SYSTEM AND METHOD FOR CONTROL OF NON-ISOLATED BIDIRECTIONAL POWER CONVERTER - Patent application

Abstract

Disclosed are embodiments including a power converter system and method including an N-phase power converter stage having an alternating current (AC) side and a direct current (DC) side, where N≧1. The system and method further includes an N-phase LC filter including one or more capacitors, one or more neutral points of each of the one or more capacitors being electrically connected to a DC negative terminal of the DC power source. A control system drives power switching elements of the N-phase power converter stage to convert the received power and output the converted power. The control system drives the power switching elements using variable frequency soft switching at a frequency of at least 20 kHz. The power converter can have a bidirectional operation operating in a traction mode to drive a motor or a charging mode to charge the DC power source.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Application No. 63 / 226,136, filed July 27, 2021, U.S. Provisional Application No. 63 / 242,840, filed September 10, 2021, U.S. Provisional Application No. 63 / 345,896, filed May 25, 2022, U.S. Provisional Application No. 63 / 351,768, filed June 13, 2022, U.S. Provisional Application No. 63 / 226,059, filed July 27, 2021, U.S. Provisional Application No. 63 / 270,311, filed October 21, 2021, and U.S. Provisional Application No. 63 / 319,122, filed March 11, 2022, each of which is incorporated by reference in its entirety herein.

[0002] STATEMENT REGARDING FEDERALLY FUNDED RESEARCH This invention was made with Government support under 1653574 awarded by the National Science Foundation. The Government has certain rights in this invention. [Background technology]

[0003] Various types of power converters are manufactured and used in many industries and situations. Exemplary power converters include alternating current (AC) to direct current (DC) rectifiers, DC-AC inverters, and DC-DC converters. An AC-DC rectifier, also called an AC / DC rectifier, converts AC power to DC power. A DC-AC inverter, also called a DC / AC inverter, converts DC power to AC power. Power converters can be used for a variety of purposes, such as rectifying AC power from an AC grid source into DC power for charging a battery, or converting DC power from a battery back to AC power to drive a motor or to supply AC power to an AC grid. Additionally, power converters can be used in or connected to electric vehicles, engine generators, solar panels, and the like in a variety of situations. Summary of the Invention

[0004] A power converter can be described in terms of power conversion efficiency, power density, and cost, among other characteristics. In general, it is desirable to have a power converter that has higher power efficiency, higher power density, and lower cost. A highly efficient power converter can convert power (e.g., AC to DC, DC to AC, and / or DC to DC) without significant energy loss. A less efficient power converter experiences higher energy losses during power conversion. Such energy losses may manifest, for example, as heat generated by the power converter while converting power. The power efficiency of a power converter, inductor, or other electronic component is expressed as a percentage between 0 and 100% and is expressed by the formula:

number

number

[0005] Energy costs, including monetary and environmental costs, continue to be a significant factor across many industries that incorporate power converters. Thus, even small increases in the power efficiency of a power converter (e.g., of a tenth of a percent) are significant and highly desirable. Similarly, reductions in materials and size of power converters are significant and highly desirable, allowing for reductions in the cost and physical space required to house the power converter in systems that incorporate the power converter.

[0006] In grid-connected power converter applications, such as electric vehicle (EV) chargers and photovoltaic (PV) power sources, leakage current and DC bus utilization are factors that affect performance. For leakage current issues, bulky line frequency transformers are typically installed to break the leakage path at the point of common coupling (PCC), which increases the cost, volume, and weight of the system. For DC bus utilization, the DC bus voltage needs to be boosted to at least twice the grid voltage swing to avoid saturation issues that bring additional switching losses and challenges to the switch voltage withstand capability.

[0007] A bidirectional power converter can be used to charge a DC power source with AC power and to drive an AC motor with DC power from the DC power source. Such a power converter may be referred to as an integrated charger when included in an electric vehicle. The integrated charger can be used both as a primary charging interface for the electric vehicle's battery and as a traction inverter to drive the electric vehicle's motor. By using a dual-purpose power converter rather than a separate charger converter and traction inverter, material costs and size can be reduced. However, compared to a dedicated power converter, a dual-purpose power converter adds complexity in designing an efficient and effective converter for both charging and traction modes. Furthermore, design factors extend beyond efficiency issues because without proper design, the power converter can shorten the life of the motor due to leakage currents and / or common mode voltages that can cause current spikes in one or more of the motor bearings, motor shaft, motor windings, and gear train, damaging and shortening the life of these components, respectively.

[0008] Some embodiments disclosed herein address these or other problems. For example, some embodiments disclosed herein relate to a non-isolated power converter having one or more of: (i) injection of multiples of N-phase harmonics for zero sequence voltage control, (ii) a cascade control system, (iii) model predictive control (MPC) for active damping to mitigate resonance, (iv) variable frequency critical soft switching (VFCSS), and (v) a modular converter block. These features may be included in the power converter embodiments independently or in any combination. For example, the power converter may include one of the above features, any two of the above features, any three of the above features, any four of the above features, or all five of the above features. Furthermore, in combination with any of these embodiments, the power converter may include at least one LC filter for each of the N phases of the power converter (N≧1), where a capacitor of each LC filter is connected to the DC bus positive terminal or the negative terminal of the power converter, and optionally, an additional capacitor of each LC filter is connected to the other of the DC bus positive terminal or the negative terminal of the power converter. These capacitors of multiple phases with a common point connected to the positive or negative terminal of the DC bus form a bypass path for zero sequence voltage control. The capacitor coupled to the DC bus positive terminal (upper capacitor) can also reduce both EMI and the total ripple current handling requirements of the power converter without increasing the total capacitance or volume. In some embodiments disclosed herein, an additional drain-source capacitor (C DS ) may be coupled between the drain and source terminals of the power switching device to slow the voltage rise during an on-off transition. This slower voltage rise may reduce switching losses in the power switching device.

[0009] Some embodiments disclosed herein include systems, methods, and other implementations (including hardware, software, and hybrid hardware / software implementations) directed to a novel modular model predictive control (MPC) method for a non-isolated N-phase DC / AC converter with zero-sequence voltage stabilization capability, for N≧1, and optionally for regulated common-mode voltage injection (e.g., third harmonic injection (THI) for three-phase systems, or multiples of any other harmonic) for the purpose of increasing the available fundamental frequency AC voltage swing for a given DC voltage. When N=1 or N=2, the DC / AC power converter is considered to be a single-phase system. When N=3, the power converter is a three-phase system, and when N>3, the power converter is referred to as a multi-phase system. Although the description herein may focus on a three-phase system, various embodiments and features described are applicable to any number of phases.

[0010] This non-isolated topology is designed to connect the common point of the three-phase LC filter capacitors to the positive / negative DC bus terminals to bypass the zero-sequence leakage current. The zero-sequence voltage MPC controller stabilizes the zero-sequence capacitor voltage to a constant, in some embodiments, approximately half of the DC bus voltage. Thus, the leakage current flowing through the grid or other coupling elements is attenuated. Furthermore, the regulated third harmonic voltage injection (THI) technique disclosed herein improves the utilization of the DC bus. By adding the third harmonic to the zero-sequence voltage MPC reference, stability and robustness are improved. Compared to conventional THI techniques, grid-connected power quality is improved since no extra harmonics are injected into the grid. Explicit MPC per phase simplifies the implementation complexity on the controller (e.g., digital signal processor (DSP)) and does not require updating angular rates in the state space matrix, which allows for offline MPC optimization. Compared to conventional proportional-integral (PI) controllers, the embodiments of the MPC controller disclosed herein provide power converter control with improved dynamic performance and control bandwidth with faster response.

[0011] Effective zero sequence voltage control (by the LC filters and control schemes disclosed herein) also helps to reduce certain bearing, shaft, motor winding, gear train, and other currents that can potentially damage and shorten the life of motors and their components (bearings, shafts, wiring, etc.). For example, currents caused by high voltage rates of change (dV / dt), especially at higher voltages (e.g., levels above 400V, above or approaching 800V, and therebetween), can cause damage to motor bearings, motor shafts, motor windings (e.g., insulation can be damaged), and gear trains (e.g., bearing currents can propagate to the gear train via electromagnetic interference (EMI) or noise, vibration, and harshness (NVH) resulting from damaged bearing race walls).

[0012] In some examples, the power converter is driven using a variable frequency critical soft switching (VFCSS) scheme, which can provide improved efficiency and reduced filter volume (i.e., improved power density) of the power converter.

[0013] In some examples, the power converter is implemented by a combination of modular converter units or modules, also called autoconverter modules (ACMs), that are coupled together like building blocks to form a power converter having desired specifications. Each ACM may include, for example, a circuit board with input / output connections (e.g., for coupling to other ACMs and a central controller) and a converter block including power switching elements and an LC filter (e.g., configured in a half-bridge configuration). In one embodiment, a non-isolated power converter system is provided that includes an N-phase power converter stage having an alternating current (AC) side and a direct current (DC) side, where N>1, an N-phase LC filter comprising one or more capacitors, where one or more neutral points of each of the one or more capacitors are electrically connected to a DC negative terminal of a DC power source, and a control system configured to drive power switching elements of the N-phase power converter stage to convert received power and to output the converted power, the control system configured to drive the power switching elements using variable frequency soft switching at a frequency of at least 20 kHz.

[0014] In one embodiment, a method for power conversion is provided, the method including: receiving input power from an alternating current (AC) side or a direct current (DC) side by an N-phase power converter stage, where N≧1, filtering on the AC side of the N-phase power converter stage by an N-phase LC filter comprising one or more capacitors, one or more neutral points of each of the one or more capacitors being electrically connected to a DC negative terminal of a DC power source, and driving power switching elements of the N-phase power converter stage by a control system to convert the input power and output a converted power, the control system being configured to drive the power switching elements using variable frequency soft switching at a frequency of at least 20 kHz.

[0015] In one embodiment, a non-isolated power converter system for an electric vehicle is provided, the system includes an N-phase power converter stage having an alternating current (AC) side and a direct current (DC) side, where N≧1, the DC side including a DC power supply terminal, an N-phase LCL filter including one or more capacitors, where one or more neutral points of each of the one or more capacitors are electrically connected to a negative DC terminal of the DC power supply terminals, and a control system configured to drive power switching elements of the N-phase power converter stage to convert input AC power received via the AC terminals to output DC power provided to the DC power supply terminals in a charging mode to charge the DC power source, and to convert input DC power received via the DC power supply terminals to output AC power provided to the AC terminals in a traction mode to drive a motor.

[0016] In one embodiment, a method of converting power for an electric vehicle is provided, the method including receiving input power from an alternating current (AC) side having AC terminals or a direct current (DC) side having DC power supply terminals by an N-phase power converter stage, where N≧1, filtering on the AC side of the N-phase power converter stage by an N-phase LC filter comprising one or more capacitors, one or more neutral points of each of the one or more capacitors being electrically connected to a DC negative terminal of the DC power supply terminals, and driving, by a control system, power switching elements of the N-phase power converter stage to convert the input AC power received via the AC terminals to output DC power provided to the DC power supply terminals in a charging mode to charge a DC power source, and to convert the input DC power received via the DC power supply terminals to output AC power provided to the AC terminals in a traction mode to drive a motor.

[0017] The above and other aspects and advantages of the present disclosure will become apparent from the following description. In the description, reference is made to the accompanying drawings, which form a part of this specification and show one or more embodiments by way of example. However, these embodiments do not necessarily represent the full scope of the invention, and therefore, reference should be made to the claims and this specification to interpret the scope of the invention. In the following description, like reference numerals are used to refer to like parts from figure to figure. [Brief description of the drawings]

[0018] [Figure 1] 1 illustrates a power converter system according to some embodiments. [Diagram 2] 1 illustrates a half-bridge power converter according to some embodiments. [Figure 3A] 1 illustrates a multi-phase power converter system according to some embodiments. [Figure 3B-C] 3B and 3C illustrate a multi-phase power converter in a charging mode and a traction mode, respectively, in accordance with some embodiments. [Figure 3D] 1 shows a simplified equivalent circuit of the parasitic capacitances of an electric machine. [Figure 4] 1 illustrates a converter system according to some embodiments. [Figure 5A-B] 13 illustrates a waveform for third harmonic injection according to some embodiments. [Figure 6] 1 illustrates a communication system for a cascade control system according to some embodiments. [Figure 7] 1 illustrates an MPC-based converter system according to some embodiments. [Figure 8] 1 illustrates a model predictive control (MPC) control system according to some embodiments. [Figure 9] 1 illustrates a state estimator according to some embodiments. [Figure 10] 1 illustrates timing diagrams and boundary conditions for soft switching according to some embodiments. [Figure 11]1 illustrates a control system for variable frequency critical soft switching according to some embodiments. [Figure 12] 1 illustrates a power converter system including model predictive control (MPC) with variable frequency critical soft switching (VFCSS) in accordance with some embodiments. [Figure 13] 1 illustrates a control system for local MPC-VFCSS control using variable continuous frequency critical soft switching (VCFCCS) according to some embodiments. [Figure 14] 1 illustrates a control system for local MPC-VFCSS control using variable discrete frequency critical soft switching (VDFCCS) according to some embodiments. [Figure 15] 1 illustrates waveforms for VCFCCS and VDFCCS control according to some embodiments. [Figure 16] 1 illustrates plots of carrier and sampling signals for VDFCCS control according to some embodiments. [Figure 17A-B] 1 shows experimental results for a power converter according to some embodiments. [Figure 18A-B] 1 illustrates an auto-converter module according to some embodiments. [Figure 19] 1 illustrates a power converter incorporating an auto-converter module according to some embodiments. [Figure 20] 1 illustrates a process for converting power using variable frequency critical soft switching according to some embodiments. [Figure 21] 1 illustrates a process for converting power in charging and traction modes according to some embodiments. [Figure 22] 10A-10C show respective efficiency plots of experimental results for power converters according to some embodiments. [Diagram 23] 10A-10C show respective efficiency plots of experimental results for power converters according to some embodiments. [Figure 24] 1 illustrates a configurable power converter architecture according to some embodiments. [Figure 25A]25 illustrates a single-phase grid application configuration of the power converter architecture of FIG. 24 according to some embodiments. [Figure 25B] 25 illustrates a single-phase grid application configuration of the power converter architecture of FIG. 24 according to some embodiments. [Figure 26A] 25 illustrates a three-phase grid application configuration of the power converter architecture of FIG. 24 according to some embodiments. [Figure 26B] 25 illustrates a three-phase grid application configuration of the power converter architecture of FIG. 24 according to some embodiments. [Figure 27A] 25 illustrates a motor application configuration of the power converter architecture of FIG. 24 according to some embodiments. [Figure 27B] 25 illustrates a motor application configuration of the power converter architecture of FIG. 24 according to some embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] One or more embodiments are described and illustrated in the following description and the accompanying drawings. These embodiments are not limited to the specific details provided herein and may be modified in various ways. Furthermore, there may be other embodiments not described herein. Also, functions performed by multiple components may be integrated and performed by a single component. Similarly, functions described herein as being performed by one component may be distributed and performed by multiple components. Furthermore, components described as performing a particular function may also perform additional functions not described herein. For example, a device or structure that is "configured" in a particular way may be configured in at least that way, but may also be configured in other ways not recited.

[0020] As used in this application, "non-transitory computer-readable medium" includes all computer-readable media, but not consisting of a transitory propagating signal. Thus, non-transitory computer-readable media can include, for example, hard disks, CD-ROMs, optical storage devices, magnetic storage devices, Read Only Memory (ROM), Random Access Memory (RAM), register memory, processor cache, or any combination thereof.

[0021] Furthermore, the phrases and terms used herein are for purposes of explanation and should not be considered as limiting. For example, the use of "comprising," "including," "containing," "having," and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. Furthermore, the terms "connected" and "coupled" are used broadly and include both direct and indirect connections and couplings and can refer to physical or electrical connections or couplings. Furthermore, the step "and / or" used in conjunction with two or more items is intended to cover the items individually and both items together. For example, "a and / or b" is intended to cover a (and not b), b (and not a), and a and b.

[0022] Disclosed herein are systems and methods relating to power converters, also referred to as voltage converters, that can provide power conversion with, among other advantages, increased power efficiency, increased power density, and / or reduced cost.

[0023] 1 illustrates a power converter system 100 according to some embodiments. The power converter system 100 includes a control system 105, a first direct current (DC) load / source 110, a power converter 115 (also referred to as a power converter stage 115), an LC filter 120, a contactor 125, a second source / load 130, a third source / load 135, and one or more sensors 140. The control system 105 includes a central controller 150 having an electronic processor 155 and a memory 157, and optionally, in some embodiments, one or more local controllers 160, each having an electronic processor 165 and a memory 167. The power converter system 100, as well as other power converter systems provided herein, may be non-isolated power converter systems. That is, the power converter system may be coupled to an AC power source (e.g., a single-phase or three-phase power grid) or an AC load (e.g., a single-phase or three-phase motor) without a transformer. The use of transformers is common in electrical circuits to provide isolation between a power converter and an AC power source or load. However, such transformers can add inefficiency and size or volume to the power converter. Thus, the power converter systems provided herein are non-isolated, also referred to as transformerless, to increase efficiency and / or reduce size of the power converter system. Because the power converter is provided without transformer isolation, the power converter may include additional features to prevent transmission of unwanted signals or currents (e.g., leakage currents) from passing between the power converter and other circuit components (e.g., DC power sources, DC loads, AC power sources, AC loads, and other structures that contact or support the power converter).

[0024] In operation, generally, the control system 105 controls the power switching elements of the power converter 115 using control signaling (e.g., pulse width modulated (PWM) signals) to convert power (i) from the DC load / power source 110, which acts as a power source, to either the second power source / load 130 or the third power source / load 135, which acts as a load (depending on the state of the contactor 125), or (ii) from the second power source / load 130 or the third power source / load 135, which acts as a power source (depending on the state of the contactor 125), to the DC load / power source 110, which acts as a load. Thus, when the DC load / power source 110 is acting as a power source for the power converter 115, the second power source / load 130 (or the third source / load 135, depending on the state of the contactor 125) is acting as a load for the power converter 115. Conversely, when DC load / source 110 serves as a load for power converter 115, second source / load 130 (or third source / load 135, depending on the state of contactor 125) serves as a source for power converter 115.

[0025] The DC load / power source 110 may be a direct current (DC) load, a DC power source, or both a DC load and a DC power source (i.e., depending on the mode of the power converter 115, it functions as a DC power source in some cases and as a DC load in other cases). In some examples, the DC load / power source 110 is a battery. In other examples, the DC load / power source 110 may be a capacitor, an ultracapacitor, a DC power source from a rectified AC power source (e.g., AC grid power converted to DC power by a diode bridge rectifier), etc. The second power source / load 130 may be an AC load, an AC power source, both an AC load and an AC power source (i.e., depending on the mode of the power converter 115, it functions as an AC power source in some cases and as an AC load in other cases), a DC load, a DC power source, or both a DC load and a DC power source (i.e., depending on the mode of the power converter 115, it functions as a DC power source in some cases and as a DC load in other cases). In some examples, the second power source / load 130 may be an electric (AC) motor, an AC generator, an AC power supply grid, a DC battery, a DC capacitor, a DC ultracapacitor, a DC power source from a rectified AC power source (e.g., AC grid power converted to DC power by a diode bridge rectifier), etc. The third power source / load 135 may be an AC load, an AC power source, both an AC load and an AC power source (i.e., depending on the mode of the power converter 115, it functions as an AC power source in some cases and an AC load in other cases), a DC load, a DC power source, both a DC load and a DC power source (i.e., depending on the mode of the power converter 115, it functions as a DC power source in some cases and a DC load in other cases). In some examples, the third power source / load 135 may be an electric (AC) motor, an AC generator, an AC power supply grid, a DC battery, a DC capacitor, a DC ultracapacitor, a DC power source from a rectified AC power source (e.g., AC grid power converted to DC power by a diode bridge rectifier), etc.

[0026] In some examples, the DC load / source 110 is a DC battery (e.g., an electric vehicle battery), the second source / load 130 is an AC grid, and the third source / load 135 is an AC motor (e.g., an electric vehicle motor). In this case, the power converter 115 can function as a bidirectional converter operating in a DC / AC traction mode (or motor mode) to drive the third source / load 135 (motor) with AC power converted from DC power from the DC load / source 110 (battery), and an AC / DC charging mode to charge the DC load / source 110 (battery) with DC power converted from AC power from the second load / source 130 (AC grid). In some other examples, the DC load / source 110 is a DC source, the second source / load 130 is an AC motor, and the third source / load 135 is not present in the system 100.

[0027] The contactor 125 is an electrically controlled switch and may be, for example, a contactor, a relay, a MOSFET, etc. In some examples of the system 100, the contactor 125 is not present, and instead the LC filter 120 is connected to both the second source / load 130 and the third source / load 135 simultaneously. However, other control techniques are employed to prevent the third source / load 130 from driving as a load (e.g., a motor) when receiving power from the second source / load 135 as a source (e.g., an AC grid).

[0028] A DC load / source of power 110 is coupled to the power converter 115 at a first (DC) side or section 111 of the power converter 115, and a second source of power / load 130 is coupled to the power converter 115 at a second (AC) side or section 112 of the power converter 115. The first side may be referred to as the input side or output side of the power converter 115 or may be referred to as the DC side of the power converter 115 depending on the mode of the power converter. The second side may be referred to as the input side or output side of the power converter or may be referred to as the AC side of the power converter 115 depending on the mode of the power converter. In some embodiments, the second side of the power converter 115 may be an AC side having single-phase AC power, three-phase AC power, or AC power having another number of phases.

[0029] In some embodiments, power converter 115 operates at high DC voltage levels. For example, during operation, the DC side of power converter 115 has a DC voltage (e.g., between the input terminals of power converter 115) of at least 200V, at least 600V, at least 800V, at least 1000V, at least 1200V, between 200V and 1200V, between 600V and 1200V, between 800V and 1200V, or another range. Such high DC voltage levels may be desirable in some situations, such as some electric vehicles. For example, some current electric vehicles (e.g., passenger cars and hybrid electric vehicles) operate with a DC bus voltage of about 200V to 400V. This DC bus voltage for passenger electric vehicles may increase in the future. Additionally, while some current electric vehicles (e.g., Class 4-8, off-road, or other larger electric vehicles) can operate with DC bus voltages in excess of 1000V, the high DC voltage levels can introduce challenges to typical power converter systems, such as increased leakage currents, increased common mode voltages, and higher rates of change of common mode voltages. These challenges can lead to resonances on the LC filter 120, shaft voltages, excessive bearing currents that can result in bearing failure (e.g., from discharge events when lubricant breakdown occurs), excessive motor shaft currents, excessive motor winding currents (e.g., insulation can be damaged), and excessive gear train currents (e.g., bearing currents can propagate to the gear train via electromagnetic interference (EMI) or noise, vibration, and harshness (NVH) resulting from damaged bearing race walls). However, the embodiments described herein can mitigate such challenges through improved LC filters and control techniques including harmonic injection, cascade controllers, MPC control, and / or variable frequency critical soft switching (VFCSS).

[0030] LC filter 120 may be referred to as an N-phase LC filter and includes an LC filter for each phase of power converter 115. Each LC filter of the N-phase LC filter may include at least an inductor and a capacitor, or at least an inductor and two capacitors, as described in further detail below (see, e.g., the description of FIGS. 2 and 3).

[0031] The sensors 140 include, for example, one or more current sensors and / or one or more voltage sensors. For example, the sensors 140 may include respective current and / or voltage sensors for monitoring the current and / or voltage of one or more of the DC load source 110, each phase of the second power source / load 130, each phase of the third power source / load 135, each phase of the LC filter 120, or other nodes or components of the power converter 115. For example, if the LC filter 120 is a three-phase LC filter, the sensors 140 may include at least three current sensors, one for sensing the current of each phase of the three-phase LC filter 120. In some embodiments, additional or fewer sensors 140 are included in the system 100. For example, the sensors 140 may also include one or more vibration sensors, temperature sensors, etc. In some examples, the control system 105 infers a characteristic (e.g., current or voltage) of the power converter 115 rather than directly sensing the characteristic. The sensors 140 may provide sensor data to the control system 105 indicative of the sensed characteristic of the system 100. Such sensor data may thus be indicative of electrical operating characteristics of system 100. In some examples, control system 105 infers or estimates a characteristic (e.g., current or voltage) at one or more nodes of power converter 115 based on sensor data from sensors 140 that do not directly sense the characteristic, but rather sense a different type of characteristic or even a different component. Further discussion of such inferences or estimates is provided below with respect to state estimation.

[0032] Input / output (I / O) interface 142 includes or is configured to receive input from one or more inputs (e.g., one or more buttons, switches, a touch screen, a keyboard, etc.) and / or includes or is configured to provide output to one or more outputs (e.g., LEDs, a display screen, a speaker, a haptic generator, etc.). Other electronic devices and / or a user may communicate with system 100, and in particular control system 105, via I / O interface 142. For example, control system 105 may receive commands (e.g., from a user or another device) for power converter system 100 indicating a target torque, a target speed, a target power level, a conversion type, etc. In response, control system 105 may drive power converter 115 to achieve the target and / or conversion type indicated by the command.

[0033] The control system 105 generally monitors the system 100, including the power converter 115 (e.g., based on sensor data from the sensors 140), receives commands (e.g., via the input / output interface 142), and controls the power switching elements of the power converter 115 using control signaling (e.g., pulse width modulated (PWM) signals) to convert power (e.g., according to the sensor data and / or commands). In some embodiments, the control system 105 includes a controller (e.g., the central controller 150) that performs this monitoring and control without additional local controllers. In other embodiments, the control system 105 is a cascaded control system that includes the central controller 150 and one or more local controllers 160. The cascaded control system can communicate real-time (e.g., each control cycle) monitoring information (e.g., sensor data) and control information between the central controller 150 and the one or more local controllers 160. In some examples, the local controllers 160 each implement model predictive control (MPC) or another regulatory control scheme (e.g., PID control, PI control, etc.). In some examples, the central controller implements a non-MPC regulation technique, such as proportional-integral-derivative (PID) control or proportional-integral (PI) control.

[0034] Each controller of the control system 105, including the central controller 150 and the local controllers 160, is an electronic controller that may include an electronic processor. Such electronic controllers may further include a memory (e.g., memory 157 or 167). The memory may be, for example, one or more of a read-only memory (ROM), a random access memory (RAM), or other non-transitory computer-readable medium. The electronic processor 155, 165 is configured to, among other things, receive instructions and data from the memory 157, 167 and execute the instructions to perform the associated controller functions described herein, including, for example, the processes described herein. For example, the memory may include control software. In some embodiments, instead of or in addition to executing software from the memory to perform the controller functions described herein, the electronic processor includes one or more hardware circuit elements configured to perform some or all of this functionality. Furthermore, although certain controllers, electronic processors, and memories may be referred to herein as respective single units, in some embodiments, one or more of these components are distributed components. For example, in some embodiments, the electronic processor includes one or more microprocessors and / or hardware circuit elements.

[0035] 2 illustrates an example of a half-bridge converter 200 that can function as the power converter 115 of the system 100 of FIG. 1. As illustrated, the converter 200 includes a DC terminal 220 (also referred to as a DC node, DC link, DC rail, etc.) having a positive DC terminal 222 and a negative DC terminal 224. The converter 200 further includes an interface terminal 225 (also referred to as an interface node) having a positive interface terminal 227 and a negative interface terminal 229. The converter 200 can operate as a bidirectional converter or a unidirectional converter (in either direction), depending on the configuration and control of the system in which it is implemented. Thus, in some examples, the DC terminal 220 can be an input terminal and the interface terminal 225 can be an output terminal (e.g., DC / DC conversion and DC / AC inversion), and in some examples, the DC terminal 220 can be an output terminal and the interface terminal 225 can be an input terminal (e.g., AC / DC rectification). Furthermore, the interface terminal 225 may be an AC input terminal (for example, for AC / DC rectification), an AC output terminal (for example, for a DC / AC inverter), or a DC output terminal (for example, for DC / DC conversion).

[0036] The converter 200 includes a DC link capacitor (C DC 1 (e.g., when LC filter 120 is an N-phase LC filter with N=1), a high-side (upper) power switching element (M1) 235 (also referred to as an upper switch or upper FET 235), a low-side (lower) power switching element (M2) 240 (also referred to as a lower switch or lower FET 240), a midpoint node 242 connecting the drain terminal of the upper switch 235 and the source terminal of the lower switch 240, and an LC filter 245. LC filter 245 is an example of LC filter 120 of system 100 of FIG. 1 (e.g., when LC filter 120 is an N-phase LC filter with N=1).

[0037] The power switching elements 235 and 240 may each be a field effect transistor (FET) having respective gate, source, and drain terminals. The FETs may be, for example, MOSFETs, silicon carbide (SiC) FETs, gallium nitride (GaN) FETs, among other types of FETs.

[0038] The LC filter 245 is connected to the switch-side inductor L SW 250 and the lower capacitor C B 255 and the upper capacitor C A 215. The switch side inductor L SW 250 is coupled between the midpoint node 242 and a filter node 260. For example, the switch-side inductor L SW One end of the lower capacitor C250 is coupled to the midpoint node 242 and the other end is coupled to the filter node 260. B 255 is coupled between the filter node 206 and the negative DC terminal 224. For example, the lower capacitor C B A first end of 255 is coupled to the filter node 260 and a second end is coupled to the negative DC terminal 224. A 215 is coupled between the filter node 260 and the positive DC terminal 222. For example, the lower capacitor C A A first end of 215 is coupled to the filter node 260 and a second end is coupled to the positive DC terminal 222 .

[0039] In some examples, LC filter 245 is an LCL filter (an LC filter with an additional inductor (L)) in which an additional (interface) inductor is coupled between filter node 260 and the positive interface terminal 227.

[0040] The upper capacitor 215 allows the ripple currents at both the input and output nodes (nodes 222, 227) of the converter 200 to be shared. Since the ripple currents on the input and output nodes have some correlation, the differential mode currents at these input and output nodes can be cancelled through this capacitance. This reduction in differential mode currents can result in improved EMI performance and reduced total capacitor ripple current when compared to a typical half-bridge converter (e.g., when the total capacitance between the two converters is held constant). Furthermore, the reduction in total capacitor ripple current can allow for a reduction in capacitor size, for example, when the capacitor ripple current drives the capacitor sizing.

[0041] The converter is connected to a drain-source capacitor, C DS 265a and 265b, each coupled across one of the switches 235, 240, respectively. In particular, a first drain-source capacitance 265a is provided across the source terminal 270a and the drain terminal 275a of the upper switch (M1) 235, and a second drain-source capacitance 265b is provided across the source terminal 270b and the drain terminal 275b of the lower switch (M2) 240. DS ) 265a-b are collectively referred to herein as drain-source capacitors (C DS )265.

[0042] Drain-source capacitor (C DS ) 265 can slow down the voltage rise during the on-off transitions of switches 235 and 240. This slower voltage rise can in turn reduce the switching losses of switches 235 and 240.

[0043] In some examples of converter 200, the upper capacitor C A 215 and the drain-source capacitor C DS One or both of these are not included in converter 200.

[0044] As previously mentioned, in some examples, power converter 200 can function as power converter 115 of system 100 of FIG. 1. In the context of power converter 115 (and thus power converter 200) implementing an AC / DC rectifier or a DC / AC inverter, power converter 200 is a single-phase power converter 200. In some examples, multiple instances of power converter 200 are paralleled to collectively function as power converter 115 of FIG. 1 to provide single-phase conversion (whether rectification or inversion) or to provide DC / DC power conversion. In some examples, power converter 115 is a multi-phase power converter (e.g., operates with three or more phases of AC power). In such examples, power converter 115 can include multiple instances of power converter 200, each associated with a phase of the AC power, each instance having a shared DC terminal 220, and each instance having an independent V interface 2, 3, 4, 7, and 12. In some of these examples, multiple instances of power converters 200 are paralleled to collectively provide power conversion for each phase, as shown in FIGS. 19-20 (e.g., two parallel power converters 200 for phase A, two parallel power converters 200 for phase B, and two parallel power converters 200 for phase C). In some examples, the particular number of parallel power converters 200 and the number of phases vary.

[0045] As used herein, converter block may refer to a half-bridge circuit as described with respect to converter 200 of Figure 2. For example, converter block 262 may include power switching elements 235 and 240, LC filter 245 (including upper capacitor 215 (if present) and additional interface inductors (if present)), their interconnection nodes (e.g., midpoint node 242, filter node 260, DC terminal 220, and interface terminal 225), and drain-source capacitor 265 (if present).

[0046] 3A illustrates a multi-phase power converter system 300 selectively coupled to an AC grid 302 or an AC motor 303 by a contactor 125. The multi-phase converter system 300 includes a multi-phase converter 304 coupled on the DC side to a DC power source 306, shown as a battery 306, and coupled to the contactor 125 (and thus to the AC grid 302 or the AC motor 303) via an LC filter 308. The multi-phase converter 304 may function as the power converter 115 of the system 100 of FIG. 1. The LC filters 308, collectively the three-phase LC filters, may function as the LC filter 120 of the system 100 of FIG. 1. With reference to FIG. 1, the battery 306 is an example of a DC load / source 110, the AC grid 302 is an example of a second source / load 130; and the AC motor 303 is an example of a third source / load 135. Returning to FIG. 3A, in operation, the converter system 300 can function as a DC / AC inverter or an AC / DC rectifier, depending on the power supply and the switching of the power switching elements.

[0047] The AC motor 303 may be, for example, a permanent magnet rotor synchronous machine, a wound field synchronous machine (WFSM), or other electric motor. The AC motor 303 may include a stator and a rotor. The stator may include multiple stator windings that can be driven with current to generate a changing magnetic field to rotate the rotor. The rotor may include permanent magnets, wound fields, or a combination of permanent magnets and wound fields (hybrid). The rotor may be coupled to a motor draft such that the motor shaft is driven in rotation (i.e., rotated) when the rotor is driven in rotation. The motor windings of the stator and / or rotor may have insulation that may be damaged due to excessive current resulting from high voltage rates of change (dV / dt), as previously discussed. The motor shaft may be supported by one or more motor bearings to support and allow rotation of the shaft. For example, the motor shaft may include a first motor bearing at the driven end of the shaft (the end of the shaft coupled to the rotor) and a second motor bearing at the distal end of the shaft. In some examples, other motor bearing arrangements are used. The motor shaft may be further coupled to a transmission or gearing (gear train) that drives an end load. For example, in an electric vehicle, the transmission may ultimately be coupled to one or more wheels to propel the vehicle. In other examples, the motor may be part of industrial equipment and cause the rotation of another load (e.g., a cutting tool, grinder, conveyor motor, etc.).

[0048] The multi-phase converter 304, also referred to as a power converter stage, includes three instances of the power converter 200 (or converter block 262) of FIG. 2, one for each phase of the AC grid 302. Each instance includes upper and lower switches 235 and 240. Although not shown, each switch 235 and 240 may include a respective drain-source capacitor coupled across them, as shown in FIG. 2. The multi-phase converter 304 is further coupled to a battery 306 via DC terminals 220, and to an AC grid 302 and an AC motor 303 via interface terminals 225a,b. More specifically, the interface terminals 225a,b include three grid connection points 225a (one for each grid phase) for connecting to the AC grid 302, and three motor connection points 225b (one for each motor phase) for connecting to the AC motor 303. The multi-phase converter system 300 includes three LC filters 308. Each LC filter 308 has a configuration similar to that of the LC filter 245 in FIG. fs,a , L fs,b , or L fs,c ) and the lower capacitor 255 (C f,a , Cf,b , and C f,c ), and upper capacitor 215 (C f,a , C f,b , or C f,c 2. The lower capacitor 255 has a neutral point 311a coupled to the negative DC terminal 224, and the neutral point 311 and the negative DC terminal 224 are a common node. Similarly, the neutral point 311b of the upper capacitor 215 is coupled to the positive DC terminal 222. The switch-side inductor 250 is coupled between the midpoint node 242 and the filter node 260. The LC filter 308 has a common-mode inductor (L CM) 312 to the AC interface terminals 225a,b. More specifically, the common mode inductor 312 is coupled between the filter node 260 and the interface terminals 225a,b. The common mode inductor 312 can help eliminate leakage currents in the system.

[0049] In the depicted example, multi-phase converter system 300 is coupled to a battery 306 and an AC grid 302. In other examples, multi-phase converter system 300 is coupled to a DC source / load other than battery 306 (e.g., a capacitor, an ultracapacitor, a DC source from rectified AC power, etc.), coupled to a different AC source / load other than grid 302 and motor 303, and / or coupled to only one of AC grid 302 or AC motor 303. Additionally, multi-phase converter system 300 includes an upper capacitor 215 for each phase, although in some examples, an upper capacitor 215 is not included.

[0050] As shown in Figures 2 and 3A, in some examples of the power converter system provided herein, the LC filter 120 (implemented as LC filter 245 in Figure 2 and 308 in Figure 3A) includes an LC filter for each phase, with the common point of each capacitor connected to the negative terminal (and / or positive terminal) of the DC bus. This connection creates a bypass path for zero sequence voltage control. By utilizing topological modifications and zero voltage control, the common mode voltage can be stabilized and leakage current can be reduced.

[0051] As mentioned above, in some examples, the contactor 125 is not present; rather, the interface terminals 225a and 225b are simultaneously coupled to both the AC grid 302 and the AC motor 303. In some examples, when AC power is received from the AC grid 302 via the interface terminal 225a (i.e., converted by the converter 304 to charge the battery 306), a torque cancellation scheme may be used to prevent the received AC power from driving the AC motor 303. That is, the power signal on the terminal 225b that may otherwise cause a motor torque in the AC motor 303 is effectively cancelled by appropriate control of the power switching elements 235 and 240 of the converter 304. The torque cancellation scheme implemented may be commonly known in the art.

[0052] 3B and 3C show circuit diagrams of the converter system 300 when installed in an electric vehicle (EV), represented by an EV chassis 350. The circuit diagrams show the converter system 300 in two different operating modes: a charging mode in FIG. 3B and a traction mode in FIG. 3C. In the charging mode in FIG. 3B, the converter 304 converts incoming AC power from the grid 302 into DC power for charging the battery 306. In this mode, the contactor 125 (shown in FIG. 3A) can be configured or controlled to connect the AC terminals 225 to the AC grid 302 and disconnect them from the AC motor 303. Thus, the AC motor 303 is not shown in the diagram of FIG. 3B (although it is still physically present in the electric vehicle). The grid inductance between the common mode inductor (LCM) and the AC grid 302 is represented by the grid inductor 352. The capacitance between the DC power source 306 or the DC terminals 220 and the EV chassis 350 is represented by the capacitor 354.

[0053] In the traction mode of FIG. 3C, the converter 304 converts the ADCC power received from the battery 306 into AC power to drive the AC motor 303, which is represented in the diagram as a three-phase inductor. In this mode, the contactor 125 (shown in FIG. 3A) can be configured or controlled to connect the AC terminals 225 to the AC motor 303 and disconnect it from the AC grid 302. Thus, although the AC grid 302 is not shown in the diagram of FIG. 3C, the grid connection point 225a is still present in the vehicle so that it can be connected to the AC grid 302. The additional capacitance between the AC motor 303 and the EV chassis 350 is represented by the capacitor 356.

[0054] In some examples, system 300 (as well as other power converter systems described herein) provides a transformerless vehicle-to-grid (V2G) or vehicle-to-everything (V2X) interface. Systems disclosed herein, including system 300, may be particularly well suited for such operation given the common mode control (discussed further below) and filtering hardware (e.g., LC filter 308) used in these systems. In some examples, system 300 is included as part of an integrated drivetrain solution (e.g., providing both traction and charging modes). In other examples, system 300 is a dedicated on-board charger (e.g., having a charging mode but not a traction mode) or a dedicated traction drive (e.g., having a traction mode but not a charging mode).

[0055] In general, system 300 can provide single-phase AC, two-phase AC, three-phase AC, or DC operation. In a V2X interface implementation, system 300 can interact with grid 302 to return power or balance grid 302, interface with photovoltaic and energy storage systems, and / or form a local microgrid, etc. Control of this V2X interface can be done in a manner similar to one or more of the control techniques described below, including the use of (1) global control at the function level, (2) local control at the device level, and (3) application interfaces (e.g., for motoring, charging, solar, three-phase operation, AC grid, etc.).

[0056] Compared to transformer-based systems, the disclosed converter systems (e.g., systems 300, 700, 1200, 1900, etc.) functioning as transformerless DC EV (bidirectional, fast) chargers eliminate the galvanic isolation stage, greatly increasing charging efficiency and providing increased power density V2G capabilities. The disclosed systems further provide a non-isolated transformerless topology that allows for the elimination of additional transformer volume and losses present in galvanically isolated topologies. The design can utilize switching frequency and filter parameters to minimize losses and volume depending on the switching device and power level selected.

[0057] In the absence of a transformer, the disclosed transformerless system and charger use other techniques to manage common mode leakage current. This is accomplished, for example, by a zero sequence voltage control technique that stabilizes the common mode voltage. The charger can include two energy conversion stages: a DC / DC converter for battery side control and a DC / AC converter for grid interface and common mode voltage control to provide charging and V2X services.

[0058] A transformerless charger may be particularly well suited for traction drivetrain integration. An integrated charger, where the traction inverter is used as the primary charging interface, may be a solution to reduce the cost and footprint of electric vehicle charging. Furthermore, integrating the disclosed transformerless bidirectional non-isolated charging technology into the drivetrain may also increase motor life and reliability by reducing bearing currents and voltages, and extend the life of the motor windings and gear train.

[0059] In electric machines, such as electric vehicle motors, the main fault points are bearing currents and voltages, motor shaft currents and voltages, excessive motor winding currents, and gear train currents. Such damaging currents and voltages are caused by common-mode voltages, v CM The embodiments disclosed herein help reduce one or more of these currents and voltages, thus preventing or mitigating such damage and extending the life of the electric machine.

[0060] Figure 3D shows a simplified equivalent circuit 360 of the parasitic capacitances of an electric machine, such as the AC motor 303 of an electric vehicle incorporating the power converter system 300 as shown in Figures 3B-3C. Using this diagram, the bearing voltage ratio (BVR), which is indicative of the bearing voltage in the system, can be calculated using the following equation:

number

[0061] The common mode voltage seen at the terminals of an electric machine with the proposed integrated charger (e.g., the terminals of the AC motor 303 connected to the motor node 225b) is composed of two components: c、CM A fixed (DC) value over v and a superimposed AC disturbance signal v representing ripples in the control, noise, etc. AC This allows the common mode voltage v CM =v c,CM +v AC is obtained.

[0062] In a well-designed control, v AC This means that the bearing voltage v b =(v c,CM +v AC ) BVR, meaning low DC offset and very low disturbances due to switching and control. Using the embodiments disclosed herein, v c、CM By keeping the lubricant's breakdown voltage below, damaging discharges caused by bearing currents can be avoided.

[0063] Bearing current (i b ) is approximated by the leakage current i lkg C wf and the bearing path. wf has a low impedance compared to the support path and therefore the bearing current i b is a portion of the leakage current. CM is controlled to be almost constant, which means that i lkg becomes smaller, so i b This means that becomes very small.

[0064] Systems 100 and 300, as well as other systems disclosed herein, are each examples of power converter systems that may incorporate various features disclosed herein, both alone and in combination. In the following sections, the disclosure describes (I) Three-Phase Converter Modeling, (II) Harmonic Injection, (III) Cascade Control System, (IV) Model Predictive Control, (V) State Estimation, (VI) Variable Frequency Critical Soft Switching, and (VI) Modular Converter Blocks. These headings are included for convenience and should not be construed as limiting.

[0065] I. Three-phase Converter Modeling In some examples provided herein, a control scheme for controlling a power converter is based on a dq0 coordinate system. By using the dq0 coordinate system as provided herein, the control scheme can utilize the zero-sequence voltage component to control the common-mode voltage. Compared to the abc coordinate system, the active / reactive power and the common-mode voltage can be independently controlled with the d, q, and zero-sequence components of the dq0 coordinate system. From the abc reference coordinate system, a coordinate system model of a three-phase converter can be derived (e.g., as shown in FIG. 3A).

[0066] The state space equation in the abc coordinate system is expressed as follows:

number

number

number

number

[0067] Because it is difficult to control a time-varying sinusoidal reference in the abc coordinate system and it is convenient to calculate the active / reactive power in the dq0 coordinate system to stabilize the zero-sequence voltage, the state-space model is transformed to the dq0 reference coordinate system for control purposes. For example, the dq0 coordinate system transformation is useful because the dq0 coordinate system can transform a time-varying sinusoidal waveform into an equivalent constant DC value. To implement control, the DC value may be easier to control than the AC value. However, conventional methods mainly utilize the dq system without considering the zero (zero sequence) component. The topology of the converter system 300, in which the common points of the AC three-phase capacitors are connected to the positive and / or negative terminals of the DC bus, allows for the extraction of the zero sequence from the abc coordinate system to the dq0 coordinate system and the control of the zero sequence voltage to half of the DC bus voltage. Thus, the common-mode voltage v cm is the zero sequence component and can be stabilized to a constant accordingly.

[0068] For a reference coordinate system transformation with zero sequence components, the abc coordinate system can be first transformed to αβ0 and then transformed to the dq0 coordinate system. From abc to αβ0, the Clarke transformation is applied as follows:

number

[0069] In the αβ0 system, the signal consists of two orthogonal sinusoidal AC waveforms in the α and β coordinate systems and a zero sequence component. The Park transformation is secondly performed to transform the αβ0 stationary reference frame to the rotating dq0 system, which is calculated as follows:

number

[0070] Based on the above Park and Clark equations for coordinate system transformation, the above state space equations can be transformed from abc to dq0 as follows:

number

number

number

number

[0071] In both traction and charging modes, the neutral can be assumed to be isolated (in the sense that the system does not form an unintentional common-mode conduction path), which means that the zero-sequence component can be neglected. In charging mode, the three-phase grid is modeled as a voltage source at the common connection point at the capacitor. In traction mode, the drive literature usually refers to the v dq The motor terminal voltage, called here v c、dq Similarly, usually i dq The motor current, called i o、dq If we change the notation of the standard PMSM model, we get

number

number

[0072] By utilizing the dq0 state-space equation and the connection of three-phase capacitors with the positive / negative terminals of the DC bus and the common point, the zero-sequence voltage can be explicitly controlled and u cm can be stabilized.

[0073] In some examples, a different rotating reference frame other than the dq0 reference frame is used by the control system 105.

[0074] II. Harmonic injection for zero sequence voltage control In some power converter applications, such as grid-connected power converters for electric vehicle (EV) chargers for photovoltaic (PV) arrays, leakage current and DC bus utilization are two factors that affect the converter's performance. To address leakage current, a bulky line frequency transformer is typically installed to break the leakage path at the point of common coupling (PCC), which increases the cost, volume, and weight of the system. To improve DC bus utilization, the DC bus voltage can be stepped up (e.g., to be at least twice the grid voltage amplitude to avoid saturation issues), which brings extra switching losses and challenges to the switch voltage withstand capability.

[0075] To address these and other issues, in some examples, harmonic signals are injected into the power converter systems provided herein, which may be non-isolated (transformerless) converters. Conventional harmonic injection involves direct injection in the duty cycle used to modulate the switching elements, which reduces control stability and robustness, can cause divergence in PWM modulation, and injects extra harmonics into the grid, reducing the power quality of the grid voltage and current. In contrast, in some examples provided herein, the systems and methods provide harmonic injection for zero sequence voltage control. The disclosed systems and methods improve DC bus utilization without reducing control stability and robustness, and without injecting additional harmonics into the grid (or other AC sources or loads).

[0076] In some embodiments, a power converter system (e.g., system 100) has a non-isolated N-phase power converter and a control system that injects multiples of N-th phase harmonics for zero-sequence voltage control. For example, for a three-phase power converter (i.e., N=3), the injected harmonics may be third harmonic injection (THI), sixth harmonic injection, etc. Furthermore, rather than injecting harmonics directly into the duty cycle for modulation, in some examples, the system injects harmonics (e.g., sinusoidal or triangular voltage signals) into a zero-sequence voltage control signal of a set of direct quadrature zero-sequence (dq0) rotating reference frame control signals. The control signals may also be referred to as rotating reference frame reference targets. This approach provides additional regulation via constraints on the dq0 rotating reference frame control signals that do not apply when harmonics are injected directly into the duty cycle for modulation. Thus, system stability and robustness may be improved compared to direct duty cycle side injection techniques.

[0077] For example, referring to FIG. 4, a power converter system 100 is shown that may be an example of the power converter system 400 of FIG. 1. As shown, the power converter system 400 is a non-isolated three-phase power converter that includes a control system 105 that includes a central controller 150 and three local controllers 160a-c (each instance of the local controller 160 of FIG. 1). Each of the local controllers 160a-c may be associated with and control a respective converter block 262a-c. The converter blocks 262a-c may be instances of the converter block 262 described with respect to FIG. 2. The local controllers 160a-c may implement a particular control scheme for performing control of the associated converter block 262a-c. For example, the local controllers 160a-c may implement model predictive control (MPC), proportional-integral (PI) control, proportional-integral-derivative (PID) control, or another type of control or regulation, as described further below. In some embodiments, rather than a cascaded control system as shown, the control system 105 does not include the local controllers 160a-c. For example, the reference voltages generated by the central controller 150 may instead be directly mapped (eg, by a look-up table) to respective duty cycle values ​​provided to respective gate drivers 402 for each power switching element of the converter.

[0078] As shown, the central controller 150 controls the power converters 304 (e.g., i L,abc ;i g,abc ,v g,abc ) electrical characteristics of a reference electrical characteristic (e.g., i g,d *;i g,q *,v g,q*) and determines the fundamental frequency (theta or θ) of the AC load / source (e.g., AC grid) coupled to terminals 225. Based on these received and determined values, central controller 150 generates control reference signals in the dq0 reference frame. Central controller 150 then transforms the control reference signals to the stationary (abc) reference frame via dq0 / abc reference frame translator 410 and aligns these control reference targets 415 (e.g., v a *,v b *, and v c *) to the local controllers 160a-c. In some examples, a notch filter 411 is provided to compensate for resonances that may be present in the system. For example, a notch filter 411 can be added at a cutoff frequency of the dq0 axis. The notch filter 411 can be designed in continuous time as follows:

number

[0079] More specifically, the central controller 150 transforms the received electrical characteristics of the power converter 304 from a stationary reference frame to a dq0 reference frame (e.g., via the abc / dq0 translator 412). The central controller 150 converts the transformed electrical characteristics into the dq0 reference frame (e.g., i g,d * to i g,d and i g,q * to i g,q ) to a reference electrical characteristic in the voltage controlled reference signal (e.g., v d * and v q For example, regulator 413 (e.g., a PI or PID controller) generates the d and q components of a voltage control reference signal (v d *) to generate the resulting d component, the reference d component and the transformed grid current value (i g,d * and i g,d) can be compared to the voltage control reference signal (v q *) to generate the resulting q-component of the reference and the transformed grid current value (i g,q * and i g,q ) can be compared to the d and q components of the voltage controlled reference signal. These d and q components of the voltage controlled reference signal are provided to a dq0 / abc translator 410. g,d * and i g,q ) may be provided to central controller 150 by I / O interface 142 (see FIG. 1) from a memory (eg, memory 157) or another source based on received user input commands.

[0080] To generate a reference electrical characteristic in the dq0 reference frame for regulators 413 and 414, central controller 150 further includes a reference characteristic block 416 that includes a current reference generator 417, a constant current / constant voltage (CC / CV) controller 418, and a selector 419. When system 400 is in traction mode, selector 419 selects the reference electrical characteristic (e.g., i o,dq When the system 400 is in charging mode, the selector 419 can select the output of the current reference generator 417 to serve as the reference electrical characteristic (e.g., i o,dq The output of the CC / CV control can be selected to function as the torque reference (e.g., an input torque command from memory 157 or a user input provided via I / O 142), the angular velocity (ω) of AC motor 303 (see FIG. 3A), and V DC (e.g., the voltage across the DC terminals 220, which may be provided by a voltage sensor of the sensor 140) as an input. The current reference generator 417 may be configured to generate a reference electrical characteristic (e.g., an output current i o,dq*) A real-time function can be implemented that maps to the torque per ampere control technique. A lookup table or function can be implemented, for example.

[0081] The CC / CV controller 418 operates on the input V DC (e.g., the voltage across the DC terminals 220, which may be provided by a voltage sensor of the sensor 140) and IDC (e.g., the current passing through the DC terminals 220, which may be provided by a current sensor of the sensor 140). The CC / CV controller 418 receives the reference electrical characteristic (e.g., the d-axis output current i for charging the battery 306) from a look-up table (e.g., populated through experimentation) or the two inputs. o,d *(see FIG. 3)) to the d-axis current reference i o,d * corresponds to the active power. In some cases, for example, when grid support is desired, the reactive power criterion i o,q * may also be added. In some examples, the CC / CV controller has two states: a constant current state where the system 400 pushes a constant current until the battery 306 has a near-peak state of charge (SOC), and a constant voltage state where the system applies a constant voltage that gradually charges the battery 306 to complete the charging cycle.

[0082] For the zero-sequence (0) reference component, the power converter system 400 uses a harmonic injector 405 (e.g., provided as part of the central controller 150). That is, the harmonic injector 405 generates the harmonic injection and provides the zero-sequence component target to the dq0 / abc reference frame translator 410.

[0083] As shown in FIG. 4, the harmonic injector 405 is a DC offset (e.g., V dc2 and 3 , the control reference target 415 receives the voltage reference V OUT output by the translator 410, which refers to the target voltage of the lower capacitor of the control block 262a-c (e.g., capacitor 255, see FIGS. 2 and 3 ), the fundamental frequency (theta or θ) of the AC section of the power converter 115, and the control reference target 415 for each phase of the power converter 115. In this example, the control reference target 415 (also referred to as the power reference target) is the voltage reference V OUT output by the translator 410, which refers to the target voltage of the lower capacitor of the control block 262a-c (e.g., capacitor 255, see FIGS. 2 and 3 ). c,a* ,V c,b* , and V c,c* Based on these characteristics, the harmonic injector 405 can calculate the zero sequence component target. Therefore, the harmonic injector 405 may also be referred to as a zero sequence reference generator. In some embodiments, the harmonic injector 405 calculates the zero sequence reference component by summing two components: (i) a DC offset and (ii) a multiple of the Nth phase harmonic injection.

[0084] The first component, DC offset, is the DC bus voltage (V dc This DC offset component of the zero sequence reference may be set to half of the DC bus voltage measurement, V. This DC offset component of the zero sequence reference ultimately prevents leakage current from flowing to the grid. That is, the zero sequence output current may be attenuated by the stabilization control of the zero sequence capacitor voltage provided by this DC offset, which is used as an input for the zero sequence voltage reference. The operating principle of the zero sequence voltage control is based on three-phase output capacitor voltage reference tracking. Specifically, in the central controller 150, the zero sequence component of the reference is set to half of the DC bus voltage measurement, V. dcThe control reference targets 415 are designed as dq / 2. This reference is combined with the dq component references from the outputs of the regulators 413 and 414 and then transformed to the abc reference frame as the control reference target 415 for the local controllers 160a-c. Thus, each of the control reference targets 415 can be composed of a sinusoidal AC component (based on the dq input to the translator 410) and a zero sequence DC component (based on the zero sequence (0) input to the translator 410). Thus, based on the control reference target 415 with the zero sequence control integrated, the local controllers 160a-c adjust the zero sequence voltage control to provide a stabilized common mode capacitor voltage and low leakage current. In some examples of the converter 300, this DC offset as a zero sequence voltage reference provides its own advantages, so the DC offset can be used to reduce harmonic injection (e.g., the output of the injector 405 reduces the DC offset (V dc 410 as a zero sequence voltage reference without the addition of

[0085] In another example, DC bus utilization can be further improved by injecting an Nth phase harmonic along with this DC offset. By injecting an Nth harmonic into the zero-sequence voltage reference (i.e., summed with the DC offset), these two components form the zero-sequence portion of the control reference target 415 for the local controllers 160a-c. Thus, the local controllers 160a-c for each phase inject the capacitor voltage (v c、abc ) to adjust the

[0086] The harmonic injector 405 may calculate the Nth phase harmonic injection multiple based on the fundamental frequency and the control reference target 415. Thus, in some embodiments, the Nth phase harmonic injection multiple may be considered as a feedback signal calculated from N previous control reference targets generated by the control system in the stationary (abc) reference frame based on previously received rotating reference frame targets. In some embodiments, the Nth phase harmonic injection multiple is a sinusoidal signal. The harmonic injector 405 may derive the sinusoidal signal based on the Nth order of the fundamental frequency of the AC voltage section of the power converter. In other embodiments, the Nth phase harmonic injection is a triangular signal. The harmonic injector 405 may derive the triangular signal based on the average of the maximum and minimum values ​​of the fundamental frequency (θ) of the AC voltage section of the power converter. An exemplary equation that the harmonic injector 405 may use to calculate the sinusoidal or triangular signal is shown below:

[0087] Sine wave injection for third harmonic injection (Sin-RTHI) may be implemented by deriving a third order grid fundamental frequency (θ) component to be superimposed on the zero sequence voltage reference. The Sin-RTHI zero sequence voltage reference can be expressed as follows: v * 0,3rd =v dc / 2+V m D 3rd sin(3ωt)

[0088] Therefore, the abc coordinate system Sin-RTHI three-phase capacitor voltage reference v * c,abc can be expressed as follows: v * c,a =v * c,a +V m D 3rd sin(3ωt)

number

number

number

[0089] By utilizing harmonic injection into the zero-sequence voltage, the peak-to-peak capacitor voltage can be reduced to improve DC bus utilization and avoid duty cycle saturation at lower DC bus voltages. Figure 5A shows simulated waveforms of the third-order, fundamental frequency, and injection capacitor voltages for one grid period for Sin-RTHI.

[0090] The triangular space vector for third harmonic injection (Tri-RTHI) may be implemented by deriving the average of the maximum and minimum grid fundamental frequency component capacitor voltages superimposed on the zero sequence voltage reference. The Tri-RTHI zero sequence voltage reference can be expressed as follows: v* 0,3rd =v dc / 2-D 3rd [max(v * c,abc )+min(v * c,abc )]

[0091] Therefore, the abc coordinate system Tri-RTHI three-phase capacitor voltage reference v * c,abc can be expressed as follows: v * c,a =v * c,a -D 3rd [max(v * c,abc )+min(v * c,abc )] v * c,b =v * c,b -D 3rd [max(v * c,abc )+min(v * c,abc )] v * c,c =v * c,c -D 3rd [max(v * c,abc )+min(v * c,abc )].

[0092] FIG. 5B shows the simulated waveforms of the third fundamental frequency and the injection capacitor voltage for one grid period for the Tri-RTHI.

[0093] As shown in Figures 5A-5B, the DC bus utilization can also be improved to avoid the problem of duty cycle saturation. To evaluate the effectiveness of the injected third harmonic in Figures 5A-5B, the voltage gain is calculated by multiplying the fundamental component capacitor voltage peak value v base and the reference modulation waveform peak value v THIIt can be defined as the ratio of

number

[0094] The maximum voltage gain of the continuous third harmonic injection method can be derived as π / 3 when the third harmonic is at the zero crossing point. Therefore,

number

[0095] By utilizing the disclosed harmonic injection techniques, the DC bus voltage can be reduced (eg, by a factor of 1.15), thus reducing the voltage stress and switching losses of the power switching elements.

[0096] In some embodiments, a control reference target 415 (here, V c,a* ,V c,b* , and V c,c* Instead of calculating the Nth phase harmonic injection multiple using a 1 / 2 Ω resistor, harmonic injector 405 may derive the Nth phase harmonic injection from a direct or indirect voltage measurement of each phase of power converter 115. For example, in the case of DC voltage measurements, harmonic injector 405 may receive an output from a respective voltage sensor of each of the N phases of power converter 115, or an output from an analog-to-digital converter (ADC) that converts the respective analog output of the voltage sensor into a digital signal indicative of a voltage measurement. As another example, in the case of indirect voltage measurements, harmonic injector 405 may receive one or more communications from local controller 160 indicating a voltage measurement of each of the N phases of power converter 115. Here, local controller 160 may measure the voltage directly and communicate the measurement to harmonic injector 405 as a voltage measurement. In both the direct and indirect examples, the voltage measurements are each calculated by measuring the capacitance (e.g., lower capacitor C) of an LC filter of each phase of power converter 115. B Or C f255) (e.g., V c,a , V c,b , and V c,c ) may also be used.

[0097] In these embodiments using direct or indirect voltage measurements, the Nth phase harmonic injection multiple can be considered a feedback signal calculated from at least N voltage measurements, including at least one voltage measurement for each phase of power converter 115. In some of these embodiments, the Nth phase harmonic injection multiple is a sinusoidal or triangular signal. The above example equations that harmonic injector 405 can use to calculate a sinusoidal or triangular signal can similarly be used to calculate a sinusoidal or triangular signal in these embodiments, with the voltage measurement signal replacing the control reference target in the equation, respectively.

[0098] Further, in some embodiments of power converter system 100, N is three and the multiple of the Nth phase harmonic injection is a third of the fundamental frequency of the AC voltage section of the power converter. However, as previously mentioned, in some embodiments, N may be another integer value and / or another multiple of the Nth phase harmonic may also be selected.

[0099] Although described with respect to system 400 of FIG. 4, the harmonic injection feature may, at least in some examples, be incorporated into the other power converter systems disclosed herein.

[0100] III.Cascade Control System In some embodiments, the power converter system includes a non-isolated N-phase power converter and a cascade control system. The cascade control system includes a central controller and at least one local controller. For example, referring to systems 100 and 400 of FIGS. 1 and 4 above, the control system 105 may be a cascade control system including a central controller 150 cascaded with one or more local controllers 160. When the control system 105 is referred to herein as a cascade control system 105, it should be understood that the control system 105 includes at least one of the optional local controllers 160 in addition to the central controller 150. The cascade control system 105 may provide, for example, resonance damping, improved dynamic performance, and / or leakage current damping capabilities. Additionally, the cascade control system 105 may improve modularity of components (e.g., facilitating the addition and removal of local controllers and corresponding converter blocks as a modular autoconverter module), as described in more detail below with respect to FIGS. 18A, 18B, and 19.

[0101] In some embodiments of the cascade control system 105, the central controller 150 provides an outer loop of control, and each of the local controllers 160 provides a separate inner loop of control. For example, the central controller 150 may implement a PI controller, a PID controller, or other tuning controller that regulates control of the power converter 115 in a rotating reference frame (e.g., a dq0 reference frame). As part of the outer loop of control, the central controller 150 generates a control reference target (e.g., target 415) based on the regulation in the rotating reference frame. The control reference target may be generated in a stationary (abc) reference frame. Additionally, the central controller 150 may provide the control reference target to the local controllers 160. The local controllers 160 may be configured to control one or more of the N phases of the power converter 115, and the control of the N phases of the power converter 115 is divided among the local controllers 160. Thus, each phase of the power converter 115 may be associated with and controlled by a particular local controller 160.

[0102] Each local controller 160 implements inner loop control via model predictive control (MPC), PI control, PID control, or another tuning technique based on a control reference target (e.g., target 415) received from the central controller 150. For example, each local controller 160 also controls the voltage (v c ) may receive a measured or estimated voltage value of the capacitor voltage (v c ) and a control reference target (e.g., v c *), each local controller 160 controls its associated converter block 262 to produce a capacitor voltage (v c), the switching of the power switching elements may be adjusted or controlled to achieve (or move towards) the desired power supply voltage (Vp) of the power converter 115. The inner loop control provided by each local controller 160 includes generating control signaling to be provided to the power switching elements of the power converter 115 (or, in FIG. 3A, converter 304). For example, with reference to FIGS. 3A-C and 4, local controller 160a provides control signaling to the power switching elements 235, 240 (M1, M2) of a first phase of the power converter 115, local controller 160b provides control signaling to the power switching elements 235, 240 (M3, M4) of a second phase of the power converter 115, and local controller 160c provides control signaling to the power switching elements 235, 240 (M5, M6) of a third phase of the power converter 115.

[0103] The central controller 150 and the local controllers 160 can communicate both monitoring information (e.g., sensor data) and control information with each other in real time (e.g., each control cycle). For example, each local controller 160 can determine and transmit to the central controller 150 electrical operating characteristics specific to one or more phases of the power converter 115 with which the local controller 160 is associated in real time. For example, with reference to FIG. 4, these electrical operating characteristics may be determined in real time based on the V c,abc ,i o,abc , and i L,abc( For example, V from the local controller 160a c,a ,i o,a , and i L,a ;V from local controller 160b c,b ,i o,b , and i L,b ;V from local controller 160c c,c ,i o,c , and i L,c) in the control reference target 415. In some embodiments, the local controllers 160 provide other electrical operating characteristics. Additionally, the central controller 150 can determine and transmit a control reference target (e.g., 415) in real time to each local controller 160. Although the control reference target 415 is shown as a voltage reference target, in some examples, the control reference target 415 can be a current reference target (e.g., i L,abc * or i o,abc *) In such an example, the local controller 160 can control the power switching elements of each phase according to the current reference target.

[0104] 6 illustrates a communication system 600 for a cascade control system such as that described above with respect to converter system 400 and other converter systems provided herein. Communication system 600 illustrates an example of communication for at least some examples of converter system 100 and converter system 400 (e.g., where n=3). For example, communication system 600 is an example of a communication system that enables the communication described above with respect to the cascade control system of FIG.

[0105] The communication system 600 includes a central controller 150 and local systems 605a-n. Each local system includes a respective local controller 160a-n and a respective local converter or converter block 262a-n (an instance of the converter block 262 described with respect to FIG. 2). The central controller 150 and the local controllers 160a-n are communicatively coupled via a communication bus 615. The communication bus 615 may include a collection of dedicated communication paths between each local controller 160 and the central controller 150, may include a shared communication path between the local controllers 160 and the central controller 150 (e.g., the communication includes addressing information to identify the intended destination device), or may include a combination thereof.

[0106] As previously mentioned, the central controller 150 and the local controllers 160 can communicate both monitoring information (e.g., sensor data) and control information with each other in real time (e.g., each control cycle). For example, the local controller 160 can g,abc ,i g,abc , and i L,abc The electrical operating characteristics, including one or more of: c,abc *,i L,abc *, or i o,abc The local controllers 160 may further generate and send PWM control signals to their corresponding converter blocks 262. The PWM control signals output by the local controllers 160 may indicate the duty cycle and / or frequency of the PWM signals driving the gate terminals of each power switching element of the converter blocks 262, or may be the PWM signals themselves. Each converter block 262 may further include a respective gate driver for driving the power switching elements of the converter block, or the gate drivers for the local converter systems 605 may be considered part of the corresponding local controller 160.

[0107] As described in more detail below, in some embodiments, a state estimator (e.g., state estimator 900 in FIG. 9) is associated with each of the local controllers and provides estimates of one or more electrical operating characteristics of the phase associated with the local controller based on sampling of other electrical characteristics of the phase. For example, the state estimator may estimate the capacitor voltage (v c,abc ) and the grid-side inductor current (i o,abc ), the switched inductor current in the phase (referred to herein as the inductor current i L,abcA Luenberger observer technique can be implemented to estimate the state estimator (also called the state estimator). The use of a state estimator can reduce the number of sensors used in the system to provide electrical characteristics to the MPC controller, thereby reducing the cost and / or size of the motor circuit.

[0108] In some embodiments, the cascade control system further incorporates one or both of MPC for harmonic injection, as previously described, or active damping to mitigate resonances, as described below.

[0109] IV. Model Predictive Control In some embodiments, a power converter system includes a non-isolated N-phase power converter and a control system that utilizes model predictive control (MPC). When used in a power converter system (e.g., systems 100 and 400), MPC can provide, for example, active resonance damping, improved dynamic performance, and / or leakage current damping capabilities.

[0110] A controller of the control system 105, such as the central controller 150 or the local controllers 160, that implements MPC may be referred to as an MPC controller. The MPC controller may be configured to determine electrical operating characteristics of the power converter 115 (e.g., characteristics of each phase of the converter), determine one or more control reference targets for the power converter 115 (e.g., targets per converter phase), and then generate control signaling based on an MPC algorithm that uses the electrical operating characteristics and the control reference targets. The control signaling may be applied to operate power switching elements of the power converter 115 to perform voltage conversion and active damping and to mitigate resonances in the filter circuit 120 of the power converter 115.

[0111] An MPC controller (or MPC controller) may implement an MPC algorithm for each phase of the power converter 115 to generate control signaling. As used herein, MPC may refer to a control algorithm that is dynamically dependent or aware of the system (e.g., implements or uses a dynamic model that represents the converter under control) and predicts input commands or reference values ​​for controlling the behavior of the system by calculations based on the electrical characteristics of the converter and the dynamic model. Thus, MPC as used herein may refer to model predictive control algorithms (as described in more detail below) in a stricter usage of the term, as well as other dynamic predictive algorithms (e.g., a linear quadratic regulator (LQR) control algorithm).

[0112] In one example, to implement the MPC algorithm for a particular phase, the MPC controller can solve a cost function using the electrical characteristics and control reference target of that phase at each control period. By solving the cost function, the MPC controller can predict future steps of control signaling to operate power switching elements to control the power of that phase of the AC voltage section of the power converter toward the control reference target. The MPC controller can then generate the control signaling for that particular stage based on the first step of the future steps of the control signal. Thus, in contrast to the PI control algorithm, the MPC algorithm derives the optimal duty cycle by linearly processing the state variables and tracking error with specific coefficients. Since no integration procedure is required in MPC, the dynamic performance of MPC can be improved compared to the PI technique with less overshoot and faster tracking speed. Furthermore, since MPC has a higher control bandwidth, the MPC controller can provide active damping terms to mitigate (reduce or eliminate) LC or LCL resonances that may otherwise be present in the filter circuit in the AC section of the power converter 115.

[0113] Figure 7 illustrates a power converter system 700 including MPC control. Converter system 700 is an example of systems 100 and 400 described above, in which local controllers 160 are implemented as MPC controllers. In particular, in Figure 7, these local controllers are identified as local MPC controllers 760a-c. Thus, the above description of system 400 of Figure 4 also applies to system 700 of Figure 7, with like numbers used for like components.

[0114] As shown in Figure 7, converter system 700 includes a control system 105, which is a specific example of the control system 705 referenced above (e.g., with respect to Figures 1 and 4). The control system 705 includes a central controller 150 and local controllers 760a-c. Although shown separately, the gate drivers 402 may also be considered part of the local controllers 760a-c. The converter system 700 is a three-phase converter configured to function as an AC / DC rectifier and / or a DC / AC inverter.

[0115] The central controller 150 may, for example, in a manner similar to that described above with reference to FIGS. 4 and 6, provide three-phase control references (three-phase capacitor voltage references v c,abc The local MPC controllers 760a-c also generate L,abc and v c,abc By adjusting the weighting coefficient between the switch-side inductor current i L,abc Adjust.

[0116] Each local MPC controller 760 performs MPC-based control for each phase in the stationary abc coordinate system. In this example, the MPC-based control involves applying a dynamic model of the corresponding converter circuit under control (e.g., converter block 262a-c of a particular phase associated with each MPC controller 760a-c). More specifically, the MPC-based control involves solving an optimization function defined based on the dynamic model to identify an (optimal) control input (or inputs). The dynamic model may include measurements or estimates of the dynamic system as well as target or reference commands. In some examples, the MPC-based control involves solving an optimization function for each control period over a finite time range to identify control inputs for each step within the time range to achieve a desired output. The control input of the first step is then applied while the other control inputs are discarded. In the next control period, the process is repeated to identify the next control input. In some examples, another MPC control algorithm is implemented.

[0117] Using local MPC controllers 760 to implement MPC-based control for each phase in the stationary abc coordinate system includes, for example: (1) the state-space matrices of the LC for each phase are simpler than the rotating dq (or dq0) system for implementing offline piecewise affine optimization code in lower-cost controller hardware (e.g., cheaper DSP controllers); (2) the explicit MPC state-space matrices for the offline optimization calculations allow the omission of the time-varying angular velocity term ω used in the calculations; and (3) the MPC for each phase for LC is more flexible in terms of modular design for extending the number of parallelized phases and other topologies, e.g., DC / DC, single-phase DC / AC converters.

[0118] In the case of MPC implementation, for each control period, the local MPC controllers 760a-c receive the sensor 140 (e.g., the switch-side inductor current (i L,abc ), capacitor voltage (v c,abc ), and AC interface current (i o,abc) from the central controller 150 and a control reference target 415 (here, a capacitor voltage reference, v * c,abc ) may be received. As previously mentioned, each of the electrical properties from sensors 140 may be directed to be sensed (e.g., by a current or voltage sensor), or one or more of the electrical properties may be inferred from another sensed electrical property (e.g., see the state estimation discussion below).

[0119] In some examples, each of the local MPC controllers 760a-c includes an offline generated piecewise affine search tree that is used to derive a duty cycle (e.g., an optimal duty cycle) for explicit MPC control. To this end, the state equation of the switch-side LC filter (e.g., LC filter 308) can be expressed as follows:

number

number

[0120] For the flexibility of implementing explicit MPC and the convenience of experimentally adjusting the DC bus voltage during testing, the last term U dc d(k) is the phase leg output voltage u x (k) can be substituted. The state space model can be expressed in standard matrix form as follows: X k+1 =AX k +Bv k +Ee k where the variables and matrices are

number

number

[0121] In the MPC formulation, the inductor current / capacitor voltage reference is

number

number

number

number

[0122] Therefore, the cost function contains two terms.

number

[0123] For the penalty cost function, Q and R represent the weighting coefficient matrices applied to the state and input values, respectively. Specifically, Q is a 2 × 2 matrix [Q 11 ,0;0,Q 22 ] column. Because the target of the local MPC controllers 760a-c is to track the output capacitor voltage reference, in some examples, the corresponding weighting factor Q 22 is the switch-side inductor current term Q 11 R is constructed to be larger (e.g., 1000 times larger) than R. R is a 1 × 1 matrix [R 11 ]. R 11 Q 22 In other examples, other weighting factors may be used.

[0124] The constraints of the MPC controller can be expressed as follows:

number

number

[0125] FIG. 8 illustrates an exemplary implementation of an MPC control system 800 that may be executed by each of the local MPC controllers 760a-c. In this example, the MPC algorithm is implemented in an explicit manner. The MPC control algorithm executed by the local MPC controllers 760a-c is represented by an MPC control block 805. Specifically, a piecewise affine (PWA) feedback law is generated offline based on preselected state space modeling and constraints. The corresponding MPC partitions 810 are then stored in the memory of each local MPC controller 760a-c to be available for online search. In each control period, the MPC control block 805 searches n regions of the PWA MPC partition 810 (at block 820) to identify an active region r based on the input 815 received by the MPC control block 805. For example, the MPC control block 805 can use a binary search tree to search and quickly find the active region r from the n regions. Furthermore, each of the n regions is associated with a respective pair of identity matrices H and K. Therefore, the applicable active region r is determined by the matrix H r and K r Then, for the active region r, the corresponding feedback law (control) matrix F r and G r(block 825) to calculate an input matrix that contains optimal input values ​​over a prediction horizon (or time window). The first value of the input matrix is ​​then output and applied to the dynamic system for MPC control, while the other input values ​​of the input matrix are discarded.

[0126] Thus, the (offline generated) MPC partitions 810 represent n regions of PWA feedback laws for the MPC control block 805 to search. During operation of the MPC control block 805 (online), the identity matrix H r and K r leads to the active region of the MPC partition 810, and the corresponding control matrix F r and G r is the optimal input value for PWM modulation (u N (k)), where uN(k)=(v dc *d(k)), where vdc is the DC bus voltage across the DC terminals 220 (see, for example, FIG. 3A), and d(k) is the duty cycle of the PWM control signal. r ,G r is derived based on the cost function and constraints described above.

[0127] At each control period, the MPC control block 805 calculates the corresponding identity matrix H r and K r input 815 (e.g., i L (k),v(k),i g (k) and v c,ref (k) where k indicates phase a, b, or c. Next, F s,c and G s,cA duty cycle d(k) is derived using a particular control matrix of and output by the control block 805 (e.g., as part of uN(k)). The output duty cycle (d(k)) may be a value between 0 and 1. The output duty cycle is provided to a dynamic system 830 representing the converter block 262 (e.g., a gate driver associated with a local MPC controller 760a-c implementing the MPC control block 805 may receive the output duty cycle).

[0128] In some examples, control system 105 includes N MPC controllers (e.g., N local controllers 160, where N>1), one for each phase of power converter 115. In some embodiments, each MPC controller receives a control reference target for the phase associated with the MPC controller from a central controller (e.g., central controller 150). In other (non-cascaded control system) embodiments, each MPC controller determines its respective control reference target locally. For example, the MPC controllers may execute a separate MPC algorithm to derive the control reference target, or may include a non-MPC algorithm (e.g., a PI control algorithm, a PID control algorithm, etc.) executed to derive the control reference target.

[0129] In some embodiments, a state estimator is associated with each of the N MPC controllers to provide an estimate of one or more electrical characteristics of the phase associated with the MPC controller based on sampling of other electrical characteristics of the phase. For example, the state estimator may estimate the capacitor voltage (v c,abc ) and the grid-side inductor current (i g,abc ), the switched-side inductor current of the phase (referred to herein as the inductor current i L,abcA Luenberger observer technique can be implemented to estimate the state estimator (also called the state estimator). The use of a state estimator can reduce the number of sensors used in the system to provide electrical characteristics to the MPC controller, thereby reducing the cost and / or size of the motor circuit.

[0130] In some embodiments, an MPC for active damping to mitigate resonance may be included in a power converter that includes one or both of a cascade control system and harmonic injection, as previously described.

[0131] V. State Estimator As described herein, in some examples, the control system 105 or controller 150, 160, 760, or 805 uses or implements a state estimator to determine one or more electrical characteristics of the corresponding converter under control. The use of a state estimator can reduce the number of sensors in the system, which can reduce sensor cost, reduce converter volume (improve power density), and / or improve control performance through noise immunity (i.e., reduced noise) compared to sensing specific electrical characteristics.

[0132] For example, with reference to the various power converter systems described herein (e.g., converter systems 100, 200, 300, 700), three variables are considered: the switch-side inductor current (i Lfs ), filter capacitor voltage (v Cf ), and the grid-side inductor current (i Lfg ) can be estimated by the other two variables. Referring back to FIG. 3A, the grid filter inductor may represent a discrete inductor or inherent inductance on the AC interface terminal 225 that couples the filter node 260 to the AC grid 302 and / or the motor 303, as the case may be. Thus, the grid filter inductor current (i Lfg ) also represents the interface current i oand are used interchangeably with reference to the state estimator 900. FIG. 9 illustrates a state estimator 900 for use with cascaded model predictive control of an LCL filter system, such as the converter 700 of FIG. 7. However, the state estimator 900 is applicable to other converters using similar principles. The state estimator 900 may be implemented by one of the controllers (e.g., controllers 150, 160, 760, 805), for example, as controller hardware or an executable software block. For example, referring to FIG. 7, the state estimator 900 may be incorporated into each of the local MPC controllers 760. An example of a state estimator 900 included within a local MPC controller is also illustrated in FIG. 12.

[0133] Specifically, the state estimator 900 calculates the capacitor voltage v Cf and the grid side inductor current i Lfg By sampling the switch-side inductor current

number

number

number

[0134] The state space equations for the discrete-time state estimator can be expressed in the standard matrix form:

number

number

number

number

number

[0135] L E is a 3×2 observer gain matrix that can be adjusted to achieve the minimum estimation error. A diagram of the state estimator is shown in Figure 9. The state observer minimizes the estimation error e(k) using the dynamic equation: e k+1 =(A E -L E C E )e k It is.

[0136] The estimated gain can be derived by:

number

number

[0137] In this particular example, a system incorporating state estimator 900 may not have a current sensor to directly sense the switch-side inductor current, but may instead rely on estimating this current value (e.g., based on a sensed voltage on the low-side capacitor and / or a sensed current in the grid-side inductor). This approach may be beneficial because it may be difficult to directly sense the switch-side inductor current with a current sensor, for example, due to noise from the proximity of the sensor to the power switching devices of the converter.

[0138] In some examples, instead of or in addition to performing state estimation based on a Luenberger observer as described above, other estimation techniques can be used, such as, but not limited to, optimization-based estimators, sliding mode estimators, and disturbance estimators.

[0139] In some embodiments, such a state estimator may be included in a power converter that includes one or more of a cascade control system, harmonic injection, or MPC-based control, as previously described.

[0140] VI. Variable Frequency Critical Soft Switching In some examples, one or more of the controllers provided herein (e.g., controllers 150, 160, 760) drive their corresponding power converter blocks 262 (e.g., forming converters 115, 200, 300, or 304) using a variable frequency critical soft switching (VFCSS) scheme. The VFCSS scheme can provide improved efficiency and reduced filter volume (i.e., improved power density) of the power converter. Soft switching allows turn-on switching losses to be replaced by turn-off switching losses, which is at least beneficial because the turn-on losses of at least some FETs (e.g., SiC FETs) are typically much larger than the turn-off losses. This VFCSS technique allows for an increase in switching frequency (e.g., 5 times) and a reduction in inductance (e.g., 20 times) while reducing FET switching losses, resulting in improved power density and efficiency.

[0141] VFCSS is implemented by varying the switching frequency to achieve a desired inductor ripple current in an LC filter (e.g., in LC filter 245 and switch-side inductor 250 of LC filter 308 in FIGS. 2 and 3) to provide a soft switching transition. The desired inductor ripple current is determined by the inductor current valley point being equal to the inductor threshold current I L,thr For a converter such as converter 200 of FIG. 2 or converter 304 of FIG. 3A, I L,thr is set according to the dead time of inductor 250 and the boundary conditions of the peak / valley inductor current, which can be derived from the output capacitance of the corresponding switching elements 235, 240. d ) and peak-valley inductor current I L,max and I L,min, and , respectively. The inductor current and dead time values ​​that result in soft switching are identified as the soft turn on switching region or regions, and the inductor current and dead time values ​​that do not result in soft switching are identified as the hard switching region or regions. The soft switching region represents the operating region where there is sufficient time and current to discharge the output capacitance of the power switching element (M1 or M2) before turning on. Analytically, these boundaries are expressed as follows: (1 / 2)I L,max T d ≦Q min ≦0, (1 / 2)I L,min T d ≧Q max ≧0 Here, Q min and Q max is the minimum discharge threshold of the switch output capacitance for soft switching.

[0142] For large positive values ​​of the DC inductor current, the threshold current level -I L,thr A large current ripple (e.g., greater than 200% or in the range of 200%-300% of the current through the inductor) is used or required to keep the valley inductor current point lower than the threshold current I. The negative inductor current discharges the upper switch output capacitance during the turn-off transition of the lower switch. Similarly, for high negative values ​​of DC inductor current, the peak inductor current point is higher than the threshold current I. L,thrA large current ripple is also used or required to ensure that the threshold current is greater than . Zero voltage switching (ZVS) of the lower switch is achieved if the output capacitance of the lower switch is fully discharged by the positive inductor current during the turn-off transition of the upper switch. In general, to achieve complete soft switching over the entire cycle (e.g., the entire grid cycle), the current ripple should be large enough to ensure a bidirectional inductor current path or the dead time should be extended. Since an unnecessarily long dead time can introduce distortion, the VFCSS adjusts the switching frequency to maintain critical soft switching over the entire cycle. The VFCSS scheme is implemented to maintain a positive threshold current during the negative part of the cycle and a negative threshold current during the positive part of the cycle. The switching frequency to achieve this for any threshold can be calculated by the following equation:

number

[0143] 11 illustrates a control system 1100 for controlling a pair of switching elements of a power converter. In particular, the control system 1100 illustrates a controller 1160 that implements an exemplary control scheme for VFCSS control of the converter block 262 (see FIG. 2). In some examples, the controller 1160 is a specific implementation of one or more of the controllers 150, 160, 750, and 760. The controller 1160 determines a reference duty cycle (d*) and a reference switching frequency (f SWThe frequency generating controller 1110 includes a duty cycle generating controller 1105 and a frequency generating controller 1110, which may be regulators for generating a reference duty cycle (d*), respectively, based on sensed (or estimated) characteristics of the power converter 210, such as current and / or voltage, provided by the sensor 140, the state estimator 900, or a combination thereof. For example, the duty cycle generating controller 1105 may implement a PID controller, an MPC controller (see, e.g., MPC control block 805), or another type of regulator. The frequency generating controller 1110 may generate a reference duty cycle (d*), respectively, based on the sensed (or estimated) characteristics of the converter block 262 and F SW *Based on the above formula to calculate the reference switching frequency (f SW *) can be generated.

[0144] The gate driver 1115 receives a reference duty ratio (d*) and a reference switching frequency (f SW *) based on the received reference values. The gate driver 1115 generates a first PWM control signal for the upper switch (M1) 235 and a second PWM control signal for the lower switch (M2) 240 based on the received reference values. For example, the gate driver 1115 generates a PWM control signal for the upper switch (M1) 235 and a second PWM control signal for the lower switch (M2) 240 at a frequency (f SW ) and generates a first PWM control signal having a duty cycle (d1) equal to the reference duty cycle (d*). Similarly, the gate driver 1115 generates a first PWM control signal having a reference switching frequency (f SW *) equal to the frequency (f SW ) and 1-d1-(T d / f SW ) and an ON edge of the second PWM control signal occurs at a time T later than an OFF edge of the first PWM control signal. d / 2, and the OFF edge of the second PWM control signal is delayed by a time T d / 2 precedes.

[0145] While FIG. 11 illustrates VFCSS control for a single phase, FIG. 12 illustrates a VFCSS scheme implemented in a three-phase MPC control based power converter. More specifically, FIG. 12 illustrates a power converter system 1200 including MPC control with variable frequency critical soft switching (VFCSS). Converter system 1200 is another example of power system 100 and is similar to systems 400 and 700 described above, except that local controller 160 is implemented as an MPC-VFCSS controller. In particular, in FIG. 12, these local controllers are identified as local MPC-VFCSS controllers 1260a-c. Thus, the above description of system 100 of FIG. 1, system 400 of FIG. 4, and system 700 of FIG. 7 also apply to system 1200 of FIG. 12, with like numbers used for like components. Further, because system 1200 is in some respects an extrapolation of the single-phase VFCSS of FIG. 11 to a polyphase system, like components are numbered similarly with the addition of a phase designation of "a," "b," or "c," as the case may be (e.g., each of the three instances of frequency controller 1110 of FIG. 11 are identified as frequency controllers 1110a, 1110b, or 1110c in FIG. 12).

[0146] As shown in FIG. 12, converter system 1200 includes a control system 1205, which is a specific example of control system 105 and is similar to control system 705 referenced above (e.g., with respect to FIGS. 1, 4, and 7). Control system 1205 includes central controller 150 and local MPC-VFCSS controllers 1260a-c. Although shown separately, gate drivers 1115a-c may also be considered part of local MPC-VFCSS controllers 1260a-c. Converter system 1200 implements a three-phase converter configured to function as an AC / DC rectifier and / or DC / AC inverter. Thus, a converter circuit (e.g., power switching elements) identified as converter 304 may include respective converter blocks 262a-c for each phase a, b, c.

[0147] The central controller 150 calculates three-phase control references (three-phase capacitor voltage references v ) in the stationary abc reference frame based on the electrical characteristics of the converter 304 from the local controllers MPC-VFCSS 1260a-c, for example in a manner similar to that described above with respect to FIGS. c,abc *)

[0148] 12, each of the local MPC-VFCSS controllers 1260a-c includes a respective MPC controller 760a-c, a respective state estimator 900a-c, and a respective frequency controller 1110a-c. The MPC controllers 760a-c may function similarly to the MPC controllers 760a-c of FIG. 7, and may include a duty cycle reference d for the phase a, b, or c corresponding to the particular MPC controller 760a-c. a *,d b *, or d c11 to provide a reference frequency fsw* for a phase a, b, or c corresponding to a particular frequency controller 1110a-c. Further examples of frequency controllers 1110a-c are described below with reference to FIGS. 13-14. Gate drivers 1115a-c may function similarly to gate driver 1115 of FIG. 11 to provide an estimate of a phase a, b, or c corresponding to a particular state estimator 900a-c based on measurements provided by sensors 140. Frequency controllers 1110a-c may function similarly to frequency controller 1110 of FIG. 11 to provide a reference frequency fsw* for a phase a, b, or c corresponding to a particular frequency controller 1110a-c. Further examples of frequency controllers 1110a-c are described below with reference to FIGS. 13-14. Gate drivers 1115a-c may function similarly to gate driver 1115 of FIG. 11 to provide a reference frequency fsw* for a phase a, b, or c corresponding to a particular frequency controller 1110a-c. abc * and reference switching frequency f SW,abc Based on the *, a PWM control signal is provided to the power switching elements of the converter 304 for phase a, b, or c corresponding to the particular gate driver 1115a-c.

[0149] In some examples, state estimators 900a-c are not provided in system 1200, and instead, each of the measurements used by MPC controllers 760a-c and frequency controllers 1100a-c are provided by direct sensing via sensors 140 (e.g., as shown in FIGS. 11 and 14). In some examples, instead of MPC controllers 760a-c, separate local controllers 160a-c (e.g., PI or PID controllers) are provided for local PWM regulation of each phase of converter 304.

[0150] 13 and 14 show the reference switching frequency f SW12 show an example of a local MPC-VFCSS controller 1260 having different control strategies for generating the frequency .times. ...

[0151] The two controllers 1360 and 1460 are implemented to achieve the key soft switching behavior for high efficiency at different types of frequencies. The continuous frequency controller 1360 derives the continuous switching frequency based on the critical soft switching boundary conditions and then directly implements the frequency value into the PWM control signal (via the gate driver 1115). The continuous frequency controller 1360 also derives the switch-side inductor current value (i Lfs,est On the other hand, the discrete frequency controller 1460 may discretize the calculated switching frequency at multiples of the sampling frequency for PWM, without using the state estimator 900 to derive the switch-side inductor current value.

[0152] FIG. 15 shows the switch-side inductor current waveform 1500 for VCF-CSS and switch-side inductor current waveform 1505 VDF-CSS, respectively. The envelopes of VCF-CSS and VDF-CSS are smooth and discretized due to the different types of switching frequencies. Both techniques can achieve important soft-switching behavior to improve efficiency. Both VCF-CSS and VDF-CSS techniques can be combined with MPC-based control to address the time-varying switching frequency, and MPC-based control can improve transient performance with less oscillations and spikes, even in the case of discretized frequency VDF-CSS techniques. Therefore, the corresponding di / dt stress on the power switching elements of the converter is low.

[0153] More specifically, referring to FIG. 13, a continuous frequency controller 1360 can be designed to calculate a desired continuous switching frequency based on the peak / valley switch-side inductor current and critical soft-switching boundary conditions. More specifically, the continuously varying switching frequency f SW,cal is the threshold current for the critical soft switching boundary condition (I th ) is derived based on the switch side inductor current ripple Δi Lfs can be calculated as follows:

number

number

number

[0154] As shown in FIG. 13, a continuous frequency control block 1310 (an example of the frequency controllers 1110a to 1110c in FIG. 12) receives i Lfs,est ,v Cf,est and i o,est and a reference duty cycle value (d*) from the MPC controller 760. Based on these received values, the continuous frequency control block 1310 determines the reference switching frequency f SW,cal The frequency controller 1310 calculates the reference switching frequency f SW,cal to the gate driver 1115.

[0155] State estimator 900 can provide a more accurate switch-side inductor current value for reference switching frequency calculation compared to direct sampling of the current (e.g., via sensor 140). For example, direct sampling via sensor 140 can result in deviation of the sampling from the true average inductor current value when the switching frequency changes, especially when critical soft switching current ripple is large. However, this deviation error can be mitigated as a result of the calculations performed by state estimator 900.

[0156] 14, discrete frequency controller 1460 includes similar components as continuous frequency controller 1360 (similarly numbered), except that it includes a discrete frequency control block 1410 instead of continuous frequency control block 1310 and state estimator 900. Discrete frequency control block 1410, like continuous frequency control block 1310, is another example of frequency controllers 1110a-c of FIG. 12. Instead of state estimator 900, discrete frequency controller 1460 (including MPC controller 760 and discrete frequency control block 1410) receives measurements from sensors 140 for associated currents and voltages.

[0157] In the discrete frequency controller 1460, the continuously varying switching frequency in the above equation is a function of the fundamental sampling frequency f SW,base The PWM signal is then further discretized into predetermined frequency bandwidth sections designed as integer multiples of f. Thus, the discretized variable switching frequency in the PWM signal is SW,base (

number

[0158] The relationship between the PWM switching carrier signal and the sampling signal (for the sensor 140) is shown in plot 1600 of FIG. SW,base 2F from SW,base , then f SW,base A variable switching frequency to is shown. The process of frequency discretization can be expressed as follows:

number

[0159] The discretized frequency may ring back and forth due to the oscillation of sampling noise during frequency change transients. After the frequency discretization process, a hysteresis loop is constructed to remove the frequency oscillation. Then, the reference discretized frequency (f SW,discrete ) to the gate driver 1115 to control the frequency of the PWM control signal to the converter 304.

[0160] Compared with VCF-CSS, VDF-CSS discretizes the switching frequency to multiple times the fundamental sampling frequency. This allows the switch-side inductor current to be sampled at the mean point of the current ripple without deviation from the correct value, as shown in Figure 16. Therefore, i Lfs Even without a state estimator to estimate V, the inductor current sampling can be accurate for critical soft-switching calculations at high current ripple.

[0161] 17A and 17B include plots 1700 and 1705, respectively, illustrating exemplary experimental results of one example of a power converter system 1200 as described herein incorporating a three-phase converter with SiC FETs (see, e.g., FIG. 3A), third harmonic injection (see, e.g., FIG. 4), a cascaded control system (see, e.g., FIGS. 4, 6, 7), an MPC-based local controller in the cascaded control system (see, e.g., FIG. 7), and variable frequency soft switching (see, e.g., FIGS. 11-14). Other examples provided do not include one or more of these features (e.g., a zero sequence voltage control reference of Vdc / 2 is provided in place of the third harmonic injection, or another local regulator is included in place of the local MPC-based control).

[0162] In Figure 17A, plot 1700 shows rate power (W) versus switching frequency (Hz) for power converter system 1200 as well as several other example systems. In Figure 17B, plot 1705 shows power density (kW / L) versus efficiency (%). As shown, in comparison to other systems, power converter system 1200 can achieve a high switching frequency and a balance of both high power density and high efficiency.

[0163] In some embodiments, a VFCSS as described may be included in a power converter that includes one or more of a cascade control system, harmonic injection, MPC-based control, or a state estimator as described herein.

[0164] VII. Modular Power Converters This section describes systems and methods for modular power converters composed of one or more modular power converter units, also referred to as autoconverter modules or power converter modules. Such autoconverter modules (ACMs) can be easily connected together for different applications and remain highly efficient power converters across different applications. As described further below, in some examples, each modular power converter may provide a single-phase multi-phase power output (e.g., in a DC / AC inverter application) or may receive a single-phase multi-phase power input (e.g., in an AC / DC rectifier application). In some examples, multiple modular power converters are coupled together in parallel for each phase of a multi-phase modular power converter. Any of the power converters previously described herein can be implemented as modular power converters based on the principles described in this section. That is, in some examples, one or more of the power converter systems 100, 400, 700, and 1200 previously described are modular power converters composed of one or more ACMs.

[0165] Referring to FIG. 18A, a modular power converter 1800 is shown having a single ACM 1805. In FIG. 18B, a modular power converter 1820 is shown having n ACMs 1805 connected in parallel. Each ACM 1805 has a DC link capacitor (C DC ), a high-side switch, a low-side switch, a midpoint node connecting the drain terminal of the high-side switch and the source terminal of the low-side switch, and an LC filter. As shown, the ACM1805 converter 200 includes a source-drain capacitor for each of the high-side and low-side switches, and the LC filter includes both the high-side and low-side capacitors, as will be described in more detail with reference to FIG. 2. In some examples, one or more of the source-drain capacitor and the high-side capacitor of the LC filter are not included in the ACM1805 converter 200. Similar to FIG. 2, the ACM1805 converter 200 further includes DC terminals 220 including a positive DC terminal 222 and a negative DC terminal 224, and interface terminals 225 including a positive interface terminal 227 and a negative interface terminal 229.

[0166] Further, each ACM 1805 may include a single printed circuit board (PCB) on which the elements of converter 200 are mounted. Additionally, although not shown in FIGS. 18A-18B, a local controller 160 (e.g., in the form of a local MPC controller 760 or a local MPC-VCSS controller 1260) may be part of each ACM 1805 and may be implemented or otherwise included on the same PCB as converter 200 for that ACM. The PCB may be represented by a dashed box around each ACM 1805. Each ACM 1805 may be of a similar size, orientation, and general configuration so that it is modular and can be interchanged in and out of the converter system with another ACM 1805.

[0167] In some examples, a modular power converter is provided, such as modular power converter 1820, that includes n ACMs 1805 coupled to each other as shown in FIG. 18B and further coupled to a central controller (e.g., central controller 150) as shown in various power converter systems of the present disclosure (see, e.g., FIGS. 4, 6, 7, and 12). As described with respect to these examples, the central controller 150 can determine target operating parameters (e.g., at a macro level) for the modular ACMs 1805 and provide these target operating parameters to local controllers of those ACMs 1805. The local controllers can then control and adjust the power switching elements of their respective ACMs 1805 in accordance with their target operating parameters.

[0168] 18B , in some examples, the n ACMs 1805 include at least two power converter modules or three power converter modules coupled in parallel such that the positive DC terminals 222 of each ACM 1805 are coupled to each other, the negative DC terminals 224 of each ACM 1805 are coupled to each other, and the negative interface terminals 229 of each ACM 1805 are coupled to each other. Additionally, the positive interface terminals 227 of the ACMs 1805 of a particular phase of the AC may be coupled to each other, or in the example of one ACM 1805 per phase, each positive interface terminal 227 may be independent of (i.e., not coupled to) any other positive interface terminals 227 of the active ACMs 1805.

[0169] In some examples, modular power converters 1800 and 1820 are AC-DC rectifiers, DC-AC inverters, or multi-mode power converters having an AC-DC rectifier mode and a DC-AC inverter mode.

[0170] In some examples of modular power converters 1800 and 1820, each local controller is configured to drive a pair of power switching elements of one or more ACMs 1805 using variable frequency critical soft switching at a frequency of at least 20 kHz, at least 40 kHz, at least 60 kHz, at least 80 kHz, at least 100 kHz, 60 kHz to 1 MHz, 100 kHz to 1 MHz, or 300 kHz to 1 MHz. In some examples, the LC filter of each of the one or more power converter modules is configured to filter an AC power signal received by the LC filter, the AC power signal having a current ripple of at least 200% of a local average current, the average current being ... Lf ) indicates the instantaneous value of the output current passing through the

[0171] In some embodiments, a process for converting power with a modular power converter is provided. For example, the process may include receiving input power by one or more power converter modules. Each of the one or more power converter modules may include a circuit board having a positive direct current (DC) terminal and a negative DC terminal, a capacitor coupled across the positive and negative DC terminals, a power switching element pair, an LC filter including a capacitor and an inductor, a local controller coupled to the power switching element pair, the positive and negative DC terminals, the capacitor, the power switching element pair, the LC filter, and the local controller, as described above. The process may further include driving, by the local controller, the power switching element pair using variable frequency soft switching to convert the input power to an output power. The process may further include communicating, by a central controller, with each local controller of the one or more power converter modules.

[0172] FIG. 19 illustrates a modular three-phase power converter 1900. The converter system 1900 is another example of the power system 100 and may incorporate elements of the systems 400, 700, and 1200 described above. Thus, the above description of similar aspects of the system 100 of FIG. 1, as well as the system 400 of FIG. 4, the system 700 of FIG. 7, and the system 1200 of FIG. 12, also applies to the system 1900 of FIG. 19, and similar numbers are used for similar components. For example, the power converter 1900 is shown as being coupled to the AC grid 302 via the grid connection point 225a, and the motor connection point 225b or the AC motor 303 are not shown. However, in some embodiments, the power converter 1900 further includes a motor connection point 225b, each coupled to an output of the common mode inductor 312, similar to the diagram of FIG. 3A. Thus, like the previously described systems (e.g., 300, 400, 700, and 1200), power converter 1900 may be a bidirectional power converter that can use AC grid power to charge a DC power source and use the DC power source power to drive an AC motor.

[0173] The modular three-phase power converter 1900 includes three ACMs 1905, one for each phase of the three-phase power converter 1900. Each ACM 1905 is generally similar to the ACM 1805 of FIGS. 18A and 18B, but includes m parallel-connected converter blocks 262 in each ACM 1905. For example, in FIG. 19, three converter blocks 262 in the ACM 1905 for phase C are labeled, although there may be additional converter blocks 262 in phase C. In FIG. 19, three converter blocks 262 are also illustrated but not labeled for phases A and B for ease of illustration. As shown, each ACM 1905 includes shared DC and interface terminals for the m converter blocks 262 that make up that particular ACM 1905. Additionally, each converter block 262 in each ACM 1905 may be associated with a local controller on the same PCB as the converter block 262. Thus, the converter 1900 may include 3×m local controllers for a one-to-one relationship with the 3×m converter blocks 262. In other examples, a local controller may control multiple converter blocks 262. The local controller may be implemented as one of the local controllers described herein, such as local controller 160, 760, or 1260. In FIG. 19, the 3×m local controllers are represented by local MPC controllers 7601-760. 3m It is implemented as:

[0174] 19 is illustrated as having m converter blocks 262 and corresponding local MPC controllers each, in some examples, the ACM 1905 is an ACM assembly that includes m ACMs 1805. In other words, each phase of the converter 1900 may include multiple ACMs 1905 connected together to form the ACM 1805. Furthermore, in some embodiments, the power converter 1900 is constructed without modular ACMs 1805 or ACMs 1905 (e.g., the circuitry may not be modularized but rather may be on multiple circuit boards, custom boards, etc.).

[0175] The modular multi-phase MPC power converter 1900 implements a converter with power modules stacked in parallel for each phase to increase the current and power rating of each phase of the converter. Each of the stacked power modules receives a control reference target (e.g., a reference voltage (v cf,abc 7 and 12, respectively, to control the converter block 262 corresponding to the particular local MPC controller.

[0176] Thus, the ACMs 1805 and 1905 described herein provide a modular power converter system in which the ACMs 1805 and / or 1905 can be used as modular building blocks to design a modular power converter that meets desired specifications in terms of number of phases, current ratings, power ratings, etc.

[0177] Although the various converter circuits provided herein are described primarily in the context of power switching element pairs including an upper switch and a lower switch, in some examples, one or more of these converters include power switching elements arranged in a multi-level switch topology (e.g., a three-level or five-level switch topology), and the power switching element pairs of each power converter module may include two or more high-side switching elements and two or more low-side switching elements.

[0178] In addition to the functions and operations of the various power converters previously described, the following are examples of the operating processes of the disclosed power converters.

[0179] In Figure 20, a process 2000 for converting power is provided. Process 2000 is described as being performed by power converter system 100 implemented as power converter system 1200 of Figure 12. However, in some embodiments, process 2000 may be implemented by another power converter system, or by power converter system 100 implementing another power converter system (e.g., converter systems 400, 700, 1900, or another system provided herein). Additionally, although the blocks of process 2000 are shown in a particular order, in some embodiments one or more of the blocks may be performed partially or wholly in parallel, may be performed in a different order than that shown in Figure 20, or may be bypassed.

[0180] In block 2005, an N-phase power converter stage (N≧1) receives input power from an alternating current (AC) side or a direct current (DC) side. For example, when operating as a DC / AC inverter, the (three-phase) power converter stage 304 of the power converter system 1200 can receive an input DC voltage from a DC power source such as a battery, a capacitor, an ultracapacitor, or a DC power source from a rectified AC power source (e.g., AC grid power converted to DC power by a diode bridge rectifier). For example, the power converter 304 of FIG. 12 is shown in more detail in FIG. 3A. In FIG. 3A, on the DC side of the power converter 304, a DC power source 306 is connected to the power converter 304 via DC terminals 220.

[0181] Furthermore, when operating as an AC / DC rectifier, the (three-phase) power converter stage 304 of the power converter system 1200 can receive an input AC voltage from an AC source, such as an AC grid or an AC generator (e.g., a motor operating in regenerative braking mode). For example, the power converter 304 of FIG. 12 is shown in more detail in FIG. 3A. In FIG. 3A, on the AC side of the power converter 304, the AC grid 302 is coupled to the power converter 304 via the AC interface terminals 225. Alternatively, the AC motor 303, which may operate as a generator during regenerative braking or may be an engine-generator, is coupled to the power converter 304 via the AC interface terminals 225.

[0182] In block 2010, an N-phase LC filter filters the AC side of the N-phase power converter stage. The N-phase LC filter includes one or more capacitors, with one or more neutral points of each of the one or more capacitors electrically connected to the DC negative terminal of the DC power source. For example, in the power system 1200 of FIG. 12, the (three-phase) LC filter 308 is an example of an N-phase LC filter that filters the AC side. The LC filter 308 of FIG. 12 is shown in more detail in FIG. 3A, at least in some examples. In FIG. 3A, the LC filter 308 includes three lower capacitors 255 and three switch-side inductors 250. The three lower capacitors 255 have a neutral connection point coupled to a neutral point 311a that is coupled to the negative DC terminal 224. In some examples, the LC filter 308 further includes three upper capacitors 215, as shown in FIG. 3A. In both cases where power converter stage 304 operates as an AC / DC rectifier and a DC / AC inverter, LC filter 308 filters the AC signal between midpoint node 242 and interface terminal 225. Further details of LC filter 308, at least in some examples, are provided above, e.g., with respect to FIG.

[0183] In block 2015, a control system (e.g., control system 1205) drives power switching elements of an N-phase power converter stage (e.g., converter stage 304) to convert an input power and output a converted power. Further, the control system drives the power switching elements using variable frequency soft switching at a frequency of at least 20 kHz. For example, to drive the power switching elements, the control system 1205 can implement a cascaded control system including a central controller 150 and N local controllers 760. As described above, the central controller 150 can determine a rotating reference frame target and generate N control reference targets 415. The local controllers 760 can receive the N control reference targets 415 and drive the power switching elements of their corresponding converter blocks 262 using control signaling according to the received control reference targets. Further, the control signaling drives the power switching elements using variable frequency critical soft switching (VFCSS), for example, as described above with respect to FIGS. 12-16. The switching frequency may be at least 20 kHz, at least 40 kHz, at least 60 kHz, at least 80 kHz, at least 100 kHz, 60-100 kHz, 60 kHz-1 MHz, 100 kHz-1 MHz, or 300 kHz-1 MHz. Higher switching frequencies are particularly useful and provide efficient, power dense systems for the disclosed power converter 304 topology and associated control techniques.

[0184] The control signaling may include a PWM control signal provided to the power switching elements 235, 240 (e.g., at a gate terminal of the switching element), a reference duty cycle (d*) indicative of the duty cycle of the PWM control signal, and / or a reference switching frequency f SW * (For example, in the case of VFCSS) may be possible.

[0185] In some examples, the local controller 760 may implement MPC, as described above with respect to Figures 7-8. In some examples, the system 1205 may further implement zero sequence control, with or without harmonic injection, for example, as described with respect to Figure 4. In some examples, the control system operates the power converter stage in a traction mode and a charging mode (e.g., at different times in time). In the traction mode, the power converter stage converts input (DC) power received from a DC power source into converted output (AC) power and drives an N-phase motor coupled to the N interface terminals with the converted output (AC) power. In the charging mode, the power converter stage converts input (AC) power into converted output (DC) power and charges a DC power source with the converted output (DC) power. The traction mode and the charging mode are further described with respect to the systems 300 and 400 of Figures 3A-3C and 4, and these descriptions are equally applicable to the other power converter systems 700, 1200, and 1900. In some examples, the central controller 150 uses the estimated electrical characteristics from the local MPC controller 760 generated by the state estimation to generate the control reference target 415. In some examples, the local MPC controller 760 estimates the electrical characteristics using the state estimation and generates control signaling for the corresponding power switching elements.

[0186] Further description of the generation of reference targets, generation of control signals, communication in a cascaded control system, power conversion, and operation of control system 1200 are provided throughout this specification with respect to Figures 12-16 and may be incorporated into process 2000. For example, to generate control signaling, local controllers 1260a-c may perform one or more of state estimation (see, e.g., state estimator 900 and the description of Figure 9), zero sequence control with or without harmonic injection (see, e.g., the description of harmonic injector 405 with respect to Figure 4), and MPC control (see, e.g., the description of local MPC controller 760 in Figures 7 and 8 and local MPC-VFCSS controller 1260 in Figure 12).

[0187] As mentioned above, although process 2000 is described with respect to converter 1200 of FIG. 12, process 2000 may similarly be performed by converters 400, 700, and / or 1900. In such a case, power converter stage 304 (present in each of these converters) may function as described above to perform block 2005, LC filter 308 (present in each of these converters) may function as described above to perform block 2010, and control system 105 or 705 of each respective converter system may perform block 2015 to drive power switching elements to convert input power to output converter power using VFCSS (e.g., as described with respect to converter system 1200 of FIG. 12).

[0188] In Figure 21, a process 2100 for converting power for an electric vehicle is provided. Process 2100 is described as being performed by power converter system 400 implemented as power converter system 100 of Figure 4. However, in some embodiments, process 2100 may be implemented by another power converter system, or by power converter system 100 implementing another power converter system (e.g., converter systems 700, 1200, 1900, or another system provided herein). Additionally, although the blocks of process 2100 are shown in a particular order, in some embodiments one or more of the blocks may be performed partially or wholly in parallel, may be performed in a different order than that shown in Figure 21, or may be bypassed.

[0189] In block 2105, an N-phase power converter stage (N≧1) receives input power from an alternating current (AC) side having AC terminals or a direct current (DC) side having DC power supply terminals. For example, when operating as a DC / AC inverter, the (three-phase) power converter stage 304 of the power converter system 400 can receive an input DC voltage from a DC power source such as a battery, a capacitor, an ultracapacitor, a DC power source from a rectified AC power source (e.g., AC grid power converted to DC power by a diode bridge rectifier). For example, the power converter 304 of FIG. 4 is shown in more detail in FIG. 3A. In FIG. 3A, on the DC side of the power converter 304, a DC power source 306 is connected to the power converter 304 via DC terminals 220.

[0190] Furthermore, when operating as an AC / DC rectifier, the (three-phase) power converter stage 304 of the power converter system 400 can receive an input AC voltage from an AC source, such as an AC grid or an AC generator (e.g., a motor operating in regenerative braking mode). For example, the power converter 304 of FIG. 12 is shown in more detail in FIG. 3A. In FIG. 3A, on the AC side of the power converter 304, the AC grid 302 is coupled to the power converter 304 via the AC interface terminals 225. Alternatively, the AC motor 303, which may operate as a generator during regenerative braking or may be an engine-generator, is coupled to the power converter 304 via the AC interface terminals 225.

[0191] In block 2110, an N-phase LC filter filters the AC side of the N-phase power converter stage. The LC filter includes one or more capacitors, with one or more neutral points of each of the one or more capacitors electrically connected to the DC negative terminal of the DC power supply terminals. For example, in the power system 400 of FIG. 4, the (three-phase) LC filter 308 is an example of an N-phase LC filter that filters the AC side. The LC filter 308 of FIG. 4 is shown in more detail in FIG. 3A, at least in some examples. In FIG. 3A, the LC filter 308 includes three lower capacitors 255 and three switch-side inductors 250. The three lower capacitors 255 have a neutral connection point coupled to a neutral point 311a that is coupled to the negative DC terminal 224. In some examples, the LC filter 308 further includes three upper capacitors 215, as shown in FIG. 3A. In both cases where power converter stage 304 operates as an AC / DC rectifier and a DC / AC inverter, LC filter 308 filters the AC signal between midpoint node 242 and interface terminal 225. Further details of LC filter 308, at least in some examples, are provided above, e.g., with respect to FIG.

[0192] In block 2115, a control system (e.g., control system 400) drives the power switching elements of an N-phase power converter stage (e.g., power converter stage 304) in a charging mode and a traction mode. For example, referring to FIGURES 3A and 3B, when in a charging mode, power converter stage 304 converts input AC power received from grid 302 via AC terminals 225a to output DC power that is provided to DC power source terminals 220 to charge DC power source 306. Referring to FIGURES 3A and 3C, when in a traction mode, power converter stage 304 converts input DC power received from DC power source 306 via DC power source terminals 220 to output AC power that is provided to AC terminals 225b to drive motor 303. Control system 400 can alternately drive power switching elements 235 and 240 of converter stage 304 in a charging mode and a traction mode.

[0193] In some examples, the control system can drive the power switching elements in a charging mode during a first period (e.g., when the power converter stage is coupled to an AC grid via the AC terminals) and drive the power switching elements in a traction mode during a second period (e.g., when the AC grid is not connected via the AC terminals). In other words, the power converter stage can operate in a charging mode at a different time than in a traction mode. The control system can determine whether to operate in a charging mode or a traction mode based on, for example, whether the grid connection point 225a is currently coupled to an active AC grid 302 and whether the control system 400 has received a user or operator command to drive the motor. For example, in the case of an electric vehicle, the central controller 150 can determine to operate in a traction mode in response to detecting no connection to the grid 302, detecting an ignition switch being enabled on the electric vehicle, and / or detecting a user torque or driving comment (e.g., depression of the vehicle's accelerator pedal). An ignition switch and an input device (e.g., an accelerator pedal) for receiving torque or drive commands may be part of an I / O interface 142 (see FIG. 1 ) coupled to the control system 400. In traction mode, the central controller 150 (e.g., via a current reference generator 417) uses a first algorithm or scheme to generate a reference electrical characteristic (e.g., i o,dq *), and in the charging mode, the central controller 150 (e.g., via the CC / CV control block 418) may use a second algorithm or scheme to generate a reference electrical characteristic (e.g., i o,dq *) can be generated.

[0194] In order for the control system 400 to drive the power switching elements in both the charging and traction modes, the central controller 150 can generate and provide a reference target 415 to the local controllers 160a-c. The local controllers 160a-c can then generate control signaling for their corresponding converter blocks 262a-c. Further description of the generation of reference targets, generation of control signals, communication in a cascaded control system, power conversion, and operation of the control system 400 are provided throughout this specification with respect to Figures 4-6 and can be incorporated into the process 2100. To generate the control signaling, the local controllers 160a-c may perform one or more of state estimation (e.g., see the description of state estimator 900 and FIG. 9), zero sequence control with or without harmonic injection (e.g., see the description of harmonic injector 405 with respect to FIG. 4), MPC control (e.g., see the description of local MPC controller 760 in FIG. 7 and FIG. 8 and local MPC-VFCSS controller 1260 in FIG. 12), and variable frequency critical soft switching (VFCSS) (e.g., see the description of local MPC-VFCSS controller 1260 and VFCSS with respect to FIG. 11-FIG. 16).

[0195] As mentioned above, although process 2100 is described with respect to converter 400 of Figure 4, process 2100 may similarly be performed by converters 700, 1200, and / or 1900. In such a case, power converter stage 304 (present in each of these converters) functions similarly to above to perform block 2105, LC filter 308 (present in each of these converters) functions similarly to above to perform block 2110, and control system 105, 705, 1205 of each respective converter system may perform block 2015 to drive power switching elements to convert input power to output converter power in traction mode (as a DC / AC inverter) and charging mode (as an AC / DC rectifier).

[0196] Experimental testing of the embodiments provided herein has shown that converter systems such as the system 300 of Figure 3A have reduced motor leakage current and shaft voltage. For example, one 11 kW prototype having the topology shown in Figure 3A has shown a 94% reduction in peak-to-peak leakage current, a 97% reduction in RMS leakage current, and a 90% reduction in peak-to-peak shaft voltage compared to a converter without the common mode inductor 312, LC filter 308, and common mode voltage control described above. More specifically, a prototype converter system incorporating the topology shown in Figure 3A, a permanent magnet synchronous motor (PMSM), and the characteristics of Table 1 (below) was tested. [Table 1]

[0197] Table 2 (below) shows the leakage current measured from experiments using different variants of the prototype in traction mode. [Table 2]

[0198] Thus, in some examples of the disclosed systems and methods, leakage current is maintained at less than 0.5 A, less than 0.4 A, and less than 0.3 A (peak-to-peak) and / or less than 30 mA, less than 25 mA, less than 20 mA, or less than 17 mA (RMS).

[0199] Furthermore, this prototype is m =T L ω m Measure the input power P in =V DC I DCBy dividing by N=1200RPM, four cases were shown to show improved efficiency in traction mode: 20kHz switching without LC filter, representing a standard traction drive; 80kHz without LC filter, which is a standard drive topology at higher switching frequencies; 80kHz with the proposed topology, which does not always achieve soft switching; and variable frequency critical soft switching implementation of the proposed topology. The results of the efficiency measurements are shown in Figure 22, and it can be seen that the variable frequency drive has the highest efficiency, being 0.6% more efficient than the 20kHz standard drive at maximum power. The peak efficiency of the PMSM is 93% according to its datasheet.

[0200] Furthermore, in charging mode, the peak efficiency was measured to be between 98.4% and 99.4%. More specifically, the efficiency of the prototype converter system in charging mode was tested under different load and line conditions. The results are shown in Figure 23, where a nominal DC voltage of 835V was applied with a variation of ±10% of the nominal AC line voltage of 400V. The peak efficiency is 99.4% and the minimum efficiency at rated power is 98.4%. Other results in the literature range from 93-95% for a 3.3kW add-on interface integrated charger with a 400V battery, 90-95% for a 6.6kW integrated charger with a 6-phase machine, and 80% for a split-phase three-phase PMSM operating at 2kW. Non-integrated on-board chargers, i.e., charging-only units, have been shown to be up to 97% efficient at 22kW, but they require a significant number of components and commercially available on-board chargers are up to 95% efficient. Thus, the disclosed topology provides net efficiency and reliability benefits in traction mode, and performs well while eliminating the need for an isolation transformer in charging mode.

[0201] Of course, this particular prototype is merely one example of a power converter that can be implemented and operated in accordance with the embodiments and examples disclosed herein.

[0202] Execution of the various techniques and operations described herein may be facilitated by an electronic controller (e.g., a processor-based computing device), such as the central controller 150, local controller 160, local MPC controller 760, local MPC VFCSS controller 1260 described herein. Such an electronic controller may include a processor-based device, such as a computing device that may include a central processing unit (CPU) or processing core. In addition to the CPU or processing core, the system includes main memory, cache memory, and bus interface circuitry. The electronic controller may include memory storage devices, such as a hard drive (solid state hard drive or other type of hard drive) or flash drive, associated with the computer system. The electronic controller may further include a keyboard, keypad, or some other user input interface, and a monitor, such as an LCD (liquid crystal display) monitor, and may be located where a user can access them.

[0203] 24 illustrates a configurable power converter architecture 2400 according to an example disclosed herein. That is, the power converter architecture 2400 can be implemented by one or more of the power converter systems described herein, including systems 300, 700, 1200, and 1900. The architecture 2400 is a hierarchical software-defined control architecture including a central control layer 2405, a local control layer 2410, and an application layer 2415. The central control layer 2405 can be similar to or an example of the central controller 105 described herein. The central control layer 2405, among other things, manages the power converters of the local layers 2410, generates and executes central voltage, current, power, torque, speed, and / or control targets and functions, and can identify the type of application to which the architecture 2400 is being applied and reconfigure accordingly (e.g., to generate appropriate control targets and execute appropriate functions). The local control layer 2410 may be similar to or an example of one or more of the local controllers 160, 760, or 1260 described herein. The local controllers may provide one or more of local voltage / current control, MPC-based control, VFCSS control, state estimation / observation, and PWM modulation, as described above. In some examples, the local controllers of the local control layer 2410 are examples of modular local control modules or ACMs, as described with respect to Figures 18A, 18B, and 19. The central control layer 2405 and the local control layer 2410 are connected via a communication bus 2420 (similar to the bus 615 of Figure 6). The application layer 2415 may include interfaces for interfacing with different current-carrying loads / power source applications, such as EV batteries, single-phase grids, three-phase grids, solar (photovoltaic (PV)) arrays, motors, etc.

[0204] The architecture 2400 is flexible and configurable for several different types of applications using the same hardware. For example, the central control layer 2405 and / or the local control layer 2410 may include the same components, but the central control layer 2405 may have a central level controller programmed (software defined) to implement a particular application. In some examples, the central level controller may include various application software packages (e.g., one per application type) present thereon, and one software package is selected (or activated) for use during the installation of a configuration step of a given application layer connected to the local control layer 2410. The particular software package included or selected on the central level controller may include one or more of the various functions shown in FIG. 24. These functions, and software packages, generally, ultimately generate the reference targets of the local controllers of the local layer 2410. The connected application layer 2415 may be one of the applications shown in FIG. 24, including, for example, a solar (PV array) application, a battery application, a three-phase grid application, a single-phase grid application, a three-phase motor application (e.g., in an electric vehicle or industrial equipment setting), etc.

[0205] 25A-25B illustrate an example of architecture 2400 of FIG. 24 implemented in a single-phase grid application configuration 2500. Configuration 2500 includes a central control layer 2505, which is an implementation of layer 2405 of FIG. 24, configured for single-phase grid control operation. Configuration 2500 further includes a local control layer 2510, which is an implementation of layer 2410 of FIG. 24, and includes two local controllers (e.g., local controllers 160, 760, or 1260). Configuration 2500 further includes an application control layer 2515, which is an implementation of layer 2415 of FIG. 24, and is a single-phase grid application. In configuration 2500, architecture 2400 is operable to inject power into the grid (e.g., inverted from a DC source such as a battery, solar (photovoltaic) array, etc.), charge or power a DC load with DC power (e.g., received from the grid and modified), or both. To perform this power conversion, configuration 2500 may be operated and controlled using the principles described herein, such as with respect to Figures 2-21 (e.g., one or more of cascade control, stabilized common mode control, harmonic injection, MPC control, VFCSS control, or combinations thereof). Figure 25B shows an example circuit diagram of configuration 2500 along with local and central level controllers.

[0206] 26A-26B illustrate an example of architecture 2400 of FIG. 24 implemented in a three-phase grid application configuration 2600. Configuration 2600 includes a central control layer 2605, which is an implementation of layer 2405 of FIG. 24, configured for three-phase grid control operation. Configuration 2600 further includes a local control layer 2610, which is an implementation of layer 2410 of FIG. 24, and includes three local controllers (e.g., local controllers 160, 760, or 1260). Configuration 2600 further includes an application control layer 2615, which is an implementation of layer 2415 of FIG. 24, for a three-phase grid application. In configuration 2600, architecture 2400 is operable to inject power into the grid (e.g., inverted from a DC source such as a battery, solar (photovoltaic) array, etc.), charge or power a DC load with DC power (e.g., received from the grid and modified), or both. To perform this power conversion, configuration 2600 may be operated and controlled using the principles described herein, such as with respect to Figures 2-21 (e.g., one or more of cascade control, stabilized common mode control, harmonic injection, MPC control, VFCSS control, or combinations thereof). Figure 26B shows an example circuit diagram of configuration 2600 including a PV array as a DC load / power source (which may be a battery, ultracapacitor, etc.) along with a local controller and a central level controller.

[0207] 27A-27B illustrate an example of architecture 2400 of FIG. 24 implemented in a three-phase motor application configuration 2700. Configuration 2700 includes a central control layer 2705, which is an implementation of layer 2405 of FIG. 24, configured for motor operation. Configuration 2700 further includes a local control layer 2710, which is an implementation of layer 2410 of FIG. 24, and includes three local controllers (e.g., local controllers 160, 760, or 1260). Configuration 2700 further includes an application control layer 2715, which is an implementation of layer 2415 of FIG. 24, for a three-phase motor application. In configuration 2700, architecture 2400 is operable to drive a motor (e.g., inverted from a DC power source such as a battery, solar (photovoltaic) array, etc.), charge or power a DC load with DC power (e.g., received from the motor and rectified). In some examples, the configuration 2700 is further coupled to an AC grid and configured to charge a DC power source with DC power (e.g., received from the grid and modified) and to inject power into the grid (e.g., inverted from the DC power source). To perform this power conversion, the configuration 2700 can be operated and controlled using the principles described herein with respect to Figures 2-21 (e.g., one or more of cascade control, stabilized common mode control, harmonic injection, MPC control, VFCSS control, or combinations thereof). Figure 27B shows an example circuit diagram of the configuration 2700 including a battery as a DC load / power source (which may be a PV array, ultracapacitor, etc.) along with a local controller and a central level controller. The configuration 2700 can provide V2G or V2X interface functionality as previously described.

[0208] The electronic controller is configured to facilitate the implementation of, for example, a power converter (e.g., by controlling switching devices of a non-isolated three-phase DC / AC power converter system). Thus, the storage device may include a computer program product that, when executed on the electronic controller (which may be a processor-based device, as previously described), causes the processor-based device to perform operations that facilitate the implementation of the procedures and operations described herein. The electronic controller may further include peripheral devices for enabling input / output functions. Such peripheral devices may include, for example, a flash drive (e.g., a removable flash drive) or a network connection (e.g., implemented using a USB port and / or a wireless transceiver) for downloading relevant content to a connected system. Such peripheral devices may also be used to download software including computer instructions for enabling the general operation of the respective system / device. Alternatively and / or additionally, in some embodiments, dedicated logic circuits such as FPGAs (field programmable gate arrays), ASICs (application specific integrated circuits), DSP processors, graphics processing units (GPUs), application processing units (APUs), etc. may be used in the implementation of the electronic controller. Other modules that may be included in the electronic controller may include a user interface for providing or receiving input and output data. The electronic controller may include an operating system.

[0209] A computer program (also referred to as a program, software, software application or code) includes machine instructions for a programmable processor and may be implemented in a high-level procedural and / or object-oriented programming language and / or in an assembly / machine language. As used herein, the term "machine-readable medium" refers to any non-transitory computer program product, apparatus and / or device used to provide machine instructions and / or data to a programmable processor (e.g., magnetic disks, optical disks, memory, programmable logic devices (PLDs)), including non-transitory machine-readable media that receive machine instructions as machine-readable signals.

[0210] In some embodiments, any suitable computer-readable medium can be used to store instructions for performing the processes / operations / procedures described herein. For example, in some embodiments, a computer-readable medium can be transitory or non-transient. For example, a non-transient computer-readable medium can include media such as magnetic media (e.g., hard disks, floppy disks, etc.), optical media (e.g., compact disks, digital video disks, Blu-ray disks, etc.), semiconductor media (e.g., flash memory, electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc.), any suitable media that is not transitory or lacks any representation of permanence in transmission, and / or any suitable tangible media. As another example, a transitory computer-readable medium can include signals, wires, conductors, optical fibers, circuits on a network, or any suitable media that is transitory and does not lack any representation of permanence in transmission, and / or any suitable intangible media.

[0211] Although certain embodiments have been disclosed in detail herein, this is done by way of example for illustrative purposes only and is not intended to be limiting with respect to the scope of the following appended claims. Features of the disclosed embodiments can be combined, rearranged, etc., within the scope of the invention to produce many more embodiments. Certain other aspects, advantages, and modifications are believed to be within the scope of the claims provided below. The presented claims represent at least some of the embodiments and features disclosed herein. Other unclaimed embodiments and features are also contemplated.

[0212] Further examples Example 1: A non-transitory computer-readable medium storing methods, apparatus, and / or processor-executable instructions for a non-isolated power converter system comprising: an N-phase power converter stage having an alternating current (AC) side and a direct current (DC) side, where N>1; an N-phase LC filter comprising one or more capacitors, where one or more neutral points of each of the one or more capacitors are electrically connected to a DC negative terminal of a DC power source; and a control system configured to drive power switching elements of the N-phase power converter stage to convert received power and to output the converted power, the control system configured to drive the power switching elements using variable frequency soft switching at a frequency of at least 20 kHz.

[0213] Example 2: The method, apparatus, and / or non-transitory computer-readable medium of Example 1, wherein the control system is a cascade control system comprising: a central controller including a processing unit, configured to determine a rotating reference coordinate system target and generate N control reference targets; and at least one local controller, each of the at least one local controller including a local processing unit, each of the at least one local controller configured to receive one of the N control reference targets and drive a portion of the power switching elements associated with the local controller in accordance with the control reference target.

[0214] Example 3: The method, apparatus, and / or non-transitory computer-readable medium of Example 1 or 2, wherein each of the at least one local controller is configured to implement model predictive control (MPC) to generate control signaling for the portion of the power switching elements to drive the portion of the power switching elements in accordance with a control reference target.

[0215] Example 4: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 3, wherein the central controller is further configured to receive at least one electrical operating characteristic from each of the at least one local controller, the electrical operating characteristic being in a stationary reference frame, transform the at least one electrical operating characteristic to a rotating reference frame, and determine direct axis (D-axis) and quadrature axis (Q-axis) components of the rotating reference frame target based on the at least one electrical operating characteristic in the rotating reference frame.

[0216] Example 5: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 4, wherein the central controller is further configured to determine a zero sequence component target of the rotating reference frame target based on a DC offset of half the DC voltage across the positive terminal of the DC power supply and the negative terminal of the DC power supply.

[0217] Example 6: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 5, wherein the central controller is further configured to determine a zero sequence component target of the rotating reference frame target based on a DC offset and a multiple of the N-phase harmonic injection.

[0218] Example 7: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 4 to 6, wherein the central controller is further configured to transform the D-axis voltage component, the Q-axis voltage component, and the zero sequence component targets to the stationary reference frame to generate N control reference targets in the stationary reference frame based on the rotating reference frame targets.

[0219] Example 8: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 7, wherein the power switching element includes a high-side element and a low-side element connected to a midpoint node for each of the N phases of the power converter stage, and the midpoint node of each of the N phases of the power converter stage is coupled to a respective LC filter of the N-phase LC filter, the LC filter including: (i) an inductor coupled between the midpoint node and a filter node of the respective LC filter; and (ii) one capacitor of the one or more capacitors of the N-phase LC filter coupled between the filter node of the respective LC filter and a negative DC terminal.

[0220] Example 9: The method, apparatus, and / or non-transitory computer-readable medium of Example 8, wherein each respective LC filter further includes a second capacitor coupled between a filter node of the respective LC filter and a positive DC terminal of the DC power source.

[0221] Example 10: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 8 or 9, further comprising an N-phase common mode inductor coupled between the filter node and the N interface terminals.

[0222] Example 11: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 10, further comprising an N-phase motor coupled to the N interface terminals.

[0223] Example 12: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 11, wherein the N interface terminals include N motor connection points for coupling to an N-phase motor and N grid connection points for coupling to an N-phase power grid.

[0224] Example 13: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 12, further including a traction mode and a charging mode, where when in the traction mode, the power converter is configured to convert DC power from the DC power source to AC power on the N motor nodes to drive the N-phase motor, and when in the charging mode, the power converter is configured to convert AC power from the N grid nodes to DC power to charge the DC power source.

[0225] Example 14: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 13, comprising a sensor configured to sense a first electrical characteristic of a first component of the N-phase LC filter selected from the group of switch-side inductors and capacitors and generate sensor data indicative of the first electrical characteristic, wherein the control system is further configured to receive the sensor data from the sensor and perform state estimation based on the sensor data to estimate a second electrical characteristic of a second component of the N-phase LC filter, different from the first component, and drive a power switching element based on the second electrical characteristic.

[0226] Example 15: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1 to 14, wherein to drive the power switching elements using variable frequency soft switching, the control system is configured to determine a switching frequency for driving the power switching elements of the converter based on electrical characteristics of the N-phase LC filter.

[0227] Example 16: A power converter module further comprising: N power converter modules, N>1, each power converter module including a positive direct current (DC) terminal and a negative DC terminal of a DC side of an N-phase power converter stage; a power switching element pair including a high-side power switching element coupled to the positive DC terminal and a low-side power switching element coupled to the negative DC terminal, the high-side power switching element and the low-side power switching element being coupled to each other at a midpoint node; and an N-phase LC filter including a capacitor of the one or more capacitors and an inductor, the inductor being coupled between the midpoint node and the capacitor, the capacitor being coupled to the inductor. 16. The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 1-15, wherein the method, apparatus, and / or non-transitory computer-readable medium includes: an LC filter coupled between a positive DC terminal of the N power converter modules and a negative DC terminal of the one or more power converter modules; one local controller of the at least one local controller configured to drive a power switching element pair, where the power switching element pair is part of a power switching element associated with the local controller; a circuit board on which the positive and negative DC terminals, the power switching element pair, the LC filter, and the local controller are located;

[0228] Example 17: A method, apparatus, and / or non-transitory computer-readable medium for a non-isolated power converter system for an electric vehicle comprising: an N-phase power converter stage having an alternating current (AC) side and a direct current (DC) side, where N>1, the DC side including a DC power supply terminal; an N-phase LCL filter comprising one or more capacitors, where one or more neutral points of each of the one or more capacitors are electrically connected to a negative DC terminal of the DC power supply terminals; and a control system configured to drive power switching elements of the N-phase power converter stage to, in a charging mode, convert input AC power received via the AC terminals to output DC power provided to the DC power supply terminals to charge the DC power source, and to, in a traction mode, convert input DC power received via the DC power supply terminals to output AC power provided to the AC terminals to drive the motor.

[0229] Example 18: The method, apparatus, and / or non-transitory computer-readable medium of Example 17, further comprising a contactor circuit including a plurality of contactors configured to selectively connect the AC terminals to either a motor connection point or an AC grid connection point.

[0230] Example 19: The method, apparatus, and / or non-transitory computer-readable medium of Examples 17 or 18, wherein the AC terminals are connected to both the motor connection point and the AC grid connection point during traction mode and during charging mode.

[0231] Example 20: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17-19, further comprising an N-phase common mode inductor coupled between the N-phase LC filter and the AC terminal.

[0232] Example 21: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 20, further comprising a motor bearing for the motor and a motor shaft driven by the motor.

[0233] Example 22: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 21, wherein the control system is a cascade control system comprising: a central controller including a processing unit, configured to determine a rotating reference coordinate system target and generate N control reference targets; and at least one local controller, each of the at least one local controller including a local processing unit, each of the at least one local controller configured to receive one of the N control reference targets and drive a portion of a power switching element associated with the local controller in accordance with the control reference target.

[0234] Example 23: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 22, wherein each of the at least one local controller is configured to implement model predictive control (MPC) to generate control signaling for the portion of the power switching elements to drive the portion of the power switching elements in accordance with a control reference target.

[0235] Example 24: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 23, wherein the central controller is further configured to receive at least one electrical operating characteristic from each of the at least one local controller, the electrical operating characteristic being in a stationary reference coordinate system, transform the at least one electrical operating characteristic to a rotating reference coordinate system, and determine direct axis (D-axis) and quadrature axis (Q-axis) components of the rotating reference coordinate system target based on the at least one electrical operating characteristic in the rotating reference coordinate system.

[0236] Example 25: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 24, wherein the central controller is further configured to determine a zero sequence component target of the rotating reference frame target based on a DC offset of half the DC voltage across the positive terminal of the DC power supply and the negative terminal of the DC power supply.

[0237] Example 26: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 25, wherein the central controller is further configured to determine a zero sequence component target of the rotating reference frame target based on a DC offset and a multiple of the N-phase harmonic injection.

[0238] Example 27: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 26, wherein the central controller is further configured to transform the D-axis voltage component, the Q-axis voltage component, and the zero sequence component targets to the stationary reference frame to generate N control reference targets in the stationary reference frame based on the rotating reference frame targets.

[0239] Example 28: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 27, wherein the power switching element includes a high-side element and a low-side element connected to a midpoint node for each of the N phases of the power converter stage, and the midpoint node of each of the N phases of the power converter stage is coupled to a respective LC filter of the N-phase LC filter, the LC filter including: (i) an inductor coupled between the midpoint node and a filter node of the respective LC filter, and (ii) one capacitor of the one or more capacitors of the N-phase LC filter coupled between the filter node of the respective LC filter and a negative DC terminal.

[0240] Example 29: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 28, wherein each respective LC filter further includes a second capacitor coupled between a filter node of the respective LC filter and a positive DC terminal of the DC power source.

[0241] Example 30: The method, apparatus, and / or non-transitory computer-readable medium of any of Examples 17 to 29, comprising a sensor configured to sense a first electrical characteristic of a first component of the N-phase LC filter selected from the group of switch-side inductors and capacitors and generate sensor data indicative of the first electrical characteristic, wherein the control system is further configured to receive the sensor data from the sensor and perform state estimation based on the sensor data to estimate a second electrical characteristic of a second component of the N-phase LC filter, different from the first component, and drive a power switching element based on the second electrical characteristic.

Claims

1. An N-phase power converter stage having an alternating current (AC) side and a direct current (DC) side, where N ≧ 1, the N-phase power converter stage, and An N-phase LC filter comprising one or more capacitors, wherein one or more neutral points of each of the one or more capacitors are electrically connected to the DC negative terminal of the DC power supply, the N-phase LC filter, and A control system configured to drive the power switching elements of the N-phase power converter stage to convert the received power and output the converted power, the control system being configured to drive the power switching elements using variable frequency soft switching at a frequency of at least 20 kHz, the control system, and Comprising, The control system is a cascade control system, and the cascade control system A central controller including a processing unit, Determining a rotating reference coordinate system target, Generating N control reference targets, The central controller configured as such, and At least one local controller, each of the at least one local controller including a local processing unit, and each of the at least one local controller Receiving one of the N control reference targets, Driving a part of the power switching elements associated with the local controller according to the control reference target, The at least one local controller configured as such, and A non-insulated power converter system comprising.

2. In order to drive a part of the power switching elements according to the control reference target, each of the at least one local controller Performs model predictive control (MPC) to generate control signaling for a part of the power switching elements, The power converter system according to claim 1, configured as such.

3. The central controller Receives at least one electrical operating characteristic from each of the at least one local controller, the electrical operating characteristic being within a stationary reference coordinate system, Converts the at least one electrical operating characteristic to the rotating reference coordinate system, Based on the at least one electrical operating characteristic in the rotation reference coordinate system, determining a direct axis (D-axis) component and a quadrature axis (Q-axis) component of the rotation reference coordinate system target. The power converter system according to claim 1, further configured as described above.

4. The central controller is Determining a zero-sequence component target of the rotation reference coordinate system target based on a DC offset that is half of the DC voltage across the positive terminal and the negative terminal of the DC power supply. The power converter system according to claim 3, further configured as described above.

5. The central controller Determining a zero-sequence component target of the rotation reference coordinate system target based on the DC offset and a multiple of the N-phase harmonic injection. The power converter system according to claim 3, further configured as described above.

6. To generate the N control reference targets in the stationary reference coordinate system based on the rotation reference coordinate system target, the central controller Converting the D-axis voltage component, the Q-axis voltage component, and the zero-sequence component target to the stationary reference coordinate system. The power converter system according to claim 5, further configured as described above.

7. The power switching element includes a high-side element and a low-side element connected to an intermediate point node for each of the N phases of the power converter stage. The intermediate point nodes of each of the N phases of the power converter stage are coupled to respective LC filters of the N-phase LC filter, and the LC filter includes (i) an inductor coupled between the intermediate point node and a filter node of each of the LC filters, and (ii) one of the one or more capacitors of the N-phase LC filter coupled between the filter node of each of the LC filters and the negative DC terminal. The power converter system according to claim 1.

8. Each LC filter further includes a second capacitor coupled between the filter node of each of the LC filters and the positive DC terminal of the DC power supply. The power converter system according to claim 7.

9. The power converter system according to claim 7, further comprising an N-phase common mode inductor coupled between the filter node and the N interface terminals.

10. The power converter system according to claim 9, further comprising an N-phase motor coupled to the N interface terminals.

11. The power converter system according to claim 9, wherein the N interface terminals include N motor connection points for coupling to an N-phase motor and N grid connection points for coupling to an N-phase power grid.

12. Further including a traction mode and a charging mode, when in the traction mode, the power converter is configured to convert DC power from the DC power source into AC power on the N motor connection points to drive the N-phase motor, when in the charging mode, the power converter is configured to convert AC power from the N grid connection points into DC power to charge the DC power source. The power converter system according to claim 11.

13. Further comprising a sensor configured to detect a first electrical characteristic of a first component of the N-phase LC filter selected from a group of switch-side inductors and capacitors and generate sensor data indicative of the first electrical characteristic, the control system receives the sensor data from the sensor, performs a state estimation based on the sensor data to estimate a second electrical characteristic of a second component different from the first component of the N-phase LC filter, and drives the power switching element based on the second electrical characteristic. The power converter according to claim 1, further configured as such.

14. To drive the power switching element using variable frequency soft switching, the control system is configured to determine a switching frequency for driving the power switching element of the converter based on the electrical characteristics of the N-phase LC filter. The power converter system according to claim 1.

15. Further comprising N power converter modules, where N>1, and each power converter module the positive direct current (DC) terminal and the negative DC terminal on the DC side of the N-phase power converter stage, A pair of power switching elements including a high-side power switching element coupled to the positive DC terminal and a low-side power switching element coupled to the negative DC terminal, wherein the high-side power switching element and the low-side power switching element are coupled to each other at an intermediate point node, the pair of power switching elements; An LC filter of the N-phase LC filter including one of the one or more capacitors and an inductor, wherein the inductor is coupled between the intermediate point node and the capacitor, and the capacitor is coupled between the inductor and the negative DC terminal, the LC filter; One of the at least one local controller configured to drive the pair of power switching elements, wherein the pair of power switching elements is part of the power switching elements associated with the local controller, the local controller; A circuit board on which the positive and negative DC terminals, the pair of power switching elements, the LC filter, and the local controller are located; including; The positive DC terminals of each of the N power converter modules are coupled to each other, and the negative DC terminals of each of the one or more power converter modules are coupled to each other; The power converter system according to claim 1.

16. Receiving input power from an AC side or a DC side by an N-phase power converter stage, where N ≧ 1, the step; Filtering at the AC side of the N-phase power converter stage by an N-phase LC filter including one or more capacitors, wherein one or more neutral points of each of the one or more capacitors are electrically connected to the DC negative terminal of a DC power supply, the step; Driving the power switching elements of the N-phase power converter stage by a control system that is a cascade control system to convert the input power and output the converted power, wherein the control system is configured to drive the power switching elements using variable frequency soft switching at a frequency of at least 20 kHz, the step; Determining a rotating reference coordinate system target by a central controller of the cascade control system; The step of generating N control reference targets by the central controller; The step of receiving, by each of at least one local controller of the cascade control system, one of the N control reference targets; The step of driving, according to the control reference target, a part of the power switching elements associated with the local controller; A method for power conversion, including the above steps.

17. For each of the at least one local controller, the step of driving a part of the power switching elements according to the control reference target is: For each of the at least one local controller, the step of implementing model predictive control (MPC) to generate control signaling for a part of the power switching elements; The method according to claim 16, including the above steps.

18. The step of receiving, by the central controller, at least one electrical operating characteristic from each of the at least one local controller, wherein the electrical operating characteristic is in a stationary reference coordinate system; The step of converting, by the central controller, the at least one electrical operating characteristic to the rotating reference coordinate system; The step of determining, by the central controller, a direct axis (D-axis) component and a quadrature axis (Q-axis) component of the rotating reference coordinate system target based on the at least one electrical operating characteristic in the rotating reference coordinate system; The method according to claim 16, further including the above steps.

19. The step of determining, by the central controller, a zero-sequence component target of the rotating reference coordinate system target based on a DC offset that is half of the DC voltage across the positive terminal and the negative terminal of the DC power supply; The method according to claim 18, further including the above step.

20. The step of determining, by the central controller, a zero-sequence component target of the rotating reference coordinate system target based on the DC offset and a multiple of the Nth-phase harmonic injection; The method according to claim 18, further including the above step.

21. The step of generating, by the central controller, the N control reference targets in the stationary reference coordinate system based on the rotating reference coordinate system target is: Step of converting the D-axis voltage component, the Q-axis voltage component, and the zero-sequence component target into the stationary reference coordinate system The method according to claim 20, comprising the above step

22. The power switching element includes a high-side element and a low-side element connected to an intermediate point node for each of the N phases of the power converter stage The intermediate point node of each phase of the N phases of the power converter stage is coupled to each of the LC filters of the N-phase LC filter, and the LC filter includes (i) an inductor coupled between the intermediate point node and the filter node of each of the LC filters, and (ii) one of the one or more capacitors of the N-phase LC filter coupled between the filter node of each of the LC filters and the negative DC terminal. The method according to claim 16

23. The method according to claim 22, wherein each LC filter further includes a second capacitor coupled between the filter node of each LC filter and the positive DC terminal of the DC power supply

24. The method according to claim 22, further comprising a step of filtering by an N-phase common mode inductor coupled between the filter node and N interface terminals

25. The step of operating in the traction mode, comprising a step of converting the input power received from the DC power supply into the converted output power by the power converter stage wherein the input power is DC power and the converted output power is AC power The step of driving an N-phase motor with the converted output power, wherein the N-phase motor is coupled to the N interface terminals The method according to claim 24, further comprising the above steps

26. The step of operating in the charging mode, comprising a step of converting the input power into the converted output power by the power converter stage, wherein the input power is AC power and the converted output power is DC power The step of charging the DC power supply with the converted output power The method according to claim 25, further comprising the above steps

27. generating, by a sensor, sensor data indicative of a first electrical characteristic of a first component of the N-phase LC filter selected from a group of a switch-side inductor and capacitors; receiving, by the control system, the sensor data from the sensor; executing, by the control system, a state estimation based on the sensor data to estimate a second electrical characteristic of a second component different from the first component of the N-phase LC filter; driving, by the control system, the power switching element based on the second electrical characteristic; The method according to claim 16, further comprising. **Claim 28** The step of driving the power switching element using variable frequency soft switching comprises: determining, by the control system, a switching frequency for driving the power switching element of the converter based on electrical characteristics of the N-phase LC filter; The method according to claim 16, comprising.

Images