Power processing and time-varying voltage profile generation

Abstract

The time-varying voltage profile generation device may include a hierarchical power structure for controlling flow among a plurality of power sources. The plurality of power sources may be coupled into a plurality of stages, each stage including one or more power sources. A plurality of switches of the time-varying voltage profile generation device may selectively couple outputs from respective ones of the plurality of stages to generate a time-varying voltage profile.

Description

[0001] Priority

[0002] This application claims priority to U.S. Provisional Patent Application No. 63 / 438,135, filed January 10, 2023, entitled “POWER PROCESSING AND TIME-VARYING VOLTAGE PROFILE GENERATION,” which is incorporated by reference herein in its entirety. TECHNICAL FIELD

[0003] The present disclosure relates generally to power processing and time-varying voltage profile generation. BACKGROUND

[0004] In recent years, green power generation and energy storage technologies have seen widespread adoption. For example, in the United States alone, several gigawatts of solar cell units were installed last year. As another example, in 2020, the United States advanced power storage installation exceeded the gigawatt threshold. Projections and current incentive systems indicate that this installation growth trend will continue in the coming years. As such, there is a growing need for systems that efficiently and economically connect green technology power nodes (e.g., power sinks and / or power sources) to the grid and efficiently and economically adapt their power output to accommodate a variety of other applications. Improvements in power adaptation technology will continue to drive industry demand. BRIEF DESCRIPTION OF DRAWINGS

[0005] Figure 1 An example power conversion device is shown.

[0006] Figure 2 An example shunt power conversion device is shown.

[0007] Figure 3 An example time-varying voltage profile generation device is shown.

[0008] Figure 4 An example voltage profile generation logic is shown.

[0009] Figure 5 An example single-switch-based time-varying voltage profile generation device is shown.

[0010] Figure 6 An example single-switch-based time-varying voltage profile generation device with an expander circuit is shown.

[0011] Figure 7 An example half-bridge-based time-varying voltage profile generation device is shown.

[0012] Figure 8An example half-bridge based time-varying voltage profile generation device with a spreader circuit is shown.

[0013] Figure 9 An example full-bridge based time-varying voltage profile generation device is shown.

[0014] Figure 10 An example parallel hierarchical structure based time-varying voltage profile generation device is shown.

[0015] Figure 11A An example single-switch based time-varying voltage profile generation device including a clamp circuit is shown.

[0016] Figure 11B An example clamp circuit is shown.

[0017] Figure 12 An example single-switch based time-varying voltage profile generation device with a faulted power converter is shown.

[0018] Figure 13A An example full-bridge time-varying voltage profile generation device including an isolator is shown.

[0019] Figure 13B An example isolator is shown.

[0020] Figure 14 An example time-varying voltage profile generation device with back-to-back insulated gate bipolar transistors is shown. DETAILED DESCRIPTION

[0021] In various cases, a power source, such as a power storage device (e.g., a battery, a fuel cell unit, or other power storage device), a solar cell unit, a wind turbine, a chemical process, or other power source, can output power at a state (e.g., voltage, wattage, current, direct current, alternating current, or other characterization metric) that does not match a target output of a system containing the power source. In various cases, there can be a mismatch between multiple power sinks (e.g., battery chargers, motors, or other power-consuming devices) connected in a unified system. In other words, a system can be heterogeneous due to the presence of various power nodes (e.g., power sources and / or power sinks) in the system.

[0022] Using batteries as an illustrative example, batteries that are uniform or otherwise not diversified (e.g., at a manufacturing, installation, or other lifecycle point) can degrade at different rates in some cases, including in cases of uniform and / or load-balanced use. Thus, a set of initially uniform batteries can degrade such that the output of the example set differs from a target output of the system. Additionally, the deviation of individual batteries in the example set from the target (or expected contribution to the target) output can vary from battery to battery. Different battery technology levels can exhibit different degradation, e.g., different battery packs can degrade differently, and degradation can vary within those battery packs, modules, and / or individual battery cells. Batteries can refer to any portion of a battery technology and / or other technology that functions as a power storage unit. For example, multiple battery packs, modules, battery cells, chargers, controllers, power converters, or other battery internal components connected via almost any set of electrical interconnections can in some cases be referred to as a single “battery.” Additionally, power storage devices such as batteries can function as power sources, power sinks (e.g., when charging), or other power nodes in various cases. Solar cell units / arrays can generate power differently due to transient and / or spatially varying irradiance profiles, cell degradation, cell interference (e.g., due to dust or other debris), or other non-uniform power generation disturbances.

[0023] As an illustrative scenario, the secondary use of retired electric vehicle (EV) battery packs (e.g., as residential backup power or other backup power) can require the installation of battery packs that have experienced degradation from use. Additionally, the capacity, power rating, and form factor of battery packs range widely for various types of vehicles. This diversity can increase with the advent of faster charging technologies and new battery chemistries. This diversity is reflected not only in secondary use battery packs for energy storage, but also in charging different vehicles within a station. But, in these rapidly changing times, the market can resist some standardization to the extent that producers who are able to employ new technologies will benefit from improvements in battery performance when technological advances outpace the benefits of standardization.

[0024] Similar tradeoffs exist between standardization and new technologies and the integration of other power nodes.

[0025] In various implementations, a system can implement a power converter for converting power of a power node to a state for use at an output port. In various implementations, full power processing (FPP) can include placing a power converter between a power node and a target port to convert power at the power node to power of the target port. In some cases, a converter can be paired with each node in a group bound to the target port. The converter can process all power from that node.

[0026] In some cases, partial power processing (PPP) can be implemented. While the number of converters can depend on (e.g., equal or close to) the number of power nodes, a PPP converter can process less than the full power at the nodes. Instead, processing can be focused on a portion of the power to adjust the power from the power nodes to the output state. In some cases, PPP can reduce the overall power processing. In some cases, PPP operation can improve efficiency relative to FPP, as PPP (i.e., using otherwise identical converters) does not process the full power of the system. As such, the inefficiency of each converter is reduced by the relative size of the portion being processed. For example, a FPP system with 5% loss processing 100% of the power would lose 5% of the system power. A PPP configuration using the same converters processing 10% of the power would lose 0.5%. Other efficiencies can also be obtained, such as reduced internal heating.

[0027] For example, differential power processing (DPP) can operate on a portion of the power that is different from the target state. In some cases, the power nodes can only differ by a given range (e.g., X% to Y%, where Y>X). As such, a set of power converters each capable of individually handling the maximum deviation of that range (e.g., Y%) can be sufficient to support power conversion. In some cases, the cost of power converters can scale with the processing capability of the converter. As such, a system configured to employ PPP and / or DPP can have a cost savings advantage over a FPP system. However, some FPP systems can operate without information about the current / future operating conditions of the power nodes. For example, DPP and PPP can have a range of operating tolerances that can provide a particular output. If the set of power nodes falls outside of that range (or, for example, decays to a degree outside of the range after installation), the PPP system can fail.

[0028] In some cases, statistical, empirical, and / or theoretical models can provide information about the power node conditions. For example, a model of battery decay as a function of use and / or time can provide a distribution of states for a given population of secondary use batteries. As such, such a model can provide predictive information for a set of batteries drawn from that population.

[0029] For example, a particular group (or other set) of power nodes can exhibit diversity for one or more reasons, such as attenuation, model type, or other diversity factors. Diversity models (including models generated from power node characterizations, statistical models, or other models of power node performance) can be used to provide information about expected characteristics of power nodes selected from the particular group. Additionally, using diversity models can allow the group to be divided into defined portions. The defined portions can be statistical portions, such as percentile ranges, individual node assignments, assignments based on characterizations, or other groupings. Once divided into portions, the portions can be treated specifically, such that electrical coupling to members of the portion can be specific to characteristics of the power node portion. Thus, a system using diverse power nodes can predict size requirements for power converters. As such, if the amount of conversion capacity needed is reduced, then because of the uncertainty, a power converter with lower conversion capacity can be used.

[0030] Thus, a system capable of handling a collection of power nodes with conditions estimated by models can allow relatively robust performance to employ blind measurement and / or limited characterization implementations, while not requiring detailed characterization of individual power nodes in the collection. Additionally, a system capable of model reference correction can allow power processing systems to be constructed more uniformly, rather than relying on interconnections and power converter units that are highly specific to a collection of power nodes.

[0031] In various implementations, a sparse collection of power converters (e.g., a group, a tier (with a hierarchical relationship to another collection of power converters), multiple tiers within the collection itself, or other configurations) can be selected to correct reference model estimates of power node variations of a collection of power nodes. The sparse collection can include a number of power converters that depends on power node differences as estimated by the models. Thus, in some cases, the number of power converters in the sparse collection can be less than the number of power nodes served by the power converters. As an illustrative example scenario, an example model can estimate that a collection of nine power nodes (selected from a group of power nodes governed by the model) can be (on average) interconnected to three power converters in order to adjust among the power nodes. In this illustrative example scenario, the three power converters can rebalance output / input from the individual power nodes to ensure specific power. In some cases, the three power converters can handle a range of inputs to allow for uncertainty associated with selecting a limited number of power converters from the group. The distribution of the limited number of power converters selected from the group can not necessarily align overall with the distribution of the group.

[0032] In some cases, power nodes can be connected to the system and operate without individual characterization. The model can be a single node for evaluating the condition of the power node. These nodes can be connected together and assumed to operate within a certain tolerance of the model estimate. In some cases, the model can be a single node for evaluating the condition of the power node. These nodes can be connected together and assumed to operate within a certain tolerance of the model estimate.

[0033] In some cases, characterization such as voltage level output, power node new specification, and / or other information that can be measured without changing the power node (or cost comparable to the power handling system itself) can be performed. In some cases, the handling system can include characterization elements such as voltage testing capabilities. In some cases, the characterization can be used for initialization, dynamic configuration, and / or other configuration of the system. The characterization can be used to facilitate interconnection of power nodes that approximates the estimate (e.g., expected value) of power node difference provided by the diversity model.

[0034] In some cases, the correction of target power from the diversity model to the uniform model can be performed in stages. In various implementations, the sparse set can be implemented as one or more sparse layers, where power handling can proceed sequentially layer by layer. In some cases, power handling at a sparse converter can proceed after power conversion at one or more dense power converter sets and provide adjustment earlier in the series (in terms of current flow) than other power conversions that can be performed (e.g., for another power node connected later in the series). As such, in some cases, the layers can be defined by a local order (e.g., from dense to sparse) that does not necessarily coincide with current flow across the range of devices.

[0035] In some cases, between one or more sparse layers and the power nodes, the system can include a dense power converter set (which can include one or more dense layers). In some cases, the dense layers can be used to correct for uncertainty in deviation from a central value (or other target value) for individual power nodes from a particular portion of the population of power nodes. For example, a particular set of installed batteries (power nodes) can have a secondary use battery that is decaying below expectation for its particular portion of the population, and another battery that is decaying beyond expectation. Further, the state of all of these batteries can continue to change over time during this secondary use installation. The dense power converter set can adjust the power of the batteries to more closely match the central value that the model can predict. The sparse power converter set can then correct the model distribution to the target power of the uniform model correction. In some cases, the dense set can include a number of power converters proportional to (e.g., equal to, one less than, or otherwise dependent on) the total number of power nodes.

[0036] In various implementations, the deviation of individual batteries from model estimates can be less (on average) in magnitude than the correction from the model to the target power. As such, the processing capacity of power converters in the dense set can be less than the processing capacity of power converters in the sparse set. In some cases, the cost of power converters can scale with power processing capacity. Thus, in various implementations, a tiered system with a dense set of power converters and a sparse set of power converters can have more power converters than a PPP system (as discussed above). The number of power converters in the dense set can be similar to the total number of power converters in a PPP system. However, in some cases, the processing capacity of individual converters in the dense set of converters can be less than the processing capacity of individual power converters in a PPP system. For example, the capacity of individual power converters of a PPP system can be more similar to the power processing capacity of the sparse set of converters. As such, while having more power converters, the cost of a tiered system can still be lower than a PPP system of similar performance (which already has a lower cost than a FPP system of similar performance).

[0037] In various implementations in which a set of power nodes is subdivided into multiple portions using a diversity model, a sparse tier can be specifically constructed and used to generate a uniform model corrected target power by correcting from a central value (or other intermediate value generated by corrections from the dense tier). The dense tier of power converters can be specifically selected to correct for variations within defined portions of the population of power nodes. For example, a device can include multiple different power node connection ports that are coupled to the dense tier of power converters. Each connection port can be coupled to one or more dense tier power converters that are specifically selected to correct for expected variations within a defined portion of the population of power nodes. Further, the number of connection ports dedicated to each portion of the population (e.g., across one or more multi-port devices) can scale with the relative size of the portion within the population. For example, a portion that covers half of the population of power nodes can have half the total number of ports of a device that uses power nodes of the population that are configured to correct for differences within the population. In some cases, the defined portions can be selected to simplify such determinations. As an illustrative example, a set of power conversion devices can be designed to have 12 ports, each of which is configurable to support a particular portion of a population of power nodes. The population can then be divided into 12 different portions of at least approximately equal size. In some cases, the defined portions can overlap (or partially overlap). As such, a particular power node can be located within the definition of two or more different portions. The diversity model can provide an expected range of different power flows supported by each portion.

[0038] Reference is now made to Figure 1FIG. 1 shows an example power conversion device (PCD) 100. The example PCD 100 includes a plurality of power node connection ports 111-119. Each connection port can be configured to support power conversion of a defined portion of a power node group of power nodes.

[0039] The diversity model can provide different portions of characteristics. For example, the diversity model can provide a central value of the expected power flow (such as a mean, a median, a value selected to facilitate conversion in combination with other central values, or other values). For example, the diversity model can provide an expected range of the power flow of the defined portion. In some cases, the defined portion can be defined based on power flow values. However, other characteristics can be used. For example, power node age, power node operating voltage, power node internal resistance (e.g., battery resistance or other internal resistance), power storage device cycle count, power node current, or other characteristics. In some cases, the population can be statistically defined (e.g., based on a percentile of an expected distribution due to power node age, cycle count, or other factors). As such, membership of a particular power node within any particular portion of the population can not be fully discernible. Thus, in some cases, the ports can be configured for different portions, and then a power node can be coupled to a particular port based on a best guess and / or a best fit membership assignment. As an illustrative example, a particular PCD can have four ports tuned to different quartiles of the entire population of power nodes. Upon PCD commissioning, the power nodes can be partially characterized, e.g., an operating voltage of each power node can be measured. Then, based on the partial characterization, the power nodes can be assigned based on a ranking of the characterized values. For example, in a best fit port assignment scheme, the lowest measured operating voltage can be assumed to best fit the port of the lowest quartile, including in cases where the lowest measured operating voltage can imply membership of another quartile. In a best guess scheme, the measured characteristics can be used to estimate membership. For example, the lowest measured operating voltage can be assigned to the quartile for which the actual measured voltage value most strongly indicates, without regard to a ranking relationship with other power nodes characterized with that power node at the time of its installation.

[0040] The PCD 100 also includes a node interconnect 140 between the plurality of power node connection ports 111-119. The node interconnect 140 can be configured to couple the power node connection portions 111-119 in a parallel or series configuration. In some cases, one or more series port strings can be coupled in parallel with other individual ports. The PCD 100 also includes an interconnect 130 between the plurality of power node connection ports 111-119 and a sparse set of power converters 141, 142, 144. The sparse set is used to adjust power at different points to ensure a final unified model corrected power at port 150.

[0041] As discussed below, the interconnections can include dynamic switching to support reconfiguration of connections over time. The switching can allow for changing power converter-power source connections after initial setup, e.g., connections due to non-uniform decay among power sources. In some cases, dynamic reconfiguration can be applied in response to different usage conditions. For example, when power is flowing outward from the ports, the ports 111-119 can be switched so that they are coupled in series. This can correspond, for example, to discharging of a coupled battery during operation. However, when power is flowing inward to the ports, the ports 111-119 can be switched so that they are coupled in parallel. This can correspond, for example, to charging of a coupled battery.

[0042] The layer interconnections 130 can include a set of dense power converters 131-139 for providing a first level of adjustment to the power node connection ports 111-119 according to a center value provided by a model (e.g., to partially power process model deviation power). In some cases, such adjustment can include differencing and / or partial conversion to an intermediate value that is selected with reference to the center value from the diversity model, but is different from the referenced center value. For example, the intermediate value can include a value that corresponds to a sum of multiple center values together, a difference between two center values, or other target value with reference to a center value. In some cases, the intermediate value can be a center value from the diversity model. Model deviation power can include a portion of power that deviates from the center value provided by the diversity model. The set of dense power converters 131-139 can be connected into one or more layers (as discussed below, the one or more layers are sparse sets 141, 142, 144 within a hierarchy). The total number of layers in the power converter hierarchy can include the number of layers of the dense set power converters 131-139 plus the number of layers of the sparse power converter sets.

[0043] The layer interconnections 130 also include passive connections (e.g., parallel, series, capacitive, inductive, power conversion, and / or other interconnections) to assist with the adjustment. As such, the layer interconnections 130 do not necessarily connect the power node connection ports one-to-one with the dense layer power converters. For example, multiple series connection nodes can be used to estimate a desired operating voltage before connection to a power converter. As such, power from multiple node connection ports can be processed by a single converter. In some cases, a complex electrical system can be referred to, depicted, or reduced (via equivalent circuit) to a single node and / or a single node connection port for simplicity of analysis and / or demonstration. In various implementations, a connection port can be permanently wired to a particular power node. As such, the port can include a power interface for power flow out of and / or into the power node, regardless of whether the coupling nature of the interface is permanent or temporary.

[0044] In various implementations, the sparse set 141, 142, 144 can be fed by the interconnect 130 (and the dense power converter set). The sparse set can provide partial power processing to adjust power according to model referenced intermediate values (e.g., via adjustment approximations of the interconnect 130) to ensure uniform model corrected target power at the target port 150. In other words, the sparse power converter set performs partial power processing on power (e.g., using taps as individual points within the PCD) to obtain a power format used by the system powered by the power source.

[0045] Figure 2 An example parallel PCD 200 is shown. In the example parallel PCD 200, the power node connection ports 211-219 are coupled in parallel to the target port 250 and the individual sparse layer converters 251, 252, 254. The dense layer converters 231-238 can be coupled between the power node connection ports 211-219 and the sparse layer converters 251, 252, 254 using parallel and / or series connections.

[0046] In various example parallel partial power processing architectures, a virtual layer of power converters 241-249 can be used to allow analysis based on circuit duality. For purposes of circuit analysis, the virtual layer of power converters 241-249 can allow the power nodes to be treated as equivalent "current sources" rather than "voltage sources." Thus, a series circuit can be reformed into a parallel circuit by adding such power converters. However, rather than providing a physical converter to provide this "zero" order conversion, the contribution of this virtual layer of power converters 241-249 is factored into the operation of the dense layer converters 231-238. In some implementations, use of circuit duality can facilitate dual mode implementations. Thus, a PCD operating in series in one mode can be converted into a parallel circuit using the virtual layer of power converters 241-249 when operating in a second mode. Thus, adjustments to the operation of the dense layer converters 231-238 can be determined based on the virtual power conversion requirements when dynamically switching between series and parallel operation. Thus, in some cases, this virtualization can allow for simplified dual mode operation.

[0047] An example dynamically switched PCD can include metering circuitry that can perform characterization at the power node connection ports 111-119. For example, the metering circuitry can characterize voltage, power storage capacity (e.g., via charge-discharge cycle voltage patterns, power flow within a charge-discharge cycle, or other cycle measurements), internal resistance, power flow, current flow, cycle count, power node age, or other power node behaviors.

[0048] The switching circuitry can include processing hardware to determine when a switching condition occurs. The switching condition can include a predetermined condition for which a particular interconnection layout is assigned. For example, the switching condition can include one or more threshold values for one or more characterization values. In an illustrative example, the switching condition can include a PCD exceeding a particular charge cycle count, and / or an age relative to a reference point, such as initial installation. The switching condition can include a change in operating mode. For example, the switching condition can include a reversal of power flow from the power node connection ports 111-119 (or other indication of a change from a discharge mode to a charge mode). For example, the switching condition can include a determination that the power node connection ports 111-119 have transitioned from an initial non- diversity state to a diversity state (e.g., an initial uniform power node now exhibiting a state of different behavior, such as a decaying state).

[0049] When a switching condition is determined to have occurred, the switching circuitry can switch the interconnections 130, 140 to conform to an interconnection layout consistent with the determined switching condition.

[0050] For example, the switching circuitry 504 can recouple the power node connection ports. For example, in a device that interconnects multiple batteries, the aging batteries can decay at different rates. One or more of the power node connection ports can be coupled to power converters that are sized to handle more severe decay than other ones of the power node connection ports. As such, the switching circuitry can recouple to the ports to dedicate a particular sized converter to the battery that is measured by the metering circuitry to decay most severely. Similarly, the battery that decays least severely can be switched to a power converter that is specifically sized for lower decay. Initially, the non-diversity state of the batteries can allow any battery to be equally served by any power node connection port, despite their coupled power converters being different sizes.

[0051] Figure 3An example time-varying voltage profile generation device (TVPD) 300 is shown. The example time-varying voltage profile generation device 300 includes a plurality of power node stages 311-319. Each of the power node stages 311-319 can include one or more power nodes (e.g., such as a battery, a solar cell unit, a wind turbine, a charging battery with reverse current flow, and / or other power node types). The power node stages 311-319 can be coupled to a hierarchical power converter structure 340 that includes power converters grouped as one or more layers. The hierarchical power converter structure 340 can distribute power among power nodes within a stage and / or across multiple stages depending on the structure of the hierarchical power converter structure 340. In other words, each stage (or at least some stages) can have its own hierarchical power converter structure, and / or one or more hierarchical power converter structures can interconnect different stages. Multiple nested power converter structures can be used. The hierarchical power converter structure 340 can include, for example, any of the various hierarchical power conversion devices discussed above and / or other hierarchical power converter structures, including a single-layer power converter structure. A single-layer power converter can include the functionality of any individual layer in any of the PCDs discussed above (e.g., 100, 200). Additionally or alternatively, a single-layer power converter structure can combine the functionality of multiple stages. For example, a single-layer power converter structure can be dense in terms of converter quantity, but can also correct to and for a power converter diversity model. Additionally or alternatively, layers can be partitioned according to the power distribution functions performed. For example, one or more layers can operate to support DC-DC power conversion, e.g., to distribute power among power sources. For example, one or more layers can operate to distribute power among different switches, e.g., to support power distribution during switching functions. Distribution of power can occur within a signal power node stage (e.g., single stage), and / or across different power stages of the TVPD (e.g., across stages). Other power redistribution schemes can be used. After power flow redistribution, outputs associated with each of the stages 311-319 can be selectively output coupled through switches 320. The switches 320 can include various switching modules, such as single switches, half-bridge modules, full-bridge modules, or other switch types.

[0052] The power nodes can include different types of power nodes, such as batteries, solar cell units, electrochemical power stacks, wind turbines, fuel cell units, fuel generators, and / or other power node types.

[0053] In various implementations, the number of concurrently selectively coupled stages can be used to control the magnitude of the output voltage. Changing the number of concurrently coupled stages can be used to change the magnitude of the output voltage. The polarity of the voltage can be controlled using a switching module (e.g., for half-bridge and / or full-bridge configurations), an unfolder, and / or a polarity selection switch, which can be used to selectively couple a power node in a negative polarity coupling configuration or a power node in a positive polarity coupling configuration. Various types of switches can be used within the module, such as transistors, bipolar transistors, field effect transistors, mechanical switches, and / or other switch types.

[0054] The outputs of the switches 320 can be coupled to various output components 360, such as isolators, clamping circuits, unfolder circuits, and / or other output circuits. The output components 360, in turn, can be coupled to a load (e.g., such as a device, a grid, and / or other system).

[0055] Figure 4 An example voltage profile generation logic (VPGL) 400 is shown. The example VPGL 400 can control the operation of the TVPD 300 and can be implemented in circuitry. The VPGL 400 can obtain a target time-varying voltage profile (TVVP) for generation (402). For example, the TVVP can include a target output. For example, the target output can include an alternating current (AC) input of a grid. In some cases, the AC input can include a voltage profile having a particular phase, amplitude, and frequency for a particular periodic function, such as a sine wave. In some cases, the TVVP can be determined via a static switching protocol, a dynamic control switching protocol, a programmable input, a profile input, one or more regulatory guidelines / rules, or other profile source.

[0056] Once the target output of the TVVP is obtained, the VPGL 400 can cause the switches 320 to selectively couple the individual power node stages 311-319 with timing to generate the TVVP (404). When the TVVP is generated, the VPGL 400 can cause the hierarchical power converter structure 340 to redistribute power (406) (e.g., via DC-DC conversion among the power nodes and / or among the switches to support the TVVP generation).

[0057] Figure 5An example single-switch-based TVPD 500 is shown. A single switch-based TVPD 500 can use individual switches 502 for each switching level used by the TVPD 500 to generate voltage levels within a TVVP. Each switch 502 can be coupled to one or more power nodes 504. However, one or more layers of the PCD 506 can distribute power among the nodes 504 (e.g., via DC-DC power conversion) and / or the switches 502 (e.g., to obtain a switching level target output). In the example TVPD 500, the switches and power nodes are shown as coupled in series. However, other configurations can be used. For example, one or more of the power nodes can be coupled in parallel. Each of these nodes can include one or more series and / or parallel coupled power sources and / or power sinks. The PCD 506 can use parallel and / or series coupling to implement a layered structure.

[0058] In the example single-switch-based TVPD 500, during operation, one switch is turned on at a time to achieve each different output voltage level of the TVVP generation. The single-switch-based TVPD 500 includes a first array of power nodes 510 for a single polarity voltage output and a second array of power nodes 520 for the other polarity. Relying on a single switch can reduce conduction losses of the TVPD 500 during operation. The single switch can receive the full operating current and voltage of the device and can be rated to handle the full power output.

[0059] Figure 6 An example single-switch-based TVPD 600 is shown that includes a spreader circuit 650. The spreader circuit 650 can be paired with the single-switch-based TVPD 600 to selectively invert the polarity of the system output. In various implementations, the spreader circuit 650 can allow the single-switch-based TVPD 600 to produce two output polarities without relying on two arrays of power nodes to produce opposite polarities. Almost any full-bridge switch can be implemented as a spreader circuit. During operation, the single-switch-based TVPD 600 can use three switches to operate for each TVVP level. There can be one switch 502 turned on within the array of power nodes 610, while two switches 652 can be used within the spreader circuit. The single switch 502 in the array of power nodes 610 can receive the full operating current of the device and can be rated to handle the full current. The two switches within the spreader circuit can receive 50% (on average) of the total.

[0060] In various implementations, various inverting circuits can be used (e.g., instead of the spreader circuit 650) to invert the output polarity of the TVPD 600.

[0061] Figure 7An example half-bridge based TVPD 700 is shown. In this example half-bridge based TVPD 700, the half-bridge switching modules 702 of the half-bridge based TVPD 700 allow individual switching stages of the half-bridge based TVPD 700 to selectively contribute to the output of the TVPD 700. Thus, the half-bridge switching modules 702 within the polarity arrays 710, 720 can be turned on in any order and in any number. The voltage load of each half-bridge module is the load contributed by the corresponding stage of the TVPD 700. As such, the stress on the half-bridge switching modules 702 of the TVPD 700 can be relatively less than the stress on the single switch 502 of the TVPD 500 for a given operating output. In some cases, switching modules of the same speed can be less expensive for the TVPD 700 than for the TVPD 500, as lower voltage rated switches can be used.

[0062] Additionally or alternatively, as the half-bridge switching modules 702 can be activated in any order, some embodiments can distribute the duty cycle to different power stages such that each power stage spends a selected amount of time turning on / off the load, rather than a particular power stage in the array always turning on first and off last. This distribution can enable various schemes, such as equal average load distribution, load time targets for individual stages, maximum average time between switching operations (e.g., for a particular switch and / or all switches), or other switching optimization schemes.

[0063] In the example half-bridge based TVPD 700, similar to the single switch TVPD 500, two polarity power node arrays 710, 720 are used. Two polarity selector switches 712, 722 are used to selectively activate the corresponding arrays 710, 720. The half-bridge based modules 702 can be activated independently of the arrays 710, 720. As such, the two additional selector switches 712, 722 are used to select the active polarity array 710, 720. The active selector switches 712, 722 can receive the full output power of the TVPD 700 during operation, and can be rated accordingly.

[0064] In some cases, a TVVP can be inverted at a frequency that is lower than the frequency of the sampling rate of the TVVP. As an illustrative scenario, a sine wave can be sampled at 600Hz, but inverted at a 120Hz rate. Thus, the period of the signal can be 60Hz, while the sampling resolution of the entire signal is 600Hz. Regardless, the speed of the selector switches can be five times lower than the speed of the half-bridge switching modules 702. Other sampling rates and inversion rates can be used.

[0065] Figure 8An example half-bridge based TVPD 800 with a spreader circuit 850 is shown. Similar to the TVPD 600, the half-bridge based TVPD 800 can be implemented using a spreader circuit 850 (and / or another inverter circuit) to selectively invert the output of the TVPD 800 without relying on two polarity arrays. Thus, in the example half-bridge based TVPD 800, a single polarity array 810 can be used when implementing the spreader circuit 850.

[0066] Figure 9 An example full-bridge based TVPD 900 is shown. The full-bridge switching module 902 can be used to individually selectively activate the output of each switching stage of the selectively polarized example full-bridge based TVPD 900. Thus, the stages of the example full-bridge based TVPD 900 can be activated in any order and in any polarity.

[0067] Figure 10 Example parallel layered structure based TVPDs 1010, 1020, 1030 are shown. Parallel power nodes and PCD arrangements can be used for various implementations using parallel-series circuit duality. Such parallel / series layered arrangements 1012, 1022, 1032 can be used with single switch 1010, half-bridge 1020 and 1030 TVPDs. In various implementations, the nested layered structure 1002 can include nested TVPDs to create selectable voltage outputs for each stage of the TVPD 1010, 1020, 1030. For example, the stages of the bridge based TVPD 1020, 1030 can include nested layer single switch TVPDs that are capable of time-varying output at the stage level. In some cases, the nested layers can include single switch TVPDs 1010 and / or bridge based structures within the bridge based TVPD 1020, 1030. When a full-bridge structure is nested within another TVPD at the stage level, it can also change its own polarity output. Thus, a single switch TVPD 1010 or half-bridge TVPD can have selectively polarized outputs without the need for a spreader circuit, polarity selector switch and / or multiple polarity arrays.

[0068] Figure 11A An example single switch based TVPD 1100 including a clamping circuit 1170 is shown. The clamping circuit can be used to maintain a particular voltage output for a selected period of time. The clamping circuit 1170 can be used to eliminate switching deadband times and / or other transient waveforms that can occur due to switching actions or other TVPD operations. While the clamping circuit 1170 is shown in a single switch TVPD 1100 configuration, the clamping circuit can be used with half-bridge and full-bridge switching systems. Figure 11B Alternative example clamping circuit structures 1198, 1199 that can be implemented are shown.

[0069] Figure 12 An example single-switch-based TVPD 1200 with a fault power converter 1280 is shown. In this example single-switch-based TVPD 1200, the fault power converter 1280 can be used to compensate for a fault in one or more power stages 1203 of the TVPD 1200. A power node 1204 at a power stage 1203 can fail during operation. The failure can cause the power node 1204 to stop providing power while active. In response to the power stopping during its active state, a fault diode 1282 can bypass the power stage 1203 and the fault power converter 1280 can activate to compensate for the stop. Although not shown, the TVPD 1200 can include multiple fault diodes (e.g., for each power stage). Although not shown, multiple power converters and / or a hierarchical structure can be used as the fault power converter. For example, a nested TVPD with selectable time-varying outputs can be used to flexibly compensate for various faults of different degrees (e.g., multiple simultaneous power stage failures).

[0070] Figure 13A An example full-bridge TVPD 1300 including isolators is shown. During operation, a full-bridge switching module 1302 can implement a set of two switches in the module to "chop" the output of the switching stage (e.g., to turn the output on and off at a selected rate greater than the sampling frequency of the TVVP). The chopping of the voltage output creates a high frequency component within the signal, which can allow the power stage 1303 output to pass through an isolator 1390 for the power stage 1303. The isolator 1390 can prevent power (e.g., without the high frequency component) from flowing back to the power node, thereby preventing backflow system damage. Individual isolators 1390 can be tuned to different high frequency components to allow isolation between power stages. Thus, different types of power stages that can normally interfere with each other can be used together via isolation. Because of the set of switches, the full-bridge module is used for chopping, an expander circuit 1350 can be used. In some cases, a half-bridge switching module can be used for selective activation, chopping, and polarity selection. Figure 13B Various example isolator configurations 1392, 1394, 1396, 1398 are shown. In various implementations, a passive isolated rectifier 1392 can be used. In various implementations, an active isolated rectifier 1394 can be used. In various implementations, a passive isolated rectifier with a center-tapped winding 1396 can be used. In various implementations, an active isolated rectifier with a center-tapped winding 1398 can be used.

[0071] Figure 14An example TVPD 1400 is shown with back-to-back insulated gate bipolar transistors (IGBTs) 1460. The back-to-back IGBTs 1460 can be used as switching modules (single switch modules as shown) to selectively activate various power stages 1420 supported by the PCD 1410. The example TVPD includes a buck converter 1480 for fault tolerance, a clamping circuit 1470 for transient mitigation, and an unfolder circuit 1450 for polarity selection.

[0072] Various example implementations have been included for illustration. Other implementations are possible. Table 1 shows various examples.

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093] The present disclosure has been described with reference to specific examples which are intended to be illustrative only and not limiting of the present disclosure. Changes, additions and / or deletions can be made to these examples without departing from the spirit and scope of the present disclosure.

[0094] The above description is given for clearness of understanding only, and no unnecessary limitations should be understood therefrom, as modifications can be made by those skilled in the art necessary within the scope of the disclosure, without departing from the spirit of the disclosure.

Claims

1. A device comprising: Multiple power nodes; One or more layers of power converters configured to control power flow distribution among the plurality of power nodes to generate a respective target voltage output for each of the plurality of power nodes, the one or more layers of power converters comprising: a dense layer comprising a first number of power converters proportional to a total number of the plurality of power nodes and configured to perform less than full power processing for at least one of the power nodes; as well as A time-varying voltage profile generator including a plurality of switches is configured to selectively apply the respective target voltage outputs of the plurality of power nodes to generate a predetermined time-varying voltage profile.

2. The apparatus of claim 1, wherein the predetermined time-varying voltage profile comprises a stepwise generation of the target voltage profile via additive and / or subtractive application of individual target voltage outputs. 3 . The apparatus of claim 1 , wherein power flow is controlled using a cross-level power converter structure configured to distribute power flow from at least some of the plurality of power nodes. 4 . The apparatus of claim 3 , wherein the cross-stage power converter structure couples at least some of the plurality of power nodes in parallel and / or in series.

5. The apparatus of claim 1 , wherein the one or more layers of power converters include a DC conversion layer configured to perform DC-DC power conversion and / or a switching layer configured to perform switching-level power conversion to support time-varying voltage profile generation.

6. The apparatus of claim 1, further comprising an expander circuit for selectively inverting the time-varying voltage profile.

7. The apparatus of claim 1, wherein the plurality of switches comprises transistors, bipolar transistors, field effect transistors, and / or mechanical switches.

8. The apparatus of claim 1, wherein the plurality of switches comprises one or more back-to-back insulated gate bipolar transistors (IGBTs).

9. The device of claim 1, wherein the plurality of switches comprises half-bridge switching modules and / or full-bridge switching modules.

10. The apparatus of claim 1, wherein the time-varying voltage profile generator is configured to selectively switch respective ones of the plurality of power nodes in a defined sequence to generate a time-varying voltage profile.

11. The apparatus of claim 1 , wherein the time-varying voltage profile is characterized by an output application frequency and an isolation frequency, the isolation frequency being selected to pass the time-varying voltage profile through the isolator, the isolation frequency being greater than the output application frequency.

12. A device comprising: a plurality of power nodes grouped into a plurality of levels; One or more layers of power converters configured to control power flow distribution among the plurality of power nodes to control respective target voltage outputs of each of the plurality of stages, the one or more layers of power converters comprising: a dense layer comprising a first number of power converters proportional to a total number of the plurality of power nodes and configured to perform less than full power processing for at least one of the power nodes; as well as A time-varying voltage profile generator including a plurality of switches is configured to selectively apply the respective target voltage outputs to generate a predetermined time-varying voltage profile.

13. The apparatus of claim 12 , wherein power flow is controlled using a single-stage power converter structure configured to distribute power flow from power converters within a single one of the plurality of stages and / or a cross-stage power converter structure configured to distribute power flow from power nodes among different ones of the plurality of stages.

14. The apparatus of claim 13, wherein at least one of the power converter structures couples at least some power nodes in parallel.

15. The apparatus of claim 13, wherein at least one of the power converter structures couples at least some power nodes in series.

16. The apparatus of claim 13, wherein: The plurality of power nodes include power nodes of a plurality of power source types; and At least one of the multi-layer power converter structures distributes power among at least some power nodes having different power source types.

17. A device comprising: a plurality of power nodes grouped into a plurality of levels; For at least some of the respective levels in the plurality of levels: A multi-layer power converter configured to control power flow distribution among the plurality of power nodes to control respective target voltage outputs of the stages, the multi-layer power converter comprising: a dense layer comprising a first number of power converters proportional to a total number of the plurality of power nodes; and a sparse layer comprising at least one power converter having a power converter rating greater than a power converter rating of any power converter in the dense layer; and A time-varying voltage profile generator including a plurality of switches is configured to selectively apply the respective target voltage outputs of the plurality of stages to generate a predetermined time-varying voltage profile.

18. The apparatus of claim 17, wherein the multi-layer power converter includes at least one fault converter configured to maintain power flow during a fault event associated with one or more power nodes of the plurality of power nodes.

19. The apparatus of claim 18, wherein the fault event comprises a sudden cessation of power flow from one or more of the plurality of power sources.

20. The apparatus of claim 17, further comprising an isolator for use between the plurality of switches and a load receiving a predetermined time-varying voltage profile, wherein: The isolator includes an active rectifier, a passive rectifier and / or a center-tapped rectifier.

Images