rpm] and a gap of 1/100 [mm] (a slight rubbing state) (see FIG. 13). From this result, it was confirmed that the acceleration spectrum was modulated by the rotation frequency when rubbing occurred.

In addition, the present inventors have a Kepstram waveform, an autocorrelation waveform and an autocorrelation waveform (see FIG. 14), and a rotation speed when the rotation speed is 1200 [rpm] and the gap is 15/100 [mm] (in a non-contact state). When 1200 [rpm], gap 3/100 [mm] (not in contact), Kepstram waveform autocorrelation waveform and intercorrelation waveform (see FIG. 15), rotation speed 1200 [rpm], gap 1/100 [mm ] (Minor rubbing state) waveform of Kepstram Autocorrelation waveform and intercorrelation waveform (see FIG. 16), rotation speed 1200 [rpm], gap 0/100 [mm] (rubbing occurrence state) Waveforms Each of the autocorrelation waveform and the intercorrelation waveform (see FIG. 17) was also examined. Since the rotation speed of the rotation shaft 2 is 1200 [rpm], the rotation frequency is 20 Hz, and therefore the rotation period is 50 [ms]. In the kefrenity value, autocorrelation value and cross-correlation value in the cepstrum waveform under the condition where slight rubbing and rubbing occurred, a peak was confirmed in the portion corresponding to the rotation frequency fr (20 Hz in the case of this embodiment) (Fig. 16, see FIG. 17).

Subsequently, the inventors examined the ratio of the output level at the time of abnormality to the output level at the time of normal at a rotation speed of 1200 [rpm] (see FIG. 18). In both the slight rubbing state and the rubbing (contact) occurrence state, the output ratio (output level at abnormal time / output level at normal time) in the kefrenity level, autocorrelation and cross-correlation is the output ratio (O) due to other parameters. It was confirmed that it was larger than the output ratio of / A value). Similarly, it was confirmed that even when the rotation speed was 2800 [rpm], the output ratios in the kefrenity level, autocorrelation, and cross-correlation were larger than those in other conventional methods (see FIG. 19).

Subsequently, the inventors stated that (A) gap 2/100 (light rubbing state), (B, C) gap 0/100 (rubbing state), and (D) gap 15/100 (non-contact state). The Kefrenity trend and cepstrum were examined for each of the above (see FIG. 20). The cepstrum at the cursor position shown in the kefrenity trends of each state (A) to (D) (see the upper row of each of FIGS. 20 (A) to (D)) is shown in the lower row of FIGS. 20 (A) to 20 (D), respectively. Shown. These change from (A) a slight rubbing state to a state in which the casing is brought closer to the rotating shaft (B, C) to generate a rubbing state, and then the casing is separated from the rotating shaft and (D) not in contact. It is the result of the test. It can be seen that the level of kefrency in the rotation cycle of the rotation shaft changes in correlation with the contact state between the rotation shaft and the casing. From the above, it was confirmed that a slight rubbing state can be detected by monitoring the kefrenity value after the cepstrum calculation. Similar detection results have been obtained in autocorrelation analysis and cross-correlation analysis.

The inventors attempted to apply cepstrum, autocorrelation, and cross-correlation analysis methods in vibration measurement during commissioning of large marine diesel engines.

Generally, in the case of a diesel engine such as a diesel engine, vibrations that are subjected to amplitude modulation due to operation such as opening and closing of an intake valve and an exhaust valve and a combustion explosion are generated even during normal operation (see FIG. 21). Since each of the vibrations at this time is periodically generated in each rotation cycle, the envelope spectrum is generated by the rotation frequency and its higher-order components (see FIG. 22).

On the other hand, in the cepstrum, autocorrelation, and cross-correlation analysis, the diesel engine is not affected by noise during operation, and the peak of the characteristic value corresponding to the rotation cycle is not seen or extremely small under normal conditions (see FIG. 23). .. A slight rubbing occurred in the main bearing (plain bearing) at about 85 rpm when the rotation speed was changed during the test run. In the cepstrum, autocorrelation, and cross-correlation analysis results at this time, it was confirmed that the characteristic value peak exists in the rotation period indicating the occurrence of rubbing (see FIG. 24).

As for the vibration acceleration waveform at this time, a waveform subjected to amplitude modulation is also obtained, and it is difficult to distinguish it from the normal state (see FIG. 25). In the envelope spectrum as well, the occurrence of the rotation frequency and its higher-order components is confirmed as in the normal case, and it is difficult to discriminate the difference between normal and abnormal (see FIG. 26).

From the results of Examples 1 and 2 above, the inventors obtained or confirmed the following findings.
(1) Abnormal / normal in the characteristic value quantified by a predetermined method from a plurality of peak information generated at the rotation frequency interval of the shaft generated on the acceleration spectrum and the rotation speed of the shaft during slight rubbing of the slide bearing. It was confirmed that the output ratio (ratio of the output level in the abnormal state to the output level in the normal state) is larger than the output ratio when analyzed by other parameters. Therefore, it is possible to improve the diagnostic accuracy of the slide bearing 1 by using this characteristic value. Further, by monitoring the characteristic value, it is possible to detect a slight rubbing state, which was difficult or impossible in the past, at an early stage by using only the piezoelectric accelerometer.
(2) Even in a device in which it is difficult to detect a rubbing abnormality due to noise during conventional operation, particularly a diesel engine, it is possible to accurately detect a slight rubbing phenomenon without being affected by noise during operation.

The present invention is suitable for application to the diagnosis of plain bearings in a diesel engine.

1 ... Main metal (plain bearing)
2 ... Rotating shaft 10 ... Diagnostic device 11 ... Rotation speed detection sensor 12 ... Accelerometer 13 ... Monitoring device 14 ... Reporting device 15 ... Arithmetic processing device (arithmetic device)
20 ... Diesel engine 21 ... Outer wall 22 ... Reinforcing ribs (solid members)

Claims (9)

In a device for diagnosing an abnormality caused by a contact phenomenon inside a slide bearing of a rotating shaft of a diesel engine.
The rotating shaft is a crankshaft which is a main shaft.
The diagnostic device is
A rotation speed detection sensor that detects the rotation speed of the rotation shaft, and
An acceleration sensor attached to the surface of the housing of the diesel engine to detect the acceleration of the rotating shaft during vibration, and an acceleration sensor.
Waveform data representing the acceleration of vibration detected by the acceleration sensor is converted into waveform data in the frequency domain by Fourier conversion, and predetermined signal processing in which the waveform data in the frequency domain and the rotation speed of the rotation axis are combined. A plurality of peak information generated at rotation frequency intervals in the waveform data of the frequency domain is extracted, a characteristic value is obtained from the peak information by a calculation device having a predetermined peak value as a characteristic value, and the characteristic value is determined. A monitoring device that monitors whether or not the threshold value has been exceeded, and determines that an abnormality has occurred in the sliding bearing when the characteristic value exceeds the threshold value.
When the monitoring device determines that an abnormality has occurred in the plain bearing, a notification device that outputs the determination result to the outside, and
With
The housing has reinforcing ribs that hold the slide bearings and propagate vibrations generated by the rotation of the rotating shaft.
On the surface of the housing, a region is formed in which the degree of distance attenuation of the vibration is relatively small due to the propagation of the vibration by the reinforcing ribs.
The acceleration sensor is a diagnostic device for a slide bearing of a diesel engine, which is attached to a portion of the surface of the housing where a reinforcing rib is provided between the surface and the slide bearing. The diagnostic device for a slide bearing of a diesel engine according to claim 1, wherein the acceleration sensor has the same number as the plurality of the slide bearings. The diagnostic device for a plain bearing of a diesel engine according to claim 2, wherein each of the accelerometers is arranged equidistant from the plain bearing to be detected. The diagnostic device for a slide bearing of a diesel engine according to any one of claims 1 to 3, wherein the acceleration sensor is attached to the surface of the reinforcing rib formed on the radial outer side of the slide bearing. The diagnostic device for a slide bearing of a diesel engine according to any one of claims 1 to 4, wherein the acceleration sensor is arranged on a side in which the peripheral surface of the rotating shaft rotates from the lower side to the upper side in the vertical direction. .. The diagnosis of the slide bearing of the diesel engine according to any one of claims 1 to 5, wherein the acceleration sensor is attached to the surface of the housing of the diesel engine at a position closest to the slide bearing. apparatus. The diagnostic device for a plain bearing of a diesel engine according to claim 6, wherein the accelerometer is arranged in the horizontal direction of the plain bearing. Wherein in the surface of the housing, the acceleration sensor at a position in front SL sliding bearing and an extension of a line connecting the reinforcing ribs are arranged, the diagnostic device of the sliding bearing of the diesel engine according to claim 7. The diagnostic device for a plain bearing of a diesel engine according to any one of claims 1 to 8, wherein the acceleration sensors are arranged at equal intervals along the axial direction of the rotating shaft.

Images