Determining switching loss energies in power electronic switching converters
The method addresses inaccuracies in power loss estimation by determining multiple switching energy values during a control signal period, improving control accuracy and efficiency of electronic power circuits.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- SCHAEFFLER TECHNOLOGIES AG & CO KG
- Filing Date
- 2025-09-09
- Publication Date
- 2026-04-16
AI Technical Summary
Existing methods for determining power losses in power semiconductors fail to account for switching pulse suppression, leading to inaccuracies in loss estimation during high modulation levels, particularly in block commutation.
A method that determines multiple switching energy values during a control signal period, considering switching pulse suppression, to improve accuracy in power loss estimation by accounting for conduction and switching losses in power semiconductor switches.
Enhances the accuracy of power loss determination, enabling precise control of electronic power circuits to operate efficiently without damage, even at high modulation levels.
Smart Images

Figure DE2025100854_16042026_PF_FP_ABST
Abstract
Description
[0001] Description
[0002] Method for determining performance data of electronic power circuits
[0003] The present invention relates to a method for determining performance data of electronic power circuits. The invention also relates to a computing unit configured and programmed to execute the method.
[0004] In the prior art, power electronics are used to control electric traction drives or for energy conversion, for example in electric drives, whereby the power electronics are controlled or commutated by means of modulation methods, for example based on space vector pulse-width modulation (SVPWM). Within the power electronics, power semiconductor switches, such as MOSFETs (metal oxide field-effect transistors) or IGBTs (bipolar transistors with insulated gates), are used, which are commutated or switched on or off depending on the position of the space vector.
[0005] The commutation processes, or switching on and off, can follow a predefined mathematical pattern determined by a control algorithm and the modulation method, thus switching the operating state of the power semiconductors between conducting and non-conducting operation. Furthermore, with increasing torque and / or speed at a load to be switched, an increasingly higher effective phase voltage must be set. For this purpose, the modulation level of the power semiconductors, and thus the maximum duty cycle of the power electronics, is increased by a controller controlling the power semiconductors. To achieve very high torques and / or speeds, it may be necessary to increase the modulation level so much that individual phases are not switched for extended periods, with the duty cycle of these phases being 0 or 1.In this respect, one can speak of switching pulse suppression of the power semiconductors, whereby the load is temporarily connected to the positive (1) or negative (0) reference potential. For example, the maximum modulation level can exhibit block commutation, in which the commutation pattern follows the electrical frequency, thus achieving the maximum modulation level. Particularly during block commutation, the switching frequency corresponds to the electrical frequency, even if the switching frequency theoretically specified by the control unit can be significantly higher.
[0006] Prior art methods exist for determining the power losses of power semiconductors, which calculate average losses over one electrical period of a drive signal to the power semiconductors. However, these methods assume that all phases are continuously switched over one electrical period and that no switching pulse suppression occurs, meaning that switching pulse suppression is not considered when determining the power losses. Therefore, when using such methods, it can only be assumed that switching losses occur in the power semiconductors at all times during an electrical period.In particular, in the case of an excessive modulation level, for example duty cycle >= 1 or duty cycle <= 0, for example due to an increase in the modulation level into a non-linear range of the power semiconductors, the individual phases are no longer switched for the duration of the excess, which leads to a reduction in switching losses.
[0007] The object of the present invention is therefore to provide a technology that is more advanced than the prior art. In particular, it aims to improve the accuracy of determining power loss during excessive duty cycles.
[0008] This problem is solved by articles with the features according to the independent claims. Advantageous embodiments are the subject of the dependent claims.
[0009] A method for determining performance data of electronic power circuits is disclosed. Furthermore, the method includes determining several switching energy values during one period of a control signal of the electronic power circuit.
[0010] Furthermore, the procedure includes determining the performance data, which exhibit a switching cycle energy value determined on the basis of the multiple switching energy values and a number of determinations of the multiple switching energy values.
[0011] The electronic power circuit can, for example, include one or more power semiconductor switches. These power semiconductor switches can be designed as MOSFETs (metal oxide field-effect transistors) and / or IGBTs (insulated-gate bipolar transistors).
[0012] The multiple switching energy values can, for example, each be determined at several points in time during the period. Alternatively or additionally, each of the multiple switching energy values can have an instantaneous switching energy value at each of the respective points in time of the electronic power circuit. For example, two or more switching energy values can each be determined at two or more points in time during the period. For example, the number of determinations of the multiple switching energy values can depend on the period and / or the modulation level of a control signal of the electronic power circuit.
[0013] The switching cycle energy value can be determined, for example, once during the period.
[0014] The disclosed method can be used, for example, to control the electronic power circuit with a modulated control signal. Particularly preferably, the disclosed method can be used during switching pulse suppression of the modulated control signal, as occurs, for example, in discontinuous pulse-width modulation (DPWM), overmodulation (OM), or block commutation (synchronous PWM (SyPWM)). In this context, for example, the distribution of conduction losses between two or more power semiconductor switches of the electronic power circuit within the electrical period can be dependent on the modulation type.In particular, this distribution can occur when using SyPWM modulation (block commutation), where, for example, conduction losses occur only in the first power semiconductor switch during the first half of the electrical period and only in the second power semiconductor switch during the second half of the period. Alternatively or additionally, the multiple switching energy values can represent a thermal ripple in the two or more power semiconductor switches.
[0015] Therefore, the disclosed method can be used to determine performance data, particularly preferably, during operation of the electronic power circuit in the range of high modulation levels, especially in the nonlinear range of overmodulation, i.e., during switching pulse suppression, by determining multiple switching energy values. This allows, for example, higher accuracy in determining the performance data compared to loss estimators that, for instance, only determine the performance data once per period.
[0016] The performance data determined by the disclosed method can, for example, be provided to a control unit for controlling the electronic power circuit, so that the control unit can generate or determine any control signals based on the performance data in order to control the electronic power circuit. This allows the electronic power circuit to be preferably controlled based on the performance data, so that the electronic power circuit can, for example, be operated in such a way that it can deliver a higher power output without being damaged or destroyed.
[0017] Alternatively or additionally, the disclosed method can, for example, be implemented as software. In particular, the disclosed method can be implemented as a computer method.
[0018] Furthermore, the switching energy values can be determined based on a comparison of switching energy table data with a provided current value of the electronic power circuit, a provided voltage value of the electronic power circuit and / or a provided temperature value of the electronic power circuit.
[0019] Particularly preferably, the switching energy values can be determined on the basis of a comparison of switching energy table data with a provided current value of the electronic power circuit, a provided voltage value of the electronic power circuit and a provided temperature value of the electronic power circuit.
[0020] The switching energy values or switching losses can be provided, for example, as a three-dimensional table in the switching energy table data. Furthermore, the provided current value can be, for example, the drain current ID of one or more power semiconductor switches, particularly MOSFETs. Furthermore, the provided voltage value can be, for example, a battery voltage UDC applied across a drain-collector path of one or more power semiconductor switches, particularly MOSFETs. Furthermore, the provided temperature value Tjc can be, for example, the temperature of one or more power semiconductor switches, particularly MOSFETs.
[0021] Furthermore, the determination of one of the several switching energy values can be carried out synchronously with a clock frequency value of the control signal.
[0022] The clock frequency value can, for example, be provided by a control unit executing the disclosed method. In particular, the control unit can be operated using the clock frequency value or used by the control unit as a working clock. Furthermore, the clock frequency value can also be referred to as a fast software cycle.
[0023] Alternatively or additionally, the fast software cycle To can be synchronized with the clock frequency fsw of a pulse width modulation for controlling the electronic power circuit. Alternatively or additionally, the turn-on times Ton and / or turn-off times Toff of one or more power semiconductor switches can be specified in the software cycle TO by the control unit, in particular a current controller and / or a torque controller for the one or more power semiconductor switches.
[0024] Furthermore, the determination of one of the several switching energy values can only be carried out during a switching operation of the electronic power circuit.
[0025] Furthermore, the performance data may also include a conduction loss value of the electronic power circuit.
[0026] Furthermore, the performance data may also include a temperature value of the electronic power circuit, which is determined based on the switching cycle energy value, the conduction loss value and / or a clock frequency value of the control signal.
[0027] Furthermore, the procedure may also include the provision of performance data.
[0028] Furthermore, a computing unit for determining the performance data, which is designed and equipped to execute the disclosed method, is disclosed.
[0029] The present invention is described in detail below with reference to the figures. These show:
[0030] Fig. 1 shows an exemplary first step of a method for determining performance data of electronic power circuits; and
[0031] Fig. 2 shows an exemplary second step of a method for determining performance data of electronic power circuits.
[0032] The present invention is described below with reference to preferred embodiments and the figures. However, this description of the embodiment should not be considered exhaustive.
[0033] Figure 1 shows an exemplary first step of a method 100 for determining performance data of electronic power circuits. The method 100 comprises determining several switching energy values 110 during one period of a control signal of the electronic power circuit. Furthermore, the method 100 comprises determining the performance data 120, which includes a switching cycle energy value determined on the basis of the several switching energy values and a number of determinations of the several switching energy values.
[0034] Fig. 2 shows an exemplary second step of a method 100 for determining performance data of electronic power circuits.
[0035] Method 100 comprises determining five switching energy values 110.1, 110.2, 110.3, 110.4, 110.5 during a period 130 of a control signal of the electronic power circuit. The determination of one of the several switching energy values 110.1, 110.2, 110.3, 110.4, 110.5 is performed synchronously with a clock frequency value of the control signal. Furthermore, the determination of one of the several switching energy values 110.1, 110.2, 110.3, 110.4, 110.5 can only be performed during a switching operation of the electronic power circuit.
[0036] Furthermore, the procedure 100 includes determining two performance data points 120.1 and 120.2, which have a switching cycle energy value determined based on the multiple switching energy values and the number of times these values are determined. The determination of each of the performance data points 120.1 and 120.2 is performed only once during the period 130.
[0037] The switching energy values are determined based on a comparison of switching energy table data with a provided current value, a provided voltage value, and / or a provided temperature value of the electronic power circuit. Furthermore, the power data preferably includes a conduction loss value for the electronic power circuit. Additionally, the power data preferably includes a temperature value for the electronic power circuit, which is determined based on the switching cycle energy value, the conduction loss value, and / or a clock frequency value of the control signal. The power data also includes a temperature value for the electronic power circuit, which is determined based on the switching cycle energy value, the conduction loss value, and / or a clock frequency value of the control signal.
[0038] Furthermore, method 100 can also include providing the performance data. This performance data can, for example, be provided to a control unit for controlling an electronic power circuit. The electronic power circuit can, for instance, be configured to drive an electric machine. Alternatively or additionally, the electric machine can be used within a vehicle. Accordingly, the control unit can be configured for use in a vehicle. Particularly preferably, the control unit can control the electronic power circuit based on the performance and control data.
[0039] Reference symbol list
[0040] 100 methods for determining performance data
[0041] 110 Determining multiple switching energy values
[0042] 120 Determining the performance data
[0043] 130 Period duration of the control signal
Claims
Claims 1. Method (100) for determining performance data of electronic power circuits, comprising: Determining several switching energy values (110) during one period of a control signal of the electronic power circuit; and Determining the performance data (120), which has a switching cycle energy value determined on the basis of the multiple switching energy values and a number of determinations of the multiple switching energy values.
2. Method (100) according to claim 1, wherein the switching energy values are determined on the basis of a comparison of switching energy table data with a provided current value of the electronic power circuit, a provided voltage value of the electronic power circuit and / or a provided temperature value of the electronic power circuit.
3. Method (100) according to claim 1 or 2, wherein the determination of one of the several switching energy values is performed synchronously to a clock frequency value of the control signal.
4. Method (100) according to one of claims 1 to 3, wherein the determination of one of the several switching energy values is carried out only during a switching operation of the electronic power circuit.
5. Method (100) according to any one of claims 1 to 4, wherein the performance data further includes a conduction loss value of the electronic power circuit.
6. Method (100) according to claim 5, wherein the performance data further includes a temperature value of the electronic power circuit, which is determined on the basis of the switching cycle energy value, the conduction loss value and / or a clock frequency value of the control signal.
7. Method (100) according to any one of claims 1 to 6, wherein the method (100) further comprises providing the performance data.
8. Computing unit for determining the performance data, which is designed and configured to execute the method (100) according to any one of claims 1 to 7.
Citation Information
Patent Citations
Method and apparatus for determining the power loss of semiconductor components of a converter
DE102021212287B3
Analyzing an operating condition of a power converter
EP3627675A1
Method for managing cell and time slot information in wireless network
KR1020240100211A