JPS6216329B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method for monitoring journal bearings in rotating machines, and in particular, to monitor abnormal conditions in multi-span shaft journal bearings applied to large rotating machines such as steam turbines. and related to monitoring methods to prevent accidents.

[Conventional technology]

As is well known, large rotating machines such as steam turbines are constructed with multi-span shaft systems, and are supported by a plurality of bearings. By the way, steam turbines and generator rotors have a total weight of over 200T in large machines, and are rotating bodies at high speed (3000 to 3600RPM), so it is desirable that the bearings that support them be strong and always maintained in normal condition. Needless to say, numerous experiences have shown that burnout of the bearing metal, excessive or insufficient bearing load, etc. are direct causes that induce excessive vibration and rubbing of the rotor, which can lead to rotor scattering accidents. There is.

The main causes of this bearing load variation are as follows: (i) The bearing pedestal is exposed to heat due to the influence of ambient temperature (for example, heat radiation from the high-pressure turbine side, heat transfer, or influence from nearby steam piping). transform.

(ii) The bearing stand is displaced or tilted in the vertical direction due to pressure changes within the turbine casing.

(iii) The turbine pedestal will settle unevenly due to aging.

(iv) Bearing metal wears out.

etc. are possible. In either case, this phenomenon occurs because the turbines are supported by a large number of bearings.In particular, in recent large steam turbines, the upper and lower limits of bearing loads are close to each other, so the accident rate due to load fluctuations is high. expensive. Conventional measures to deal with abnormal conditions caused by these factors are (a) Careful alignment when connecting the rotors.

(b) Measure alignment changes during regular inspections.

(c) Bearing inspection (oil gap, bearing parallelism,
Check for abnormalities and make adjustments as necessary.

(d) Continuously monitor the bearing oil supply/drainage temperature and backing metal temperature.

(e) Monitor oil supply pressure and issue an alarm or trip the turbine in the event of an abnormality.

etc.

[Problem that the invention seeks to solve]

However, since these methods do not directly monitor changes in bearing load, it has been impossible to monitor changes during operation.

On the other hand, in recent large-capacity turbines, unstable vibrations such as oil whips have become a problem due to a decrease in shaft rigidity. Increasing the surface pressure of the bearing has been adopted as a measure to prevent this oil whip, but the high surface pressure inevitably leads to deterioration of the lubrication characteristics during low-speed operation, which often leads to bearing burnout. Particularly in steam turbines that perform turning operations for a long period of one week or more, the lubrication condition during low-speed operation must be managed to prevent the metal wipe phenomenon (hereinafter referred to as wipe) that can lead to metal burnout. It is important to detect wipes early and prevent them from occurring, but as mentioned above, the high surface pressure makes it difficult to manage and monitoring methods are insufficient, so the risk of metal burnout accidents cannot be avoided at all costs. I didn't get it.

Furthermore, as described above, since the axis changes due to changes in alignment, the outer peripheral surface of the bearing is machined into a spherical surface so that the bearing follows the axis, thereby forming an alignment mechanism. By the way, this alignment structure is effective against inclination when stationary, but it does not necessarily work effectively when an oil film is formed, and therefore the relative inclination of this shaft and bearing ( The risk of a bearing burnout accident due to uneven contact (hereinafter referred to as uneven contact) was not avoided in any way.

In addition, for journal bearings with multi-span shafts, the load distribution is set in advance to be approximately equal between each bearing, but due to deformation of the bearing stand or uneven settlement of the pedestal, the load distribution is unbalanced. Eventually, there was a problem that the load exceeded the allowable load.

The present invention has been made in view of the above points, and provides a bearing monitoring method that monitors the amount of bearing inclination, metal wipe, and bearing load during bearing operation, detects abnormalities at an early stage, and prevents accidents. It is about providing.

[Means for solving problems]

The above problem is solved by detecting the pressure of the lubricating oil film formed on the sliding surface of the journal bearing from multiple locations on the sliding surface, calculating the bearing tilt amount and bearing load from the detection signals, and then calculating the bearing metal temperature. This is solved by calculating the metal wipe from the difference between the temperature and the oil supply temperature.

[Effect]

According to the above method, the oil film pressure, bearing metal temperature, and oil supply temperature can be detected during bearing operation, and the amount of tilt of the bearing, bearing load, and metal wipe can be calculated from these detected values, so that abnormal conditions of the bearing can be detected. can be monitored.

〔Example〕

Hereinafter, one embodiment of the present invention will be explained in detail with reference to the drawings.

In FIG. 1, a bearing 2 that supports a journal 1 has a spherical seat 3 on its outer peripheral surface, and has a structure in which alignment can be adjusted using an adjustment ring 4. In a multi-span shaft system, three or more such bearings are arranged in the axial direction. Further, a Babbitt metal 5 is provided on the bearing sliding surface 9 of each bearing.
Pressure measurement holes 10 with a diameter of about 2 to 5 mm are bored in this bearing sliding surface 9 at at least four points, two in the axial direction for each bearing, and two in the axial direction, and communicate with the pressure measurement holes 10. A pressure propagation hole 11 is provided in the bearing end face 8, a pressure sensor 12 is installed on the bearing end face 8, and a temperature sensing element 7 such as a thermocouple is buried in the Babbitt metal near the pressure measurement hole. Furthermore, one or more acoustic sensors 13 are installed on the bearing end face 8. Next, a block diagram from the measuring element to the bearing diagnostic device 18 is shown in FIG. That is, pressure sensor 12, temperature sensor 7
The oil film pressure signal and metal temperature signal detected by the processor 18 are input to the signal input device (hereinafter referred to as PI/O) 17, and the acoustic signal detected by the acoustic sensor 13 is input to the preamplifier 19 and the main amplifier 2.
0 and counter 21 to the PI/O 17. On the other hand, operating conditions such as rotation speed and turbine load 23
(It is always measured in normal thermal power plants.)
The bearing oil supply temperature 24 is also input to the PI/O 17 in a separate system.

Next, FIG. 4 is a flowchart showing an overview of diagnosis by the diagnostic device 18. The diagnosis program contains bearing specifications in advance, such as bearing type, journal diameter,
Design standard values such as effective width, design load, oil supply and discharge temperature, etc. are memorized. This is shown in step 51. Next, when the diagnostic routine starts, data acquisition is performed in step 52. As for the data, the detection signals and operation data from each sensor 7, 13, 23, 24 are inputted via the PI/O 17 as described above, and the diagnostic routine program ends the data acquisition. Next, the process moves to step 53. That is, for metal wipes, it is sufficient to focus only on the low journal rotational speed range, and it is sufficient to diagnose load fluctuations and uneven contact in the medium to high speed range. Therefore, in step 53, the journal rotation speed is determined, and if it is in a low speed range, the process moves to step 54, a metal wipe diagnosis routine. After repeating the diagnosis at set time intervals in step 54, the process moves to a predictive diagnosis routine in step 55, where future metal wipes are predicted based on changes in the diagnostic routine over time, and the results are displayed in step 60. In addition, when the journal rotation speed is determined to be in the medium-high speed range in step 53, it is possible to
The uneven contact state of the bearing is determined by the determination routine 6. The determination of uneven contact will be described in detail later, but it is determined based on the magnitude relationship of the axial pressure difference. If it is determined that there is a one-sided hit in the determination routine in step 56, the process moves to the one-sided hit diagnosis routine in step 57,
Diagnose the amount of tilt of the bearing. After this uneven contact diagnosis routine is completed, and if uneven contact is not determined in step 56, the process moves to the load diagnosis routine in step 58 to diagnose the bearing load. After repeating the diagnostic routines in steps 57 and 58 at set time intervals, the abnormality prediction routine in step 59 predicts bearing abnormalities due to load fluctuations and tilt amounts over time, and the results are used to issue an alarm or alarm in step 60. display,
At the same time, data is recorded to predict lifespan based on changes over time.

The outline of the diagnosis within the diagnostic device 18 has been described above, and each diagnosis and prediction routine shown in double boxes in FIG. 4 will be described in more detail below along with specific data.

FIGS. 5 to 10 show the results of an experiment in which the metal wipe phenomenon was reproduced using a bearing equipped with the above-mentioned sensor. FIG. 5 shows the relationship between metal temperature and waste oil temperature depending on the presence or absence of metal wipe, with time on the horizontal axis and temperature on the vertical axis.
The solid line in the figure shows the metal temperature, and the broken line shows the drain oil temperature. The journal was kept at a constant rotation, metal wipe was generated under various conditions, and this was confirmed using a torque meter. "Metal wipe occurrence" in the diagram
indicates this. As can be seen from FIG. 5, when a metal wipe occurs, the metal temperature noticeably appears. That is, the temperature rises rapidly and exceeds a certain standard. On the other hand, the temperature of the drain oil does not change clearly even if metal wipe occurs.

Next, Fig. 6 shows the measurement results of the acoustic sensor that were measured at the same time. As in FIG. 5, the horizontal axis shows time, and the vertical axis shows the intensity (energy value) of the acoustic signal, the frequency of occurrence (counts/second) of acoustic signals exceeding a certain standard, and the integrated value (count). In the figure, the solid line shows the energy value, the bar graph shows the frequency of occurrence, and the broken line shows the cumulative count number. As in the case of Figure 5, when a metal wipe occurs, the energy value increases, and the frequency of occurrence increases.
The cumulative count also increases rapidly, indicating that the metal wipe is progressing. FIGS. 7 and 8 show acoustic signal waveforms, with FIG. 7 showing the waveform at normal times and FIG. 8 showing the waveforms when a wipe occurs. It can thus be seen that there is a clear difference in the acoustic signal strength.

On the other hand, Figure 9 shows the area where metal wipes occur. The horizontal axis shows the bearing average surface pressure (Kgf/cm ² ), the vertical axis shows the friction coefficient, and the average surface pressure was changed using oil temperature as a parameter. Due to the change in the coefficient of friction when
This is the result of examining the critical surface pressure for wipe generation. The solid line on the right in the diagram is the oil supply temperature of 30℃, the solid line in the center is 40℃,
The solid line on the left indicates 50°C. Furthermore, each symbol filled in, that is, the shaded area indicates that a metal wipe has occurred. Therefore, it can be seen from this figure that even if the average surface pressure is constant, it is possible to prevent the occurrence of metal wipe by controlling the oil supply temperature.

Incidentally, the oil supply temperature is maintained at approximately 32°C at startup, but it is not always possible to maintain the oil supply temperature at 32°C when the speed is lowered than the rated speed.

FIG. 10 shows an example of this. In other words, during rated operation, the oil supply temperature is kept constant at 45 to 46 degrees Celsius, and when the speed drops below the rated speed, the amount of water in the oil cooler is increased to lower the oil temperature, but if the speed slows down too quickly. As shown by the broken line, the engine reaches a low speed range before the oil temperature is sufficiently lowered, and enters the metal wipe generation area shown by diagonal lines, where metal wipe may occur. Therefore, it is necessary to monitor the oil supply temperature in each rotation range during speed reduction.

FIG. 11 is a block diagram showing details of the metal wipe diagnostic routine shown in steps 54 and 55 of FIG. In the figure, 101 is a rotation speed determination device, which determines whether the rotation speed R of the journal shown at 103 is faster or slower than the reference value R ₀ =800 rpm shown at 112, and when RR _{is 0} , the rotation speed from each detector is determined. The detection signal is input into the arithmetic unit. Therefore, in the medium and high speed range where R>R ₀ , 103
The rotation speed R shown by is input to the differentiator 107, and then the changeover switch 135, 13
The signal is inputted to a sign/negative determiner 132 that operates 6. Further, the rotation speed R of 103 is inputted to the reference oil supply temperature calculator 108 via the changeover switch 135 by operating the positive/negative determiner 132. 1
The bearing oil supply temperature T _O indicated by 02 is determined by the positive/negative determiner 132.
The subtracter 133 is entered via the changeover switch 136 operated by the subtractor 133. Then, a fuel oil temperature signal T corresponding to the rotation speed R is output from the reference oil temperature calculator 108 mentioned above.
_C is simultaneously input to the subtracter 133, and a difference signal between the two is input to the comparator 109. The rotation speed R is input to the differentiator 140 and the allowable temperature setter 134, and the allowable temperature signal T _A is input to the comparator 9 and compared with the difference signal, and the result is judged by the judge 110 and the display/alarm device 128.

Further, the bearing oil supply temperature T _O of 102 is calculated from the calculation reference value T _L from the oil supply temperature setting device 129 and the comparator 13.
0, and if T _O exceeds the reference value T _L , the display alarm device 128
It is designed to transmit an output signal to that effect.

On the other hand, in the rotation speed determiner 101, when the rotation speed is _RR0 , the contacts 137 to 139 are closed. Therefore, the metal temperature signals T ₁ to T ₄ input to the four bearing metal temperature sensors 7 to 104 are as follows:
After inputting the bearing oil supply temperature signal T _O to the subtracter 111 and calculating the temperature deviation signal ε _1-4 =T _1-4 −T _O , the signal ε _{1 −4} is input to the temperature setting device 114.
The comparator 113 compares the reference temperature ε ₀ with the reference temperature ε 0 from the . When any of these deviation signals ε _{1 - 4} exceeds the reference temperature ε ₀ , the wipe determiner 15 determines that a bearing wipe has occurred and inputs the result to the display/alarm device 128 . Furthermore, the interpolation signals ε _1-4 from the subtractor 111 are also branched and input to the differentiator 116, and a wipe prediction is performed in the wipe prediction determination unit 117 based on the tendency of increase or decrease of the maximum interpolation signal ε _MAX . . That is, if the differential value dε _MAX /dt tends to increase, occurrence of bearing metal wipe is predicted.

On the other hand, the signal from the acoustic sensor 16 is 105
At 106, the acoustic signal energy value γ is detected, and at 106, the acoustic signal occurrence probability γ _o is detected. When RR is ₀ in the rotation speed determiner 101, the contacts 138 and 139 are closed, so the signal γ is compared with the reference signal γ ₀ from the acoustic energy setting device 120 in the comparator 118, and when RR is ₀ , the signal γ is wiped. The determiner 119 determines that it is a metal wipe and transmits it to the display/warning device 128. Similarly, signal γ
The comparator 121 compares _o with the reference signal n from the acoustic signal generation frequency setter 122, and if γ _o n, the wipe determiner 123 determines that it is a metal wipe and transmits it to the display/warning device 128. Furthermore, even if these signals are γγ ₀ or γ _o <n, the signals γ and γ _o are differentiators 124 and 1, respectively.
26, the rate of change dγ/dt and dγ _o /dt are calculated, and depending on the increase/decrease tendency, for example, if these are in an increasing trend, the wipe prediction determination unit 125, 127
The wipe prediction of the bearing metal is performed and the prediction is transmitted to the display/warning device 128.

The above is the configuration of the bearing abnormality diagnosis device regarding metal wipes. Next, the operation of the above-mentioned diagnostic device will be explained using a flowchart. 11th
12 and 13, first, the rotation speed determiner 101 compares the turbine rotation speed R with the reference rotation speed R ₀ , that is, determines whether it is a low speed region or a high speed region, and R<R _{0 .} (low speed range), metal temperature
T ₁₄ is input to the subtracter 111, and 61 in FIG.
Then, the temperature deviation signal ε ₁₄ =T ₁₄ −T ₀ is calculated as shown in FIG. Next, the comparator 113 compares the temperature with the reference temperature ε ₀ , and when any of ε ₁₄ becomes ε ₁₄ >ε ₀ , the wipe determiner 115 determines that a metal wipe has occurred. are shown in 62 and 63 of the flowchart. In addition, when the rotation speed R is R>R ₀ , the acoustic signal energy value γ
and acoustic signal occurrence frequency γ _o are each calculated by the comparator 11.
8,121 and compared with reference values γ ₀ and n. As shown at 65 in the flowchart of FIG. 12, if the condition γ>γ ₀ or γ _o n is satisfied, the wipe determining devices 119 and 123 determine that the wipe is a metal wipe.

Next, the prediction determination of metal wipe will be explained. Temperature deviation signal ε ₁₄ of metal temperature is reference temperature ε
Even if it is less than ₀ , that is, ε ₁₄ < ε ₀ ,
It predicts the possibility of metal wipes occurring in the future. That is, as shown in 67 of the flowchart among the outputs T ₁₄ of the subtractor 111, the maximum value ε _MAX of the temperature deviation is input to the differentiator 116 to calculate the increase/decrease trend of the temperature deviation Δε _MAX = dε _MAX /dt, and wipe The prediction judge 117 judges the future metal wipe when Δε _MAX > 0. Therefore, even if the metal temperature deviation is ε ₁₄ < ε ₀ at the present moment, Δε _MAX > 0 as shown in 69 of the flowchart. In this case, the prediction of metal wipe occurrence will be determined at a future point in time.

Similarly, the acoustic signal energy value γ and the acoustic signal occurrence frequency γ _o are currently γ<γ ₀ and γ _o <
Also in the case of n, the signals γ and γ _o are input to the differentiators 124 and 1 in order to predict the occurrence of metal wipe in the conventional manner.
26 and the increase/decrease trend of the signal value as shown in 68 and 69 of the flowchart, that is, Δγ=d
γ/dt, dγ _o =dγ _o /dt is calculated, and when Δγ > 0 or Δγ _o > 0, the wipe prediction determination unit 125,
127, the occurrence of metal wipe is to be determined at a future point in time.

Furthermore, in the bearing abnormality diagnosis device, if the acoustic signal energy value γ and the acoustic signal occurrence frequency γ _o are γγ ₀ or γ _o n at the current time, the wipe determination device 11
Even if it is determined to be a metal wipe in 9,123, it is possible to predict whether metal wipes are on the verge of disappearing at some point in the future. That is, in this case, when Δγ and Δγ _o are Δγ < 0 and Δγ _o < 0 in the calculation results of the differentiators 124 and 126, the wipe prediction determination unit 125 ,1
At step 27, it is possible to determine whether the metal wipe has disappeared.

As is clear from the above description, it is possible to diagnose bearing abnormalities such as metal wipes from the metal temperature of the bearing sliding surface, bearing oil supply temperature, and acoustic signals.

Next, the details of the one-sided hit diagnosis routine shown as process 57 in FIG. 4 will be explained. First, as shown in Fig. 14, when the journal 1 and the lower half bearing 2 are in a state of uneven contact, if the measured values of the axial pressure of the oil film are p ₁ and p ₂ and the pressure difference between them is Δp, then the uneven contact occurs. As the amount Δh increases, the pressure difference Δp also increases.
Therefore, by monitoring this pressure difference Δp, it is possible to know the tilt state of the bearing.

Figure 15 shows this in more detail with a bearing diameter of 406 mm.
mm, and the oil film pressure distribution in a partial contact state at a rotational speed of 3000 rpm, it is clearly seen that the oil film pressure distribution changes as the bearing partial contact amount Δh increases. Further, FIG. 16 shows the relationship between the dimensionless pressure difference Δp and the dimensionless inclination amount Δ in FIG. 15. From the figure, the relationship between Δ and Δ can be expressed as the following relational expression: Δ={a ₁ Δp+a ₂ Δp ² +a ₃ Δp ³ +a ₄ Δp ⁴ }N〓(L/D)〓 (1) Therefore, by measuring the pressure difference Δp, it is possible to determine the uneven contact state of the bearing. By the way, Figure 15 shows the pressure distribution when the shaft hits one side in the vertical direction.
In reality, uneven contact does not necessarily occur only in the vertical direction; for example, uneven contact in the horizontal direction may occur due to miscoupling, or uneven contact may occur in combination in the horizontal and vertical directions. Therefore, in order to determine the direction of uneven contact, it is necessary to measure the oil film pressure at at least four points.

As mentioned above, it is possible to detect the oil film pressure using at least four pressure sensors 12 and use the pressure difference Δp to accurately determine the uneven contact state of the bearing. Determining abnormalities cannot be said to be complete. In other words, in large rotating machines such as steam turbines,
It is no exaggeration to say that slight imbalances like the one mentioned above always occur. However, even if uneven contact occurs, it cannot be said that there is an abnormality as long as the bearing is sliding without any problem. On the other hand, most bearing burnout accidents are caused by white metal melting as the temperature rises. In other words, the softening temperature of white metal is approximately 130°C, but to give an example of a bearing that caused uneven contact, when a shaft diameter of 483 mm and a total width of 250 mm were rotated at 2000 rpm, an inclination of Δh = 14% occurred. The maximum metal temperature exceeds 100℃, which shows that even a slight uneven contact has a large effect on bearing burnout. Therefore, measuring oil film pressure alone is not sufficient; by combining this with bearing metal temperature measurement, it becomes possible to monitor bearing abnormalities more reliably. Furthermore, by installing the temperature sensing element 7 near the bearing sliding surface 9, a monitoring device with short detection time and good responsiveness can be obtained.
Since the temperature sensing element 7 of this embodiment is installed at the same position as the oil film pressure measurement hole 10, by measuring the temperature difference ΔT of the bearing metal as shown in FIG. It is also possible to estimate the amount of partial contact as well as the direction. Therefore, by detecting the oil film pressure and the bearing metal temperature, it is possible to more accurately grasp the bearing condition. Additionally, the pressure sensor 12 and the temperature sensing element 7 are normally installed near the bottom of the lower half bearing 2 due to the thickness of the oil film, but if damage occurs to the upper half bearing due to alignment changes due to aging of the pedestal, etc. Taking this into consideration, by adding the acoustic detection element 13 that detects the normal sound caused by the contact between the journal 1 and the bearing to the above-mentioned measurement elements, it becomes possible to completely monitor the sliding abnormality of the bearing.

Next, referring to FIG. 18, a bearing abnormality diagnosing device relating to uneven contact diagnosis will be explained in detail using a block diagram. First, the oil film pressure calculation circuit will be described. Signals from the four oil film pressure measurement elements 12 are converted into oil film pressure values p ₁ to _{p 4} by the pressure converter 50 and then input to the calculation unit 51. Arithmetic unit 51
The axial pressure difference of the oil film pressure signal at Δp ₁ = p ₁ −
Calculate p ₂ , Δp ₂ = p ₃ −p ₄ . The obtained pressure differences Δp ₁ and Δp ₂ are input to the calculator 52, and the fourteenth
As shown in the figure, the amount of inclination Δh is calculated based on the equation (1) shown earlier than the relationship between the pressure difference Δp and the amount of inclination Δh. A setter 53 is provided for selecting a reference value h ₀ according to operating conditions for the amount of inclination Δh, and a comparator 5 determines the magnitude relationship between the reference value h ₀ and the calculated Δh.
It is compared by 4. And this result is judger 5
5, and when Δh>h ₀ , an abnormal state of uneven contact in the bearing is determined. The result determined by the determiner 55 is displayed by the display/alarm device 44, and if it is determined to be abnormal, it is input to the metal temperature determination circuit and compared with the metal temperature result.

Next, the metal temperature calculation circuit will be explained. Metal temperature signal value T ₁ detected from four thermocouples 7
~T ₄ is compared with the signal value T ₀ of the bearing oil supply temperature shown at 24 in the calculator 32, and the temperature rise value ε ₁₄ = T ₁₄ −T ₀
Calculate. The temperature increase value ε ₁₄ is compared by a comparator 34 with a reference value signal ε _T set by a setter 33 in accordance with the rotation speed R, etc., which is an operating condition value of the turbine. Then, the comparison result of the comparator 34 is determined by the determiner 35 to be abnormal when ε ₁₄ >ε _T , and is displayed on the display/alarm device 44 . Further, for the metal temperature signal values T _{1 to 4} , the axial temperature difference ΔT ₁ =T ₁ −T ₂ , ΔT ₂ =T ₃ −T ₄ is calculated by the calculator 27, and the temperature difference T ₁₂ from the setting device 45 is calculated. The comparator 58 compares the magnitude relationship with the reference value β ₀ , and the determiner 59
is input. The determiner 59 determines whether or not there is an abnormality with respect to the temperature difference ΔT ₁ , ₂ , and the determiner 55 of the pressure determination circuit described above
Compare with the results from . In other words, if the pressure judgment circuit determines that there is an abnormality, and the metal temperature is determined to be normal or has not reached an abnormal state, it is determined that the sensor is abnormal, and if the metal temperature result is also abnormal. is defined as the degree of abnormality and is displayed on the display/alarm device 44.

In this way, by comparing the pressure measurement results and the metal temperature measurement results and determining an abnormality, it is possible to determine whether the measurement element is malfunctioning.

Furthermore, the results determined to be abnormal by the metal temperature determination circuit are input to the acoustic signal circuit and compared with the results from the acoustic signal.

Finally, the acoustic signal circuit will be explained.As shown in FIG. 18, the detection signal from the acoustic detection element 13 consists of an acoustic energy signal value 48 and a signal value 49 indicating the frequency of acoustic signal generation.
A setting device 3 sets reference values γ ₀ and n for the energy value 48 and occurrence frequency 49, respectively.
8 and 42 are installed respectively, and the comparator 37
The comparator 41 compares the magnitude relationship of the acoustic signals, and the results are sent to the determiner 39 and the determiner 43.
is input. The determiners 39 and 40 determine an abnormality based on the acoustic signals γ and γ _o , and further determine the magnitude of the abnormality based on the result from the metal temperature determination circuit.

In other words, if the metal temperature judgment circuit determines that there is an abnormality and no abnormality is recognized in the acoustic signal value, it is determined that the degree of abnormality is determined, that is, the shaft and bearing are tilted significantly but are not in contact yet. . If an abnormality is also found in the acoustic signal circuit, the degree of abnormality is determined, that is, the shaft and bearing are in complete contact and the metal is burnt out.

In this way, by combining the three measurement elements of oil film pressure, metal temperature, and acoustic signals, we can diagnose the uneven contact state of the bearing and its size.
Abnormalities in bearings can be accurately monitored.

19 to 21 show the determination flow of the pressure determination circuit, metal temperature determination circuit, and acoustic signal circuit. That is, in FIG. 19, the axial pressure differences Δp ₁ =p _{1 -p 2 and} Δp ₂ =p ₃ -p ₄ are calculated in step 150 from the measured oil film pressure values p ₁ to p 4 . Here, the pressure values at two points Δp ₁ and Δp ₂ are calculated in order to determine the direction of inclination.

The pressure differences Δp ₁ and Δp ₂ are then compared with reference values ε _P1 and ε _P2 in step 151, and the reference values ε _P1 , ε _{P2 are} determined.
If it is smaller, it is determined that there is no abnormality and the display system 155 displays it. If the pressure difference is larger than the reference value, then in step 152 the inclination amount Δh and inclination direction are calculated from equation (1), and Δh is compared with the reference value Δh ₀ in step 153. do.
If the amount of inclination Δh is smaller than the reference value Δh ₀ , it is not abnormal, but it is displayed that the bearing is tilted.
If Δh is larger than the reference value, it is determined to be abnormal and an alarm is issued at step 154, and the determination result is transmitted to the metal temperature determination circuit at step 156. This determination result is also displayed on the 155 display system.

Next, FIG. 20 shows the determination flow of the metal temperature determination circuit h. The metal temperatures T ₁ to _{T 4} are first compared with the oil supply temperature signal T ₀ in step 160, and the temperature rise values ε _{1 to 4} = T _{1 to 4} − T ₀ is calculated. This temperature rise value ε _{1 to 4} is the reference value ε _T and process 1
61, and even one of them is ε
If it is larger than _T , an abnormality is displayed at 162. Next, in step 163, temperature differences _{ΔT 1 =} T ₁ −T ₂ and ΔT ₂ =T ₃ −T ₄ are calculated from the metal temperatures T 1 to T ₄ . And this temperature difference ΔT ₁ and ΔT ₂ is the process 164
are compared with the reference value β ₀ for each bearing piece. If the reference value β is smaller than ₀ , it is determined that there is no abnormality, and this is transmitted to the display system 155. In addition, if it is large, it is input as an abnormality to the next step 165, but in this step 165 it is compared with the judgment result of the oil film pressure judgment circuit, and if both judgment results are abnormal, the abnormality level is determined in step 166. 155 display systems. Also, if the two judgment results are different, 16
7, it is determined that the sensor is abnormal and it is displayed on the display system 155. Further, in the case of an abnormality, the signal is transmitted to the acoustic signal circuit as shown in step 168. Further, FIG. 21 shows the determination flow of the acoustic signal circuit, in which the acoustic signal is converted into the acoustic energy value γ and the frequency of occurrence γ _o in step 170, and then
71, compared with the respective reference values γ ₀ and n,
If it is larger than the reference value, such as γ>γ ₀ or γ _o >n, it is also compared in step 172 together with the metal temperature determination result. If it matches the metal temperature judgment result, an alarm will be issued as an abnormality level of 175, and if it does not match, it will be displayed on the display system 155 as an abnormality level of 173 or a sensor abnormality of 174. There is.

As is clear from the above, according to this embodiment, it is possible to determine the amount of tilt of the bearing and abnormal conditions of the bearing, such as uneven contact of the bearing, from the oil film pressure of the bearing. Further, it is possible to determine whether the bearing is in an abnormal state based on the temperature difference between the metal temperature of the bearing and the oil supply temperature, or based on the temperature difference between metal temperatures at multiple points. Furthermore, abnormal conditions such as uneven bearing contact can be determined from the acoustic signal of the bearing.

Next, the contents of the bearing load diagnosis routine will be explained.
First, the bearing load calculation method and the calibration of the standard calculation value will be described in detail. Figure 22 shows the relationship between oil film pressure P and bearing average surface pressure P _n , where shaft diameter D =
254φmm, bearing width L=152mm, C=0.66mm, L/D
= 0.6 bearing. That is, it can be seen that the average surface pressure P _n is expressed by P _n =KP 〓 (2) K: proportionality constant. (Although α varies somewhat depending on the measurement position, it is between 0.8 and 0.85.) Further, the relationship between the rotational speed N and the above-mentioned K is shown in FIG. The proportionality constant K has a relationship of K∝N ^0.17 , and ⁱⁿ the end, P _n becomes P _n = ^{K'N 0.17} P ^{0. 8} (3). Furthermore, the relationship between L/D and P _n is P _n ∝(L/
^{D ) From the} _fact ^{that it has} been proven ^that The values vary somewhat depending on the bearing shape and measurement position, so it is necessary to memorize the set values for each bearing.Also, Figures 22 and 23 show the values under a constant oil supply temperature of 46°C, but changes in the oil supply temperature ^{Considering the influence} _of ^{viscosity due to} _{_} ^{_ _} Calculated. By the way, if the bearing is always in a normal state, the bearing average surface pressure can be calculated using equation (5) by measuring the oil film pressure at one point, but if the bearing is tilted as mentioned above, it cannot be calculated at one point. It is possible.

The oil film pressure in the state of uneven contact is quite different from that in the normal state as shown in Figure 15, but as can be seen in Figure 15, if you pay attention to the oil film pressure at a position L/4 from both end faces of the bearing, for any amount of inclination. Also in P ₁ ,
It has been found that the average value of P ₂ almost matches the normal P ₁ (=P ₂ ). Therefore, even in the case of uneven contact, the bearing load can be calculated by substituting P=(P ₁ +P ₂ )/2 into equation (5).

On the other hand, in a shaft system supported by multiple bearings such as a steam turbine, the flying height of the shaft differs depending on adjacent bearings. Alternatively, a horizontal force F _H may act on the bearing (hereinafter referred to as horizontal force F _H ) due to miscoupling or the like. Figure 24 shows the oil film pressure distribution when this horizontal force F _H is applied.
Depending on the direction of the horizontal force, the oil film pressure increases or decreases compared to normal conditions, making it impossible to calculate the load using equation (5). Therefore, two or more oil film pressure measurement points P ₃ and P ₄ are provided in the rotational direction, and the horizontal force is calculated from the oil film pressure ratio ₁ = P ₁ /P ₃ , ₂ = P ₂ /P ₄ in each operating state. do. The horizontal force F _H is expressed as F _H =K ₀ {log( ₁ + ₂ )}(L/D) ¹ N ² (6). On the other hand, if the magnitude of the horizontal force F _H is known, the magnitude of the horizontal force F _H and the oil film pressure directly below is the second
As shown in Figure 5, there is an almost proportional relationship, so correct the oil film pressure to the normal oil film pressure when no horizontal force is acting, and calculate the charge using equation (5). Furthermore, the 25th
In the figure, D=254 mm, L/D=0.9, and 3000 rpm.

Therefore, in order to calculate the bearing load under all operating conditions of the bearing, oil film pressures at at least four points are required.

FIG. 26 shows a block diagram of the load diagnosis routine described above. The oil film pressure signal 220 detected by the pressure sensor 12 is sent to a comparator 300 using the result of the uneven hit diagnosis routine 57 shown in FIG.
If it is diagnosed in the diagnostic routine 57 that there is an uneven hit, pressure correction is performed by the corrector 301. Next, the calculation unit 302 calculates _the horizontal force F _H using equation (6). Then, this F _H is set to a certain reference value 303 and the comparator 3
04, and if _FH is large, the pressure corrector 305 corrects the oil film pressure value to the normal oil film pressure value. Thereafter, the bearing load is calculated by the calculator 306 using equation (5), and it is diagnosed whether the load is within the allowable value 307. If the load exceeds the allowable value, the
Displays abnormalities. Further, these results are recorded in the data filing device 308 to predict aging and aging.

It should be noted that the present invention is not limited to the above-mentioned embodiments, and any other rotating machine may be used as long as it is a bearing that is likely to cause the above-mentioned abnormal condition.Moreover, the greater the number of measurement points, the more accurate the monitoring can be, so the number of bearings that can be measured may be selected as appropriate. Of course, various modifications can be made depending on the purpose.

〔Effect of the invention〕

As described above, the present invention has the effect of realizing a bearing monitoring method that can prevent accidents by monitoring a wide range of bearing operating conditions and diagnosing abnormal conditions. .

[Brief explanation of the drawing]

FIG. 1 is a cross-sectional view of a journal bearing to which a bearing monitoring device according to an embodiment of the present invention is installed, FIG. 2 is a partial cross-sectional view showing the vicinity of the bearing sliding surface in FIG. 1, and FIG.
The figure is a block diagram of a bearing monitoring device that is an embodiment of the present invention, FIG. 4 is a flowchart showing the diagnostic content of the bearing monitoring device shown in FIG. 3, and FIG. 5 is a metal temperature status diagram during metal wiping. , Figure 6 is a diagram of the acoustic signal generation situation during metal wipe, Figures 7 and 8
The figure is a diagram of acoustic signal fluctuations, Figure 9 is a diagram of the relationship between bearing average surface pressure and friction coefficient during metal wipe, and Figure 1
Figure 0 is a diagram of the bearing oil supply temperature state during metal wipe, Figure 11 is a block diagram of metal wipe diagnosis which is part of the bearing monitoring device which is an embodiment of the present invention.
Figures 2 and 13 are flowcharts showing the contents of the metal wipe diagnosis in Figure 11, Figure 14 is an explanatory diagram of the bearing uneven contact state, Figure 15 is a diagram of the oil film pressure situation during uneven contact, and Figure 16. 17 is a diagram showing the relationship between the oil film pressure difference and the amount of inclination, FIG. 17 is a diagram showing the relationship between the oil film thickness and metal temperature when the bearing is tilted, and FIG. 18 is a part of a bearing monitoring device that is an embodiment of the present invention. 19 to 21 are flowcharts showing the contents of the uneven contact diagnosis shown in FIG. 18. FIG. Figure 23 is a diagram of the relationship between the journal rotational speed and the proportionality constant K at the average surface pressure P _n = KP, the second
Figure 4 is a bearing oil film pressure distribution diagram under the action of horizontal force, Figure 25 is a diagram showing the relationship between horizontal force and oil film pressure directly below, and Figure 26 is a part of a bearing monitoring device that is an embodiment of the present invention. FIG. 3 is a block diagram of bearing load diagnosis. 1... Journal, 2... Bearing, 7... Temperature sensor, 12... Pressure sensor, 13... Acoustic sensor, 23... Operating conditions, 24... Bearing oil supply temperature, 54... Metal wipe diagnosis, 57... ...Uneven contact diagnosis, 58... Bearing load diagnosis, 101... Rotation speed determiner, 110, 115, 117, 119, 12
3,131...determiner, 109,118,12
1,130...Comparator, 117,125,127
...Wipe prediction determiner, 108, 134... Arithmetic unit, 51, 52... Arithmetic unit, 34, 54... Comparator, 35, 55... Judgment device, 57... Arithmetic unit, 5
8... Setting device, 59... Judgment device, 300... Judgment device, 301, 305... Corrector, 302, 306
...Arithmetic unit.

Claims (1)

[Claims] 1. The oil film pressure of lubricating oil is measured from multiple points on the sliding surface of a journal bearing, and the temperature of the bearing metal near the sliding surface and the oil supply temperature to the bearing are measured, and the oil film pressure at these multiple points is measured. Calculate the mutual pressure difference from the pressure signal, calculate the amount of inclination of the bearing based on the pressure difference, monitor the bearing piece contact, calculate the bearing metal wipe from the temperature difference between the metal temperature and the bearing oil supply temperature, A method for monitoring a multi-span shaft type journal bearing, characterized in that the metal wipe is monitored, and the bearing load is further calculated from the oil film pressure signal to monitor the bearing load. 2. A patent claim characterized in that the oil film pressure is detected from a plurality of points separated in the rotational direction, a horizontal force acting on the bearing is calculated based on the oil film pressure, and a bearing load is calculated by taking the horizontal force into consideration. range 1
Method for monitoring multi-span shaft journal bearings as described in Section 1. 3 Calculate the pressure difference in the direction of the rotation axis from the oil film pressure signals at multiple points separated by a certain distance along the rotation axis direction, and compare this pressure difference with the pressure setting value to monitor the uneven contact state of the bearing. A method for monitoring a multi-span shaft type journal bearing according to claim 1 or 2, characterized in that: 4. Measuring the oil film pressure of lubricating oil from multiple points on the sliding surface of the journal bearing, calculating the mutual pressure difference from the oil film pressure signals at the multiple points, and calculating the amount of tilt of the bearing based on the pressure difference, In addition, the bearing metal temperature near the bearing sliding surface and the oil supply temperature to the bearing are measured, the bearing metal wipe is calculated from the temperature difference between the metal temperature and the oil supply temperature, and the bearing load is calculated from the oil film pressure signal. A multi-span journal bearing characterized in that an acoustic signal is detected from the bearing, the detected acoustic signal is compared with an acoustic setting value, and the condition of the bearing is monitored from the amount of inclination of the bearing, metal wipe, and bearing load. monitoring method. 5. The method for monitoring a multi-span shaft journal bearing according to claim 4, wherein the detected acoustic signal is an acoustic energy signal. 6. The method for monitoring a multi-span shaft journal bearing according to claim 4 or 5, wherein the detected acoustic signal is an acoustic signal occurrence frequency signal of a reference level or higher. 7. The multi-span shaft system according to claim 4, further comprising calculating the rate of change of the temperature difference between the metal temperature and the bearing oil supply temperature, and predicting the bearing metal wipe according to the direction of increase or decrease. How to monitor journal bearings.