A soft-switching converter conducted electromagnetic interference source spectrum prediction method

Abstract

The present application relates to a kind of soft-switching converter conducted electromagnetic interference source spectrum prediction method, comprising: obtaining the time-domain waveform of soft-switching converter switch tube drain-source voltage, using S-shaped wave respectively fitting the rising edge and falling edge of time-domain waveform, obtain the noise source time-domain waveform after fitting;Based on fitting waveform, extract rising time, falling time, rising initial time, falling initial time, switching frequency and the multiplicative factor obtained by continuous differentiation such as time-domain characteristic parameter;The preset frequency spectrum amplitude envelope curve inflection point prediction expression and frequency spectrum amplitude envelope curve prediction expression are substituted into time-domain characteristic parameter, and the noise source spectrum amplitude envelope curve and its inflection point are obtained by calculation.This application can be compatible with symmetric and asymmetric switch waveform, can truly reflect the actual spectrum characteristics of soft-switching converter noise source, with high prediction accuracy, fast operation speed, can be used to guide the optimization design of conducted electromagnetic interference suppression measures.

Description

Technical Field

[0001] This invention relates to the field of power electronics, and in particular to a method for predicting the spectral amplitude envelope curve and inflection point of the spectral amplitude envelope curve of a conducted electromagnetic interference source in a soft-switching converter. Background Technology

[0002] With the development of power electronics technology and the upgrading of switching devices, the switching frequency and power density of power electronic devices are constantly increasing, leading to increasingly serious electromagnetic interference problems. The switching devices in power electronic converters generate high voltage and current change rates during operation. Nodes with high-frequency potential transitions in the circuit will generate common-mode conducted interference through parasitic capacitance to ground, while the switching component in the converter's input current will cause differential-mode conducted interference. This electromagnetic noise can be conducted through the device ports, polluting not only the power grid but also affecting the normal operation of other equipment in the system.

[0003] Interference sources, coupling paths, and affected sources together constitute the three essential elements of electromagnetic compatibility (EMC) problems. The design and implementation of conducted EMC suppression measures mainly revolve around the interference source and the coupling path. The basic approaches include cutting off the coupling path and reducing interference source emissions, leading to various measures, among which EMI filters are the most widely used suppression method. When designing EMI filters, their insertion loss and cutoff frequency must be determined based on the noise spectrum and relevant EMC standards. Therefore, predicting the noise spectrum can aid in the design of conducted EMC suppression measures; furthermore, the prediction expression can reveal the relationship between the time-domain waveform and the noise spectrum, providing a basis for actively shaping the spectrum by adjusting time-domain characteristic parameters.

[0004] In power electronic converters, the changing drain-source voltage of the switching transistor is generally considered the primary source of conducted electromagnetic interference (EMI) noise. In hard-switching converters, the drain-source voltage of the switching transistor can be approximated as a trapezoidal wave. However, for converters implementing soft switching, the switching transistor achieves zero-voltage or zero-current switching under resonance, and the rising and falling edges of its drain-source voltage waveform are approximated as sine waves. Due to the different time-domain waveforms, their corresponding spectral contents also differ. N. Oswald et al., in their paper "Analysis of Shaped Pulse Transitions in Power Electronic Switching Waveforms for Reduced EMI Generation" published in IEEE Transactions on Industry Applications, Vol. 47, No. 5, pp. 2154-2165 in 2011, provided spectral prediction expressions for symmetrical conducted EMI noise sources in hard-switching and soft-switching converters, assuming that the rising and falling edges are symmetrical waveforms. However, in actual switching waveforms, due to differences in drive and the influence of parasitic parameters, the rising and falling edges are often asymmetrical, i.e., the rise time is not equal to the fall time. Currently, D. Han et al., in their article "Comparative Analysis on Conducted CMEMI Emission of Motor Drives: WBG Versus Si Devices" published in IEEE Transactions on Industrial Electronics, Vol. 64, No. 10, pp. 8353-8363 in 2017, have provided a spectral prediction expression for the asymmetric conducted electromagnetic interference noise source of hard-switching converters. Compared to symmetrical trapezoidal waves, the asymmetric trapezoidal wave spectrum has an additional inflection point f. c Correspondingly, an additional frequency band decreases at a rate of -20dB to 40dB / dec is added. However, there are currently no literature reports on the prediction of the conducted electromagnetic interference noise source spectrum for asymmetric soft-switching converters. This makes it impossible to quickly and accurately predict the noise spectrum using analytical expressions when designing conducted EMI suppression schemes for soft-switching converters, hindering effective noise assessment and parameter optimization in the initial design phase. Summary of the Invention

[0005] The purpose of this invention is to provide a method for predicting the spectrum of conducted electromagnetic interference sources in soft-switching converters, which is compatible with both symmetrical and asymmetrical switching waveforms and aims to more realistically reflect the actual spectral characteristics of noise sources in soft-switching converters within the frequency range of conducted electromagnetic interference (150kHz-30MHz).

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter includes the following steps:

[0008] S1. Obtain the time-domain waveform of the conducted electromagnetic interference source of the soft-switching converter;

[0009] S2. Based on the obtained time-domain waveform of the interference source, use an S-shaped wave to fit the rising and falling edges of the time-domain waveform of the interference source to obtain the fitted time-domain waveform of the noise source.

[0010] S3. Continuously differentiate the time-domain waveform of the noise source fitted in step S2 until a waveform composed of impulse functions appears. Determine the amplitudes of the four impulse functions within the rising edge duration as the multiplicative factor A. m The amplitudes of the four impulse functions during the duration of the falling edge are determined as the multiplicative factor B. m ;

[0011] S4. Based on the fitted time-domain waveform of the noise source in step S2, extract time-domain feature parameters, including rise time. Initial time of ascent descent time Initial descent time Switching frequency f s ;

[0012] S5. Substitute the extracted time-domain feature parameters into the preset prediction expressions for the spectrum amplitude envelope curve and the inflection point of the spectrum amplitude envelope curve of the soft-switching converter noise source to calculate the spectrum amplitude envelope curve and its inflection point of the soft-switching converter noise source, as follows:

[0013] Substitute the extracted time-domain feature parameters into the preset expression for predicting the inflection point of the spectrum amplitude envelope curve of the soft-switching converter noise source:

[0014]

[0015] Among them, f A1 f A2 f B1 f B2 All are inflection points;

[0016] Substitute the extracted time-domain feature parameters into the preset expression for predicting the amplitude envelope curve of the soft-switching converter noise source spectrum:

[0017]

[0018] in, The expression for the spectral amplitude envelope curve, f sHere, f is the switching frequency, and f is the current frequency value being calculated. A1 f A2 f B1 f B2 A is the inflection point calculated by formula (1). m With B m Multiplicative factor and The effects of the rising and falling edge time-domain characteristics on the spectrum are represented respectively, and their expressions are related to the inflection point f. A1 f A2 f B1 f B2 The expression needs to be selected based on the frequency band being calculated.

[0019] As a preferred technical solution of the present invention: in step S1

[0020] The time-domain waveform of the conducted electromagnetic interference source of the soft-switching converter was obtained through modeling and simulation, as follows:

[0021] Beforehand, the parasitic parameters of the components in the soft-switching converter are measured using analysis equipment, and a simulation model containing the parasitic parameters is established.

[0022] Input the measured parasitic parameters into the simulation software to perform time-domain transient simulation;

[0023] By directly observing the voltage and current waveforms of the switching transistors through simulation, the time-domain waveform of the conducted electromagnetic interference source in the soft-switching converter can be obtained.

[0024] As a preferred technical solution of the present invention: In step S2:

[0025] The S-shaped wave is composed of three smooth curves: a quadratic curve S1, a linear curve S2, and a quadratic curve S3.

[0026] As a preferred technical solution of the present invention: In step S3:

[0027] The multiplicative factor A m With B m The extraction process is as follows:

[0028] The fitted noise source time-domain waveform is continuously differentiated to obtain its first, second, and third-order differential waveforms. In the third-order differential waveform, four impulse functions appear within the rising edge duration, and their amplitudes are determined to be the multiplicative factor A. m Four impulse functions appear during the duration of the falling edge, and their amplitudes are determined to be the multiplicative factor B. m .

[0029] As a preferred technical solution of the present invention: In step S4:

[0030] The rise time The initial rise time is the total duration of the rising edge of the time-domain waveform of the noise source. The falling time is the duration of the quadratic curve S1 segment in the fitted S-shaped wave with rising edge. The total duration of the falling edge of the time-domain waveform of the noise source, the initial falling time The duration of the S1 segment of the quadratic curve in the S-shaped wave with the falling edge obtained by fitting.

[0031] As a preferred technical solution of the present invention: In step S5:

[0032] The spectral amplitude envelope curve prediction is a piecewise function based on four inflection point frequencies, and the segmentation points are determined by the four inflection point frequencies obtained from the spectral amplitude envelope curve inflection point prediction expression.

[0033] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0034] 1. This invention performs spectrum prediction based on asymmetrical drain-source voltage waveforms, fully considering the actual operating conditions where the rise time and fall time are not equal and the rise edge and fall edge shapes are asymmetrical. Compared with traditional prediction methods based on symmetrical waveforms, it can more realistically reflect the actual spectrum characteristics of the noise source of the soft-switching converter. In particular, it can accurately predict the spectrum inflection points and corresponding attenuation frequency bands added due to waveform asymmetry, which significantly improves the prediction accuracy.

[0035] 2. The calculation process of this invention is based on analytical expressions, which eliminates the need for complex numerical calculations and results in fast operation. Attached Figure Description

[0036] Figure 1 This is an overall flowchart of the present invention;

[0037] Figure 2 This is a schematic diagram of continuously differentiating the fitted time-domain waveform of the noise source in this invention;

[0038] Figure 3 This is a graph showing the prediction results of the spectral envelope curve in an example of the present invention. Detailed Implementation

[0039] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0040] like Figure 1 As shown, the present invention proposes a method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter, comprising the following steps:

[0041] S1. Obtain the time-domain waveform of the conducted electromagnetic interference source of the soft-switching converter;

[0042] This application obtains the time-domain waveform of the conducted electromagnetic interference source of the soft-switching converter through modeling and simulation, as follows:

[0043] Beforehand, the parasitic parameters of the components in the soft-switching converter are measured using analysis equipment, and a simulation model containing the parasitic parameters is established.

[0044] Input the measured parasitic parameters into the simulation software to perform time-domain transient simulation;

[0045] By directly observing the voltage and current waveforms of the switching transistors through simulation, the time-domain waveform of the conducted electromagnetic interference source in the soft-switching converter can be obtained.

[0046] S2. Based on the obtained time-domain waveform of the interference source, use an S-shaped wave to fit the rising and falling edges of the time-domain waveform of the interference source to obtain the fitted time-domain waveform of the noise source.

[0047] The S-shaped wave is composed of three smooth curves: a quadratic curve S1, a linear curve S2, and a quadratic curve S3.

[0048] S3. Continuously differentiate the time-domain waveform of the noise source fitted in step S2 until a waveform composed of impulse functions appears. Determine the amplitudes of the four impulse functions within the rising edge duration as the multiplicative factor A. m The amplitudes of the four impulse functions during the duration of the falling edge are determined as the multiplicative factor B. m ;

[0049] Among them, the multiplicative factor A m With B m The extraction process is as follows:

[0050] The fitted noise source time-domain waveform is continuously differentiated to obtain its first, second, and third-order differential waveforms. In the third-order differential waveform, four impulse functions appear within the rising edge duration, and their amplitudes are determined to be the multiplicative factor A. m Four impulse functions appear during the duration of the falling edge, and their amplitudes are determined to be the multiplicative factor B. m .

[0051] S4. Based on the fitted time-domain waveform of the noise source in step S2, extract time-domain feature parameters, including rise time. Initial time of ascent descent time Initial descent time Switching frequency f s ;

[0052] Among them, rise time The initial rise time is the total duration of the rising edge of the time-domain waveform of the noise source. The falling time is the duration of the quadratic curve S1 segment in the fitted S-shaped wave with rising edge. The total duration of the falling edge of the time-domain waveform of the noise source, the initial falling time The duration of the S1 segment of the quadratic curve in the S-shaped wave with the falling edge obtained by fitting.

[0053] S5. Substitute the extracted time-domain feature parameters into the preset prediction expressions for the spectrum amplitude envelope curve and the inflection point of the spectrum amplitude envelope curve of the soft-switching converter noise source to calculate the spectrum amplitude envelope curve and its inflection point of the soft-switching converter noise source, as follows:

[0054] Substitute the extracted time-domain feature parameters into the preset expression for predicting the inflection point of the spectrum amplitude envelope curve of the soft-switching converter noise source:

[0055]

[0056] Among them, f A1 f A2 f B1 f B2 All are inflection points;

[0057] Substitute the extracted time-domain feature parameters into the preset expression for predicting the amplitude envelope curve of the soft-switching converter noise source spectrum:

[0058]

[0059] in, The expression for the spectral amplitude envelope curve, f s Here, f is the switching frequency, and f is the current frequency value being calculated. A1 f A2 f B1 f B2 A is the inflection point calculated by formula (1). m With B m Multiplicative factor and The effects of the rising and falling edge time-domain characteristics on the spectrum are represented respectively, and their expressions are related to the inflection point f. A1 f A2 f B1 f B2 The expression needs to be selected based on the frequency band being calculated.

[0060] The spectral amplitude envelope curve prediction is a piecewise function based on four inflection point frequencies, and the segmentation points are determined by the four inflection point frequencies obtained from the spectral amplitude envelope curve inflection point prediction expression.

[0061] The present invention will be further described below with reference to specific embodiments.

[0062] In this embodiment, the noise source switching frequency f s =50kHz, using an S-shaped wave to fit the time-domain waveform of the noise source, the fitted waveform is as follows. Figure 1 As shown at the top. The rise time is... =0.4μs, fall time =0.6μs; the duration of the first segment of the quadratic curve of the S-shaped wave used to fit the rising edge of the noise source, i.e., the initial rise time. =0.12μs, used to fit the duration of the first segment of the quadratic curve of the S-shaped wave at the falling edge of the noise source, i.e., the initial falling time. =0.18μs.

[0063] By continuously differentiating the fitted waveform, its first-order, second-order, and third-order differential waveforms are obtained, as follows: Figure 2 As shown, in the third-order differential waveform, four impulse functions appear during the rising edge duration, and their amplitudes are determined by the multiplicative factor A. m =1.07*10 10 Four impulse functions appear during the duration of the falling edge, and their amplitudes are determined as the multiplicative factor B. m =4.76*10 9 .

[0064] Time-domain feature parameters , , , f s and multiplicative factor A m With B m Substituting the inflection point prediction expression of the noise source amplitude envelope curve of the soft-switching converter preset in this invention, we can obtain:

[0065]

[0066] Based on the inflection point frequency calculated above, further substituting the time-domain characteristic parameters, multiplicative factors, and inflection point frequency into the preset expression for predicting the amplitude envelope curve of the soft-switching converter noise source spectrum in this invention yields:

[0067] .

[0068] Based on the calculated envelope curve prediction expression, the following plots were obtained: Figure 3 The predicted envelope curve shown (dashed line in the figure) is used to verify the prediction accuracy. At the same time, a Fast Fourier Transform (FFT) analysis is performed on the time-domain waveform of the noise source to obtain the actual spectrum curve (solid line in the figure).

[0069] Depend on Figure 3 As can be seen, the predicted envelope curve and the spectrum curve obtained by FFT analysis have a high degree of agreement, indicating that the method of the present invention can effectively and accurately predict the spectrum amplitude envelope curve of the conducted electromagnetic interference noise source of the soft-switching converter.

[0070] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any modifications or equivalent changes made based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter, characterized in that, Includes the following steps: S1. Obtain the time-domain waveform of the conducted electromagnetic interference source of the soft-switching converter; S2. Based on the obtained time-domain waveform of the interference source, use an S-shaped wave to fit the rising and falling edges of the time-domain waveform of the interference source to obtain the fitted time-domain waveform of the noise source. S3. Continuously differentiate the time-domain waveform of the noise source fitted in step S2 until a waveform composed of impulse functions appears. Determine the amplitudes of the four impulse functions within the rising edge duration as the multiplicative factor A. m The amplitudes of the four impulse functions during the duration of the falling edge are determined as the multiplicative factor B. m ; S4. Based on the fitted time-domain waveform of the noise source in step S2, extract time-domain feature parameters, including rise time. Initial time of ascent descent time Initial descent time Switching frequency f s ; S5. Substitute the extracted time-domain feature parameters into the preset prediction expressions for the spectrum amplitude envelope curve and the inflection point of the spectrum amplitude envelope curve of the soft-switching converter noise source to calculate the spectrum amplitude envelope curve and its inflection point of the soft-switching converter noise source, as follows: Substitute the extracted time-domain feature parameters into the preset expression for predicting the inflection point of the spectrum amplitude envelope curve of the soft-switching converter noise source: ； Among them, f A1 f A2 f B1 f B2 All are inflection points; Substitute the extracted time-domain feature parameters into the preset expression for predicting the amplitude envelope curve of the soft-switching converter noise source spectrum: ； in, The expression for the spectral amplitude envelope curve, f s Here, f is the switching frequency, and f is the current frequency value being calculated. A1 f A2 f B1 f B2 A is the inflection point calculated by formula (1). m With B m Multiplicative factor and The effects of the rising and falling edge time-domain characteristics on the spectrum are represented respectively, and their expressions are related to the inflection point f. A1 f A2 f B1 f B2 The expression needs to be selected based on the frequency band being calculated.

2. The method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter according to claim 1, characterized in that, In step S1, The time-domain waveform of the conducted electromagnetic interference source of the soft-switching converter was obtained through modeling and simulation, as follows: Beforehand, the parasitic parameters of the components in the soft-switching converter are measured using analysis equipment, and a simulation model containing the parasitic parameters is established. Input the measured parasitic parameters into the simulation software to perform time-domain transient simulation; By directly observing the voltage and current waveforms of the switching transistors through simulation, the time-domain waveform of the conducted electromagnetic interference source in the soft-switching converter can be obtained.

3. The method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter according to claim 1, characterized in that, In step S2: The S-shaped wave is composed of three smooth curves: a quadratic curve S1, a linear curve S2, and a quadratic curve S3.

4. The method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter according to claim 1, characterized in that, In step S3: The multiplicative factor A m With B m The extraction process is as follows: The fitted noise source time-domain waveform is continuously differentiated to obtain its first, second, and third-order differential waveforms. In the third-order differential waveform, four impulse functions appear within the rising edge duration, and their amplitudes are determined to be the multiplicative factor A. m Four impulse functions appear during the duration of the falling edge, and their amplitudes are determined to be the multiplicative factor B. m .

5. A method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter according to claim 1 or 3, characterized in that, In step S4: The rise time The initial rise time is the total duration of the rising edge of the time-domain waveform of the noise source. The falling time is the duration of the quadratic curve S1 segment in the fitted S-shaped wave with rising edge. The total duration of the falling edge of the time-domain waveform of the noise source, the initial falling time The duration of the S1 segment of the quadratic curve in the S-shaped wave with the falling edge obtained by fitting.

6. The method for predicting the spectrum of conducted electromagnetic interference sources in a soft-switching converter according to claim 1, characterized in that, In step S5: The spectral amplitude envelope curve prediction is a piecewise function based on four inflection point frequencies, and the segmentation points are determined by the four inflection point frequencies obtained from the spectral amplitude envelope curve inflection point prediction expression.

Images