Electromagnetic noise analysis device and electromagnetic noise analysis method

Abstract

The present invention provides an electromagnetic noise analysis device that: is capable of performing electromagnetic noise analysis of an arbitrary observation point on a circuit during electromagnetic noise analysis, of a frequency axis, that takes into consideration various operation conditions, such as current dependency and control, which have conventionally been difficult to express except by using a time axis; and is capable of achieving both shortening of analysis time and improvement of analysis accuracy. The present invention is characterized by comprising: a noise spectrum analysis unit for executing frequency axis noise analysis in which each switch is expressed by a voltage source or a current source of the frequency axis for an arbitrary observation point of a power conversion circuit in which at least two switches are connected between power sources; a voltage source / current source noise source parameter generation unit for generating a parameter table for expressing a switching operation by a voltage source or a current source of the frequency axis, the switching operation corresponding to an operation condition including the power source voltage, the gate resistance of each switch, or a current; a worst condition extraction unit for extracting a current amount and a phase relationship in which a noise amount can be a worst condition during an operation on the time axis of each switch as based on control; and a worst condition noise source model generation unit for setting a noise source parameter for the worst condition using an extraction result by the worst condition extraction unit.

Description

Electromagnetic noise analysis device and electromagnetic noise analysis method

[0001] The present invention relates to a configuration of an electromagnetic noise analysis device and a method thereof, and in particular to a technique that is effective when applied to electromagnetic noise analysis on the frequency axis.

[0002] As systems and devices become increasingly electronic and electrified in various industrial fields, social infrastructure, and automotive applications, it is important to operate electronic systems and devices safely and securely and continue to provide value.

[0003] Therefore, in the design process of electronic systems and equipment, EMC simulation (EMC: Electromagnetic Compatibility) is being used to prevent rework through advance verification and to shorten and streamline design time by narrowing down the scope of verification.

[0004] Furthermore, EMC simulation is also used for analysis when product specifications or parts are changed, and it is necessary to establish an EMC analysis process that uses highly accurate and efficient analysis technology.

[0005] Background art in this technical field includes, for example, a technique such as that disclosed in Patent Document 1. Patent Document 1 discloses a noise analysis device that can easily analyze noise in a device to be analyzed.

[0006] Furthermore, Patent Document 2 discloses a noise analysis device that can quickly and accurately derive the calculation results of observed noise even when the on-period length and off-period length of a semiconductor element change.

[0007] JP 2020-24574 A International Publication No. 2023 / 223684

[0008] In EMC verification, noise can be predicted analytically using a noise analysis model based on an equivalent circuit. Electromagnetic noise analysis using EMC simulation can be performed in two ways: time-axis analysis, which operates a device model, and frequency-axis analysis.

[0009] Accurate time-axis noise analysis is possible depending on the device model, but the analysis time is very long and convergence deteriorates as the model scale increases, posing challenges for high-frequency analysis (50 MHz and above).

[0010] On the other hand, while frequency-axis noise analysis can be performed in a relatively short time, there are issues with the reproducibility of operating conditions associated with physical phenomena, such as ringing caused by current dependency.In addition, noise is often evaluated as a peak value over a certain period of time, and there are also issues with setting the worst-case conditions in frequency-axis noise analysis.

[0011] In the above-mentioned Patent Document 1, the above-mentioned problems in the frequency axis analysis are not sufficiently considered, and there is room for improvement in terms of noise prediction accuracy.

[0012] Furthermore, in the above-mentioned Patent Document 2, the analysis method uses a noise transfer function, so there is room for improvement in terms of targeting any observation point on the circuit.

[0013] Therefore, the object of the present invention is to provide an electromagnetic noise analysis device and electromagnetic noise analysis method that are capable of analyzing electromagnetic noise at any observation point on a circuit in frequency-axis electromagnetic noise analysis that takes into account various operating conditions such as current dependency and control, which have previously been difficult to express without a time axis, and that are capable of both shortening the analysis time and improving the analysis accuracy.

[0014] In order to solve the above problems, the present invention is characterized by comprising: a noise spectrum analysis unit that performs frequency-axis noise analysis targeting an arbitrary observation point of a power conversion circuit in which at least two switches are connected between power sources, and expressing each switch as a voltage source or a current source on the frequency axis; a voltage source / current source noise source parameter generation unit that generates a parameter table for expressing switching operation as a voltage source or a current source on the frequency axis according to operating conditions including at least one of a power supply voltage, a gate resistance of each of the switches, and a current; a worst-case condition extraction unit that extracts phase relationships and current amounts that may result in the worst-case condition for the amount of noise in the operation of each of the switches on the time axis based on control; and a worst-case condition noise source model generation unit that sets noise source parameters under the worst conditions using the extraction results by the worst-case condition extraction unit.

[0015] The present invention is also characterized by including the steps of: (a) performing a frequency-axis noise analysis on an arbitrary observation point of a power conversion circuit in which at least two switches are connected between power sources, and expressing each switch as a voltage source or a current source on the frequency axis; (b) generating a parameter table for expressing switching operation as a voltage source or a current source on the frequency axis according to operating conditions including at least one of a power supply voltage, a gate resistance of each of the switches, and a current; (c) extracting a phase relationship and a current amount that may result in the worst condition for the amount of noise in the operation of each of the switches on the time axis based on control; and (d) setting noise source parameters under the worst condition using the extraction results from step (c).

[0016] According to the present invention, it is possible to perform electromagnetic noise analysis on the frequency axis, taking into account various operating conditions such as current dependency and control, which have previously been difficult to express without a time axis, and it is possible to realize an electromagnetic noise analysis device and electromagnetic noise analysis method that can analyze electromagnetic noise at any observation point on a circuit, and that can achieve both a reduction in analysis time and an improvement in analysis accuracy.

[0017] This makes it possible to analyze, for example, the noise characteristics of any observation point on a circuit with high accuracy and in a short time for operating conditions using control signals such as PWM, for a power conversion device consisting of multiple phases.

[0018] Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0019] 1 is a diagram showing a schematic configuration of an electromagnetic noise analysis device according to an embodiment of the present invention. FIG. 2 is a diagram showing a schematic function of a worst-case condition extraction unit 2 of FIG. 1. FIG. 3 is a diagram showing a schematic function of the worst-case condition extraction unit 2 and the worst-case condition noise source model generation unit 3 of FIG. 1. FIG. 4 is a diagram showing a schematic function of the worst-case condition extraction unit 2 and the worst-case condition noise source model generation unit 3 of FIG. 1. FIG. 5 is a diagram showing a schematic function of the noise spectrum analysis unit 4 of FIG. 1. FIG. 6 is a diagram showing a schematic function of the voltage source / current source noise source parameter generation unit 5 of FIG. 1. FIG. 7 is a diagram showing a schematic function of the voltage source / current source noise source parameter generation unit 5 of FIG. 1. FIG. 8 is a diagram showing a schematic function of the voltage source / current source noise source parameter generation unit 5 of FIG. 1. FIG. 9 is a diagram showing a schematic function of an electromagnetic noise analysis method using the electromagnetic noise analysis device of FIG. 1.

[0020] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the drawings, the same components are designated by the same reference numerals, and detailed description of overlapping parts will be omitted.

[0021] An electromagnetic noise analysis apparatus and an electromagnetic noise analysis method according to an embodiment of the present invention will be described with reference to FIGS.

[0022] FIG. 1 is a diagram showing a schematic configuration of an electromagnetic noise analysis device 1 according to this embodiment.

[0023] As shown in FIG. 1 , the electromagnetic noise analysis device 1 of this embodiment mainly comprises a worst-case condition extraction unit (phase / current) 2, a worst-case condition noise source model generation unit 3, a noise spectrum analysis unit 4, a voltage source / current source noise source parameter generation unit 5, an equivalent circuit model generation unit 6, and an analysis result determination unit 7.

[0024] 1 shows an example in which the electromagnetic noise analysis device 1 includes storage devices (memories) 8a, 8b, and 8c, but the storage devices (memories) 8a, 8b, and 8c may be installed externally and connected to the electromagnetic noise analysis device 1 by wired or wireless communication. The same applies to storage devices (memories) 8d, 8e, and 8f, which will be described later.

[0025] The worst condition extraction unit (phase / current) 2 acquires control data of various electronic systems and devices stored in the storage device (memory) 8 a, and extracts the phase relationship and current amount that may cause the worst condition of noise amount in the operation on the time axis of each switch of the electronic system / device (especially between power supplies) based on the control.

[0026] In addition, the worst condition extraction unit (phase / current) 2 acquires device data (measurement data, specifications, etc.) of various electronic systems and equipment stored in the storage device (memory) 8b, and generates ideal waveform data based on the device data.

[0027] The worst-condition noise source model generation unit 3 sets the worst-condition noise source parameters using the extraction results by the worst-condition extraction unit (phase / current) 2 and the parameter table generated by the voltage source / current source noise source parameter generation unit 5.

[0028] Based on the noise source parameters under the worst conditions set by the worst-condition noise source model generation unit 3 and the equivalent circuit model generated by the equivalent circuit model generation unit 6, the noise spectrum analysis unit 4 performs frequency-axis noise analysis on an arbitrary observation point of a power conversion circuit in which at least two switches are connected between power supplies of various electronic systems and devices, expressing each switch as a voltage source or a current source on the frequency axis.

[0029] The voltage source / current source noise source parameter generation unit 5 acquires device data (measurement data, specifications, etc.) of various electronic systems and devices stored in the storage device (memory) 8b and design / structure data of various electronic systems and devices stored in the storage device (memory) 8c, and generates a parameter table for expressing switching operations as voltage sources or current sources on the frequency axis according to operating conditions including at least one of the power supply voltages of the various electronic systems and devices, the gate resistances of each switch, and the currents, based on the ideal waveform data generated by the worst condition extraction unit (phase / current) 2, the device data, and the design / structure data.

[0030] The equivalent circuit model generation unit 6 acquires design and structural data of various electronic systems and devices stored in the storage device (memory) 8c and generates an equivalent circuit model. The equivalent circuit model generated by the equivalent circuit model generation unit 6 is used in the frequency axis noise analysis by the noise spectrum analysis unit 4, thereby further improving the accuracy of the analysis.

[0031] The analysis result judgment unit 7 judges the results of the frequency axis noise analysis performed by the noise spectrum analysis unit 4 by comparing them with a predetermined threshold value. In this case, the threshold value may be a peak value, a quasi-peak value, or an average value. In this case, an offset is reflected in the execution result and a converted value is compared. Furthermore, depending on the resolution band width of the spectrum analyzer, an offset is reflected if necessary and a converted value is compared.

[0032] The determination result by the analysis result determination unit 7 is stored in a storage device (memory) 8c. The analysis result determination unit 7 compares the execution result of the frequency axis noise analysis by the noise spectrum analysis unit 4 with a predetermined threshold value and determines whether the result exceeds the threshold value (regulation value), and feeds back the result to the design and structure data. For example, the noise filter configuration is changed and reanalyzed so that the result falls below the threshold value (regulation value).

[0033] Fig. 2 is a diagram showing the function of the worst-case condition extraction unit 2 in Fig. 1. In Fig. 2, (1) and (2) respectively indicate the timing of maximum current recovery switching and the timing of two-phase simultaneous switching (three-phase simultaneous switching under 0 A conditions).

[0034] 2, the worst condition extraction unit 2 generates an ideal analysis model (without parasitic parameters) based on the control data of various electronic systems and devices stored in a storage device (memory) 8a and the design and structural data of various electronic systems and devices stored in a storage device (memory) 8c, and obtains ideal waveform data (voltage and current) corresponding to the control data (PWM). The ideal waveform data (voltage and current) is stored in a storage device (memory) 8d.

[0035] FIG. 3 is a diagram schematically showing the functions of the worst condition extraction unit 2 and the worst condition noise source model generation unit 3 in FIG.

[0036] As shown in Figure 3, the worst-case condition extraction unit 2 extracts the maximum current recovery switching in the upper and lower arms of each phase at the timing of the maximum current recovery switching (1). Furthermore, at the timing of the two-phase simultaneous switching (three-phase simultaneous switching under the 0 A condition) (2), the unit extracts the total current and the time difference below the threshold for simultaneous switching of two or more phases. The unit then extracts the combination of the switching pattern with the largest total current (recovery priority) and the smallest time difference as the worst-case condition. In other words, the worst-case conditions are extracted using the control data and ideal waveform data (voltage and current).

[0037] FIG. 4 is a diagram schematically showing the functions of the worst-case condition extraction unit 2 and the worst-case condition noise source model generation unit 3 in FIG. 1, and shows an example of switching at the timing of recovery switching of the maximum current (1).

[0038] As shown in FIG. 4, a voltage source is set for the transistor to be turned on, and a current source is set for the transistor to be recovered.

[0039] FIG. 5 is a diagram schematically illustrating the functions of the worst-case condition extraction unit 2 and the worst-case condition noise source model generation unit 3 in FIG. 1, and shows an example of switching at the timing of (2) two-phase simultaneous switching (three-phase simultaneous switching under 0 A conditions).

[0040] As shown in Figure 5, the difference in switching timing between each phase is reflected by the phase difference, similar to the timing of the maximum current recovery switching in (1). Also, a voltage source is set for the transistor to be turned on, and a current source is set for the transistor to be recovered.

[0041] FIG. 6 is a diagram schematically showing the function of the noise spectrum analysis unit 4 in FIG.

[0042] As shown in FIG. 6, one of the worst conditions is analyzed.

[0043] Alternatively, by taking the peak value envelope of each of the worst conditions, the peak noise level on the frequency axis can be obtained as an output. In this case, by identifying the worst condition for each frequency, it is possible to shorten the time required for reanalysis.

[0044] Furthermore, the worst conditions under which the effects of the design change are expected may be reanalyzed depending on the design change location. By reanalyzing the worst conditions under which the effects of the design change are expected, accuracy is improved.

[0045] However, if the design and structure data are changed during the reanalysis, the worst conditions may be different, so the analysis may be performed again for all conditions.

[0046] FIG. 7 is a diagram schematically showing the function of the voltage source / current source noise source parameter generating unit 5 in FIG.

[0047] The parasitic parameter and passive element extraction unit (including gate resistor) 13 acquires design and structure data of various electronic systems and devices stored in the storage device (memory) 8c, and extracts parasitic parameters and passive elements of the various electronic systems and devices based on the design and structure data.

[0048] The device model generation unit 14 acquires device data (measurement data, specifications, etc.) of various electronic systems and devices stored in the storage device (memory) 8b, and generates a device model based on the device data.

[0049] The switching characteristic analysis model generation unit 15 generates a switching characteristic analysis model based on the parasitic parameters and passive elements extracted by the parasitic parameter and passive element extraction unit (including gate resistance) 13 and the device model generated by the device model generation unit 14. The switching characteristic analysis model generation unit 15 generates a circuit (for example, a double pulse test) that can evaluate the switching characteristics of a device according to operating conditions (current, voltage, gate resistance, etc.).

[0050] The noise source parameter extraction unit 16 extracts voltage source noise source parameters and current source noise source parameters based on the switching characteristics analysis model generated by the switching characteristics analysis model generation unit 15 and the ideal waveform data (voltage and current) stored in the storage device (memory) 8d, and stores them in the storage device (memory) 8e and the storage device (memory) 8f, respectively.

[0051] Note that some or all of the processing of each of the above sections may be replaced by actual testing, and the overall parameter table may be generated by supplementing the results of testing and analysis under some conditions.

[0052] FIG. 8 is a diagram showing the function of the voltage source / current source noise source parameter generating unit 5 in FIG. 1, and shows the definition of each parameter in the time axis waveform.

[0053] As shown in Figure 8, the ringing component is decoupled to generate an ideal trapezoidal waveform. For the voltage waveform, the ringing component is decoupled to generate an ideal trapezoidal waveform. For the current waveform, it is generated from only the recovery component.

[0054] FIG. 9 is a diagram schematically showing the function of the voltage source / current source noise source parameter generating unit 5 in FIG. 1, and shows an example of spectral characteristics.

[0055] As shown in Figure 9, the current and voltage source noise source parameters use an approximation of the Fourier transform formula of an ideal trapezoidal wave signal. Note that the graph on the right of Figure 9 shows an example where the rise time and fall time are constant. Alternatively, the envelope of the fast Fourier transform (FFT) of the trapezoidal wave may be used.

[0056] FIG. 10 is a diagram schematically showing an electromagnetic noise analysis method using the electromagnetic noise analysis device 1 of FIG.

[0057] As shown in FIG. 10 , first, the worst condition extraction unit (phase / current) extracts the phase relationship and current amount that may cause the worst condition of noise amount in the operation of each switch on the time axis of the control-based electronic system / device (particularly between power supplies).

[0058] Furthermore, the noise source model generator 12 sets the voltage source and current source noise source parameters generated by the voltage source and current source noise source parameter generator 5 in the voltage source and current source model according to the worst conditions.

[0059] Next, the noise spectrum analysis unit performs frequency axis noise analysis.

[0060] As a result, it is possible to obtain a more accurate LISN voltage AC analysis result (LISN: artificial mains network) as shown in FIG.

[0061] As shown in FIG. 10, when comparing the analysis results for two conditions, where the maximum current is 20 A and 400 A, the method of the present invention, which is based on a voltage source and a current source, expresses the difference in noise level due to the current dependency between 20 A and 400 A (this is particularly noticeable in ringing above 10 MHz), and thus improves the analysis accuracy, compared to the conventional method based on a voltage source.

[0062] As described above, the electromagnetic noise analysis device 1 of this embodiment is equipped with: a noise spectrum analysis unit 4 that performs frequency-axis noise analysis, targeting an arbitrary observation point in a power conversion circuit in which at least two switches are connected between power sources, and expressing each switch as a voltage source or a current source on the frequency axis; a voltage source / current source noise source parameter generation unit 5 that generates a parameter table for expressing switching operation as a voltage source or a current source on the frequency axis according to operating conditions including at least one of the power supply voltage, the gate resistance of each switch, and the current; a worst-case condition extraction unit (phase / current) 2 that extracts phase relationships and current amounts that may result in the worst-case condition for the amount of noise in the operation of each switch on the time axis based on control; and a worst-case condition noise source model generation unit 3 that sets noise source parameters under the worst conditions using the extraction results by the worst-case condition extraction unit (phase / current) 2.

[0063] This makes it possible to achieve both shorter analysis times and improved analysis accuracy in frequency-axis electromagnetic noise analysis, which takes into account various operating conditions such as current dependency and control, which were previously difficult to express without a time axis.

[0064] The present invention is not limited to the above-described embodiments, but includes various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

[0065] 1...electromagnetic noise analysis device, 2...worst condition extraction unit (phase / current), 3...worst condition noise source model generation unit, 4...noise spectrum analysis unit, 5...voltage source / current source noise source parameter generation unit, 6...equivalent circuit model generation unit, 7...analysis result determination unit, 8, 8a, 8b, 8c, 8d, 8e, 8f...storage device (memory), 9...AC motor, 10...smoothing capacitor, 11...equivalent circuit model generation unit, 12...noise source model generation unit, 13...parasitic parameter / passive element extraction unit (including gate resistor), 14...device model generation unit, 15...switching characteristic analysis model generation unit, 16...noise source parameter extraction unit.

Claims

1. An electromagnetic noise analysis device comprising: a noise spectrum analysis unit that performs frequency-axis noise analysis on an arbitrary observation point of a power conversion circuit in which at least two switches are connected between power sources, and expresses each switch as a voltage source or a current source on the frequency axis; a voltage source / current source noise source parameter generation unit that generates a parameter table for expressing switching operation as a voltage source or a current source on the frequency axis according to operating conditions including at least one of the power supply voltage, the gate resistance of each of the switches, and the current; a worst-case condition extraction unit that extracts phase relationships and current amounts that may result in the worst-case condition for the amount of noise in the operation of each of the switches on the time axis based on control; and a worst-case condition noise source model generation unit that sets noise source parameters under the worst conditions using the results of extraction by the worst-case condition extraction unit.

2. An electromagnetic noise analysis device according to claim 1, further comprising an equivalent circuit model generation unit that acquires design and structure data stored in a storage device and generates an equivalent circuit model, wherein the noise spectrum analysis unit performs the frequency axis noise analysis based on the noise source parameters under the worst conditions set by the worst condition noise source model generation unit and the equivalent circuit model generated by the equivalent circuit model generation unit.

3. An electromagnetic noise analysis device according to claim 2, characterized in that it analyzes any one of the worst conditions.

4. An electromagnetic noise analysis device according to claim 2, characterized in that it takes an envelope of peak values ​​of analysis results under a plurality of worst conditions.

5. An electromagnetic noise analysis device according to claim 1, further comprising an analysis result determination unit that determines the results of the frequency axis noise analysis performed by the noise spectrum analysis unit by comparing them with a predetermined threshold value that has been set in advance.

6. An electromagnetic noise analysis device according to claim 5, characterized in that, if the result of frequency axis noise analysis performed by said noise spectrum analysis unit exceeds said predetermined threshold, the result is fed back to the design and reanalyzed.

7. An electromagnetic noise analysis method including the following steps: (a) performing a frequency-axis noise analysis on an arbitrary observation point of a power conversion circuit in which at least two switches are connected between power sources, and expressing each switch as a voltage source or a current source on the frequency axis; (b) generating a parameter table for expressing switching operation as a voltage source or a current source on the frequency axis according to operating conditions including at least one of the power supply voltage, the gate resistance of each of the switches, and the current; (c) extracting the phase relationship and current amount that may result in the worst-case condition of noise amount in the operation of each of the switches on the time axis based on control; and (d) setting the worst-case condition noise source parameters using the extraction results from step (c).

8. An electromagnetic noise analysis method as set forth in claim 7, further comprising the step of (e) acquiring design and structure data stored in a storage device and generating an equivalent circuit model, wherein in step (a), the frequency axis noise analysis is performed based on the noise source parameters under the worst conditions set in step (d) and the equivalent circuit model generated in step (e).

9. The electromagnetic noise analysis method according to claim 8, characterized in that any one of the worst conditions is analyzed.

10. The electromagnetic noise analysis method according to claim 8, wherein the envelope of peak values ​​of analysis results under a plurality of worst conditions is taken.

11. An electromagnetic noise analysis method according to claim 7, further comprising: (f) a step of judging the results of the frequency axis noise analysis performed in step (a) by comparing them with a predetermined threshold value that has been set in advance.

12. An electromagnetic noise analysis method according to claim 11, characterized in that, if the result of the frequency axis noise analysis in step (a) exceeds the predetermined threshold, the result is fed back to the design and reanalyzed.

Images