Transformer vibration reduction method based on particle damping

Abstract

The invention relates to a transformer vibration reduction method based on particle damping, which comprises the following steps of: finding out the maximum vibration deformation position of a transformer by using modal analysis, arranging a hole-shaped cavity on a radiating fin of the transformer, and filling particles in the hole-shaped cavity; the vibration reduction amplitude values corresponding to the multiple sets of design parameters are obtained through simulation, iteration is conducted on the design parameters by establishing a linear regression model, and the design parameters with the optimal vibration reduction effect are obtained on the premise that the structural rigidity is met; the hole-shaped cavities in the cooling fins extend from the sides, away from the transformer, of the cooling fins to the transformer, and the cooling effect of the cooling fins cannot be affected. A discrete element method and a finite element method are used for conducting simulation analysis on various particle dampers of different design sizes. And after a result is obtained, establishing a regression model by utilizing data obtained by simulation, optimizing parameters, and carrying out iterative simulation to obtain an optimal scheme.

Description

Technical Field

[0001] The invention relates to the technical field of vibration reduction and noise reduction, and in particular to a transformer vibration reduction method based on particle damping. Background Art

[0002] As an indispensable part of the power system, the transformer's working characteristics cause it to generate vibration noise during operation. Transformer noise is mainly caused by the vibration of the transformer body, and its root lies in the iron core and winding. The reason why the iron core generates noise is that the silicon steel sheets that constitute the iron core produce tiny size changes due to the magnetostrictive effect under the action of the alternating magnetic field. Magnetostriction causes the iron core to vibrate periodically with the change of the excitation frequency. In the transformer, the vibration generated by the iron core and winding is transmitted to the oil tank through the transformer oil and the body support and positioning components, and is superimposed with the vibration noise generated by the cooling system to radiate noise to the outside world.

[0003] At present, the vibration reduction methods for transformer boxes include installing soundproof covers, achieving noise reduction through soundproof materials, or installing vibration reduction and noise reduction equipment on the transformer. In the prior art, sound-absorbing coating materials are used to absorb the noise that spreads outward, and rubber vibration-damping pads are used between the iron core and the supporting structure, and vibration absorbers are installed on the surface of the transformer box. The above methods can suppress or prevent the propagation of vibration noise, but covering the transformer box with damping materials or installing other vibration-damping devices will affect the heat dissipation of the transformer to a greater or lesser extent and the engineering workload is large. There are also problems such as complex structure, high processing costs and poor durability.

[0004] Particle damping technology is a nonlinear passive vibration reduction technology that reduces vibration by filling damping particles in an additional cavity or reserving appropriate space inside the vibration reduction structure. As the vibration reduction body vibrates, particles in the cavity collide and rub against each other and against the inner wall of the cavity, thereby consuming vibration energy and achieving the effect of vibration reduction. In addition, particle dampers have the characteristics of simple structure, easy installation, no change to the original structure, adaptability to harsh environments and high economy.

[0005] For example, the Chinese patent with publication number CN113936902A discloses a method for reducing vibration and noise of a power equipment box based on particle damping, in which at least one hollow wall with an opening is opened in the power equipment box; particles and oil are filled in the hollow wall; the vibration and noise of the power equipment box can be effectively reduced;

[0006] For example, the Chinese patent with publication number CN117390915A discloses a pipeline vibration reduction design method and system based on a mixed-size particle damper, which performs three-dimensional modeling of the pipeline, uses finite element software to perform modal and static analysis on the pipeline model, obtains the maximum deformation position of the pipeline, and determines the installation position of the mixed-size particle damper; designs the shape parameters of the mixed-size particle damper based on the pipeline size; and determines the optimal mixed particle group of the mixed-size particle damper;

[0007] The transformer box is the only way for vibration noise to be transmitted to the outside world. The radiator fins are installed outside the transformer box. The vibration noise of the transformer box will eventually act on the radiator fins. The design and installation method of the fins directly affect the heat conduction efficiency between them and the box. Therefore, it is necessary to design particle damping for vibration reduction while maintaining the heat dissipation effect and stiffness of the heat sink fins. Summary of the invention

[0008] In view of the deficiencies in the prior art, the present invention provides a transformer vibration reduction method and a construction method based on particle damping to solve the above problems.

[0009] The present invention provides the following technical solutions:

[0010] A transformer vibration reduction method based on particle damping comprises the following steps:

[0011] Carry out 3D modeling of the transformer, import the 3D model into the finite element software, use modal analysis to find the maximum vibration deformation position of the transformer, and determine the opening position of the hole-shaped cavity;

[0012] The hole-shaped chamber is opened on the heat dissipation fins of the transformer, and particles are filled in the hole-shaped chamber;

[0013] Apply sinusoidal excitation to the transformer by combining discrete element and multi-body dynamics software co-simulation;

[0014] The vibration reduction amplitudes corresponding to the multiple sets of design parameters are obtained through simulation, wherein the design parameters include the diameter, depth and spacing of the hole-shaped chambers, and the diameter of the particles;

[0015] By establishing a linear regression model to iterate the design parameters, the design parameters with the best vibration reduction effect are obtained while meeting the structural stiffness.

[0016] Preferably, the parameters of the hole-shaped chamber are a diameter of 2.8-3.2 mm, a depth of 176-185 mm, and a spacing of 48-50 mm.

[0017] Preferably, the particles are spherical particles with a diameter of 1.5-2.5 mm.

[0018] Preferably, the particles are particles that are not affected by a magnetic field, and the material types include any one of tungsten carbide, zirconium oxide, and aluminum alloy.

[0019] Preferably, the particle filling rate in the hole-shaped chamber is 70%-90%.

[0020] Preferably, the hole-shaped chamber is sealed by a bolt column.

[0021] Preferably, a 100 Hz sinusoidal excitation is applied to the transformer by combining discrete element and multi-body dynamics software for joint simulation.

[0022] Preferably, the finite element method is used to calculate whether the modified heat sink fins meet the structural stiffness during operation.

[0023] Preferably, the hole-shaped chamber extends from a side of the heat dissipating fin facing away from the transformer toward the transformer.

[0024] Preferably, the linear regression model is expressed as:

[0025]

[0026] in, is the predicted value, X is the feature matrix, and β is the regression coefficient;

[0027] The loss coefficient of the weighted linear regression model is expressed as:

[0028]

[0029] Where n is the number of samples, ω is the weight of the i-th sample, yi is the target value, X i is the i-th row of the feature matrix.

[0030] The present invention has the following beneficial technical effects:

[0031] The present invention designs the transformer fins to reduce vibration and noise. The heat dissipation fins installed on the transformer box are used to drill holes inside the fins to fill particles. The damping particles have non-magnetic properties and are not affected by the magnetic field around the transformer. The friction and collision energy between particles and walls and between particles is used to reduce the vibration energy transmitted by the transformer to the outside world, thereby achieving the purpose of reducing vibration and noise.

[0032] The present invention reduces vibration and noise of the transformer without changing the main structure of the transformer. Compared with the traditional method, it does not require much change to the transformer body, reduces the engineering workload, and reduces the cost of noise and vibration reduction.

[0033] Extending the hole-shaped chamber on the heat sink fin from the side of the heat sink fin away from the transformer to the transformer will not reduce the substantial heat dissipation area of ​​the heat sink fin; at the same time, there will be no uneven airflow distribution or local heat accumulation due to inappropriate hole openings, and since the particles are made of non-magnetic conductive material, the particle temperature will not increase due to the magnetorheological effect around the transformer.

[0034] The discrete element method and finite element method were used to simulate and analyze particle dampers of various design sizes. After obtaining the results, the data obtained from the simulation were used to establish a regression model, optimize parameters, and iterate the simulation to obtain the optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of the present invention;

[0036] Figure 2 It is a cross-sectional mark of the hole-shaped cavity processed by the heat dissipation fin of the transformer of the present invention;

[0037] Figure 3 It is a simulation iteration flow chart of the present invention;

[0038] Figure 4 The present invention corresponds to Figure 2 Three views of the chamber with section markings;

[0039] Figure 5 is a kinetic energy diagram of particles at different filling rates in an example of the present invention;

[0040] Figure 6 It is a particle energy consumption diagram under different particle materials in the example of the present invention.

[0041] 1. Transformer; 2. Heat sink fin; 4. Hole-shaped chamber; 5. Particles; 6. Stud. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] A transformer vibration reduction method based on particle damping, such as Figure 1-Figure 6 As shown: The method comprises the following steps:

[0044] S1. A three-dimensional model is established based on the dimensions of the transformer 1 box, and the model is imported into the finite element software for modal analysis to find out the maximum deformation of the transformer 1 and mark the relevant heat sink fins 2;

[0045] S2. For the marked heat sink fin 2, a hole-shaped cavity 4 is obtained by drilling. Holes are drilled on the heat sink fin 2 and filled with particles 5 to reduce the vibration and noise of the transformer. However, since the heat sink fin 2 is embedded in the surface of the transformer 1 box, it is similar to a cantilever beam system. It can be seen that the vibration at the end of the heat sink fin 2 is the largest. Figure 2 As shown in the direction of the BB arrow in the figure, the hole-shaped chamber 4 extends straightly from the end side of the heat sink fin 2 (the side away from the transformer 1) to the side of the transformer 1, thereby maximizing the vibration reduction effect; the particles 5 with a diameter of 2 mm are filled, and the filling rate of the particles 5 has a great influence on the vibration reduction effect. The optimal filling rate needs to be determined, and finally the opening of the hole-shaped chamber 4 is sealed with the stud 6.

[0046] There are some technical difficulties in applying particle damping to transformer box fin vibration reduction. First, it is necessary to process the hole-shaped chamber 4 additionally to determine the specific design of the hole-shaped chamber 4, including the hole diameter, hole depth and hole spacing of the arrangement. The size of the hole spacing will not only affect the structure of the heat sink fin 2, but also affect the particle damping vibration reduction effect. When the spacing is too large, the particle vibration reduction effect is not obvious, and when it is too small, it will have a greater impact on the stiffness of the heat sink fin 2. The processing depth of the hole also has a great influence on the vibration reduction effect of the particle damper. The second is to determine the parameters of the filled particles 5. Since some metal particles will be disturbed by the magnetic field of the transformer, the particles 5 need to be made of materials with high density and non-magnetic conductivity. In addition, the specific pore size, material and filling rate of the particles 5 in the hole-shaped chamber 4 need to be determined. Different particles 5 have different material densities. The size of the density of the particles 5 determines the amount of energy that the particles 5 can carry when they are stimulated to move. Therefore, it is necessary to determine the specific material of the filled particles 5 through simulation. When the filling rate of particles 5 is small, particles 5 do not collide sufficiently with the hole-shaped chamber 4 during vibration. If the filling rate is too large, the space for particles 5 to move in the hole-shaped chamber 4 will be reduced, thereby reducing the momentum exchange and collision energy consumption between particles 5 and the structure.

[0047] The pore size, pore depth and pore spacing of the pore chamber 4, and the pore size and filling rate of the particles 5 have different weights on the final vibration reduction effect. Therefore, a linear regression model with weights is introduced.

[0048] The opening of the heat sink fin 2 will affect its structural stiffness. It is necessary to consider whether the heat sink fin 2 after modification meets its structural stiffness during operation. The cross-sectional inertia after punching is complicated and cumbersome to calculate due to the complexity of the modification, so the finite element method is used for calculation. If it does not meet the structural stiffness standard of the heat sink fin 2, it is necessary to ensure that the vibration reduction requirements of the transformer 1 box are met, and the size of the hole cavity 4 is recalculated. The hole depth or hole diameter or hole spacing of the hole cavity 4, as well as the hole diameter or filling rate of the particles 5 are appropriately changed, as long as the vibration reduction requirements are met.

[0049] S3. Apply 100 Hz sinusoidal excitation to the transformer 1 by combining discrete element and multi-body dynamics software for joint simulation to simulate the vibration of the transformer 1.

[0050] S4. Data iteration and convergence lead to the optimal particle damper design size / size range.

[0051] After in-depth analysis using discrete element and multi-body dynamics software, the vibration reduction results are obtained. Table 1 shows a training data set of particles 5 with a diameter of 2 mm and a filling rate of 60%;

[0052] Table 1 Training dataset

[0053] Drilling depth / mm Aperture / mm Hole spacing / mm Vibration reduction effect on heat sink fins / % 140 2.4 20 42.5 150 2.6 40 48.3 160 2.8 60 46.3 170 3.0 80 47.7 180 3.2 100 49.6 190 3.4 120 46.2

[0054] Repeat the above steps to simulate and obtain the training data corresponding to different particle 5 diameters, filling rates, drilling depths, hole diameters and hole spacings of different hole-shaped chambers 4. Calculate the structural stiffness of each of the modified heat sink fins 2 using the finite element method. Set the minimum structural stiffness of the heat sink fin 2 according to the design requirements.

[0055] The linear regression model is trained using matlab with multiple sets of training data as input, parameters are optimized with the best vibration reduction effect as the goal, and the optimal design parameters are obtained through iterative simulation convergence; in this process, the structural stiffness of the heat sink 2 is limited to be greater than the set value. The drilling depth of the hole-shaped chamber 4 does not exceed the height of the heat sink 2, the hole diameter does not exceed the width of the heat sink 2, and the hole spacing is less than the length of the heat sink 2; the outer diameter of the particle 5 is less than the hole diameter of the hole-shaped chamber 4.

[0056] Linear Regression Model:

[0057]

[0058] in, is the predicted value, X is the feature matrix, and β is the regression coefficient.

[0059] The weight of a sample may change as other factors change. By dynamically adjusting the weight, the model can be more adaptable to such changes, improving the model's prediction accuracy and generalization ability.

[0060] The loss coefficient of the weighted linear regression model is expressed as:

[0061]

[0062] Where n is the number of samples, ω is the weight of the i-th sample, and y i is the target value, X i is the i-th row of the feature matrix.

[0063] The filling rate can also be determined by comparing the particle kinetic energy when the filling rate increases from 60% to 100% to determine the optimal particle filling rate.

[0064] In order to reduce calculations, after calculating the optimal design parameters of the heat sink fins 2, the finite element method can be used to calculate whether the modified heat sink fins 2 meet the structural stiffness requirements during operation.

[0065] The energy loss of particles when filled with three materials that are not affected by the magnetic field, namely tungsten oxide, zirconium oxide and aluminum alloy, is obtained by using discrete element simulation to determine the material of the particles, and the opening of the hole-shaped chamber 4 is sealed with a stud 6;

[0066] The force on the particles 5 in the hole-shaped chamber 4 of the heat dissipation fin 2 is as follows:

[0067] Normal force F n is the normal overlap δ n The function is obtained by the following formula:

[0068]

[0069] Where: E* and R* represent the equivalent elastic modulus and equivalent contact radius respectively. Where E i , ν i , R i and E j , ν j , R j Respectively represent the Young's modulus, Poisson's ratio and radius of particles i and j that are in contact with each other.

[0070] The normal and tangential damping forces are:

[0071]

[0072] in, is the damping ratio, is the normal stiffness, is the tangential stiffness (where G* is the equivalent shear modulus, and G i , G j are the shear moduli of the contact particles. is equivalent mass.

[0073] Embodiment 1:

[0074] Take a 10kV transformer fin as an example. The heat sink fin 2 of this model is 550mm long, 6mm wide and 232mm high. A row of hole-shaped chambers 4 are bored in the heat sink fin 2 with a width of 6mm. The hole-shaped chamber 4 has a diameter of 3mm, a depth of 180mm and a hole spacing of 50mm. There are 9 hole-shaped chambers 4 in a heat sink fin 2. When the vibration transmitted from the inside of the transformer 1 acts on the heat sink fin 2, the collision and friction of the particles 5 inside the heat sink fin 2 dissipate the vibration energy generated by the operation of the transformer 1. The particles 5 are made of tungsten oxide material and are not affected by the magnetorheological effect of the transformer 1 leaking to the surrounding area to generate additional heat. The tungsten oxide material is heat-resistant and can stably function under high temperature conditions even if the temperature rises due to the heat generated by the friction and collision of the particles.

[0075] The discrete element method is used to simulate and analyze the particles 5 in the heat sink fin 2. When the vibration frequency is 100 Hz, the particle kinetic energy of the hole-shaped cavity 4 at different particle filling rates is obtained as follows: Figure 5 As shown, it can be seen that when the filling rate is 70%-90%, the particles have greater kinetic energy, among which the filling rate of 80% is the best, so the filling rate of one embodiment of the heat sink fin 2 of the present invention is 80%. Under the premise of filling rate of 80%, the energy loss diagrams of the three materials are as follows Figure 5 As shown, it can be found that the energy dissipation effect is best when filled with tungsten carbide particles.

[0076] In this example, a stud 6 is used to seal the outlet of the hole-shaped chamber 4 to prevent the particles 5 from leaking out. The diameter of the particles 5 is 2 mm, and the particle filling rate is 80%.

[0077] Table 2 Vibration reduction rate under different particle sizes

[0078] Particle size / mm Vibration reduction rate / % 1 30.5 1.5 45.3 2 58.9 2.5 47.5 3 34.2

[0079] Embodiment 2:

[0080] On the basis of the first embodiment: the simulation is performed with the diameter of the hole-shaped chamber 4 as the only variable, and it is found that the vibration reduction rate is better when the diameter of the hole-shaped chamber 4 is 2.8-3.2 mm.

[0081] Embodiment three:

[0082] On the basis of the first embodiment: the simulation is performed with the depth of the hole-shaped chamber 4 as the only variable, and it is found that the vibration reduction rate is better when the depth of the hole-shaped chamber 4 is 176-185 mm.

[0083] Embodiment 4:

[0084] On the basis of the first embodiment: the simulation is performed with the spacing between the hole-shaped chambers 4 as the only variable, and it is found that the vibration reduction rate is better when the spacing between the hole-shaped chambers 4 is 48-51 mm.

[0085] The above-mentioned embodiments only express the specific implementation of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A transformer vibration reduction method based on particle damping, characterized in that: The following steps are involved: Carry out 3D modeling of the transformer, import the 3D model into the finite element software, use modal analysis to find the maximum vibration deformation position of the transformer, and determine the opening position of the hole-shaped cavity; The hole-shaped chamber is opened on the heat dissipation fins of the transformer, and particles are filled in the hole-shaped chamber; Apply sinusoidal excitation to the transformer by combining discrete element and multi-body dynamics software co-simulation; The vibration reduction amplitudes corresponding to the multiple sets of design parameters are obtained through simulation, wherein the design parameters include the diameter, depth and spacing of the hole-shaped chambers, and the diameter of the particles; By establishing a linear regression model to iterate the design parameters, the design parameters with the best vibration reduction effect are obtained while meeting the structural stiffness.

2. A transformer vibration reduction method based on particle damping according to claim 1, characterized in that: The parameters of the hole-type chamber are diameter 2.8-3.2 mm, depth 176-185 mm and spacing 48-51 mm.

3. A transformer vibration reduction method based on particle damping according to claim 2, characterized in that: The particles are spherical particles with a diameter of 1.5-2.5 mm.

4. The transformer vibration reduction method based on particle damping according to claim 3 is characterized in that: The particles are particles that are not affected by a magnetic field, and the material types include any one of tungsten carbide, zirconium oxide, and aluminum alloy.

5. The transformer vibration reduction method based on particle damping according to claim 3 is characterized in that: The particle filling rate in the hole-shaped cavity is 70%-90%.

6. The transformer vibration reduction method based on particle damping according to claim 1, characterized in that: The hole-type chamber is sealed by bolt columns.

7. The transformer vibration reduction method based on particle damping according to claim 1, characterized in that: A 100Hz sinusoidal excitation is applied to the transformer by combining discrete element and multi-body dynamics software for joint simulation.

8. The transformer vibration reduction method based on particle damping according to claim 1, characterized in that: The finite element method is used to calculate whether the modified heat sink fins meet their operating structural stiffness.

9. The transformer vibration reduction method based on particle damping according to claim 1, characterized in that: The hole-shaped chamber extends from a side of the heat dissipation fin away from the transformer toward the transformer.

10. The transformer vibration reduction method based on particle damping according to claim 1, characterized in that: The linear regression model is expressed as: in, is the predicted value, X is the feature matrix, and β is the regression coefficient; The loss coefficient of the weighted linear regression model is expressed as: Where n is the number of samples, ω is the weight of the i-th sample, and y i is the target value, X i is the i-th row of the feature matrix.

Images