JPH0527331B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to an apparatus for diagnosing abnormal vibrations of rotating machines, and in particular detects abnormalities of unbalanced vibrations that are synchronized with the rotational speed and occur in the shaft systems of large rotating machines such as power plants. Related to diagnostic equipment.

[Background of the invention]

In recent thermal and nuclear power plants,
As capacity increases, the demand for improved reliability of rotating machines such as turbines and generators, which are the main equipment, is becoming stronger. Furthermore, in order to supply power smoothly, it is necessary to quickly perform maintenance and inspection after an abnormality occurs in equipment.

There are various causes and phenomena for abnormalities in rotating machines, but shaft vibration is particularly common, and there is concern that shaft vibration may lead to serious accidents. Furthermore, since the rotating shaft systems used in power generation plants, such as turbines and generators, are long and large-scale systems, a lot of time is spent locating the point where an abnormality occurs. Against this background, as a means of improving the reliability of rotating machines, it is possible to detect abnormalities in the rotating shaft at an early stage from the vibrations generated in the rotating shaft, and to quickly diagnose the type of abnormality and the location where it occurs. There is a strong need for the development of vibration diagnostic technology. In particular, many of the abnormal vibrations that occur in the rotating shaft include vibrations that are synchronized with the rotational speed, that is, unbalanced vibrations, and it is important to diagnose the location where the unbalance occurs, which is the source of this unbalanced vibration.

As a conventional method for diagnosing the location of unbalance,
A method has been reported in which the relationship between rotational speed and journal vibration is plotted on polar coordinates over the entire operating range, and the unbalance position is estimated from the characteristics of the pattern. However, this method has disadvantages in that the pattern in which the unbalance occurs is limited, the accuracy of the occurrence position is not necessarily sufficient, and the monitored speed range is wide.

[Purpose of the invention]

An object of the present invention is to provide a rotating machine capable of diagnosing each occurrence position and amount of unbalance at a certain rotational speed even when abnormal unbalanced vibrations that occur in a rotating machine are synchronized with the rotational speed at a plurality of locations. An object of the present invention is to provide an abnormal vibration diagnosis device.

[Summary of the invention]

In order to achieve this object, the present invention provides a vibration analysis means for extracting an unbalanced vibration component synchronized with the rotational speed from the detection signals of the rotational speed detection means and the vibration detection means, and a vibration analysis means for extracting an unbalanced vibration component synchronized with the rotational speed, a first storage means for storing in advance the normal journal vibration corresponding to each rotational speed; and a first storage means for storing the unbalanced vibration component determined to be abnormal. a first calculating means for calculating an abnormal unbalance vibration component by subtracting the normal journal vibration corresponding to the rotational speed read from the rotor; a second storage means that stores in advance an influence coefficient representing the journal vibration when the balance amount is added; and a position where an abnormal unbalance occurs and the amount of the unbalance are determined from the abnormal unbalance vibration component and the influence coefficient. a second calculation means; and a third calculation for obtaining a residual abnormal unbalanced vibration component by subtracting the abnormal unbalanced vibration found by the second calculation means from the abnormal unbalanced vibration component found by the first calculation means. and determining means for determining whether the residual abnormal unbalanced vibration component obtained by the means is equal to or greater than a predetermined value, and the determined means determines that the residual abnormal unbalanced vibration component is equal to or greater than the predetermined value. The present invention is characterized in that the abnormal unbalance occurrence position and the amount of unbalance of this residual abnormal unbalanced vibration component are determined by the same means as in the case of the abnormal unbalanced vibration component.

[Embodiments of the invention]

Hereinafter, the present invention will be explained in detail based on illustrated embodiments.

FIG. 1 is a block diagram of an embodiment in which the present invention is applied to a steam turbine power plant. In this figure, 1 and 2 are turbine rotors, 3 is a generator rotor, and 4 is a shaft support that supports each of these rotors. 5 is a vibration sensor for detecting journal vibration received by each bearing 4; 6 is a rotation speed detector for detecting the rotation speed of the rotors 1 to 3;
The detected signal is also used as a phase reference signal for unbalanced vibration analysis. 7 is an amplifier that amplifies the detection signal of the vibration sensor 5; 8 is a rotation speed detector 6;
and a vibration analyzer 9 which inputs the output of the amplifier 7 and extracts an unbalanced vibration component synchronized with the rotation speed.
10 is a determination device that compares the unbalanced vibration component extracted by the vibration analyzer 8 with a predetermined value and determines that it is abnormal when the unbalanced vibration component is greater than the predetermined value. A storage device 11 stores the number of unbalanced vibration components of the rotational speed determined to be abnormal by the determiner 9 and the rotational speed read out from the storage device 10 (the unbalanced vibration component is determined to be abnormal). 12 is a vector computing unit for extracting only the abnormal vibration component due to unbalance occurring in any of the rotors 1 to 3 from the normal journal vibration of the rotors 1 to 3;
Any one of the axial positions (virtual unbalance positions) of
A storage device that stores in advance the influence coefficient representing the journal vibration of each bearing (each measurement position) when a unit unbalance amount is added to each point; 13 is an abnormality at each measurement position extracted by the vector calculator 11; A computing unit that calculates the maximum amplitude value of the unbalanced vibration component and the maximum amplitude value of the influence coefficient at each measurement position of the rotation speed read from the storage device 12; 14 is the abnormal unbalanced vibration component and influence; A computing unit normalizes each amplitude of the abnormal unbalanced vibration component and the influence coefficient at each measurement position so that the maximum amplitude value of each coefficient becomes 1; An arithmetic unit that sequentially calculates the difference between each amplitude of the influence coefficient and the root-sum-mean-square (R _ns value) thereof; 16 is the output of the arithmetic unit 14, which is the normalized abnormal unbalance vibration at each measurement position; A pattern recognizer that sequentially compares the component amplitude pattern and each influence coefficient amplitude pattern (at each virtual unbalance position) to determine the most similar influence coefficient pattern; 17 is the output of the arithmetic unit 14; The amplitude pattern of the abnormal unbalanced vibration component at each normalized measurement position and each amplitude pattern of the influence coefficient, each difference and its root mean square sum (R _ns value) which is the output of the calculator 15, and the pattern recognizer 16 is a display device that shows the determination results; 18 is a display device that displays the influence coefficient (read from the storage device 12) at the virtual unbalance position of the pattern determined to be most similar by the pattern recognizer 16 and the normalized output of the vector calculator 11; 19 is a computing unit that calculates the residual abnormal unbalance vibration component by removing the unbalance amount generated in the rotor and the abnormal unbalance vibration component due to this unbalance amount from the abnormal unbalance vibration component that is not present; 19 is the residual abnormal unbalance vibration component; 20 is a display device that displays the determination result of the determiner 19.

FIG. 2 is a flowchart showing the flow of processing by the abnormal vibration diagnosis apparatus shown in FIG. 1. Before explaining FIG. 2, symbols will be explained to simplify the explanation.

i: Measurement position number (i = 1, 2, ..., I) j: Rotation speed number (j = 1, 2, ..., J) k: Virtual unbalance position number (k = 1, 2, ...
..., K) u ^j _i : Unbalance vibration at abnormal time at the j-th rotation speed of measurement position i U ^j _i : Unbalance vibration at normal time at the j-th rotation speed at measurement position i v ^j _i : Abnormal unbalance only Initial unbalanced vibration component (initial abnormal unbalanced vibration component = u ^j _i −
U ^j _i ) V ^j _i : Normalized v ^j _i β ^j _ik : Vibration at the j-th rotational speed of measurement position i when unit unbalance amount is added to virtual unbalance position k (influence coefficient) B ^j _ik : Normalized β ^j _ik | | : Vibration amplitude w ^j _k : Unbalance amount existing at virtual unbalance position k ε ^j _i : Residual abnormal unbalance vibration component E at j-th rotation speed at measurement position i ^j _i : normalized ε ^j _i Now, suppose that an abnormal vibration occurs at the measurement position i at the j-th rotation speed while the rotors 1 to 3 are rotating, this abnormal vibration is detected by the vibration sensor 5 and the amplifier 7 is amplified. Then, from this amplified abnormal vibration signal, the rotational speed ω from the rotational speed detector 6, and the phase reference signal, the unbalanced vibration u ^j _i synchronized with the rotational speed is analyzed by the vibration analyzer 8, and this is analyzed by the determiner 9. If it is determined to be abnormal, the unbalanced vibration u ^j _i determined to be abnormal and the normal unbalanced vibration read from the storage device 10
The vector calculation unit 11 performs vector calculation according to the following equation (1) from U ^j _i to extract the initial abnormal unbalanced vibration component v ^j _i .

v ^j _i =u ^j _i −U ^j _i (1, 2,..., I) (1) where v ^j _i is a vector quantity having phase and amplitude information.

Next, the maximum amplitude of the initial abnormal unbalanced vibration component v ^j _i and the influence coefficient β ^j _ik read out from the storage device 12 by the calculator 13 |vj|=|v ^j _i | _nax ...(2) |β ^j _k ｜=｜β ^j _ik ｜ _nax ...(3) is obtained. Furthermore, the arithmetic unit 14 calculates |V ^j _i |=|v ^j _i |/|vj|...(4) |B ^j _i |=|β ^j _ik |/|β ^j _k |...(5) is performed, and the amplitudes of v ^j _i and β ^j _ik are normalized, thereby |V ^j _i | _nax =1 ...(6) |β ^j _ik | _nax =1 ...(7) can get.

The arithmetic unit 15 calculates the difference ^Δv j _ik between the amplitude of the initial abnormal unbalanced vibration component |V ^j _i | normalized using equations (4) and (5) above and the amplitude of the influence coefficient |B ^j _ik | Calculated using the following equation (8), Δv ^j _ik = |V ^j _i | − | B ^j _ik | ...(8) Furthermore, from the Δv ^j _ik calculated using the equation (8), using the following equation (9), The root mean square sum of squares (R _ns value) R ^j _k is obtained.

Here, the amplitude of the initial abnormal unbalanced vibration component |
If V ^j _{i |} ^and the amplitude of the influence coefficient |B ^j _ik The mean square root R ^j _k also becomes zero. At this time, the abnormal unbalance that has occurred in the rotor is at rotation speed j and the unbalance position is point k.

Next, in the pattern recognizer 16, the amplitudes of the initial abnormal unbalanced vibration components |V ^j _i | are arranged in descending order with the measurement position i as a parameter, and the amplitude of the influence coefficient |B ^j _ik | with respect to the virtual unbalance position k are arranged in ascending order using the measurement position i as a parameter. At this time, a virtual unbalance position k where the distributions of the amplitudes |V ^j _i | and |B ^j _ik | in the rotor axial direction (measurement position) are similar in descending order of vibration is determined to be the abnormal unbalance occurrence position. Note that if the same determination result is obtained at a plurality of virtual unbalance positions k, the abnormal unbalance position is determined based on the magnitude of the R _ns value obtained by the calculator 15.

The above will be specifically explained based on FIG. 3.

Now, the vibration measurement positions are set at six points #1 to #6, and if excessive unbalanced vibration occurs at measurement position #2 and it is determined that abnormal vibration has occurred, abnormal unbalance has occurred somewhere in the rotating body. It can be assumed that it did. Therefore, at each vibration measurement position, the normal unbalanced vibration is subtracted (vector subtraction) from the abnormal unbalanced vibration that has occurred to extract the abnormal unbalanced vibration component due only to the abnormal unbalanced vibration that has occurred.
Among the extracted abnormal unbalanced vibration components (vector values with amplitude and phase angle), focusing only on the vibration amplitude and arranging them in descending order of magnitude, #2,
#1, #3, #4, #6, #5, record the order at the abnormal unbalance UA on the horizontal axis of Figure 3, and further adjust # so that the maximum amplitude becomes 1. Divide the amplitudes of #1 to #6 by the amplitude of #2.

From the pattern representing the magnitude relationship of the abnormal unbalance vibration amplitude at each vibration measurement position, it is estimated where on the rotating body the abnormal unbalance has occurred.

Assume that the positions of abnormal unbalance that occur in the rotating body are (1), (2), (3), ..., (K) in the axial direction. For this purpose, for example, a unit unbalance amount is added to the virtual unbalance position (1), and unbalance vibrations at positions #1 to #6 are calculated. At this time, if the vibration amplitude of #1 shows the maximum, use this value to increase the maximum amplitude to 1.
Dividing the vibration amplitudes of #1 to #6 so that
The same process is repeated for other virtual unbalanced positions. As a result, the pattern shown in FIG. 3 can be obtained.

From FIG. 3, it can be seen that the virtual unbalance position having the same amplitude pattern as the abnormal unbalance is (2). As a result, the abnormal unbalance occurrence position can be estimated as (2).

Next, a case will be described in which there is a deviation between the amplitude pattern at each measurement position when there is an imbalance at the virtual unbalance position and the pattern due to the abnormal imbalance that has occurred. This is effective when there are a plurality of similar patterns to the amplitude pattern due to virtual unbalance compared to the amplitude pattern due to abnormal unbalance, and it is difficult to determine the amplitude pattern. A specific example will be explained below.

As described above, the vibration amplitude is normalized to a maximum of 1 when a unit unbalance amount is added to the virtual unbalance position or when an abnormal unbalance occurs. Using this normalized vibration amplitude, the vibration amplitude due to abnormal unbalance and the vibration at vibration measurement positions #1 to #6 for virtual unbalance positions (1), (2), (3), ..., (K) are calculated. The difference with the amplitude is calculated and shown in the box in FIG. If the virtual unbalance position is the same as the abnormal unbalance occurrence position, #1 to #6
The difference between is 0. In Figure 3, the measurement position where the maximum amplitude occurs at virtual unbalance positions (2) and (3) is #2, which is the same position as the abnormal unbalance, so it is 0 (because the maximum value is set to 1). ).
These differences are obtained for each measurement position, and the Rms value is calculated at each virtual unbalance position and displayed at the top as shown in FIG. This virtual unbalance position with a small Rms value can be estimated as the abnormal unbalance occurrence position. Thereby, it is possible to estimate the position closest to the abnormal unbalance from a plurality of virtual unbalance positions similar to the amplitude pattern caused by the occurrence of the abnormal unbalance.

The result of pattern recognition by the pattern recognizer 16 is displayed on the display device 17 as shown in FIG. In Fig. 3, the numbers 1 to 6 are in the order of the magnitude of the amplitude, for example, 1 indicates the largest amplitude, and the numbers inside indicate the difference Δv ^j _ik , but in the figure only zero of Δv ^j _ik is shown. However, the description of other items is omitted.

The amount of unbalance occurring at the abnormal unbalance position determined by the pattern recognizer 16 is calculated by the calculating unit 18. The contents of the calculation at this time will be explained in detail below.

Vibration a due to the amount of unbalance w ^j _k that occurred at the abnormal unbalance position determined by the pattern recognizer 16
^j _i is expressed as a= ^j _i =β ^j _ik w ^j _k (10). Here, let ε ^j _i be the difference between the initial abnormal unbalanced vibration component v ^j _i extracted by the vector calculator 11 and the vibration a ^j _i due to the unbalance amount w ^j _k , that is, the residual abnormal unbalance vibration component. The relationship holds true.

ε ^j _i =v ^j _i −a ^j _i =v ^j _i −β ^j _ik w ^j _k ……(11) Here, in order to obtain the appropriate amount of unbalance w ^j _k , the following evaluation function is introduced. do.

J= _I 〓 ⁱ⁼¹ ｜ε ^j _i ｜ ² ...(12) Unbalance amount w ^j _k that minimizes this evaluation function J
is the amount of unbalance determined by the pattern recognizer 16 to have occurred at the virtual unbalance position k at the rotation speed j. This unbalance amount w ^j _k is calculated according to the following equation (13).

∂J/∂w ^j _k =0 ...(13) That is, by solving the simultaneous equations given by equation (13) by the calculator 18, the unbalance amount w ^j _k
can be found.

Once the unbalance amount w ^j _k is determined, the arithmetic unit 18 further calculates the equations (10) and (11) to determine the residual abnormal unbalance vibration component ε ^j _i .

If the abnormal unbalance that occurs in the rotor is at one location in the axial direction of the rotor, the residual abnormal unbalance vibration component ε ^j _i will be extremely small, but if it exists at multiple locations in the axial direction of the rotor, A considerable residual abnormal unbalance vibration component ε ^j _i remains.

Therefore, the determiner 19 determines whether the residual abnormal unbalance vibration component ε ^j _i is equal to or greater than a predetermined value, and if it is minute and less than the predetermined value, the abnormal unbalance that has occurred in the rotor is at one location in the axial direction of the rotor. The abnormal unbalance position and unbalance amount are determined by the pattern recognizer 16 and the arithmetic unit 18, respectively.

On the other hand, when the residual abnormal unbalance vibration component ε ^j _i is large and exceeds a predetermined value, it means that another abnormal unbalance still remains in one of the rotors, so this residual abnormal unbalance vibration component ε ^j _i is again replaced with the initial abnormal unbalanced vibration component v ^j _i and the above-mentioned processing after the vector calculator 11 is repeated. This makes it possible to determine the positions and amounts of unbalance at a plurality of locations where the rotor is abnormally unbalanced.

Based on FIG. 4, a method for determining the amount of abnormal unbalance will be specifically explained.

The pattern recognizer shows the location of abnormal unbalance (3)
When it is determined that this is the case, the arithmetic unit 18 calculates the unbalance amount (w ₁ , θ ₁ ) and the residual abnormal unbalance vibrations ε ₁ , ε ₁ , . . . , ε ₁ at each vibration measurement position. If _the maximum amplitude α ₁ is small among these _{ε 1} , ε _{1 , .} However, if _the maximum vibration α ₁ is large, the residual abnormal unbalance vibrations ε ₁ , ε _{1 ,} . By repeating these steps, the amount of unbalance generated in the rotating body is determined as (w ₂ , θ ₂ ), (w ₃ , θ ₃ ), etc. When the residual abnormal unbalance vibration becomes smaller than the predetermined value, End the repeat. This makes it possible to estimate the position and amount of abnormal unbalance occurring in at least one location in the rotating body.

The determination result of the determiner 19 is displayed on the display device 20 as shown in FIG. 4, for example, and shows the abnormal unbalance position k that has occurred in the rotor and the amount of unbalance (including the size w and the angle θ relative to the reference position).
It is now possible to monitor In FIG. 4, (a) is the first determination result, (b) is the second determination result, and (c) is the third determination result.

Note that in equation (8) above, only the amplitude was subtracted, but the phase angle having the maximum value of the amplitude of the initial abnormal unbalanced vibration component and the amplitude of the influence coefficient is taken as zero degrees, and the phase angle is normalized to obtain the vector. By performing the calculation and then calculating the equation (9), it is possible to determine the abnormal unbalance occurrence position with higher accuracy.

Furthermore, in the above embodiment, in order to determine the position where abnormal unbalance occurs, the R _ns value is used in combination with the distribution of the amplitude of the abnormal unbalance vibration component and the amplitude of the influence coefficient at each measurement position. , it is also possible to determine both of these independently depending on the case.

〔Effect of the invention〕

As explained above, according to the present invention, the abnormal unbalance occurrence position and the amount of the unbalance are determined from the abnormal unbalance vibration component and the pre-stored influence coefficient representing the relationship between the unbalance and vibration. Therefore, it is possible to accurately diagnose the location of abnormal unbalance occurring during operation and the amount of unbalance at a certain rotation speed, and
Furthermore, the residual abnormal unbalance vibration component is determined, and it is determined whether this residual abnormal unbalance vibration component is equal to or higher than a predetermined value. Assuming that the residual abnormal unbalanced vibration component is replaced with the initial abnormal unbalanced vibration component and the occurrence position and amount of unbalance are determined by the same method, it is possible to It is also possible to accurately diagnose the respective occurrence positions and amount of unbalance, and maintenance and inspection work of the rotating machine can be performed easily.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an abnormal vibration diagnosis device according to an embodiment of the present invention, Fig. 2 is a flowchart showing the flow of processing by the diagnosis device, and Figs. 3 and 4 are display examples on each display device. FIG. 1 to 3...Rotor, 5...Vibration sensor, 6...
Rotation speed detector, 8... Vibration analyzer, 9, 19...
Judgment device, 10, 12... Storage device, 11... Vector computing unit, 13 to 15, 18... Computing unit, 16
...Pattern recognizer, 17,20...Display device.

Claims (1)

[Scope of Claims] 1. An apparatus for diagnosing an abnormality of unbalanced vibration synchronized with the rotation speed, comprising means for detecting the rotation speed of a rotor and means for detecting journal vibration of a plurality of bearings supporting the rotor. a vibration analysis means for inputting the detection signals of the rotational speed detection means and the vibration detection means to extract an unbalanced vibration component synchronized with the rotational speed, and a vibration analysis means for determining whether the unbalanced vibration component extracted by this means is abnormal a first storage means for storing in advance the normal journal vibration of the bearing corresponding to each rotational speed; a first calculation means for calculating an abnormal unbalance vibration component by subtracting the normal journal vibration corresponding to the rotational speed read from the first storage means; and an arbitrary virtual unbalance position in the axial direction of the rotor. a second storage means for storing in advance an influence coefficient representing the journal vibration of the bearing when a unit unbalance amount is added to one point of the abnormal unbalance vibration component obtained by the first calculation means; and a second calculation means for determining the abnormal unbalance occurrence position and the amount of unbalance from the influence coefficient read from the second storage means, and the abnormal unbalance vibration component determined by the first calculation means. a third calculating means for subtracting the abnormal unbalanced vibration obtained by the second calculating means from the residual abnormal unbalanced vibration component to obtain a residual abnormal unbalanced vibration component; and determining means for determining whether the residual abnormal unbalanced vibration component is greater than or equal to a predetermined value, the abnormal unbalance occurrence position and the amount of unbalance of the residual abnormal unbalanced vibration component. An apparatus for diagnosing abnormal vibrations of a rotating machine, characterized in that the abnormal vibration diagnosing device for a rotating machine is configured to obtain the abnormal unbalanced vibration component using the same means as in the case of the abnormal unbalanced vibration component. 2. In claim 1, the second calculation means calculates the abnormal unbalance occurrence position of the abnormal unbalanced vibration component from the abnormal unbalanced vibration component and the influence at each vibration measurement position of the vibration detection means. An abnormal vibration diagnosis device for a rotating machine, characterized in that the diagnosis is made from the order of the magnitude of each amplitude of the coefficients. 3. In claim 1, the second calculation means calculates the abnormal unbalance occurrence position of the abnormal unbalanced vibration component by calculating the amplitude of the abnormal unbalanced vibration component at each vibration measurement position of the vibration detection means. and the magnitude of the root mean square of the sum of squares of the differences in the amplitudes of the influence coefficients. 4. Claim 2, further comprising display means for displaying the magnitude relationship between the amplitudes of the abnormal unbalance vibration component and the influence coefficient as a two-dimensional distribution of the virtual unbalance position and the vibration measurement position. An abnormal vibration diagnosis device for rotating machines, which is characterized by: 5. Claim 3, further comprising display means for displaying each difference between the amplitude of the abnormal unbalanced vibration component and the amplitude of the influence coefficient at each of the vibration measurement positions, and the root mean square of the sum of squares thereof. An abnormal vibration diagnosis device for rotating machines characterized by: