Method and device for providing a behavioral model for simulating electromagnetic interference emissions from an electrical or electronic component

Description

Technical area

[0001] The invention relates to the field of simulating the electromagnetic compatibility of electrical or electronic components. In particular, the present invention relates to providing EMC behavior models (EMC: electromagnetic compatibility) for simulating electromagnetic interference emissions from components in order to predict the behavior of the component in a real system or environment. Technical background

[0002] In practice, the EMC behavior of electrical or electronic components in a system is simulated to determine whether, on the one hand, trouble-free operation of the system is possible and, on the other hand, whether the exposure of the system's environment to electromagnetic radiation does not exceed the specified limits. A key feature of the EMC behavior model is that only the behavior of the component is described, without disclosing relevant information about the component's structure. This allows the know-how in the component to be protected, and at the same time, a simulation model is provided for the component that adequately describes the component's behavior with regard to electromagnetic interference emissions.

[0003] Document CN 107783040 A discloses a method for modeling electromagnetic interference of a DC motor, wherein the common-mode admittance parameters and differential admittance parameters are recorded and determined accordingly. The model is determined by measuring a first scattering parameter matrix and by measuring the current variables of a current source relative to the target. The scattering parameters are measured using a network analyzer in a loaded state.

[0004] The paper SUN BINGYAO ET AL: "Common-Mode EMI Unterminated Behavioral Model of Wide-Bandgap-Based Power Converters Operating at High Switching Frequency", IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, Vol. 7, No. 4, December 1, 2019, pages 2561-2570 discloses the use of a behavioral model to predict common-mode emissions in an energy range of 150 kHz to 30 MHz.

[0005] The paper HILLENBRAND P ET AL: "Frequency domain EMI simulation and resonance analysis of a DCDC converter", 2016 INTERNATIONAL SYMPOSIUM ON ELECTROMAGNETIC COMPATIBILITY - EMC EUROPE, IEEE, September 5, 2016, pages 176-181, discloses a frequency domain model for predicting electromagnetic interference from an inverter. In this model, the inverter's nonlinear and time-varying semiconductor switches are replaced by voltage sources characterized by measurements taken during inverter operation.

[0006] The paper QIAN LIU ET AL: "Modular-Terminal-Behavioral (MTB) Model for Characterizing Switching Module Conducted EMI Generation in Converter Systems," IEEE TRANSACTIONS ON POWER ELECTRONICS, INSTITUTE OF ELECTRICAL AND ELECTRONICS ENGINEERS, Vol. 21, No. 6, November 1, 2006, pages 1804-1814, discloses a new frequency-domain behavioral model for characterizing electromagnetic interference noise waves in an inverter. The electromagnetic interference emission of a switching module under specific switching conditions is modeled using an equivalent Norton network.

[0007] The paper M. REUTER ET AL: "Emulation of conducted emissions of an automotive inverter for filter development in HV networks", 2013 IEEE INTERNATIONAL SYMPOSIUM ON ELECTROMAGNETIC COMPATIBILITY, 1 August 2013, pages 236-241 discloses the emulation of emissions of an inverter for a hybrid vehicle, wherein the emulated component is tested as a voltage source in combination with an emulation network of the inverter.

[0008] The paper RAMANUJAN ABHISHEK ET AL: "Developing a universal exchange format for Integrated Circuit Emission Model - Conducted Emissions", 2015 10TH INTERNATIONAL WORKSHOP ON THE ELECTROMAGNETIC COMPATIBILITY OF INTEGRATED CIRCUITS (EMC COMPO), IEEE, November 10, 2015, pages 252-257 describes an emission model with a data exchange format. Disclosure of the invention

[0009] According to the invention, a method for providing an EMC behavior model for simulating an electromagnetic interference emission of an electrical or electronic component with linear and time-invariant behavior according to claim 1 and a corresponding device, a method for simulating an EMC behavior of an electrical or electronic component with linear and time-invariant behavior and a simulation device according to the independent claims are provided.

[0010] Further embodiments are specified in the dependent claims.

[0011] According to a first aspect, a method, in particular a computer-implemented method, is provided for providing a behavioral model, in particular an EMC behavioral model, in the form of model data for simulating an electrical or electronic component described by the behavioral model in a simulation device, wherein the behavioral model describes the behavior at one or more electrical terminals of the component by describing the electrical behavior at each of the terminals of the component by a terminal equivalence circuit as the effect of complex, frequency-dependent electrical quantities of passive electrical components, in particular admittances or impedances, and of a source element, in particular a voltage and / or a current source, wherein the behavioral model is created by measuring an impedance / admittance matrix in the passive,unloaded state of the terminals using a network analyzer and by determining the complex, frequency-dependent electrical quantities of the source elements by measuring the currents and / or voltages at the terminals under a single load condition.

[0012] The complete simulation of electromagnetic interactions of a component, such as an electric motor, a spark plug, a radio receiver unit, and the like, in a system typically requires detailed knowledge of the component's functional and parasitic elements. Simulation models would therefore have to account for all electromagnetic coupling effects within the component and between the component and the rest of the system. However, this requires a complex development process, especially if the internal structure of the component is completely or partially unknown.

[0013] EMC behavioral models can significantly simplify the development process and the process of evaluating the use of a component within a system, especially when the component exhibits linear and time-invariant behavior. Such behavioral models use a terminal equivalent circuit that represents the behavior of the component at its terminals, rather than modeling all interference sources and coupling effects within the component. Conventional approaches, for example, use an equivalent circuit based on Norton or Thévenin. For example, Thévenin theory uses impedance elements combined with current sources to model the behavior of a component's terminal.

[0014] To create the behavioral model, conventional measurements with different load conditions are laborious and usually result in numerically poorly conditioned models with only a limited range of validity. Furthermore, the measurements of the load conditions must exhibit correct phase relationship to each other. These problems scale with the number of connections to be modeled, making this method unsuitable for creating behavioral models with more than one connection.

[0015] One way to describe an EMC behavior model is the Touchstone format. The Touchstone format is traditionally known as a text file format for storing network parameters, usually from measurement data. Scattering parameters, admittance parameters, impedance parameters, hybrid h-parameters, and hybrid g-parameters can be described in the Touchstone format. The Touchstone format can also be used to describe the behavior of an electrical or electronic component with regard to its EMC behavior for all electrical connections of the component based on admittances and the associated current sources.

[0016] The above method describes an alternative approach that allows the creation of a behavioral model without applying different load conditions. The core of the above method is the measurement of an impedance / admittance matrix of the modeled connections using a passive impedance measurement, in which the component to be characterized is not in operation. This admittance / impedance measurement can be performed easily and accurately, for example, using a network analyzer that provides a corresponding impedance / admittance matrix. Therefore, for the subsequent determination of the electrical behavior of the source elements assigned to the connections in the behavioral model, the measurement of currents or voltages is also required for only one load condition.

[0017] A fundamental advantage of using the EMC behavior model is that it only contains data related to the behavior at the component's terminals, allowing the EMC behavior model to be shared between different parties without disclosing system expertise. Using the terminal equivalent circuit as the file format for the EMC behavior model enables the definition of passive network parameters such as scattering, admittance, impedance, and hybrid parameters.

[0018] Furthermore, the behavior model can define the terminals with respect to a ground potential and with respect to the other terminals.

[0019] Furthermore, only the behavior model can be used to simulate the behavior of the component.

[0020] In particular, the behavior model can be provided in a Touchstone format, which is a common format for describing a behavior model.

[0021] Furthermore, the behavioral model can describe a real component with a number of electrical connections and provide the same number of virtual connections for the simulation.

[0022] According to one embodiment, the electrical size of the source element of each terminal can be determined frequency-dependently using an oscilloscope by converting the recorded time signals into the frequency domain (for example, using a Fourier transform).

[0023] Furthermore, the measurement of the electrical size of the passive electrical components of the terminal equivalent circuit in the unloaded state can be carried out for each terminal simultaneously.

[0024] Alternatively or additionally, the electrical size of the source elements of the terminal equivalent circuit can be measured using a load admittance for each terminal, in particular simultaneously.

[0025] According to a further aspect, a device for providing a behavior model, in particular an EMC behavior model, in the form of model data for simulating an electrical or electronic component described by the behavior model in a simulation device is provided, wherein the behavior model describes the behavior at one or more electrical terminals of the component by describing the electrical behavior at each of the terminals of the component by a terminal equivalence circuit as the effect of complex, frequency-dependent electrical variables of passive electrical components, in particular admittances or impedances, and of a source element, in particular a voltage and / or a current source.The device is designed to determine the behavior model by measuring an impedance / admittance matrix in the passive, unloaded state of the terminals using a network analyzer and by determining the complex, frequency-dependent electrical quantities of the source elements by measuring the currents and / or voltages at the terminals under a single load condition. Such a device enables automatic or semi-automatic measurement of the component, in particular through a computer implementation of the method described above for providing the behavior model.

[0026] According to a further aspect, a method is provided for simulating a component using a behavior model associated with the component, wherein the behavior of the component is simulated using the behavior model determined according to the above method.

[0027] According to a further aspect, a simulation device is provided for simulating a component based on a behavioral model associated with the component, wherein the simulation device is configured to simulate the behavior of the component using the behavioral model determined according to the above method. Such a simulation device enables the user to perform computer-implemented EMC simulation of the component by utilizing the provided behavioral model. Brief description of the drawings

[0028] Embodiments are explained in more detail below with reference to the attached drawings. They show: Figure 1 shows a graphical representation of a system for measuring a component to determine a terminal equivalent circuit; Figure 2 shows a flowchart illustrating a method for providing an EMC behavior model; Figure 3 shows an arrangement for measuring the current sources for the terminal equivalent circuit; and Figure 4 shows a flowchart illustrating a method for applying an EMC behavior model and simulating an electrical or electronic component based on the EMC behavior model. Description of embodiments

[0029] In Figure 1An electrical or electronic component 1 with two terminals A1 and A2 and an internal auxiliary circuit is shown to illustrate the general concept of the terminal equivalent circuit and the corresponding EMC behavior model. A model component is to be created for the electrical or electronic component 1 by measuring it. This model corresponds to a simulation model for the electrical or electronic component 1.

[0030] In the example described here, electronic component 1 has two terminals, designated A1 and A2 in the model component. In general, the component can have any number of terminals. Each terminal A1, A2 is defined with respect to a ground potential GND. The electrical or electronic component 1 to be modeled exhibits linear and time-invariant behavior at the terminal being modeled. Component 1 can be active or passive; in the following, it is assumed that terminals A1, A2 are active.

[0031] The behavior of the two terminals is characterized by determining an associated impedance / admittance matrix and the characteristics of the current sources for each of the terminals A1, A2. In the example described, the resulting circuit could be as shown in Fig. 3are described. In the following, admittances and current sources are assumed to be elements of the terminal equivalent circuit.

[0032] According to the previous approach, the behavior at terminals A1 and A2 is characterized by connecting various external loads to the respective terminals A1 and A2. Each of the load conditions generates currents and voltages at all terminals. The number of required load conditions therefore depends on the number of terminals. In the time domain, values ​​exist for each time step. In the frequency domain, complex values ​​exist for all frequencies. To determine the required impedances / admittances and current sources, a system of equations must be solved using the data from all load conditions.

[0033] To create the entire behavior model, this approach would require complex values ​​for all frequency points of interest. Since the phase positions of the currents and voltages are also relevant, the phase relationships of the quantities at the terminals to each other, as well as to each other with respect to the load conditions used for the measurement, would have to be known. In practice, this phase relationship is rarely known. This approach would also require a significant change in the load conditions, which can often lead to undesirable nonlinear behavior changes of the measured component. Therefore, for the reasons described above, this approach is only useful for models.

[0034] Components typically exhibit a mixture of linear time-invariant and nonlinear or time-varying behavior at their terminals. For the modeling described below, it is assumed that the equivalent admittances of the terminal equivalent circuit exhibit only linear time-invariant behavior and can be described with appropriate elements. This is possible because the passive linear time-invariant structures, such as passive EMI filters near the component terminals, usually represent the most important influencing factor on the equivalent impedance / admittance.

[0035] In order to avoid the disadvantages of the above conventional approach, a method is provided for model creation, which is described below in conjunction with the flow chart of the Figure 2 is described in more detail.

[0036] In step S1, the component's terminals are measured using a network analyzer (VNA) 2 to determine an admittance / impedance matrix of all modeled terminals in a passive state of component 1. For the exemplary creation of a model with the two terminals A1 and A2, this can be represented as admittances Y1, Y2, and Y12.

[0037] The VNA measurement is performed for all terminals of component 1 simultaneously and provides scattering parameters that can be transformed into admittance parameters and that correspond to or can be converted into the admittance elements of the Thévenin equivalent circuit.

[0038] A network analyzer 2 is generally used to measure the scattering parameters (S-parameters), i.e., the wave size of reflection and transmission at a component's electrical connections as a function of frequency. Network analyzers are used for everything from determining the transmission characteristics of filters or amplifiers, to measuring complex transmission lines. In a network analyzer, the component to be measured is fed by measurement generators (transmitters) built into the network analyzer, and the resulting signal changes are simultaneously recorded by measurement receivers in the network analyzer. This allows the measurement of the transmitted and received signals to be designed as relative measurements, and the measured values ​​can be related to each other.With a network analyzer, the component is not measured under load with a measuring device that behaves as passively as possible, but the component is supplied by the network analyzer on its own and its properties are measured.

[0039] The network analyzer 2 can therefore determine the admittance / impedance matrix of the connections of component 1.

[0040] After this step, the current sources I of the equivalent circuit are the remaining unknown elements.

[0041] In step S2, the magnitude of a current source assigned to the respective terminal A1, A2 is determined by simultaneous measurement in a single load condition. For this purpose, the components are actively operated, for example by applying loads. Y load in Fig.3The resulting voltages or currents V A1 or I A1, or V A2 or I A2, are measured at terminals A1 and A2. From these quantities, together with the measured passive impedance / admittance matrix and the known external loads, the desired source elements of the model can be calculated. This ensures stable model generation because only one load state is required, regardless of the number of terminals.

[0042] To measure the current sources for the terminal equivalent circuit, an arrangement as shown in Figure 3 shown. The measurement involves connecting component 1 to a load admittance 3 and an oscilloscope 4.

[0043] A successful calculation of the equivalent current sources requires linearly independent potentials, which can be measured using the oscilloscope 4. In this way, the corresponding current sources for each terminal can be determined.

[0044] In step S3, a behavior model corresponding to the terminal equivalent circuit is defined based on the known admittances and the current sources determined in step S2, each of which has a frequency-dependent behavior.

[0045] A behavior model created in this way can be represented in a description format, such as the Touchstone format.

[0046] Figure 4 shows an example application of the proposed behavioral model using a flow chart.

[0047] In step S11, an EMC behavior model is provided by a manufacturer.

[0048] After the behavioral model has been transmitted, it can be used to simulate the EMC behavior of the component described by the model component of the behavioral model.

[0049] In step S12, the simulation is performed and the frequency response of the component defined by the behavior model is determined.

Claims

1. Method, in particular computer-implemented method, for providing a behavioural model, in particular a behavioural model for electromagnetic compatibility, that is to say EMC behavioural model, in the form of model data for the simulation of an electrical or electronic component (1) described by the behavioural model in a simulation device, wherein the behavioural model describes the behaviour at one or more electrical connections (A1, A2) of the component (1) by virtue of the electrical behaviour at each of the connections (A1, A2) of the component being described by a connection equivalent circuit as the effect of complex, frequency-dependent electrical variables of passive electrical components, in particular admittances or impedances, and of a source element, in particular a voltage and / or a current source, characterized in that the behaviour model is determined by measuring (S1) an impedance matrix or admittance matrix in the passive, unloaded state of the connections using a network analyser (2) and by determining (S2) the complex, frequency-dependent electrical variables of the source element by measuring the currents and / or voltages at the connections (A1, A2) under a single load condition.

2. Method according to Claim 1, wherein the behaviour model defines the connections with respect to an earth potential and with respect to the other terminals.

3. Method according to either one of Claims 1 and 2, wherein only the behaviour model is used to simulate the behaviour of the component (1).

4. Method according to any one of Claims 1 to 3, wherein the behaviour model is provided in a touchstone format.

5. Method according to any one of Claims 1 to 4, wherein the behaviour model describes a real component having a number of electrical connections and provides the same number of virtual connections for the simulation.

6. Method according to any one of Claims 1 to 5, wherein the complex, frequency-dependent electrical variable of the source element of each connection (A1, A2) is determined in a frequency-dependent manner with the aid of an oscilloscope (4), in particular by converting the recorded time signals into the frequency domain.

7. Method according to any one of Claims 1 to 6, wherein the electrical variable of the passive electrical components of the connection equivalent circuit in the unloaded state is measured for each connection simultaneously.

8. Method according to any one of Claims 1 to 7, wherein the electrical variable of the source element of the connection equivalent circuit is measured for each connection, in particular simultaneously, with the aid of a load admittance (3).

9. Apparatus for providing a behavioural model, in particular a behavioural model for electromagnetic compatibility, that is to say EMC behavioural model, in the form of model data for the simulation of an electrical or electronic component (1) described by the behavioural model in a simulation device included in the apparatus, wherein the behavioural model describes the behaviour at one or more electrical connections (A1, A2) of the component (1) by virtue of the electrical behaviour at each of the connections (A1, A2) of the component (1) being described by a connection equivalent circuit as the effect of complex, frequency-dependent electrical variables of passive electrical components, in particular admittances or impedances, and of a source element, in particular a voltage and / or a current source, characterized in that the apparatus is designed to determine the behaviour model by measuring an impedance matrix or admittance matrix in the passive, unloaded state of the connections using a network analyser (2) and by determining the complex, frequency-dependent electrical variables of the source element by measuring the currents and / or voltages at the connections (A1, A2) under a single load condition.

10. Computer-implemented method for simulating a component using a behaviour model associated with the component, wherein the behaviour of the component is simulated with the aid of the behaviour model determined in accordance with the method according to any one of Claims 1 to 8 on an apparatus according to Claim 9.

11. Computer program which codes the method according to Claim 10.

12. Machine-readable storage medium on which a computer program according to Claim 11 is stored.