JP5736634B2 - Evaluation method of magnetostriction or noise of magnetic steel sheet for three-phase transformer core excited by sine wave - Google Patents

Description

  The present invention relates to a magnetostriction evaluation method for an electrical steel sheet for estimating noise generated by an iron core of a transformer.

  Transformers emit noise when energized, which causes deterioration of the living environment of the local residents and the working environment at the substation. Therefore, when ordering a transformer, an upper limit value of the noise level is provided as a required specification, and it is strongly required that the noise level of the finished product does not exceed the upper limit value. In order to satisfy this requirement, when manufacturing a transformer, it is necessary to study various noise-related technologies and techniques from the design stage and apply them appropriately to the product.

One of the noise generation sources of transformers is an iron core, and one of the causes of the iron core noise is the magnetostriction phenomenon of electrical steel sheets. As shown in FIG. 1, when an electromagnetic steel sheet is magnetized with an alternating current, the external dimension of the electromagnetic steel sheet changes with the change in the strength of the magnetization. Generally, the length of the steel sheet changes. It is often expressed as This causes the iron core to vibrate, and the vibration propagates to an external structure such as a transformer tank and is emitted as noise.
Therefore, in order to make the noise level of the transformer below the upper limit value, it is important to select and use an electromagnetic steel sheet having an appropriate magnetostriction characteristic. The magnetostriction displacement waveform amplitude shown in FIG. 1 is often used as the magnetostriction characteristic. In this method, an electromagnetic steel plate is AC-excited as a single cut plate sample, the amplitude of a vibration displacement waveform due to magnetostriction appearing at that time is obtained, and this electromagnetic steel plate is evaluated.

As an example of a measuring apparatus used for this purpose, there is an example shown in Patent Document 1. It is considered that the smaller the amplitude, the smaller the noise. Non-patent document 1 shows an example in which the noise level of an actual transformer is reduced by using an electromagnetic steel sheet having a reduced magnetostriction amplitude.
However, noise may not be reduced only by reducing the magnetostriction amplitude. This is because the amplitude of vibration is not directly proportional to the noise level of the sound. The principle is shown below.

First, sound is a vibration phenomenon of air particles, but the magnitude of the sound is proportional to the vibration speed and is not directly proportional to the amplitude. Therefore, since the amplitude of magnetostriction does not directly correlate with sound, it is necessary to evaluate the magnetostriction vibration speed.
Moreover, although the noise of a transformer contains many harmonics so that the example which frequency-analyzed in the nonpatent literature 1 is shown, the magnetostriction waveform also contains many harmonics. However, since the sensitivity of human hearing changes with the frequency of sound, it is necessary to perform auditory correction when evaluating magnetostriction as sound.

  In order to solve the above problems, the method of Patent Document 2 has been proposed. That is, the magnetostriction to be measured is treated as a velocity waveform, and is further subjected to frequency analysis to perform auditory correction and obtain the level. By using this method, the magnetostriction can be further evaluated by converting it to a value close to the noise level of the transformer.

Japanese Patent No. 1874250 Japanese Patent No. 3456742

IEEJ Technical Report No. 616, “Current Status and Trends of Static Noise Countermeasure Technology,” IEEJ, 1996

Although the method disclosed in Patent Document 2 is basically effective, there is a problem in accuracy when a slight difference is evaluated. The following experiment was conducted for the verification.
Two types of magnetic steel sheets X and Y having different magnetostrictive characteristics were prepared, and the magnetostrictive characteristics of the samples taken from each were measured, and a three-phase three-legged iron core shown in FIG. The values for evaluation were obtained from the measured magnetostriction characteristics using the method of Patent Document 2, and the difference between the materials X and Y was calculated and shown in Table 1. Furthermore, the difference between X and Y was also obtained for the measured noise level of the transformer, and is shown in Table 1. These measurements and calculations were performed at two levels of magnetic flux densities of 1.5T and 1.7T.

In this result, the material X is reduced by about 2 dB in terms of magnetostriction, but the material X is increased by about 2 to 3 dB in the transformer noise, resulting in contradictory results. The noise difference of 2 to 3 dB is a small difference, but when such a difference is small, the method of Patent Document 2 indicates that the magnitude of the transformer noise between different materials may not be evaluated correctly. Yes.
From the above, it can be seen that there is room for improvement in the method of Patent Document 2.

The problem to be solved by the present invention is to present this improvement method.
A three-phase voltage is applied to the three coils wound around the three-phase three-legged core shown in FIG. 2, and the magnetic flux density waveform measured at the positions indicated by A, B, and C in FIG. 2 is shown in FIG. . The voltage applied to the coil is adjusted so that the average magnetic flux density of the iron core is 1.7T. Since the voltage waveform applied to the coil is a sine wave, the magnetic flux density waveform of the entire iron core is also a sine wave.

However, as shown in FIG. 3, the magnetic flux density waveforms of A, B, and C are significantly distorted and include a large amount of harmonics. This is because the three-phase three-legged iron core is a three-phase three-phase core grouped into one iron core, and the magnetic flux distribution changes complicatedly with time.
Since magnetostriction changes the deformation displacement depending on the magnetic flux density, it is considered that when the magnetic flux density waveform changes from a sine wave, the magnetostrictive waveform also changes. However, magnetostriction measurement is generally performed under the condition that the magnetic flux density is a sine wave. Therefore, even if the material is evaluated by the normal magnetostriction measurement result, it does not match the actually measured noise.

The gist of the present invention is as follows.
(1) A local magnetic flux density waveform in a three-phase transformer core when excited by a sine wave, an amplitude ratio A ₃ / A ₁ in the following formula 1 within a range of 0.05 to 0.2, a phase angle θ is set within a range of 50 ° (0.873 rad) to 70 ° (1.222 rad) , estimated by calculation using the following formula 1 , and the magnetic flux density waveform is the excitation voltage waveform of the magnetostriction measuring device. is generated in the electromagnetic steel samples by controlling, by actually measuring the magnetostriction at that time, and wherein the TURMERIC line evaluation of the magnetostrictive or noise of the three-phase transformers electrical steel sheet, the three-phase transformer which is excited by a sine wave Evaluation method for magnetostriction or noise of electrical steel sheets for ceramic cores .
In Equation 1, b is the magnetic flux density waveform (T), f is the excitation frequency (Hz), t is the time (sec), A ₁ and A ₃ are the amplitudes (T) of the fundamental wave and the third harmonic wave, θ Represents a phase angle (rad).
b = A ₁ cos (2πft) + A ₃ cos {3 (2πft + θ)} (1)

  The conventional method for evaluating the magnetostriction of the electrical steel sheet related to the transformer noise has been performed using an excitation waveform, that is, a magnetic flux density waveform as a sine wave. However, when the transformer core is viewed locally, the magnetic flux density waveform is distorted, and if the location is different, the waveform changes. Therefore, the present invention performs magnetostriction evaluation with a distorted magnetic flux density waveform that is actually generated in the local part of the iron core.

The magnetic flux density waveform may be obtained by actual measurement or calculation. However, since the waveform distortion tends to be constant for the three-phase iron core, a waveform suitable for it may be used. The following experiment was conducted in order to grasp the tendency. The iron core shown in FIG. 2 was actually produced, and search coils were installed at 78 locations in the iron core, and the magnetic flux density waveform in the steel sheet rolling direction at each portion was measured. These data were frequency-analyzed and separated into a fundamental wave and a harmonic wave, and A ₃ / A ₁ and phase angle θ obtained from A ₁ and A ₃ in the above equation 1 were obtained. Sections appropriately divided for these binary values are set, the number of waveforms corresponding to each section condition is determined, and the results of the frequency distribution are shown in FIG. 4 and FIG.

With regard to the amplitude ratio A ₃ / A _1, it can be seen from FIG. 4 that most of the waveforms fall within the range of 0.05 to 0.2, and the occurrence of waveforms outside that range is unlikely. As for the phase difference θ, it can be seen from FIG. 5 that most of the waveforms are in the range of 50 ° to 70 °, and it is unlikely that a waveform outside the range is generated. A waveform that meets this condition may be created and used for magnetostriction measurement or estimation.

  This method improves the correlation between magnetostriction evaluation results and transformer noise. Specifically, when the transformer noise is estimated using the result obtained by this method, the estimation can be performed with higher accuracy than when the conventional method is used. Moreover, when judging the superiority, inferiority, or suitability of a plurality of types of electrical steel sheets, the method according to the present invention can be used to make a more accurate evaluation than the conventional method.

The figure which shows a magnetic flux density waveform and the magnetostriction displacement waveform corresponding to it. The figure which shows a three-phase three-legged iron core. The figure which shows the magnetic flux density waveform measured in the position of A, B, C of the iron core of FIG. It shows the distribution of A ₃ / A ₁ 3 Sagami mind actually measured number of magnetic flux density waveform. The figure which shows distribution of (theta) of many magnetic flux density waveforms measured with the three-phase iron core. The figure which shows the estimation method of the magnetostriction displacement waveform which generate | occur | produces with arbitrary magnetic flux density waveforms. The figure which shows a three-phase five-legged iron core.

  In order to carry out the present invention, it is first necessary to know the magnetic flux density waveform of the local part of the transformer core. As one method for this purpose, there is a method in which two small-diameter holes are formed in a magnetic steel sheet at a local area of an actual iron core, a coil is formed therethrough through a lead wire, and an induced voltage waveform is measured to know a magnetic flux density waveform. . As another method, there is a method of knowing the magnetic flux density waveform of the iron core part by applying electromagnetic field numerical analysis based on the finite element method or the like to the iron core.

Alternatively, using Equation 1, the amplitude ratio A ₃ / A ₁ is set from one point to a plurality of points in the range of 0.05 to 0.2, and the phase difference θ is set from one point to a plurality of points in the range of 50 ° to 70 °. A magnetic flux density waveform may be created. In this method, it is also necessary to determine the maximum magnetic flux density of the waveform, but it is desirable to set it to 1 to 1.15 times the average magnetic flux density of the entire iron core from the knowledge so far.

  In order to obtain a magnetostriction waveform when excited by the magnetic flux density waveform obtained by these methods, there is a method using a magnetostriction measuring apparatus shown in FIG. Specifically, an excitation power source for the apparatus that outputs set digital waveform data as a voltage waveform is used. As the waveform data used there, data obtained by converting the magnetic flux density waveform obtained by the above method into a voltage waveform by differentiation is used. By this method, a magnetostrictive waveform is obtained with the magnetic flux density waveform generated locally at the transformer core as an excitation condition.

  Moreover, an estimation method can be used as another method for obtaining a magnetostrictive waveform based on a magnetic flux density waveform of the transformer core local area. As a method for expressing a magnetostrictive waveform, there is a so-called butterfly loop in which a Lissajous figure is drawn by taking a magnetic flux density on the horizontal axis and a magnetostrictive displacement on the vertical axis. If the maximum magnetic flux density is the same, the shape of the butterfly loop does not change greatly even if the magnetic flux density waveform changes. Therefore, an arbitrary magnetic flux as shown in FIG. 6 is used by using a butterfly loop measured by excitation with a sine wave. The magnetostrictive displacement waveform can be obtained by tracing the butterfly loop according to the time change of the density waveform.

The magnetostrictive waveform obtained by the above-described actual measurement method or estimation method may be evaluated by its amplitude, but more preferably, the magnetostrictive waveform is converted to a value close to the noise level by using the method disclosed in Patent Document 2. It is a method to do.
In advance, if the actual transformer noise data corresponding to the magnetostriction evaluation value obtained by the above method is accumulated and the correlation between them is obtained quantitatively, the above method is applied to any electrical steel sheet. Thus, the transformer noise can be accurately estimated. In addition, it is possible to determine which one of the plurality of types of electromagnetic steel sheets will result in the lowest transformer noise.

As an example of the present invention, a three-phase three-legged iron core shown in FIG. While measuring the noise level of each iron core, a single plate sample is produced from each material, and magnetostriction is measured by the method of Patent Document 1. These measurements were performed at a magnetic flux density of 1.5T and 1.7T.
First, magnetostriction was measured under sinusoidal excitation conditions, and the method of Patent Document 2 was applied to the result. As shown in Table 2, the magnitude between the materials X and Y is the magnitude of the measured noise level of the transformer. The situation was reversed and correct evaluation was not possible. Next, magnetostriction was measured using the three types of magnetic flux density waveforms shown in FIG. 3 measured at positions A, B, and C in FIG. 2, and the method shown in the present invention was applied. The magnitude between the materials in the result coincides with the actually measured noise level, which shows the effectiveness of the present invention.

As an example of the present invention, a three-phase five-legged iron core shown in FIG. While measuring the noise level of each iron core, a single plate sample is produced from each material, and magnetostriction is measured by the method of Patent Document 1. These measurements were performed at a magnetic flux density of 1.5T and 1.7T.
First, the magnetostriction was measured under sinusoidal excitation conditions, and the method of Patent Document 2 was applied to the result. As shown in Table 3, the magnitude between the materials P and Q was reversed from the measured noise level of the transformer. As a result, correct evaluation was impossible. Next, magnetostriction was measured using two types of magnetic flux density waveforms measured at positions D and E in FIG. 7, and the method shown in the present invention was applied. The magnitude between the materials in the result coincides with the actually measured noise level, which shows the effectiveness of the present invention.

As an example of the present invention, a three-phase three-legged iron core shown in FIG. While measuring the noise level of each iron core, a single plate sample is produced from each material, and magnetostriction is measured by the method of Patent Document 1. These measurements were performed at a magnetic flux density of 1.5T and 1.7T.
First, magnetostriction was measured under sinusoidal excitation conditions, and the method of Patent Document 2 was applied to the result. As shown in Table 4, the magnitude between the materials X and Y is the magnitude of the measured noise level of the transformer. The situation was reversed and correct evaluation was not possible.

Next, a method of setting a magnetic flux density waveform using Equation 1 as an application example of the present invention was used. Each coefficient of Equation ₁ was set such that A ₁ was 1.917 T, A ₃ was 0.1917 T, and θ was 60 ° at a magnetic flux density of 1.5 T. Under these conditions, A ₃ / A ₁ is 0.1, and the maximum magnetic flux density is 1.725 T. At a magnetic flux density of 1.7 T, A ₁ was 2.25 T, A ₃ was 0.45 T, and θ was 60 °. Under these conditions, A ₃ / A ₁ is 0.2, and the maximum magnetic flux density is 1.8T.
The magnetic flux density waveform was calculated using them and used for excitation to measure magnetostriction. The magnitude between the materials in the result coincides with the actually measured noise level, which shows the effectiveness of the present invention.

Claims (1)

The local magnetic flux density waveform in the three-phase transformer iron core when excited by a sine wave, the amplitude ratio A ₃ / A ₁ in the following equation ₁ is in the range of 0.05 to 0.2, and the phase angle θ is 50 Each is set within the range of ° (0.873 rad) to 70 ° (1.222 rad) , estimated by calculation using the following formula 1 , and the magnetic flux density waveform is controlled to the excitation voltage waveform of the magnetostriction measuring device. is generated in the electromagnetic steel samples by, by actually measuring the magnetostriction at that time, and wherein the TURMERIC line evaluation of the magnetostrictive or noise of the three-phase transformers electrical steel sheet, the three-phase transformer core for which is excited by a sine wave Evaluation method of magnetostriction or noise of electrical steel sheet.
In Equation 1, b is the magnetic flux density waveform (T), f is the excitation frequency (Hz), t is the time (sec), A ₁ and A ₃ are the amplitudes (T) of the fundamental wave and the third harmonic wave, θ Represents a phase angle (rad).
b = A ₁ cos (2πft) + A ₃ cos {3 (2πft + θ)} (1)

Images