Method and computing system for simulating a power circuit

Abstract

The present disclosure provides a computer-implemented method for simulating a power circuit, comprising: - determining a Modified Nodal Analysis (MNA) model of the power circuit; - converting the MNA model into a Laplace model that describes the relationship between a circuit waveform and a power supply waveform by a matrix of Laplace functions; - determining a characteristic frequency of the power circuit based on a matrix of a Laplace model; - determining a transfer function from one circuit waveform to each of the other circuit waveforms based on the matrix of the Laplace model; - determining weighting coefficients of the characteristic function of the characteristic frequency based on the initial state of the passive energy components of the power circuit and based on the transfer function; - predicting a circuit waveform of a power circuit based on the characteristic frequency, the transfer function and the weighting coefficient; The present invention relates to a computer-implemented method including:

Description

[Technical Field]

[0001] The present disclosure relates to the simulation of power circuits, and more particularly to a method and computing system for simulating power circuits in a Modified Nodal Analysis (MNA) framework. Priority is claimed to European Patent Application No. 23305759.5, filed May 12, 2023, the contents of which are incorporated herein by reference.

[0002] The present disclosure finds particularly advantageous application in, but by no means limited to, the simulation of electronic circuits. [Background technology]

[0003] Simulation of electronic circuits may be used, for example, during the design stage of an electronic circuit to identify a design that achieves desired specifications through simulation, which design may then be used to manufacture an actual electronic circuit ("real" as compared to a hypothetical electronic circuit).

[0004] Electronic circuit simulation may also be used, for example, to control and / or monitor existing (real) electronic circuits. In such applications, the simulated electronic circuit corresponds to a "digital twin" of the existing electronic circuit and is used to predict the evolution of the state of the existing electronic circuit over time. Based on such predictions, it is possible, for example, to adjust the control of the existing electronic circuit (e.g., adjust the switching times of switches in the existing electronic circuit) and / or to detect malfunctions in the existing electronic circuit, for example, by comparing the predictions with measurements of the actual state of the existing electronic circuit. In such applications, the electronic circuit simulation needs to run faster than real time to enable real-time control adjustments and / or real-time malfunction detection.

[0005] An electronic circuit typically comprises components (resistors, capacitors, inductors, machines, switches, etc.) arranged according to a circuit plan that includes branches interconnected via nodes connected to one or more power sources (i.e., voltage or current sources). In the following, the states of the branches (i.e., the currents flowing through the branches) and nodes (i.e., the voltages at each node) are referred to as circuit waveforms,

number

number

number

number

number

number

number

number

number

[0006] Such electronic circuits may be modeled using the MNA framework, in which the electronic circuit is described by a set of Kirchhoff's laws that relate the voltage at each node to the current flowing in each branch. In the MNA framework, the electronic circuit is modeled using the circuit waveforms

number

number

number

number

number

number

[0007] The main advantage of the MNA framework is that it

number

number

number

[0008] Electronic circuit simulation may also be performed using a state-space analysis (SSA) framework, where matrices relate vectors of energy states to their derivatives and power sources. The energy state vectors include waveforms that drive the accumulation of energy in passive energy components (e.g., capacitor voltages, inductor currents) of the electronic circuit.

number

number

number

number

[0009] In the SSA framework, the energy state vector

number

number

number

number

number

number

number

[0010] The problem with the SSA framework is that it

number

number

number

number

number

number

number

number

number

number

number

number

[0011] The present disclosure aims to improve this situation. In particular, the present disclosure aims to address at least some or all of the limitations of the prior art mentioned above by proposing a solution for automatically predicting circuit waveforms of any physical system that can be modeled in an MNA framework by automatically solving an MNA model (which can be automatically generated) without any approximation. [Means for solving the problem]

[0012] To this end, according to a first aspect, the present disclosure provides a computer-implemented method for simulating an electric power circuit, the electric power circuit comprising components arranged according to a circuit plan with branches interconnected via nodes, the states of each of the branches and nodes corresponding to circuit waveforms of the electric power circuit, the electric power circuit being powered by one or more power sources; - determining a Modified Nodal Analysis (MNA) model of the power circuit, the MNA model describing the relationship between the circuit waveform, derivatives of the circuit waveform, and the power supply waveform; - converting the MNA model into another model called a Laplace model, which describes the relationship between the circuit waveform and the power supply waveform by a matrix of Laplace functions; - determining a characteristic frequency of the power circuit based on a matrix of a Laplace model; - determining a transfer function from one circuit waveform, called the reference circuit waveform, to each of the other circuit waveforms based on the matrix of the Laplace model; - determining weighting coefficients of the characteristic function of the characteristic frequency based on the initial state of the passive energy components of the power circuit and based on the transfer function for the reference circuit waveform; - predicting a circuit waveform of a power circuit based on a characteristic frequency, a transfer function, and a weighting factor; The present invention relates to a method comprising:

[0013] The proposed solution therefore relates to a computer-implemented simulation of an electric power circuit. In this disclosure, a "electric power circuit" corresponds to any interconnection of components (especially impedances), including power sources and passive energy components. In practice, such electric power circuits may represent a variety of different physical systems. Of course, as mentioned above, such electric power circuits may also represent electronic circuits (in which case the power source is the source of electrical power).

[0014] However, other physical systems may be represented by such power circuits. For example, a thermal network may be represented by such a power circuit (in this case, the power source is the heat source). In fact, the heat sources, thermal impedances, and temperature levels of the thermal network are equivalent to the current sources, impedances, and voltage levels of the electronic circuit. For example, a simulation of the power circuit representing the thermal network may be used to estimate the junction temperatures of the power dies included in the electronic circuit. In such an example, the power losses due to the conduction and switching of electronic currents through each power die behave like heat sources flowing into the thermal impedance network between the power die and the baseplate. The estimated junction temperatures may serve as inputs to the control of the electronic circuit. For example, the switching frequency and / or current levels may be adjusted to comply with certain thermal constraints. The junction temperatures estimated by the digital twin (power circuit) compared to measurements can be used to monitor the degradation of the power die interconnects, which is useful for condition-based maintenance.

[0015] Other physical systems that may be represented by such power circuits include, for example, reluctance networks, mechanical systems, hydraulic systems, pneumatic systems, and the like.

[0016] The proposed method solves the MNA equation system (i.e., the MNA model) by using the Laplacian operator to replace the matrix of the MNA model with a single matrix of Laplace functions that directly link the circuit waveforms to the waveforms of one or more power sources. By processing this matrix of the Laplace model, the characteristic frequencies of the power circuit can be determined without having to identify an SSA model of the power circuit.

[0017] By focusing on one of the circuit waveforms, considered the reference circuit waveform, it can be shown that the reference circuit waveform can be formally expressed, without approximation, as a linear combination of characteristic functions (i.e., exponential functions) at (complex) characteristic frequencies combined with terms derived from the power supply waveform. Each of the other circuit waveforms can also be formally expressed, without approximation, as a different linear combination of the same characteristic functions at the same characteristic frequencies combined with terms derived from the power supply waveform. Processing the Laplace model matrix also allows for the identification, without approximation, of transfer functions that allow the weighting factors used in the linear combination of any circuit waveform to be determined from the weighting factors used in the linear combination of the reference circuit waveform.

[0018] The weighting factors used in the linear combination of the reference circuit waveforms can be determined based on the known initial conditions of the power circuit (as in the SSA framework), i.e., based on the initial states of the passive energy components of the power circuit (components that can store energy, i.e., passive energy components that are capacitors and inductors in the case of electronic circuits) and based on the initial states of the power waveforms. The transfer function can then be used to determine the weighting factors for each of the other circuit waveforms in the power circuit.

[0019] All circuit waveforms are formally represented and characterized so that they can be calculated without any approximation and with any relation to the initial conditions of the power circuit.

number

[0020] In certain embodiments, the method according to the first aspect may further comprise one or more of the following optional features, considered alone or in any technically possible combination:

[0021] In certain embodiments, the characteristic frequencies and transfer functions are determined based on triangularization of the matrix of the Laplace model.

[0022] In certain embodiments, the transfer function is determined by transforming the matrix of the Laplace model into a transformed matrix that is diagonal except for a single column that has no null components, where the transfer function is determined based on said single column of the transformed matrix.

[0023] In a particular embodiment, the method includes determining a filtered power supply waveform based on triangularization of a matrix of a Laplace model and based on the power supply waveform, and a circuit waveform is predicted based on the filtered power supply waveform.

[0024] In certain embodiments, the method comprises the steps of: - controlling an existing physical system represented by the simulated power circuit or controlling another existing physical system coupled to the existing physical system represented by the simulated power circuit; - designing a physical system to be manufactured; - monitoring an existing physical system by comparing measured circuit waveforms on the existing physical system with predicted circuit waveforms; - testing an existing control device to be used to control the physical system represented by the simulated power circuit. using the predicted circuit waveform to perform at least one of:

[0025] In certain embodiments, the simulated power circuit represents an electronic circuit and / or a thermal network.

[0026] In certain embodiments, the method includes an initialization phase and a simulation phase, the initialization phase being performed before the simulation phase; The initialization phase includes determining the MNA model, converting the MNA model to a Laplace model, determining the characteristic frequency and determining the transfer function; The simulation phase includes determining weighting coefficients and predicting the circuit waveforms of the power circuit.

[0027] In certain embodiments, the initialization phase includes calculating the inverse of a matrix describing the relationship between the weighting coefficients of the reference circuit waveform and the states of the passive energy components of the power circuit, the matrix to be inverted being dependent on the transfer function.

[0028] In certain embodiments, the initialization phase is performed independently of the operation of the physical system represented by the simulated power circuit, and the simulation phase is performed in parallel with the operation of the physical system to control or monitor said physical system in real time.

[0029] According to a second aspect, the present disclosure relates to a computer-implemented method for simulating a power circuit, the power circuit comprising a component including at least one switch, the at least one switch having different states defining different states of the power circuit, in each state the power circuit being divided by the at least one switch into one or more power sub-circuits, each power sub-circuit being simulated by using a method according to any one of the embodiments of the present disclosure.

[0030] In certain embodiments, the method according to the second aspect may further comprise one or more of the following optional features, considered alone or in any technically possible combination:

[0031] In certain embodiments, the method comprises: - identifying a power circuit and the current state of one or more power subcircuits that make up the power circuit in a current state; - determining an initial state of a passive energy component of the power circuit at a previous (latest) switching time of at least one switch, the initial state of the passive energy component at the previous switching time corresponding to a state of the passive energy component reached at the previous switching time by the power circuit in a previous state before the previous switching time; - predicting a circuit waveform of the power circuit by simulating each of the one or more power subcircuits in a current state of the power circuit based on the initial state of the passive energy components at a previous switching time; Includes.

[0032] In certain embodiments, the method includes an initialization phase and a simulation phase, the initialization phase being performed before the simulation phase; the initialization phase includes, for each different power subcircuit of the power circuit, determining an MNA model, transforming the MNA model into a Laplace model, determining a characteristic frequency and determining a transfer function; The simulation phase includes determining weighting coefficients and predicting the circuit waveforms of the power circuit.

[0033] According to a third aspect, the present disclosure relates to a computer program product comprising instructions that, when executed by at least one processor, configure the at least one processor to perform a method according to any one of the embodiments of the present disclosure.

[0034] According to a fourth aspect, the present disclosure relates to a computer-readable storage medium comprising instructions that, when executed by at least one processor, configure the at least one processor to perform a method according to any one of the embodiments of the present disclosure.

[0035] According to a fifth aspect, the present disclosure relates to a computing system comprising at least one memory and at least one processor, wherein the at least one processor is configured to execute a method according to any one of the embodiments of the present disclosure. [Brief explanation of the drawings]

[0036] The present disclosure will be better understood on reading the following description, given by way of in no way limiting example and made with reference to the figures in which: [Figure 1] FIG. 1 is a schematic diagram of an exemplary power circuit. [Figure 2] FIG. 2 is a diagram illustrating the main steps of an exemplary embodiment of a method for simulating a power circuit. [Figure 3] FIG. 1 is a diagram of a first Gaussian pivot applied to a matrix of a Laplace model of a power circuit. [Figure 4] FIG. 10 is a diagram of a second Gaussian pivot applied to a matrix of a Laplace model of a power circuit. [Figure 5]FIG. 1 illustrates an initialization phase and a simulation phase of an exemplary embodiment of a simulation method. [Figure 6] 1 is a schematic diagram of a decomposition of a power circuit into power sub-circuits based on the blocking and / or passing states of the switches of the power circuit. DETAILED DESCRIPTION OF THE INVENTION

[0037] In these figures, the same reference numbers from one figure to another indicate the same or similar elements. For clarity, elements shown are not drawn to scale unless otherwise noted.

[0038] Additionally, the order of steps depicted in these figures is provided for illustrative purposes only and is not meant to limit the present disclosure as it may apply to the same steps performed in different orders.

[0039] FIG. 1 illustrates a schematic diagram of an example of a power circuit 10. As shown in FIG. 1, the power circuit 10 is composed of branches 12 interconnected via nodes 11. The power circuit 10 includes components (not shown) powered by one or more power sources (not shown), and each branch 12 of the power circuit 10 includes one or more components and / or one or more power sources. The components include impedances, some of which are passive energy components, i.e., components that can store energy received from a power source. The arrangement of the components and power sources within the power circuit 10 is called a circuit plan.

[0040] As noted above, such a power circuit 10 may represent a variety of different physical systems, and the physical system represented by the simulated power circuit 10 may be, for example, an electronic circuit, a thermal network, a reluctance network, a mechanical system, a hydraulic system, a pneumatic system, etc.

[0041] In the following, the simulated power circuit is considered, without limitation, to represent an electronic circuit. Accordingly, the power sources of the power circuit 10 correspond to power sources, i.e., one or more voltage sources and / or one or more current sources. The components include impedances (resistors, capacitors, inductors), some of which are passive energy components (capacitors and inductors). The power circuit 10 may also include other components, such as transformers, machines, switches, voltage- or current-controlled power supplies, etc.

[0042] For example, the simulated power circuit 10 may represent a DC / DC boost chopper circuit. Other non-limiting examples of electronic circuits that may be represented by power circuit 10 include DC / DC, DC / AC, AC / AC circuits such as voltage source converters, current source inverters, or any combination of two or more of these electronic circuits, such as a DC / DC circuit feeding a DC / AC circuit.

[0043] FIG. 2 shows in schematic form the main steps of a method 20 for simulating the power circuit 10 .

[0044] Simulation method 20 is executed by a computing system (not shown). In a preferred embodiment, the computing system comprises one or more processors (which may belong to the same computer or different computers) and one or more memories (which may belong to the same computer or different computers). The one or more processors may include, for example, a central processing unit (CPU), a digital signal processor (DSP), a field programmable gate array (FPGA), an application-specific integrated circuit (ASIC), etc. The one or more memories may include any type of computer-readable volatile and non-volatile memory (magnetic hard disk, solid-state disk, optical disk, electronic memory, etc.). The one or more memories may store a computer program product in the form of a set of program code instructions executed by the one or more processors to perform all or part of the steps of simulation method 20. In other words, the computing system comprises a set of means constituted by software (specific computer program products) and / or hardware (CPU, DSP, FPGA, ASIC, etc.) to perform the steps of simulation method 20.

[0045] 2, the simulation method 20 includes a step S20 of determining an MNA model of the power circuit 10. As described above, the MNA model is a system of equations that describes the relationship between the circuit waveforms (i.e., the states of each of the branches 12 and nodes 11 of the power circuit 10), the derivatives of the circuit waveforms, and the power supply waveforms (i.e., the states of the power supply for each branch 12 and each node 11).

[0046] For example, a power circuit 10 that has full connectivity (i.e., the power circuit 10 cannot be divided into two or more independent power circuits) and has one node 11 that is grounded (i.e., connected to electrical ground);

number

number

number

number

number

number

number

number

number

number

number

number

number

[0047] As shown in FIG. 2, the simulation method 20 simulates the MNA model as a single matrix of Laplace functions.

number

number

number

number

number

number

number

number

[0048] As shown in FIG. 2, the simulation method 20 uses the matrix of the Laplace model

number

number

number

number

number

number

number

number

[0049] In some embodiments, the characteristic frequency is determined by the matrix of the Laplace model

number

number

number

number

number

number

number

[0050] In the last dimension, [Equation 5] is the circuit waveform

number

number

[0051] The pivoting configuration allows

number

number

number

[0052] [Equation 6] is the characteristic polynomial

number

number

number

number

[0053] Therefore, in step S22, the characteristic frequency of the power circuit 10

number

number

number

number

[0054] Therefore, the characteristic frequency of the power circuit 10 is expressed by the matrix of the Laplace model

number

number

[0055] As shown in FIG. 2, the simulation method 20 uses the matrix of the Laplace model

number

[0056] By focusing on one of the circuit waveforms, which is considered as the reference circuit waveform, it can be shown that said reference circuit waveform can be formally expressed without approximations as a linear combination of characteristic functions (i.e., exponential functions) of the (complex) characteristic frequencies combined with terms derived from the power supply waveform.

[0057] In fact, for example, the last circuit waveform

number

number

number

number

number

number

number

number

number

[0058] It should be noted that any circuit waveform can be used as the reference circuit waveform. In particular, the circuit waveform vector

number

[0059] Therefore, the purpose of step S23 is to identify transfer functions that allow each of the other circuit waveforms of the power circuit 10 to be expressed based on the reference circuit waveform.

[0060] In some embodiments, the transfer function is expressed as the matrix of the Laplace model in Equation 5:

number

number

number

number

number

number

number

number

[0061] Therefore, unlike the conventional lower-to-higher decomposition, the second Gaussian pivot transform generates a transfer function from the reference circuit waveform to the other circuit waveform,

number

number

number

[0062] In fact, the standard vector (

number

number

number

[0063] Based on [Equation 8], the circuit waveform

number

number

number

number

number

[0064] Therefore, each of the other circuit waveforms is the power supply waveform

number

number

number

number

number

[0065] As shown in FIG. 2, the simulation method 20 calculates the weighting coefficients of the reference circuit waveform.

number

[0066] Weighting Factor

number

number

number

number

number

[0067] energy state vector

number

number

number

number

number

[0068] The initial state of the passive energy component, i.e.

number

number

number

[0069] Using [Equation 11], the matrix

number

number

number

number

number

number

number

number

[0070] As shown in FIG. 2, the simulation method 20 calculates the characteristic frequency

number

number

number

number

number

number

number

number

number

number

number

number

number

[0071] As can be seen in [Equation 15], the circuit waveform

number

number

number

number

[0072] As mentioned above, the filtered power waveform vector

number

number

number

number

number

number

[0073] For example, the power supply waveform

number

number

number

number

[0074] According to another example, the power supply waveform

number

number

number

number

number

number

number

[0075] According to another example, the power supply waveform

number

number

number

[0076] Also, although some processing steps (e.g., steps for determining characteristic frequencies and transfer functions) may be computationally expensive in some cases, they depend only on the circuit design of the power circuit 10 and can be performed offline without real-time constraints. Second, processing steps that depend on the operating conditions of the power circuit 10 (e.g., determining weighting coefficients for reference circuit waveforms and predicting circuit waveforms) have limited computational complexity because they can use data determined offline. Therefore, processing steps that depend on the operating conditions of the power circuit 10 can be performed faster than real-time conditions, even for complex power circuits.

[0077] Thus, in some embodiments, as shown in FIG. 5, the simulation method 20 may be decomposed into an initialization phase 21 and a simulation phase 22, with the initialization phase 21 being performed before the simulation phase 22.

[0078] Typically, the initialization phase 21 is a step that does not depend on the operating conditions of the power circuit 10, e.g., - Determine the MNA model (i.e., the matrix

number

number

number

number

number

number

number

number

number

[0079] In some cases, the initialization phase 21 may be necessary to calculate the weighting coefficients, since this may be computationally expensive and only needs to be calculated once at every point, e.g., the matrix

number

number

[0080] As mentioned above, the initialization phase 21 does not depend on the operating conditions of the power circuit 10 (i.e., it does not depend on the initial states of passive energy components and the power circuit waveforms), and therefore can be performed independently of the operation of the physical system represented by the power circuit 10 and does not need to be performed in real-time conditions regardless of the intended use of the simulation.

[0081] Typically, the simulation phase 22, which corresponds to the dynamic portion of the simulation, includes steps that depend on the operating conditions of the power circuit 10, e.g., -Weighting factor

number

number

number

number

number

[0082] As noted above, simulation phase 22 may be performed under real-time conditions or at speeds faster than real-time conditions, if required for the application under consideration (e.g., when used to control and / or monitor in real time an existing physical system represented by power circuit 10). Indeed, the computational complexity of prediction step S25 (and simulation phase 22) may vary depending on the time at which the circuit waveforms must be predicted.

number

[0083] In some cases, power circuit 10 may also include one or more switches (e.g., diodes, transistors, etc.) that typically include a blocking (aka open) state (current flow is blocked in branch 12) and a conducting (aka closed) state (current can flow in branch 12).

[0084] A power circuit 10 including one or more switches can be analyzed according to the respective states of the switches. In practice, when a switch is blocked (i.e., it transitions from a conducting state to a blocking state), the power circuit 10 can be divided into two power subcircuits. Therefore, as shown in FIG. 6, the topology of the power circuit 10 dynamically evolves according to the respective states of one or more switches. The respective states of the one or more switches define different states of the power circuit 10. For a given switching state, switches in the blocking state are removed from the power circuit 10, and switches in the passing state are replaced with shortcut current paths (assuming the switches are ideal switches). Thus, the power circuit 10 is decomposed into one or several independent power subcircuits constructed only with passive components and power sources. Each different power subcircuit can be simulated using the simulation method 20.

[0085] FIG. 6 shows a schematic diagram of an example of a power circuit 10 including two switches 13. Part a) of FIG. 6 shows the state of the power circuit 10 when both switches are in a conducting state, which includes only one power subcircuit 10-1 (corresponding to the entire power circuit 10). Part b) of FIG. 6 shows the state of the power circuit 10 when one of the switches is in a conducting state and the other is in a blocking state, which includes only one power subcircuit 10-2, which is different from the power subcircuit 10-1 of part a). Part c) of FIG. 6 shows the state of the power circuit 10 when the states of the switches are reversed relative to part b), which includes two power subcircuits 10-3 and 10-4, which are different from the power subcircuits 10-1 and 10-2. Portion d) of Figure 6 represents the state of power circuit 10 when both switches are in the off state, with power subcircuit 10-4 the same as in portion c) and power subcircuit 10-5 different from power subcircuits 10-1, 10-2, 10-3, and 10-4. Thus, power circuit 10 can be decomposed into different power subcircuits 10-1 through 10-5 depending on the switch states, and each different power subcircuit 10-1 through 10-5 can be simulated as described above in connection with Figure 2.

[0086] In effect, the switching states define different power sub-circuits of the power circuit 10, each power sub-circuit comprising a subset of the passive energy components of the power circuit 10. If the power circuit 10 is divided into multiple power sub-circuits, each power sub-circuit may contribute a fraction of the total number of circuit waveforms.

number

number

number

number

number

number

[0087] Therefore, each

number

number

number

number

number

[0088] Finally, the circuit waveform vector

number

number

[0089] The initial conditions considered for the passive energy component are:

number

number

number

[0090] Therefore, the circuit waveform

number

number

number

[0091] If the simulation method 20 includes an initialization phase 21 and a simulation phase 22 as described above, the initialization phase 21 may be performed for all power subcircuits that may be active during the simulation when changing the states of the switches of the power circuit 10. These potentially computationally expensive tasks are performed only once. The simulation phase 22 is performed by independently simulating each active power subcircuit in the current state of the power circuit 10 (the current state of the power circuit 10 is completely defined by the respective states of the switches of the power circuit 10). As described above, at any time

number

number

number

[0092] For example, a simulation of the power circuit 10 may be used to control an existing physical system represented by the simulated power circuit 10. For example, if the power circuit 10 represents an existing electronic circuit including switches (e.g., an H-bridge, etc.), the predicted circuit waveforms may be used to adjust the switching times of the switches of the existing electronic circuit. For example, a control device controlling the switches of an existing physical system may desire to place the switches in a blocking state when the current through the switches becomes null, achieving zero-current switching to reduce commutation losses in the switches. The timing of when the current becomes null can be predicted using the proposed simulation method 20. Similarly, the timing of when the switches should be returned to a conducting state may be determined using the proposed simulation method 20, where zero-voltage switching is achieved when the voltage across the switches becomes null, reducing commutation losses. According to another non-limiting example, a control device controlling the switches of an existing physical system may use simulation predictive control to determine which switches should be placed in a different state next and when there is an incentive to improve, for example, predetermined key performance indicators (e.g., reduced losses, reduced capacitor voltage ripple, reduced inductor current ripple, etc.). The proposed simulation method 20 can predict accurate results faster than real-time conditions, and therefore key performance indicators can be further improved by using simulation predictive control than by using simple models as in the case of model predictive control (MPC).

[0093] A simulation of the power circuit 10 may also be used to control another existing physical system coupled to the existing physical system represented by the simulated power circuit 10 (e.g., simulating an existing thermal network to estimate the junction temperature of a power die of an existing electronic circuit, and using the estimated junction temperature to control the existing electronic circuit, e.g., to prevent the junction temperature from reaching a critical value).

[0094] By another non-limiting example, a simulation of power circuit 10 may be used to monitor an existing physical system represented by simulated power circuit 10. For example, a predicted circuit waveform may be compared to measurements of the circuit waveform performed on the existing physical system, and a malfunction of the existing physical system can be detected, e.g., if the predicted circuit waveform differs significantly from the measured one. For example, if simulated power circuit 10 is a digital twin of an existing DC / DC boost chopper circuit with sensors on the DC bus capacitors, a malfunction of the capacitors can be detected, e.g., when the current ripple measured by the sensors exceeds the current ripple of the corresponding predicted (simulated) circuit waveform.

[0095] By another non-limiting example, a simulation of the power circuit 10 may be used to design a physical system to be manufactured. In such cases, the simulated power circuit 10 represents a physical system that does not yet exist (as opposed to real-time control and / or monitoring cases that deal with existing, i.e., actual, physical systems), and the goal of the simulation is to identify a power circuit 10 that exhibits a desired predicted circuit waveform through simulation, possibly by simulating multiple power circuits 10 with different designs (e.g., different circuit plans and / or different components). Once a simulated power circuit 10 with the desired predicted circuit waveform, referred to as an optimal power circuit 10, is identified, a physical system may be manufactured based on the design of the optimal power circuit 10 found through simulation.

[0096] By another non-limiting example, a simulation of power circuit 10 may be used to test an existing control device used to control the physical system represented by the simulated power circuit 10. For example, an existing control device to be used to control a switch in an electronic circuit (existing or not) may be tested by simulating the electronic circuit to generate the predicted circuit waveforms (e.g., at ADC sampling events) that are fed to the existing control device to verify that the control device generates the desired control signal in response to the predicted circuit waveforms. This corresponds to a “simulation-in-the-loop” configuration for testing a control device. For example, if the simulated power circuit 10 represents a DC / DC boost chopper circuit, then a simulation of power circuit 10 can be used to verify that the control signals generated by the control device actually enable stabilization of the (predicted) voltage of a bus capacitor to a desired regulation level. This can also be used for more complex DC / DC, DC / AC, and AC / AC circuits, such as voltage-source converters, current-source inverters, etc. Because simulations can be performed faster than real time using the present disclosure, the proposed solution therefore enables real-time verification of control devices that would otherwise require expensive and bulky “hardware-in-the-loop” test platforms.

[0097] It should be noted that the present disclosure is not limited to the above exemplary embodiments, and modifications of the above exemplary embodiments are also included within the scope of the present disclosure.

[0098] For example, the present disclosure has been provided primarily by considering power circuits that represent electronic circuits (existing or otherwise). However, as noted above, the proposed solution may be used to simulate power circuits 10 that represent a variety of different physical systems. For example, the physical system represented by power circuit 10 may be, for example, an electronic circuit, a thermal network, a hydraulic system, etc., or any combination thereof. Generally speaking, the proposed solution may be used to simulate power circuits 10 that represent any physical system (existing or otherwise) that can be represented in an MNA framework.

Claims

1. A computer-implemented method (20) for simulating an electric power circuit (10), the electric power circuit comprising components arranged according to a circuit plan comprising branches (12) interconnected via nodes (11), the states of each of the branches and the nodes corresponding to a circuit waveform of the electric power circuit, the electric power circuit being powered by one or more power sources, the method comprising: - determining (S20) a Modified Nodal Analysis (MNA) model of the power circuit, the MNA model describing the relationship between the circuit waveforms, the derivatives of the circuit waveforms and the power supply waveform; - converting the MNA model into another model called a Laplace model, which describes the relationship between the circuit waveform and the power supply waveform by a matrix of Laplace functions (S21); - determining (S22) a characteristic frequency of the power circuit based on the matrix of the Laplace model; - determining (S23) a transfer function from one circuit waveform, called the reference circuit waveform, to each of the other circuit waveforms based on said matrices of said Laplace model; - determining (S24) for said reference circuit waveform weighting coefficients of the characteristic function of said characteristic frequency based on the initial state of the passive energy components of said power circuit and based on said transfer function; - predicting the circuit waveform of the power circuit based on the characteristic frequency, the transfer function and the weighting coefficients (S25); A computer-implemented method (20) comprising:

2. The method (20) of claim 1, wherein the characteristic frequencies and the transfer functions are determined based on triangularization of the matrix of the Laplace model.

3. 3. The method of claim 2, wherein the transfer function is determined by transforming the matrix of the Laplace model into a transformed matrix that is diagonal except for a single column that has no null components, and the transfer function is determined based on the single column of the transformed matrix.

4. 4. The method (20) of claim 2 or 3, comprising determining a filtered power supply waveform based on the triangularization of the matrix of the Laplace model and based on the power supply waveform, wherein the circuit waveform is predicted based on the filtered power supply waveform.

5. The following steps, - controlling an existing physical system represented by the simulated power circuit or controlling another existing physical system coupled to the existing physical system represented by the simulated power circuit; - designing a physical system to be manufactured; - monitoring an existing physical system by comparing measured circuit waveforms on said existing physical system with said predicted circuit waveforms; - using the predicted circuit waveforms to perform at least one of the steps of testing an existing control device to be used to control the physical system represented by the simulated power circuit.

5. The method (20) of any one of claims 1 to 4, comprising:

6. The method (20) of any one of claims 1 to 5, wherein the simulated power circuit represents an electronic circuit and / or a thermal network.

7. an initialization phase (21) and a simulation phase (22), the initialization phase being executed before the simulation phase; the initialization phase includes the determination of the MNA model, the transformation of the MNA model into the Laplace model, the determination of the characteristic frequencies and the determination of the transfer function; the simulation phase includes the determination of the weighting coefficients and the prediction of the circuit waveforms of the power circuit, The method (20) according to any one of claims 1 to 6.

8. 8. The method (20) of claim 7, wherein the initialization phase includes calculating an inverse of a matrix describing a relationship between the weighting coefficients of the reference circuit waveform and states of the passive energy components of the power circuit, the matrix being inverted being dependent on the transfer function.

9. 9. The method (20) of claim 7 or 8, wherein the initialization phase is performed independently of the operation of a physical system represented by the simulated power circuit, and the simulation phase is performed in parallel with the operation of the physical system to control or monitor the physical system in real time.

10. 10. A computer-implemented method for simulating an electric power circuit, the electric power circuit comprising a component including at least one switch, the at least one switch having different states defining different states of the electric power circuit, in each state the electric power circuit being divided into one or more power sub-circuits by the at least one switch, each power sub-circuit being simulated by using the method (20) of any one of claims 1 to 9.

11. - identifying the current state of said power circuit and said one or more power sub-circuits that make up said power circuit in their current state; - determining an initial state of the passive energy component of the power circuit at a previous switching time of the at least one switch, the initial state of the passive energy component at the previous switching time corresponding to a state of the passive energy component reached at the previous switching time by the power circuit in a previous state before the previous switching time; - predicting the circuit waveform of the power circuit by simulating each of the one or more power sub-circuits in the current state of the power circuit based on the initial state of the passive energy component at the previous switching time; The method of claim 10, comprising:

12. an initialization phase and a simulation phase, the initialization phase being executed before the simulation phase; the initialization phase comprises, for each different power subcircuit of the power circuit, the determination of the MNA model, the transformation of the MNA model into the Laplace model, the determination of the characteristic frequency and the determination of the transfer function; the simulation phase includes the determination of the weighting coefficients and the prediction of the circuit waveforms of the power circuit, 12. The method according to claim 10 or 11.

13. A computer program product comprising instructions that, when executed by at least one processor, configure the at least one processor to perform the method of any one of claims 1 to 12.

14. A computer-readable storage medium comprising instructions that, when executed by at least one processor, configure the at least one processor to perform the method of any one of claims 1 to 12.

15. A computing system comprising at least one memory and at least one processor, said at least one processor configured to execute the method of any one of claims 1 to 12.

Images