CLASSIFICATION OF THE NOISE BEHAVIOR OF FERROMAGENTIC MATERIALS

Description

FIELD OF THE INVENTION

[0001] The invention relates to a method for classifying the noise behavior of ferromagnetic materials, in particular of electrical steel used for the manufacture of an electrical machine, wherein a coil is subjected to alternating voltage and is arranged on the material to be classified in order to magnetize the material, wherein the magnetization current present in the coil is measured over time, and the square of the magnetization current is determined and used to determine the measure of the noise development.

[0002] The coil is positioned so close to the material to be classified that magnetization of the material is achieved. STATE OF THE ART

[0003] When ferromagnetic materials are exposed to a temporally varying magnetic field, a change in the material's length occurs, known as magnetostriction. If the ferromagnetic materials are used as laminations to manufacture a core for a power transformer, the change in length of the laminations induced by the alternating voltage causes noise, which is often undesirable.

[0004] Therefore, before being installed in an electrical machine, the electrical steel sheet is tested for its magnetostriction behavior. Magnetostriction can be determined using direct and indirect measurement methods.

[0005] Direct methods include measurements using strain gauges, capacitive sensors, and optical methods such as laser interferometers. These methods directly measure the change in length of the material during defined magnetic states. Indirect methods, such as the FMR (ferromagnetic resonance method), Becker-Kersten method, or SAMR (small-angle magnetization rotation) method, are based on the Villari effect. In contrast to direct methods, these methods apply mechanical stress in the form of compression or tension to the material, resulting in magnetostriction.

[0006] All direct and indirect methods require an additional measurement setup with sufficiently accurate components to determine magnetostriction. For example, the technical report IEC / TR 62581 Edition 1.0 1020-08 describes a direct optical method used with the standardized Epstein gage (IEC 60404-2) and single-sheet gage (IEC 60404-3) to determine magnetic properties.

[0007] The Epstein gauge, also known as an Epstein frame, is constructed similarly to a transformer. It consists of a uniformly distributed outer primary winding (also called the magnetizing winding) that encloses the inner secondary winding (the measuring winding). The strips of soft magnetic steel to be measured are placed in pairs, opposite each other, in the coils, creating four equally long legs with the same cross-section. By measuring the current in the primary winding and the voltage in the secondary winding, the magnetic field strength and the magnetic flux density in the samples can be determined.

[0008] In WO 2011 / 012159 A1, a classification of the noise emission of entire electrical steel coils is carried out by directly measuring the magnetostriction induced by an excitation winding using acceleration sensors.

[0009] The paper "Diagnosing Winding and Core Condition of Power Transformer by Vibration Signal Analysis" by Buyung Sofiarto Munier, et al., Condition monitoring and Diagnosis (CMD) 2012 International Conference On, IEEE, 23 September 2012, pages 429 to 432, presents a method for classifying the noise behavior of ferromagnetic materials.

[0010] Further prior art is US 8,849,587 B1. PRESENTATION OF THE INVENTION

[0011] It is an object of the invention to provide a method for classifying the noise behavior of ferromagnetic materials, in particular electrical steel sheets, which does not require a measuring setup for mechanical or optical quantities.

[0012] This object is achieved by a method according to claim 1. Further embodiments of the method are contained in the dependent claims.

[0013] Claim 1 comprises the step of applying alternating voltage to a coil and placing it on the material to be classified to magnetize it. According to the invention, the method comprises at least the following further steps: that the magnetizing current present in the coil is measured over time, and that the square of the magnetizing current is determined and used to determine the measure of noise generation.

[0014] It has been discovered that the magnetizing current of a coil correlates with magnetostriction. There is a linear relationship between the time course of the magnetizing current i0(t) squared and the time course of magnetostriction, i.e., the change in length λ(t), over wide ranges of the magnetic flux density of grain-oriented electrical steel: λ t ∼ i 0 t 2

[0015] A relationship between the peak-to-peak value of the magnetostriction λpp and the peak-to-peak value of the no-load current i0pp 2<, which depends on the magnetic flux density B, was also found: λpp = c * f B * i 0 pp 2

[0016] In this respect, in one embodiment of the invention, it can be provided that the peak-to-peak value of the temporal course of the square of the magnetizing current is used to determine the measure of the noise development.

[0017] This means that a measure of the magnetostriction or change in length can be derived solely from the magnetizing current and that, without an additional measuring setup with measuring sensors, a classification of electrical steel with regard to magnetostriction can be carried out using the magnetizing current.

[0018] Thus, it is sufficient to measure the magnetizing current, which is already done for measuring power loss using standardized methods, such as an Epstein frame or a single-sheet meter. The excitation winding of WO 2011 / 012159 A1 could also be used for the method according to the invention.

[0019] According to the invention, it is proposed that the primary winding of an Epstein measuring device is used as the coil.

[0020] Alternatively, the primary winding of a single-sheet measuring device can be used as the coil.

[0021] The method according to the invention does not require additional measuring devices such as strain gauges, laser interferometers, or sensors such as acceleration sensors. A complex measurement setup is therefore unnecessary.

[0022] By conducting comparative measurements for electrical steel sheets with a known relationship between the square of the magnetizing current and the magnetostriction, the factors c * f(B) can be determined, allowing the magnetostriction to be calculated from the magnetizing current measurements for unknown electrical steel sheets. From the known magnetostriction for known electrical steel sheets, a classification of noise generation can be made, meaning that a specific measure of noise generation can be assigned to a specific value of magnetostriction.

[0023] If the magnetizing current is then measured for an unknown electrical steel sheet and its square is determined, the magnetostriction can be calculated using the formulas given above and, based on the known relationships between magnetostriction and noise generation, a measure of the noise generation can be assigned to the calculated magnetostriction. SHORT DESCRIPTION OF THE FIGURE

[0024] To further explain the invention, reference is made in the following part of the description to the figure, from which further advantageous details and possible areas of application of the invention can be seen. The figure is to be understood as an example and is intended to illustrate the nature of the invention, but in no way restricts or even exhaustively represents it. It shows: Fig. 1a diagram showing the course of the square of the current and the magnetostriction over time. WAYS OF IMPLEMENTING THE INVENTION

[0025] To carry out a method, any coil connected to a voltage source and an ammeter can be used. For example, the excitation winding of WO 2011 / 012159 A1 can be used, which is connected to an ammeter. This would have the advantage that an entire coil of electrical steel, i.e., an entire coiled bundle of electrical steel, can be tested at once. According to the invention, an Epstein or single-sheet measuring device is used, where the primary winding is used.

[0026] Fig. 1 shows the curve of the square of the measured magnetizing current i0(t) in a coil (see dashed line), as well as the curve of the magnetostriction λ(t), see solid line, over time. This data can be obtained, for example, by measuring the current in the excitation winding of WO 2011 / 012159 A1 and by measuring the acceleration sensors of WO 2011 / 012159 A1. For better comparability, the value of the square of the magnetizing current is plotted, and both curves are scaled to a common peak-to-peak value with corresponding factors c*f(B).

[0027] From Fig. 1 it can be seen that both curves have a very similar course, whereby the periodicity of the curve is due to the frequency of the alternating voltage applied to the coil.

[0028] In the method according to the invention, the magnetizing current i0(t) would be measured, its square determined, and from this curve, the distance between the highest point, 0 in Fig. 1, and the lowest point, -4.4 in Fig. 1, would be determined. The distance here is 4.4. Based on comparative measurements, it might already have been determined, for example, that this distance should not be greater than 4 to avoid excessive noise.

[0029] However, one could also calculate the squared current value using the formula λpp = c * f B * i 0 pp 2 For known values ​​of c and f(B), calculate the magnetostriction λpp and compare this with known magnetostriction values ​​for which the associated noise generation is known. Then, a statement can be made as to whether the calculated magnetostriction value still results in acceptable noise generation because it is low, or whether the calculated magnetostriction value is associated with an unacceptably high magnetostriction.

Claims

1. Method for classifying the noise behaviour of ferromagnetic materials, in particular of electrical steel sheet which is used for the production of an electric machine, wherein an alternating voltage is applied to a coil and the coil is arranged on the material that is to be classified in order to magnetize the material, wherein - the magnetization current present in the coil is measured over time, and - the square of the magnetization current is determined and used to determine the extent of the noise development, wherein the primary winding of an Epstein measuring device or of a single-sheet measuring device is used as the coil.

2. Method according to Claim 1, wherein the peak-to-peak value of the temporal profile of the square of the magnetization current is used to determine the extent of the noise development.