JPS63179227A - Method for determining weight and fitting position of balance weight of rotary body - Google Patents

Abstract

PURPOSE:To accurately determine the weight and fitting positions of balance weights by measuring unbalance vibration quantity generated by unbalanced weights added on a rotary body at plural positions at the same time at respective measuring points. CONSTITUTION:Plural trial weights (TW) 21-24 whose weight values are already known are added in grooves 10-13 of high-, intermediate-, and low- pressure turbine rotors 1-3 and a generator rotor 4 which are in balance correction surfaces 6-9 perpendicular to a rotating shaft 5 and concentric with the rotating shaft 5. Further, balance correction surfaces are formed on the opposite surfaces of the respective rotors, and similar grooves are formed. Then TWs are added in those grooves and an effective number N of TWs added effectively at N positions generate an unbalanced state intentionally. In this state, the rotor 4 is rotated to perform trial operation only once and L unbalanced vibration amplitudes based on the unbalanced state are measured as unbalanced vibration quantities by vibration detectors 16 and 17 at L vibration measurement positions on the rotating shaft 5 at K kinds of dangerous speeds which differ in degree.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention determines the weight and mounting position of a counterweight to be attached to a rotating body such as a rotor of a turbine generator in order to eliminate unbalance of the rotating body. Regarding how to.

[Conventional technology]

A rotating body such as a rotor of a turbine generator in a thermal power plant is constructed by connecting several individual rotors such as high-pressure, intermediate-pressure, and low-pressure turbine rotors and a generator rotor.

These unbalances of single rotors are removed at the manufacturing factory, but due to the size of the balancing test machine and the space for testing at the manufacturing factory, the balance test for the connected rotors must be carried out after installation on site. Implemented. Even if the unbalance of each individual rotor has been eliminated, the rotor material is susceptible to unbalance due to processing errors in the couplings that connect them, assembly errors during coupling, heating of the turbine rotor during operation, and heat generation of the power generation rotor. Unbalances occur due to shaft bending caused by uneven temperature rise due to non-uniformity of the temperature, so after installation, an on-site balancing test run is conducted to eliminate these unbalances.

In the balance test run, first rotate the rotor,
The initial hand balance vibration amount based on the unbalance is measured at each vibration measurement point on the rotation axis of the rotor.Next, an unbalance weight with a known weight called a trial weight is attached to an appropriate position on the rotor, and the unbalance is intentionally adjusted. The operation is performed under different conditions, and the amount of unbalanced vibration is measured at the same vibration measurement point as above. This is called a trial operation, and from the difference between the unbalanced vibration amount from the above trial operation and the initial hand-balanced vibration amount obtained at the beginning, the weight of the added trial weight and the mounting position determine the amount of vibration at each vibration measurement point. From this information, the weight and mounting position of the counterweight to be added to eliminate the imbalance can be calculated.

The above calculation generally uses a method called the influence coefficient method. The influence coefficient is defined as the amount of vibration at each vibration measurement point based on the unit weight of the installation location of the trial weight or counterweight. In calculations using this influence coefficient, the amount of vibration is a complex quantity whose absolute value is the magnitude of the amount of vibration, and the phase difference with respect to the vibration of the reference point on the rotation axis is the angle of deviation.9Weight is a complex quantity whose magnitude is the absolute value of the amount of vibration. Each value is treated as a complex quantity whose argument is the mounting angle with respect to the reference plane including the central axis of rotation of the rotating body.Therefore, the influence coefficient defined as the amount of vibration also takes the magnitude of the amount of vibration as the absolute value and is based on the above criteria. It is a complex quantity whose argument is the phase difference with respect to the vibration of a point. As described later, this influence coefficient is determined through trial operation, and the balance weight is calculated as a complex quantity by applying the condition that the vibration amount given by the product of this influence coefficient and the weight of the counterweight is balanced with the initial hand balance vibration amount. By calculating the weight of the counterweight, the value of the weight of the counterweight can be obtained as its absolute value, and its mounting angle, that is, its mounting position can be obtained as its deflection angle.

FIG. 4 is a diagram schematically showing the state of the rotor during trial operation according to the prior art for determining the influence coefficient.

High-pressure, intermediate-pressure, and low-pressure turbine rotors 1 schematically shown
．． 2.3 and the generator rotor 4, respectively, there is a counterbalancing surface 6, 7.8.9 perpendicular to the axis of rotation 5, each having a circumferential groove 10 co-centered with the axis of rotation 5.
．． 11.12°13 is the setting. Turbine rotor 1.
2.3 and the opposite surface (not shown) of the generator row 4 are also balancing correction surfaces, each marked with the reference numeral 10.
．． 11゜1.2) A groove similar to 13 is provided.0 In the trial operation shown in the figure, groove 1 of the high pressure turbine rotor 1 is provided.
A trial weight 14 is attached to the rotary shaft 5, and vibration detectors 16 and 17 are provided at vibration measurement points near a plurality of bearings 15 schematically indicated by triangles with respect to the rotating shaft 5. In FIG. 4, a vibration measurement point equipped with a vibration detector 16 is used as a reference point, and the vibration detector 16 measures the vibration amplitude as the amount of vibration at the reference point, and the vibration detector 17 measures the vibration amplitude at each vibration measurement point and the reference point. The phase difference with respect to the vibration at each point is measured to obtain the vibration amplitude as a complex quantity at each vibration measurement point. The complex quantity obtained by subtracting the initial vibration amplitude measured at the corresponding vibration measurement point during the initial operation from the measured vibration amplitude and dividing the value by the weight of the trial weight 14 is the complex quantity obtained by subtracting the initial vibration amplitude measured at the corresponding vibration measurement point during initial operation. This is an influence coefficient indicating the influence on the measurement point.

Next, operation is performed by changing the mounting position of the trial weight 14 or by remounting the trial weight 14 on another rotor, and determining the influence coefficient for each measurement point of the mounting point using the same procedure. The above procedure is repeated to determine the influence coefficient at the trial weight attachment position on each rotor. If trial weights are attached to each balance correction surface on both sides of the rotor, or if multiple trial weights are attached to one balance correction surface, one equivalent trial weight is assumed as a composite effect. , let the equivalent weight and the equivalent mounting position be the weight and mounting position of the trial weight.

[Problem that the invention seeks to solve]

According to the above conventional technology, in order to obtain the necessary influence coefficient, it is necessary to perform the same number of trial operations as the number of trial weights to be added. However, large turbine generators require a high fuel cost of about 10 million yen for one trial operation. Therefore, from the standpoint of reducing test costs and saving energy, it is desirable to have as few trial runs as possible, and for this reason, the number of trial runs has recently been severely limited. It is becoming difficult to determine the coefficient.

On the other hand, by using influence coefficients obtained from calculations for the calculation model of the target rotating body and past trial runs of the same model, we compensate for the shortfall due to the reduction in the number of trial runs. There is a way.

However, the influence coefficient theory applied when deriving the influence coefficient using a calculation model assumes that a proportional relationship, that is, linearity, exists between the weight of the added weight and the vibration amplitude at each measurement point. On the other hand, in an actual machine, the above-mentioned relationship has non-linearity, and the application of the influence coefficient derived from the calculation model is limited to the range where linearity holds, so it is not practical.

Also, among the bearings and bearing frames that support each rotor, those near the high-pressure turbine rotor 1 and intermediate-pressure turbine rotor 2, which are particularly hot, are heated by the radiant heat from the casings of the turbine rotors, and the bearings and bearing frames that support the rotors are heated by the radiant heat from the casings of the turbine rotors. There is a large temperature difference between the two.

Therefore, the resulting difference in thermal expansion causes the alignment of the bearing mount to vary from the state when the turbine generator is installed. Since the radiant heat from the casing varies depending on the state of the heat insulation treatment applied to the casing depending on the conditions at the time of installation, the above-mentioned change in alignment varies depending on the installation location. Furthermore, this alignment is also influenced by the ground at the installation location and changes before and after installation. As described above, the bearings of turbines installed at different locations exhibit different fluctuations in the alignment of the pedestals, so the load on the bearings differs depending on the location where they are installed. The load on bearings is an important factor that affects the spring characteristics and damping characteristics of the oil film, and this causes completely different vibration response characteristics even for the same model. In this way, there is a considerable difference between the true influence coefficient calculations obtained by trial operation of the target rotating body and the influence coefficients obtained by similar rotary bodies, and the method of obtaining influence coefficients using conventional technology The problem is that the required balance cannot be achieved.

It is an object of the present invention to solve the above-mentioned problems and provide a method for obtaining a highly reliable influence coefficient through a single trial run.

[Means for solving problems]

This invention deals with the amount of unbalanced vibration measured in one operation with a plurality of trial weights as unbalanced weights attached to a rotating body at the same time, and the influence of each weight attachment point on the measurement point of the amount of unbalanced vibration. Difference from the approximate hand balance vibration amount calculated by adding the initial hand balance vibration measurement value at that vibration measurement point to the vibration amount as the sum of the products obtained by multiplying the approximate value given as a coefficient by the weight of each trial weight. is equal to the sum of products obtained by multiplying the difference between the true influence coefficient and the approximate influence coefficient by the weight of each of the trial weights, and the difference between the influence coefficient of the tool and the approximate influence coefficient is calculated as an unknown quantity. Then, by finding this unknown amount from the measured unbalanced vibration amount, the calculated approximate hand-balanced vibration amount, and the weight i of the trial weight and correcting the approximate influence coefficient, the influence coefficient necessary for one trial operation number is determined. The goal is to obtain all of the following. That is, the amount of initial hand-balanced vibration is measured at a plurality of vibration measurement points on the rotating shaft of the rotating body 6, and unbalance weights with known weights are simultaneously added to the plurality of points on the rotating body 6 to measure each of the vibrations. The amount of unbalanced vibration is measured at a point, and the approximate amount of unbalanced vibration obtained by adding the measured initial amount of unbalanced vibration to the product of the approximate influence coefficient given the approximate value and the weight of the unbalanced weight. The unbalanced vibration amount error is obtained by subtracting it from the measured unbalanced vibration amount, and the unbalanced weight is further added to the influence coefficient error of the unknown quantity given as the difference between the true influence coefficient and the approximate influence coefficient. The product multiplied by the weight of is subtracted from the unbalanced vibration error and squared, and the value of the influence coefficient error that satisfies the least square condition that minimizes the sum of the squared products obtained for each vibration measurement point is determined, An influence coefficient is obtained in addition to the approximate influence coefficient. The weight and mounting position of the counterweight are calculated from the obtained influence coefficient and the initial hand balance vibration amount measurement value.

[Effect]

When a rotating body is operated with multiple unbalanced weights attached at the same time, the amount of unbalanced vibration measured at each vibration measurement point is the sum of the amount of unbalanced vibration caused by each unbalanced weight, and the amount of unbalanced vibration at that vibration measurement point. The amount of initial hand balance vibration is added. On the other hand, when calculating the amount of vibration using the influence coefficient, the amount of unbalanced vibration for each vibration measurement point is the initial hand-balanced vibration measured by the sum of the products of the influence coefficient and the weight for each of the unbalanced weights. Calculated as adding the amount. Therefore, if the approximate amount of hand-balanced vibration is calculated by giving the influence coefficient of 4g1 (IlI), the unbalance, which is the difference between the measured unbalanced vibration amount at each vibration measurement point and the calculated approximate hand-balanced vibration amount, is calculated. The vibration error is equal to the product of the influence coefficient error, which is the difference between the true influence coefficient and the approximate influence coefficient, multiplied by the weight of the unbalanced weight, and the sum of the products for each unbalanced weight.Then, the influence coefficient error is taken as an unknown. , for each vibration measurement point, the product of the influence coefficient error multiplied by the weight of the unbalanced weight is subtracted from the unbalanced vibration error, squared, and the influence coefficient that satisfies the least squares condition where the sum of the squared products is minimized. Find the error. The influence coefficient error that satisfies the above condition is an approximate solution to the condition that the unbalanced vibration error is equal to the sum of the influence coefficient error multiplied by the weight of the unbalanced weight. The true influence coefficient is obtained by making corrections to the shadow space coefficient error approximation influence coefficient.The balance weight is calculated by calculating the relational expression that the product of this influence coefficient and the weight of the counterweight cancels the initial hand balance vibration amount. Find the weight of.Vibration amount 9
Both weight and influence coefficient are treated as complex quantities,
The determined weight of the counterweight is also a complex quantity determined by the size of the weight and the mounting angle with respect to the reference plane, so the mounting position can also be determined from this.

[Embodiments of the invention]

Prior to detailed description of this invention, the principle of the method of this invention will be explained first.

As already mentioned, there is nonlinearity between the rotation speed and vibration amplitude, so if the rotation speed is different, vibrations with completely different characteristics can be observed even at the same vibration measurement point. Therefore, vibration measurements are usually carried out in the vicinity of critical speeds of different orders.

If the number of vibration measurement points is L, and the type of critical speed selected as the rotational speed for measurement is KIllll, then KL-M vibration amplitude measurements are obtained in one operation, and these are different M It can be treated as having been measured at individual vibration measurement points.

The unbalanced vibration amplitude measured when the rotating body is operated with one trial weight added to each of N locations is given by the following.

(g) −(vo) + (α) (W)
(1) Here, [ε] is the unbalanced vibration amplitude εI
+ '1+...+'' Unbalanced vibration vector with N as an element.

[V0] is the initial hand balance vibration amplitude V 111 +
v68°...+ Initial hand-balanced vibration vector with vOM as an element, [α] is the influence coefficient α of the n-th weight attachment point on the m-th unbalanced vibration amplitude measurement point
, a is an influence coefficient matrix with elements, [W] is the weight of the trial weight W I, W ! +..., W.

are trial weight vectors whose elements are given below.

Here, T is a symbol that means transpose of each vector,
Each vector is a column vector, and the elements of the above vectors and matrices are all complex numbers, as already mentioned.

On the other hand, the approximate hand-balanced vibration amplitude calculated using the approximate influence coefficient obtained by giving an approximate value to the influence coefficient for the above-mentioned rotating body is given by the following.

(E) −(vs) + (A) (W)
(3) Here (E) is the approximate hand-balanced vibration amplitude E+, Ez.

. . , an approximate hand-balanced vibration vector whose elements are E, and [A] are approximate influence coefficient matrices whose elements are approximate influence coefficients A and 7, which are shown below.

Also, if we express equation (3) in terms of its elements, E, I = Vo - 1ΣA, , l - W, l
(5) (m-1, 2,..., M) It is a pot. Subtracting equation (3) from equation (2), [ε]
−(E) −(Δε〕 [α) −[A) −(Δα
]CΔε]−[Δα) (W) (
6) is obtained. [Δri, (Δα] are vectors and matrices shown below, respectively, and [Δε]
the unbalanced vibration error vector.

[Δα] will be referred to as an influence coefficient error matrix.

In other words, the elements of [Δ6] and [Δα] are each Δ111
-8. -E, (++-1, 2,...,M)
It is shown in (8).

(6) Solve for 2t-(Δα], and this [Δα] and (
A) from [α) − (A) + (Δα) (
Modify the approximate influence coefficient by the deduction of 10) to obtain the required influence coefficient.

This is the basis of the method according to the invention.

In solving equation (6) for [Δα], transform it to [ΔC]−[Δα) (W) −0(11),
The mth element of equation (11) is written as Δt, −ΣΔα
,,W, -0(12)(*=1.2...,M). This equation (12) shows MN pieces of Δ, which are unknown quantities. Therefore, by solving this simultaneous equation, Δα□ can be found, but it is obvious that MN>M and the number of unknowns exceeds the number of elements of the simultaneous equation, so it cannot be solved using normal methods.

Therefore, as shown below, an approximate solution is found using the least squares condition on the left side of equation (12).

That is, an evaluation function S given by the square of the left side of equation (12) is defined, and Δα□ that minimizes the sum of S is found by changing the value little by little from its initial value, Δα1. ．． If it is expressed as an absolute value a□ and an argument angle of 0.7, the above can be expressed as. Then, give appropriate initial values to a□ and θ1, change the values in small increments from these initial values to select all and θ, which minimize S, and use Δαme composed of these as the solution. . Δ obtained by being like this
Correct the recent influence coefficient matrix (A) by the influence coefficient error matrix [Δα] whose elements are α, , (10
) is used to calculate the influence coefficient matrix [α], and the influence coefficients that are its elements are obtained.

We will touch on how to determine the weight and mounting location of the device.

For a rotating body in a balanced state due to the addition of a counterweight, [ε] = 0 in equation (1), so 0 - (vo) + (α) (W)
The relationship (14) holds true. Hereafter the vector (Wl and its element WI.

W! ,..., W is related to the weight of the counterweight. Expressing equation (16) in terms of elements, 0−v, ,+Σ
α,,lW,l(15)(m-1,2,...,M) where V6m and α,,1 are known numbers and W7 is an unknown number.

v0□ α1Ill+ w+t are all complex numbers, so their real part V. IIK+α, R,, W□ and the imaginary part V. a.

α@111. Considering W IIIP respectively (15
) formula is 0=V,,x+iy,,,+Σ(”may +
iα,,, ) (WIIX + i Way)-V
olIz + l V 1laF+Σ((α, 1lxW
ll, l-a, , W, , )+1(α, □WllllI
+α−1. tt W-) )-vo, +Σ('Flax
WRx-Cramp Way) + l (Vo*y
+Σ (α−, A W□+α@nl L A) )
(16) From the condition that both the real part and the imaginary part of equation (16) are zero, W□ and W
Find □. A practical method for this is to use the condition that minimizes the unbalanced vibration in equation (1). (
16) Eq. The minimum condition for a- is W□IW, l
, can be obtained by setting the partial differentials of , respectively, to zero. That is, a o-a eP-a a-a a- aWsy, therefore, by solving equation (19), W□
.

It is possible to find W and lA.

Expression (19) is written in matrix representation as (α) ” (α) (w) + (α) ”
Cvo) -.

Therefore, the weight of the counterweight is (W) - ((α) [α] ) - Ku [α]
” (Vo), where (W), (V.)
, [α] are (W) - (WtIIW+y... , W,, W,,)'
, and [α1d] is the transposed matrix of [α].

The above (W), (V6), and (α) are the weight vector of the weight, the initial balanced vibration amplitude vector, and the influence coefficient matrix when the unbalanced vibration is expressed by dividing it into a real component and an imaginary component, and (1 ) in the equation (W), (v
o), Same as (α) and calyx.

From W7X and W□ obtained above, the weights W and l of the counterweight and the mounting angle are determined as follows.

w, −Wllll” +w+ty”
(22) y mtan −' (w, ly/W+x
) FIG. 1 schematically shows an embodiment of the present invention. Balance correction surface 6 perpendicular to the rotational axis 5 of the high-pressure, intermediate-pressure, and low-pressure turbine rotors 1.2.3 and the generator rotor 4
, 7.8.9 along the circumferential groove 10 concentric with the axis of rotation 5
．． A plurality of trial weights 21.22° 23 and 24, each having a known weight, are attached to 11, 12, and 13.

Further, the opposite surface (not shown) of each of the rotors is also provided with a balance correction surface, and grooves similar to those described above are provided, and trial weights are also added to these grooves to effectively adjust the N. The effective number N of trial weights added to the locations intentionally creates an unbalanced state.

In this state, the rotor was rotated and a one-time trial operation was carried out. Similar to the conventional technology, at different critical speeds of different orders, vibration detectors were detected at L vibration measurement points on the rotating shaft. 16 and 17, based on the unbalanced state < M-KL unbalanced vibration amplitudes ε
1. ε8. ..., ε is measured as the amount of unbalanced vibration.

In addition, prior to the above trial operation, the vibration amplitude based on the unbalance of the rotor itself was measured at the same L vibration measurement points at a critical speed of K11lf, and M=KL initial vibration amplitudes VOI+VO2,"・ve+e Measure.

On the other hand, an approximate influence coefficient A□ is given as an influence coefficient for each vibration measurement point of each of the above-mentioned trial weight attachment points. Since the number of trial weights φ is N and the number of vibration measurement points is M in real terms, the number of virtual influence coefficients is MN. As the approximate influence coefficient, an approximate value such as an influence coefficient obtained in the past for the same model or a calculated value for a calculation model of the target rotating body is given.

The above vibration amplitude measurement value s, (+*-1, 2,...
, M) Initial vibration amplitude V(111(Il+=1121 ”
', M) + trifle.

Weight complex quantity W, (n-1, 2,..., N), and approximate influence coefficient A-s (m-1, 2,...M; 4t
The flowchart in FIG. 2 shows the calculation procedure for determining the influence coefficient α, 9 from (l+2.4·+N).

First, E is calculated from the input V, , A, , l, W, , by equation (5), and then the input ε is calculated.

The difference Δ and ε between and El calculated above is determined by equation (8). On the other hand, influence coefficient error Δα6. Initial values are given to the absolute value (amplitude) a and the argument angle (phase angle) 0.7. Furthermore, the calculation of the evaluation quantity is repeated using formula (13) to find the values of a□ and θ□ that minimize S. By the way, repeating the above procedure for each of allll and θIIIR requires a huge amount of calculation. , the calculation time becomes longer. Therefore, in order to avoid this, first a□
After fixing θ1. Vary l and try to find the one that minimizes S. θ,.

The reason why a is changed first is that the influence coefficient is more sensitive to θ□ than allR, and also because a fairly good approximation value can be empirically given as the initial value of a, . θ□
After finding the minimum value of S by varying Determine allllll and θ□ that minimize S by returning the value, calculate Δα and ll from this using equation (13), and (
10) α, based on Eq.

-Aell+Δa1, approximate influence coefficient A
1° to obtain the required influence coefficient α, .

After determining αlI as described above, the complex 1w□, W, and y of the counterweight are determined by equation (19) or (21), and then the complex IW is determined by equation (22).

and the mounting angle with respect to the reference plane are calculated to determine the weight and mounting position of the counterweight.

FIG. 3 is a graph showing an example of the effect of modifying the approximate influence coefficient by the method of the present invention based on the results of simulation calculation. It has been shown that as a result of correcting the approximation influence coefficient that gives the error shown on the horizontal axis using the method of the present invention, the error shown on the horizontal axis can be reduced to the value shown on the vertical axis. The vertical axis shows the average error value for six vibration amounts assuming two types of critical speeds at three vibration measurement points. Influence coefficient error α,
, 1all 3□0III11, the minute value to be changed from the initial value given as an approximate value is the absolute value a
Regarding Mn, Δa is set to two values, 1% and 10% of the initial value, and Δθ about the argument angle θ□ is uniformly set to 5°. (a) is the error regarding the magnitude of vibration amplitude,
(b) shows the error regarding the phase difference,
In both cases, it is shown that the accuracy is improved by making the change width Δa smaller. The same may be said of Δθ.

Regarding the degree of error reduction, when Δa is set to 1% of the initial value, both vibration amplitude and phase difference are +30
% error has been reduced to 12% for vibration amplitude and 4% for phase difference. Furthermore, it goes without saying that the closer the approximate value of the influence coefficient is to the true value, the more accurate it is.

〔Effect of the invention〕

In this invention, the amount of unbalanced vibration caused by unbalanced weights added simultaneously to multiple locations on a rotating body is measured at each vibration measurement point, and an approximate influence coefficient that has been given an approximate value is used to send a response at the vibration measurement point. The unbalanced vibration amount is calculated, and the difference between the measured unbalanced vibration amount and the calculated approximate hand-balanced vibration amount is calculated as the influence coefficient error, which is the difference between the true influence coefficient and the approximate influence coefficient. From the relationship that the product multiplied by the weight is equal to the sum of each unbalance weight, the influence coefficient error corresponding to each approximate influence coefficient is obtained, and the approximate influence coefficient is corrected based on this to obtain the influence coefficient. Therefore, the influence coefficient specific to the rotating body can always be obtained, and the weight and mounting position of the counterweight can be determined with high precision. Furthermore, since unbalanced vibration measurements are used with a plurality of counterweights added simultaneously, all necessary influence coefficients can be obtained from the unbalanced vibration measurements obtained in one trial run. Therefore, there is no need to repeatedly perform trial runs, making it possible to significantly reduce test costs.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram of a rotor in a trial operation as an embodiment of the method of the present invention, Fig. 2 is a flowchart showing the procedure for calculating an influence coefficient from the measured values of the trial operation, etc., and Fig. 3 is a method of the present invention. FIG. 4 is a schematic diagram of a rotor in a trial operation according to the prior art. 1: High pressure turbine rotor, 2: Medium pressure/turbine rotor,
3: Low pressure turbine rotor, 4: Generator rotor, 5: Rotating shaft, 14.21.22.23.24F) Lyal weight, 16.17: Vibration detector. t#Nott'75 Now 4JiJh jnn Width ('
/, )(G) (b) Figure 3

Claims (1)

[Scope of Claims] 1) From the amount of vibration due to unbalance of the rotating body and the influence coefficient defined as the amount of vibration generated at the vibration amount measurement point of the rotating body by a weight of unit weight added to the rotating body, A method for determining the weight and mounting position of a counterweight to be added to the rotating body through calculations that treat the amount of vibration and the weight of the weight as complex quantities, the method comprising a plurality of vibration measurement points on the rotation axis of the rotating body. The initial unbalanced vibration amount was measured, and unbalanced weights with known weights were added simultaneously to multiple points on the rotating body, and the unbalanced vibration amount was measured at each of the vibration measurement points, and an approximate value was given. The amount of unbalanced vibration is determined by subtracting the approximate amount of unbalanced vibration obtained by adding the measured initial amount of unbalanced vibration to the product of the approximate influence coefficient and the weight of the unbalanced weight from the measured amount of unbalanced vibration. After obtaining the error, the product of the influence coefficient error of the unknown amount given as the difference between the true influence coefficient and the approximate influence coefficient multiplied by the weight of the unbalanced weight is subtracted from the unbalanced vibration error, and then squared. and an influence coefficient error value that satisfies a least squares condition that minimizes the sum of the squared values obtained for each vibration measurement point is added to the approximate influence coefficient to obtain an influence coefficient. How to determine the weight and mounting location of the counterweight. 2) In the method according to claim 1, the rotation is characterized in that the amount of vibration is a complex amount whose absolute value is the magnitude of the amount of vibration and whose argument is the phase difference with respect to the vibration at the reference point. How to determine the weight and mounting location of body counterweights. 3) In the method described in claim 1 or 2, the weight of the weight is determined by the absolute value of the weight and the angle of attachment with respect to a reference plane including the central axis of rotation of the rotating body as the deviation angle. A method for determining the weight and mounting position of a counterweight for a rotating body, which is characterized by being a complex quantity. 4) A method for determining the weight and attachment position of a counterweight of a rotating body in the method according to any one of claims 1 to 3, wherein the amount of vibration is a vibration amplitude. 5) In the method according to any one of claims 1 to 4, virtual initial values are respectively given to the absolute value and argument of the influence coefficient error, which are complex quantities, and the absolute value A method for determining the weight and mounting position of a counterweight of a rotating body, characterized in that a value that satisfies a least squares condition is found by changing the values of and declination by minute amounts from the initial values.