Transformer core loss calculation method and system considering variable loss coefficient

Abstract

The present invention proposes a transformer core loss calculation method and system considering a variable loss coefficient, comprising: considering the influence of high-frequency skin effect and dynamic hysteresis loop on core loss, establishing a variable coefficient loss calculation formula; substituting a normalization constant into the integral form of the average value of magnetic induction intensity within a cycle to obtain an equivalent frequency expression; based on the equivalent frequency expression and the variable coefficient loss calculation formula, obtaining a generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation, and calculating the core loss of a nanocrystalline high-frequency transformer based on the formula.

Description

Technical Field

[0001] The present invention belongs to the technical field of nanocrystalline high-frequency transformer core loss calculation, and in particular relates to a transformer core loss calculation method and system taking into account a variable loss coefficient. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid development of power electronics technology and the popularization of new materials such as nanocrystalline and amorphous alloys, high-frequency operation has become an inevitable trend in the development of power electronics devices. This high-frequency operation can improve overall efficiency and power density without affecting equipment operation, significantly reducing the weight and volume of magnetic components.

[0004] Nanocrystalline materials have small grain sizes, resulting in very low eddy current and hysteresis losses. Furthermore, they offer superior magnetic conductivity and high magnetization strength, making them an ideal material for high-frequency transformer cores. The excitation waveforms in power electronic devices are typically non-sinusoidal, contain high harmonic content, and operate at high frequencies.

[0005] Core loss increases significantly with the increase of operating frequency, greatly reducing the efficiency and reliability of the device.

[0006] Therefore, studying the core loss under non-sinusoidal high-frequency excitation and establishing an accurate and easy-to-calculate loss prediction model are of great significance for the optimization design of power electronic devices in terms of efficiency calculation, temperature rise prediction, etc.

[0007] Currently, there are three main methods for calculating core loss in engineering: 1) the hysteresis loss model based on physical phenomena, 2) the loss separation method based on the assumption that core losses can be separated, and 3) the empirical formula method based on fitting experimental data.

[0008] The hysteresis loss model is a method for calculating and analyzing the energy loss caused by hysteresis during repeated magnetization of ferromagnetic materials. The JA model and the Preisach model are two widely used mathematical models for simulating the hysteresis properties of magnetic materials. The JA model, proposed by David Jiles and Michael Atherton in the 1880s, is based on the theory of local equilibrium magnetization intensity and considers nonlinear and dissipative effects in the microscopic magnetization process. It can accurately simulate the hysteresis loop of ferromagnetic materials. However, the JA model makes certain simplifying assumptions during modeling, requiring the solution of a large number of differential equations in practical applications, increasing computational complexity. The Preisach model treats the hysteresis process as the superposition of a series of hysteresis operators. By adjusting the weights and switching values ​​of these hysteresis operators, complex hysteresis loops can be simulated. The Preisach model predicts the hysteresis loop and loss properties of ferromagnetic materials with high accuracy and can be extended to complex systems such as multiphase materials and composite materials. However, the Preisach model involves a large number of hysteresis operators and integral operations, resulting in slow computation and high complexity due to its reliance on extensive experimental data to determine the distribution function. These two models have many parameters that are interrelated, and the identification process is affected by many factors, so they are not applicable in engineering calculations.

[0009] The loss separation method is based on the different physical mechanisms of core loss and subdivides it into the following three categories: ① Hysteresis Losses ② Eddy-current Losses ③ Anomalous Losses. The core loss is the sum of these three, as shown in formula (1):

[0010] Ps=Ph+Pe+Pa (1)

[0011] Where P s is the total core loss, P h is the hysteresis loss, P e is the eddy current loss, P a Compared with the hysteresis model method, the loss separation method simplifies the iron loss analysis process and can more accurately calculate the various parts of the core loss, which helps to gain a deeper understanding of the mechanism and characteristics of the core loss. In high-frequency conditions, the classical eddy current loss is dominant, and the abnormal loss caused by the movement of the magnetic domain wall accounts for a small proportion and can be ignored. At this time, the loss separation model can be simplified to the Jordan model, that is,

[0012] Ps=Ph+Pe (2)

[0013] However, the loss separation model is rarely used in actual calculations because it involves many parameters and has a complex confirmation process.

[0014] The Original Steinmetz Equation (OSE) was proposed by German scholar CP Steinmetz in 1892. It is believed that the core loss is related to the excitation frequency and the peak value of the magnetic flux density, as shown in formula (3):

[0015]

[0016] Where, P v is the core loss density, W / m 3 , f is the excitation voltage frequency, Hz, and K, α, and β are the loss parameters under sinusoidal excitation, which are related to the core characteristics. Due to its simple form and relatively few design parameters, Equation (3) has gradually become the formula for calculating core loss in electromagnetic equipment such as transformers and motors. However, OSE can only calculate core loss under sinusoidal excitation. If it is applied to non-sinusoidal excitation, it must be corrected.

[0017] IGSE:

[0018] The Improved Generalized Steinmetz Equation (IGSE) states that core loss is not only related to the rate of change of magnetic induction intensity dB / dt, but also affected by the instantaneous value of magnetic induction intensity B(t), as shown in the formula:

[0019]

[0020] Where: ΔB is the peak-to-peak value of magnetic induction intensity within a magnetization period T, ΔB=B max -B min , B max 、B min are the maximum and minimum values ​​of the magnetic induction intensity in one magnetization cycle respectively. i The expression is as follows:

[0021]

[0022] Where θ is the integral angle. IGSE takes into account the magnetization process and magnetization history of the magnetic material and the influence of the magnetization reversal point on the hysteresis loop. It is suitable for the calculation of core loss under excitation containing local small hysteresis loops.

[0023] WcSE:

[0024] The Steinmetz waveform coefficient formula (WcSE) defines the waveform coefficient λ based on the ratio of the area enclosed by the magnetic induction intensity curve and the coordinate axis in a magnetization cycle under non-sinusoidal excitation to the area under sinusoidal excitation. Multiplying the waveform coefficient λ by OSE yields WcSE as shown in formula (6):

[0025]

[0026] WcSE corrects OSE based on the difference between the magnetic induction intensity curves under non-sinusoidal excitation and sinusoidal excitation. In principle, it is superior to the above methods. The greater the difference between the excitation voltage waveform and the sinusoidal waveform, the more the WcSE's advantage in accuracy is reflected.

[0027] Therefore, the hysteresis model method and loss separation method have clear principles and accurate calculations, but they require a large number of calculation parameters and the parameter identification process is complicated, making them unsuitable for practical engineering. The empirical formula method requires fewer parameters and is easy to use. It is currently the most popular calculation method in practical applications.

[0028] In addition, the inventors found in their research that as the frequency increases, the skin effect of the core becomes more obvious, and the magnetic flux density is no longer evenly distributed in the core. At high magnetic flux density, the core hysteresis loop changes, and the change in magnetic flux density causes the loss coefficient K to change accordingly. If the loss coefficient is regarded as a constant, the calculation error will gradually increase. Figure 1 The loss coefficient K is given as a function of the magnetic flux density B m From the relationship curve, it can be seen that there is a nonlinear relationship between K and B. Therefore, the current calculation method using a fixed loss coefficient cannot meet the high requirements of high-frequency transformer loss calculation, and the error is large.

[0029] Furthermore, the existing literature, "Improvement and Verification of the Calculation Method for High-Frequency Core Losses of Nanocrystalline Materials under Non-Sinusoidal Excitation," in the Journal of Electrical Engineering, Vol. 38, No. 5, March 2023, considers the influence of duty cycle, which can lead to significant deviations in the calculation results of actual high-frequency transformer losses. Furthermore, the literature proposes a correction factor to adjust the loss calculation process. Therefore, although this solution considers the duty cycle and introduces a correction factor, the selection process of the correction factor is subjective. Furthermore, the applicability of the correction factor is limited under complex operating conditions. Summary of the Invention

[0030] To overcome the above-mentioned deficiencies of the prior art, the present invention provides a transformer core loss calculation method taking into account a variable loss coefficient, which can effectively predict the core loss under high frequency and non-sinusoidal waveform excitation.

[0031] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0032] In a first aspect, a method for calculating transformer core loss taking into account a variable loss coefficient is disclosed, including:

[0033] Considering the influence of high-frequency skin effect and dynamic hysteresis loop on core loss, a variable coefficient loss calculation formula is established;

[0034] Substituting the normalization constant into the integral form of the average value of the magnetic induction intensity in one cycle, the equivalent frequency expression can be obtained;

[0035] Based on the equivalent frequency expression and the variable coefficient loss calculation formula, the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation is obtained, and the core loss of the nanocrystalline high-frequency transformer is calculated based on this formula.

[0036] As a further technical solution, when establishing the variable coefficient loss calculation formula, the loss coefficient K is changed with the magnetic flux density B. m The relationship curve is represented as a polynomial function:

[0037]

[0038] Where: D i is the coefficient of the loss coefficient K raised to the power of i, and K is obtained by fitting the measured values, where i = 0, 1, 2, 3, 4.

[0039] As a further technical solution, the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation is:

[0040]

[0041] Where Pv is the core loss density.

[0042] As a further technical solution, when the non-sinusoidal excitation is square wave excitation:

[0043] Obtain the function expression of the square wave excitation voltage in one cycle and the expression of the rate of change of the magnetic induction intensity in one cycle;

[0044] Substituting the expression of the rate of change of magnetic induction intensity within one cycle into the equivalent frequency expression, the equivalent frequency under square wave excitation can be obtained;

[0045] The square wave waveform coefficient can be obtained by calculating the ratio of the area enclosed by the magnetic induction intensity curve and the coordinate axis under square wave excitation to that under sinusoidal excitation;

[0046] Substituting the square wave waveform coefficient into the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation, the generalized Steinmetz waveform coefficient formula under square wave excitation is obtained.

[0047] As a further technical solution, the generalized Steinmetz waveform coefficient formula under square wave excitation is:

[0048]

[0049] As a further technical solution, when the non-sinusoidal excitation is triangular wave excitation:

[0050] Obtain the function expression of triangular wave excitation in one cycle;

[0051] Then obtain the equivalent frequency of the triangle wave excitation;

[0052] The waveform coefficient of the triangular wave can be obtained by calculating the ratio of the area enclosed by the magnetic induction intensity curve and the coordinate axis under triangular wave excitation to that under sinusoidal excitation;

[0053] Substituting the waveform coefficient of the triangular wave into the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation, the generalized Steinmetz waveform coefficient formula under triangular wave excitation is obtained.

[0054] As a further technical solution, the generalized Steinmetz waveform coefficient formula under triangular wave excitation is:

[0055]

[0056] In a second aspect, a transformer core loss calculation system considering a variable loss coefficient is disclosed, comprising:

[0057] The variable coefficient loss calculation formula establishment module is configured to: consider the influence of high-frequency skin effect and dynamic hysteresis loop on core loss, and establish the variable coefficient loss calculation formula;

[0058] The equivalent frequency expression calculation module is configured to: substitute the normalization constant into the integral form of the average value of the magnetic induction intensity within one period to obtain the equivalent frequency expression;

[0059] The transformer core loss calculation module is configured to obtain the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation based on the equivalent frequency expression and the variable coefficient loss calculation formula, and calculate the core loss of the nanocrystalline high-frequency transformer based on this formula.

[0060] One or more of the above technical solutions have the following beneficial effects:

[0061] This embodiment of the sub-technical solution conducts an in-depth analysis of the core factors affecting the core loss. Based on the Steinmetz waveform coefficient formula, it is proposed to use equivalent frequency and variable loss coefficient to improve the calculation accuracy of the Steinmetz waveform coefficient formula, build a high-frequency non-sinusoidal nanocrystal magnetic property measurement platform based on a full-bridge inverter circuit, and obtain the generalized Steinmetz waveform coefficient formula (Generalized Waveform coefficient Steinmetz Equation, GWcSE) under non-sinusoidal excitation through numerical fitting. The calculated results are compared with the measured values ​​to verify the accuracy of the proposed model. It can effectively predict the core loss under high frequency and non-sinusoidal waveform excitation, providing more accurate and comprehensive theoretical guidance for research and application in related fields.

[0062] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0064] Figure 1 Schematic diagram of the relationship between K and B;

[0065] Figure 2 Schematic diagram of square wave excitation waveform and corresponding magnetic flux density waveform;

[0066] Figure 3 Schematic diagram of the triangle wave excitation waveform and the corresponding magnetic flux density waveform;

[0067] Figure 4 Schematic diagram of non-sinusoidal experimental measurement system;

[0068] Figure 5 This is a comparison chart of core loss density under 5kHz frequency square wave excitation;

[0069] Figure 6 This is a comparison chart of core loss density under 5kHz frequency triangle wave excitation. DETAILED DESCRIPTION

[0070] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0071] It should be noted that the terms used herein are for describing particular embodiments only and are not intended to limit the exemplary embodiments according to the present invention.

[0072] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0073] Example 1

[0074] This embodiment discloses a method for calculating transformer core loss taking into account a variable loss coefficient, including:

[0075] Step 1: Consider the influence of high-frequency skin effect and dynamic hysteresis loop on core loss and establish a variable coefficient loss calculation formula;

[0076] Step 2: Substitute the normalization constant into the integral form of the average value of the magnetic induction intensity within one cycle to obtain the equivalent frequency expression;

[0077] Step 3: Based on the equivalent frequency expression and the variable coefficient loss calculation formula, the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation is obtained, and the core loss of the nanocrystalline high-frequency transformer is calculated based on this formula.

[0078] In step 1 of this embodiment: considering that high-frequency transformers have high requirements for loss calculation, a variable coefficient loss calculation formula is established here to consider the influence of high-frequency skin effect and dynamic hysteresis loop on core loss. Figure 1 The curve in can be characterized by the curve fitting tool as a 4-degree polynomial function, that is,

[0079]

[0080] Where: D i (i=0, 1, 2, 3, 4) is the coefficient of the i-th power of the loss coefficient K, and K is obtained by fitting the measured value.

[0081] The specific fitting process is:

[0082] Experimental measurement: First, by actually measuring the core loss data at different frequencies and different maximum magnetic flux densities, measurements are performed under multiple different frequency and magnetic flux density conditions to ensure that the obtained data points are sufficient.

[0083] Calculation model selection: Select the GWcSE calculation model based on the actual loss distribution. If more complex situations need to be considered, more high-order or nonlinear terms can be introduced.

[0084] Data fitting: The measured loss value P core , frequency f and maximum magnetic flux density B maxSubstitute it into the loss model formula and use the numerical fitting method (least square method) to determine the unknown coefficients.

[0085] Calculation error: Calculate the theoretical loss value using the fitted formula and compare it with the actual measured value to calculate the error. If the error is large, you need to readjust the model or add more test data points to improve the fitting accuracy.

[0086] Iterative optimization: By repeatedly adjusting the fitting method and loss model, the fitting results are optimized and the accurate coefficients are finally determined.

[0087] Among them, the least squares fitting process is specifically as follows:

[0088] The least squares method obtains the optimal coefficient by minimizing the sum of squares of the errors between the theoretical calculation value and the experimental measurement value. Assume that there are N sets of experimental data (B m,i ,P new,i ), then the fitting error of each set of data can be written as:

[0089]

[0090] The coefficients D4, D3, D2, D1, and D0 are fitted by minimizing the total sum of squared errors, i.e.:

[0091]

[0092] Solve the problem using the least squares method using numerical methods (linear regression or matrix methods). Express the problem in matrix form:

[0093]

[0094] The problem can be written as a matrix equation:

[0095] Y=XD

[0096] The least squares solution is of the form:

[0097] D=(X T X) -1 X T Y

[0098] The values ​​of each coefficient can be calculated using this formula.

[0099] After obtaining the fitting coefficients, substitute them into the original formula K(B m ), we can get the final core loss calculation formula. Then we can correct it by frequency f and calculate the actual core loss P core .

[0100] The calculation method used in this embodiment's sub-technical solution does not factor duty cycle into its analysis. The waveforms calculated in this embodiment's sub-technical solution are all based on symmetrical voltage excitation, without considering the impact of duty cycle on waveforms and losses. Considering that duty cycle can lead to significant deviations in actual high-frequency transformer loss calculations, this article makes reasonable simplifications to the duty cycle.

[0101] Differences between Correction Factor and Equivalent Frequency: This embodiment's sub-technical solution introduces the concept of equivalent frequency for related loss calculations. While both correction factor and equivalent frequency are intended to compensate for or adjust losses, their calculation principles and mathematical derivations differ fundamentally.

[0102] The calculation accuracy of several steimize correction formulas under non-sinusoidal excitation was compared. Among the various correction formulas of SE, under triangular wave voltage excitation, the calculation results of MSE and IGSE are closer to the measured values, but the WcSE calculation curve is closest to the measured value curve. Under square wave voltage excitation, there will be a large deviation between the calculated values ​​of MSE and IGSE and the measured values. Under non-sinusoidal excitation, WcSE correction is more accurate than OSE correction, and the greater the difference between the excitation waveform and the sinusoidal waveform, the higher the correction accuracy of WcSE compared with other methods. An empirical formula with simple structure and high accuracy is needed to solve the problem of core loss prediction under non-sinusoidal excitation. WcSE is most suitable for calculating nanocrystalline core loss under non-sinusoidal voltage excitation. Therefore, this embodiment chooses to make improvements based on WcSE. At this time, the core calculation formula under non-sinusoidal excitation is:

[0103]

[0104] The core loss is directly related to the macroscopic remagnetization rate dM / dt, which is positively correlated with dB / dt. The empirical loss parameter frequency in formula (6) must be replaced by the physical loss parameter:

[0105]

[0106] Where B is the average value of magnetic induction intensity B in one cycle, ΔB=B max -B min , B max 、B min are the maximum and minimum values ​​of the magnetic induction intensity within a magnetization cycle, respectively.

[0107] The integral form of formula (9) is

[0108]

[0109] The normalization constant 2 / (ΔB·π 2 ) into formula (10) to obtain the equivalent frequency f eq for

[0110]

[0111] Substituting formula (11) into formula (8) yields the generalized Steinmetz waveform coefficient formula (GWcSE) under non-sinusoidal excitation:

[0112]

[0113] P v is the core loss density, λ is the waveform factor, K(B m ) is the variable loss coefficient, B m is the maximum value of magnetic induction density.

[0114] Square wave excitation waveform and corresponding magnetic flux density waveform are as follows Figure 2 The function expression of the square wave excitation voltage in one cycle is

[0115]

[0116] The expression of the rate of change of magnetic induction intensity within one cycle is:

[0117]

[0118] Substituting equation (14) into equation (11), the equivalent frequency under square wave excitation is:

[0119]

[0120] The triangle wave excitation waveform and the corresponding magnetic flux density waveform are as follows: Figure 3 As shown. The function expression of triangular wave excitation in one cycle is:

[0121]

[0122] Similarly, the equivalent frequency of triangular wave excitation is

[0123]

[0124] By calculating the ratio of the area enclosed by the magnetic induction intensity curve and the coordinate axis under square wave and triangle wave excitation to that under sinusoidal excitation, the waveform coefficient λ of the square wave and triangle wave can be obtained. squ ,λ tri They are:

[0125]

[0126]

[0127] Substituting Equations (20) and (21) into Equation (12), and substituting Equations (13) and (17) into Equations (4) and (6), respectively, we can obtain the GWcSE formula under non-sinusoidal excitation. The iron loss calculation formula under non-sinusoidal excitation is shown in Table 1.

[0128] Table 1

[0129]

[0130] The calculation requires the fitting values ​​of system parameters α, β and K as well as the frequency f and the maximum magnetic density B. m The measured value.

[0131] Verification example:

[0132] Analytical formula for iron loss calculation based on experimental measurement

[0133] To obtain the analytical expression for the Steinmetz empirical correction formula, sinusoidal experiments were conducted on ring-shaped nanocrystal cores. Using a laboratory soft magnetic material AC magnetic properties testing system, the core loss of the ring-shaped nanocrystals was measured under sinusoidal excitation at various flux densities (0.1 to 1.1 T) and frequencies (5 to 20 kHz). The measured core loss values ​​under sinusoidal excitation were fitted to the Steinmetz empirical formula, yielding the fitting parameters (see Table 2). Substituting these parameters into the formula in Table 1 yields analytical formulas for the various correction formulas under square and rectangular wave excitation.

[0134] Table 2

[0135]

[0136] Non-sinusoidal excitation core loss measurement and verification

[0137] Construction of experimental platform

[0138] This example builds a non-sinusoidal high-frequency magnetic ring dynamic magnetic characteristics test system as shown. Non-sinusoidal excitation signals such as square waves and triangle waves are generated by a DC source through a DSP-controlled inverter circuit. The silicon carbide full-bridge inverter circuit can output square waves and triangle waves with frequencies ranging from tens to hundreds of kHz. The DC blocking capacitor in the circuit can eliminate the DC component of the excitation power supply, avoid magnetic bias, and maintain the flatness of the waveform. The test platform is as follows: Figure 4 shown.

[0139] The waveform data of the primary current i1(t) and the secondary voltage u2(t) are collected by the voltage probe and the current probe, and the dynamic hysteresis loop of the magnetic ring is obtained by equations (22) and (23), and the core loss of the magnetic ring is obtained by equation (24).

[0140]

[0141] Where N1 is the number of turns on the primary side; N2 is the number of turns on the secondary side; S is the cross-sectional area of ​​the core; and l is the effective magnetic path length of the core.

[0142] Therefore, the core loss per unit volume can be calculated by integrating the area of ​​the BH loop and multiplying it by the frequency f:

[0143]

[0144] use Figure 4 The non-sinusoidal test system shown here performs no-load experiments on nanocrystalline materials to measure their core loss. DSP software controls the inverter circuit to output square and triangular waves with a frequency range of 1 to 20 kHz and a magnetic flux density range of 0.1 to 1.1 T. Simultaneously, a computer reads the actual primary current and secondary voltage in real time to obtain experimental data.

[0145] Comparison of Steinmetz modified formula under non-sinusoidal excitation

[0146] Through the sinusoidal experiment, we can get the empirical formula parameters k, α, β, and then we can get the calculation formula of OSE and its correction formula. For the parameter D in GWcSE, i By fitting the calculated values ​​with the experimental values, the parameter values ​​can be obtained. The loss measurement results of the ring nanocrystalline core under square wave and triangle wave excitation are compared with the loss prediction results calculated by the modified formula to verify the calculation accuracy of GWcSE.

[0147] The loss calculation comparison results of nanocrystalline core under square wave and triangle wave excitation are shown in the figure. Figure 5 and Figure 6 As shown, Figure 5 This is a comparison chart of nanocrystalline core under 5kHz square wave and different magnetic flux density. Figure 6 This is a comparison chart of nanocrystalline core under 5kHz triangle wave and different magnetic flux density. Figure 5It can be seen that the calculated values ​​of WcSE and IGSE at high magnetic flux density (above 0.7T) are relatively close to the experimental measured values, but the calculated values ​​at low magnetic flux density are very different from the measured values. At the same time, it can be found that the predicted calculated values ​​of square wave loss by WcSE and IGSE are very close, which also indirectly reflects that WcSE and IGSE no longer have good accuracy in predicting non-sinusoidal losses under high frequency conditions; and overall, the calculated values ​​of OSE are very different from the measured values, and the accuracy is lower than that of MSE and IGSE. Therefore, for the non-sinusoidal loss prediction formulas above 20kHz, OSE, IGSE and WcSE do not have good calculation accuracy. For GWcSE, the calculated values ​​are very close to the measured values ​​in the entire magnetic flux density range, which shows that when predicting non-sinusoidal losses, the effect of dB / dt on B cannot be ignored in order to conform to the form of the original Steinmetz empirical formula. m impact.

[0148] The loss comparison results under triangular wave excitation are as follows: Figure 6 As shown. Figure 6 It can be seen that as the magnetic flux density increases, the calculation accuracy of the core loss by OSE, IGSE, and WcSE continues to decline. Unlike the square wave, the calculated values ​​of WCSE and IGSE are no longer close, and their trends develop in two different directions: the calculated core loss value of WcSE under rectangular wave excitation is larger than the experimental value, and the increase rate continues to increase with the increase of magnetic flux density; while the calculated core loss value of IGSE under rectangular wave excitation is smaller than the experimental value. The error between the calculated value and the measured value of GWcSE is very small, and its accuracy is better than the above three methods.

[0149] By comparing various non-sinusoidal modified empirical formulas, starting from the loss principle and considering the influence of dB / dt on the loss, an improved generalized WcSE formula for high-frequency non-sinusoidal core loss calculation model is derived.

[0150] In this example, a magnetic property test system for soft magnetic materials with high-frequency non-sinusoidal excitation was built. The calculated values ​​of several empirical formulas and the new calculation model were compared and analyzed with the experimental measurement results, verifying the accuracy and practicality of the GWcSE formula given in the technical solution of the present invention.

[0151] Example 2

[0152] The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0153] Example 3

[0154] The purpose of this embodiment is to provide a computer-readable storage medium.

[0155] A computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the above method.

[0156] Example 4

[0157] The purpose of this embodiment is to provide a transformer core loss calculation system taking into account a variable loss coefficient, including:

[0158] The variable coefficient loss calculation formula establishment module is configured to: consider the influence of high-frequency skin effect and dynamic hysteresis loop on core loss, and establish the variable coefficient loss calculation formula;

[0159] The equivalent frequency expression calculation module is configured to: substitute the normalization constant into the integral form of the average value of the magnetic induction intensity within one period to obtain the equivalent frequency expression;

[0160] The transformer core loss calculation module is configured to obtain the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation based on the equivalent frequency expression and the variable coefficient loss calculation formula, and calculate the core loss of the nanocrystalline high-frequency transformer based on this formula.

[0161] Example 5

[0162] The purpose of this embodiment is to provide a computer program product containing instructions, which, when running on a computer, enables the computer to execute the methods and functions involved in any of the above embodiments.

[0163] The steps involved in the apparatus of the above embodiment correspond to those of the method embodiment 1. For detailed implementation, please refer to the relevant description of embodiment 1. The term "computer-readable storage medium" should be understood to mean a single medium or multiple media containing one or more instruction sets; it should also be understood to include any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and causing the processor to perform any method of the present invention.

[0164] Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computer device. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0165] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. The transformer core loss calculation method considering the variable loss coefficient is characterized by: include: Considering the influence of high-frequency skin effect and dynamic hysteresis loop on core loss, a variable coefficient loss calculation formula is established; When establishing the variable coefficient loss calculation formula, the loss coefficient K With magnetic flux density B m The relationship curve is represented as a polynomial function: Where: D i is the loss coefficient K of i The coefficient of the power is obtained by fitting the measured value K,i =0,1,2,3,4; Construct an equivalent frequency expression: ； Based on the equivalent frequency expression and the variable coefficient loss calculation formula, the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation is obtained, and the core loss of the nanocrystalline high-frequency transformer is calculated based on this formula. The generalized Steinmetz form factor formula under non-sinusoidal excitation is: Among them, P v is the core loss density; The generalized Steinmetz form factor formula under square wave excitation is: ； in, 、 is the loss parameter under sinusoidal excitation; The generalized Steinmetz waveform coefficient formula under triangular wave excitation is: ； in, 、 is the loss parameter under sinusoidal excitation; Fitted from the measured values K， The specific fitting process is: Experimental measurement: By actually measuring the core loss data at different frequencies and different maximum magnetic flux densities, measurements are performed under multiple different frequency and magnetic flux density conditions to ensure that sufficient data points are obtained; Calculation model selection: Select the GWcSE calculation model based on the actual loss distribution; Data fitting: The measured loss value P core ,frequency f and maximum magnetic flux density B m Substitute into the loss model formula and use numerical fitting method to determine the unknown coefficients; Calculation error: Calculate the theoretical loss value through the fitted formula and compare it with the actual measured value to calculate the error. If the error is large, you need to readjust the model or add more test data points to improve the fitting accuracy; Iterative optimization: By repeatedly adjusting the fitting method and loss model, the fitting results are optimized and the accurate coefficients are finally determined.

2. The transformer core loss calculation method considering variable loss coefficient according to claim 1 is characterized in that: When the non-sinusoidal excitation is square wave excitation: Obtain the function expression of the square wave excitation voltage in one cycle and the expression of the rate of change of the magnetic induction intensity in one cycle; Substituting the expression of the rate of change of magnetic induction intensity within one cycle into the equivalent frequency expression, the equivalent frequency under square wave excitation can be obtained; The square wave waveform coefficient can be obtained by calculating the ratio of the area enclosed by the magnetic induction intensity curve and the coordinate axis under square wave excitation to the area under sinusoidal excitation. Substituting the square wave waveform coefficient into the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation, the generalized Steinmetz waveform coefficient formula under square wave excitation is obtained.

3. The transformer core loss calculation method considering variable loss coefficient according to claim 1 is characterized in that: When the non-sinusoidal excitation is a triangle wave excitation: Obtain the function expression of triangular wave excitation in one cycle; Then the equivalent frequency of the triangle wave excitation is obtained; The waveform coefficient of the triangular wave can be obtained by calculating the ratio of the area enclosed by the magnetic induction intensity curve and the coordinate axis under triangular wave excitation to the area under sinusoidal excitation; Substituting the triangle wave waveform coefficient into the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation, the generalized Steinmetz waveform coefficient formula under triangle wave excitation is obtained.

4. A transformer core loss calculation system considering a variable loss coefficient based on the transformer core loss calculation method considering a variable loss coefficient according to any one of claims 1 to 3, characterized in that: include: The variable coefficient loss calculation formula establishment module is configured to: consider the influence of high-frequency skin effect and dynamic hysteresis loop on core loss, and establish the variable coefficient loss calculation formula; The equivalent frequency expression calculation module is configured to: construct an equivalent frequency expression; The transformer core loss calculation module is configured to obtain the generalized Steinmetz waveform coefficient formula under non-sinusoidal excitation based on the equivalent frequency expression and the variable coefficient loss calculation formula, and calculate the core loss of the nanocrystalline high-frequency transformer based on this formula.

5. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method described in any one of claims 1 to 3 are implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are performed.

Images