Optimized asymmetric DC-DC converter

Abstract

An asymmetric two-phase SRDAB DC-DC power converter apparatus having two SRDAB power conversion modules connected in parallel between an input voltage and an output voltage. The first module is optimized for operation at a minimum converter gain by selecting a first transformer turn ratio such that the first module has a near unity gain at the minimum converter gain and will take a majority of the power at the minimum gain, and the second module is optimized for operation at a maximum converter gain by selecting a second turn ratio such that the second module has a near unity gain at the maximum converter gain and will take a majority of the power at the maximum gain. The first module is operated under voltage mode control and the second module is operated under current mode control. The power converter apparatus provides flatter full-load efficiency without additional components.

Description

[0001]OPTIMIZED ASYMMETRIC DC-DC CONVERTERTECHNICAL FIELD The aspects of the disclosed embodiments relate generally to power conversion apparatus andmore particularly to series resonant dual active bridge power converters.BACKGROUND Series-resonant dual-active-bridge (SRDAB) power converters are commonly used in applications where high efficiency and electrical isolation are required. A SRDAB converter provides high efficiency due to soft switching operation and lower root-mean-square current when voltage gain is close to unity. However, converter efficiency is significantly reduced in applications where the input and output voltages vary over a wide range. Typical examples of these applications include DC-DC power converters for network power supply applications where the input voltage may vary from 38 volts to 72 volts. Various converter solutions have been purposed to improve these power converter applications. An input parallel output series (IPOS) LLC resonant converter with partial power processor and output voltage regulation capability has been purposed. However, this solution may not be able to efficiently deliver full-load power over a wide range due to the two-stage approach. Further an additional buck converter is required to regulate the output voltage thereby increasing cost and reducing power density. Use of an input series output parallel (ISOP) LLC resonant converter with PWM based partial power processor for output regulation has also been purposed. Unfortunately, the purposed solution requires inclusion of an extra boost or four switch buck-boost converter which may negatively affect the overall efficiency. A quasi two-stage isolated buck-DAB converter has been purposed to improve efficiency over a wide input voltage range. However, efficiency of the overall system may not be good due to two stage partial power processing. An ISOP hybrid DAB-LLC converter with output voltage regulation capability provides highefficiency when the majority of power is processed by the resonant LLC converter and the DABconverter is used to regulate output voltage. However, efficiency of the DAB converter may not be good over a wide range of input voltage variations, and transient response may suffer due to open-loop operation of the LLC resonant converter.Thus, there is a need for improved SRDAB converter topologies that can achieve flatter full-load efficiency over wide range variation in input and output voltages without the use ofadditional components. Accordingly, it would be desirable to provide methods and apparatusthat addresses at least some of the problems described above.SUMMARY The aspects of the disclosed embodiments are directed to asymmetric two-phase series-resonant dual-active-bridge (SRDAB) DC-DC power converter apparatus adapted to provide a flatter full-load efficiency over wide range variations in input and output voltage without additionalcomponents and while maintaining a fast-transient response. The aspects of the disclosedembodiments achieve these benefits by employing two parallel connected and asymmetrically optimized SRDAB conversion modules operated by a novel control scheme. According to a first aspect, the above and further objectives and advantages are obtained by an apparatus that includes: a first power conversion module having a series resonant dual-active- bridge DC-DC power converter topology and a second power conversion module having a series resonant dual-active-bridge DC-DC power converter topology, where the first power conversion module and the second power conversion module are coupled in parallel between an input voltage and an output voltage. When the apparatus is operating at a pre-determined minimum voltage gain, the first power conversion module is configured to produce a nominal gain, and when the apparatus is operating at a pre-determined maximum voltage gain, the second power conversion module is configured to produce the nominal gain.In a possible implementation form, the nominal gain is between zero point nine five (0.95) andone point zero five (1.05). A gain close to one minimizes conduction losses in an SRDAB converter topology.In a possible implementation form, the apparatus further includes a controller configured toreceive the input voltage, the output voltage, and an output current of the second power conversion module, and produce a first set of switch control signals and a second set of switch control signals. The first set of switch control signals are configured to operate the first power conversion module under a voltage mode control and the second set of switch control signals are configured to operate the second power conversion module under a current mode control. The use of different control modes on each phase allows both the output voltage and current to be controlled and allows control of power sharing between the modules.In a possible implementation form, the series resonant dual-active-bridge DC-DC powerconverter topology includes a transformer having a primary winding magnetically coupled to asecondary winding, a series resonant circuit having an inductor, a capacitor, and the primarywinding connected in series. A primary full bridge switching circuit is coupled to the seriesresonant circuit and configured to receive the input voltage, and a secondary full bridgeswitching circuit is coupled to the secondary winding and configured to produce an outputvoltage.In a possible implementation form, the first power conversion module includes a firsttransformer turn ratio configured to produce the nominal gain when the converter is operating at the pre-determined minimum voltage gain. The second power conversion module includesa second transformer turn ratio configured to produce the nominal gain when the converter isoperating at the pre-determined maximum voltage gain. This asymmetric optimization of each phase provides a flatter full-load efficiency when power sharing among the phases is modified at the extreme gain operating points.In a possible implementation form, operating the first power conversion module under thevoltage mode control includes: determining a first phase shift parameters based on the inputvoltage and the output voltage, comparing a reference voltage with the output voltage to produce a voltage error, applying a first compensation to the voltage error to produce a firstswitching frequency, and generating a first set of switch control signals based on the firstswitching frequency and the first phase shift parameters. Operating the second powerconversion module under current mode control includes: determining a second phase shiftparameters based on the input voltage and the output voltage, comparing a reference current with the output current to produce a current error, applying a second compensation to the currenterror to produce a second switching frequency, and generating the second set of switch controlsignals based on the second switching frequency and the second phase shift parameters. This control scheme provides a reliable and easily designed control scheme for applying the voltage mode control and current mode control to each power conversion module. In a possible implementation form, the first compensation comprises a proportional plusintegral (PI) control algorithm, and the second compensation comprises a proportional plusintegral control algorithm. PI loop compensation provides a simple and reliable approach torealizing stable loop operation with acceptably fast transient response.According to a second aspect, the above and further objectives and advantages are obtained by a method for DC-DC power conversion. The method includes coupling a first power conversion module and a second power conversion module in parallel between an input voltage and an output voltage, configuring the first power conversion module to produce a nominal gain when the DC-DC power conversion is operating at a pre-determined minimum gain,configuring the second power conversion module to produce the nominal gain when the DC-DC power conversion is operating at a pre-determined maximum gain, operating the first powerconversion module under voltage mode control, and operating the second power conversionmodule under current mode control. In a possible implementation form, the nominal gain comprises a gain between zero point nine five (0.95) and one point zero five (1.05). A gain close to one minimizes conduction losses in an SRDAB converter topology.In a possible implementation form, the first power conversion module includes a firsttransformer turn ratio, and the first transformer turn ratio is configured to produce the nominalgain when the converter is operating at the pre-determined minimum voltage gain. The secondpower conversion module includes a second transformer turn ratio, and the second transformerturn ratio is configured to produce the nominal gain when the converter is operating at the pre-determined maximum voltage gain. Asymmetric optimization of the two phases allows for aflatter full-load efficiency, and optimization of each module via selection of the transformerturn ratio provides a simple and effective approach to optimization of each phase. In a possible implementation form, operating the first power conversion module under thevoltage mode control includes: determining a first phase shift parameters based on the inputvoltage and the output voltage, generating a voltage error by comparing a reference voltage with the output voltage, generating a first switching frequency by applying a first compensationto the voltage error, and generating the first set of switch control signals based on the firstswitching frequency and the first phase shift parameters. Use of this control scheme provides simple and reliable voltage mode control for the first conversion module. In a possible implementation form, operating the second power conversion module under thecurrent mode control includes: determining a second phase shift parameters based on the inputvoltage and the output voltage, comparing a reference current with an output current of thesecond power conversion module to produce a current error, applying a second compensationto the current error to produce a second switching frequency, and generating the second set ofswitch control signals based on the second switching frequency and the second phase shiftparameters. Use of this control scheme provides simple and reliable current mode control forthe second conversion module. These and other aspects, implementation forms, and advantages of the exemplary embodiments will become apparent from the embodiments described herein considered in conjunction with the accompanying drawings. It is to be understood, however, that the description and drawings are designed solely for purposes of illustration and not as a definition of the limits of the disclosed invention, for which reference should be made to the appended claims. Additionalaspects and advantages of the invention will be set forth in the description that follows, and inpart will be obvious from the description, or may be learned by practice of the invention. Moreover, the aspects and advantages of the invention may be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the appended claims.BRIEF DESCRIPTION OF THE DRAWINGSIn the following detailed portion of the present disclosure, the invention will be explained inmore detail with reference to the example embodiments shown in the drawings, in which likereferences indicate like elements and:Figure 1 illustrates a diagram of an exemplary asymmetric two-phase series-resonant dual- active-bridge DC-DC power converter apparatus incorporating aspects of the disclosed embodiments; Figure 2 illustrates a block diagram showing an exemplary control scheme incorporating aspects of the disclosed embodiments; Figure 3 illustrates an exemplary series resonant dual active bridge DC-DC power converter topology incorporating aspects of the disclosed embodiments; Figure 4 illustrates graphs showing exemplary operating waveforms of a SRDAB converter topology for converter gains less than one incorporating aspects of the disclosed embodiments; Figure 5 illustrates graphs showing exemplary operating waveforms of a SRDAB converter topology for converter gains greater than one incorporating aspects of the disclosed embodiments; Figure 6 illustrates graphs showing the per unit normalized rms current with respect to the voltage gain of an SRDAB power converter topology incorporating aspects of the disclosed embodiments; Figure 7 illustrates graphs showing the per unit normalized rms current with respect to the voltage gain of an optimized asymmetric DC-DC converter apparatus incorporating aspects of the disclosed embodiments; Figure 8 illustrates graphs comparing performance of an SRDAB power converter topology with performance of an asymmetric two-phase SRDAB DC-DC power converter apparatus incorporating aspects of the disclosed embodiments; and Figure 9 illustrates a flow chart of an exemplary method for operating an asymmetric two-phase SRDAB DC-DC power converter apparatus incorporating aspects of the disclosed embodiments.DETAILED DESCRIPTION OF THE DISCLOSED EMBODIMENTSFigure 1 illustrates a diagram of an exemplary asymmetric two-phase series-resonant dual-active-bridge (SRDAB) DC-DC power converter apparatus 100 incorporating aspects of the disclosed embodiments. The exemplary apparatus 100 of the disclosed embodiments is directedto an improved two phase SRDAB power conversion apparatus 100 having two SRDABconversion modules 102, 104, or phases, coupled in parallel between an input voltage and anoutput voltage. The first conversion module 102 is optimized for operation near a pre-determined minimum output voltage and the second conversion module 104 is optimized foroperation near a pre-determined maximum output voltage. Regulation of both the outputvoltage Vo and current Io2 produced by the second power conversion module 104 is achievedby a controller 128 configured to operate the first conversion module 102 under voltage modecontrol and operate the second conversion module 104 under current mode control. As shown in Figure 1, in one embodiment, the first power conversion module 102 has a seriesresonant dual-active-bridge DC-DC power converter topology, and the second powerconversion module 104 has a series resonant dual-active-bridge DC-DC power convertertopology. The first power conversion module 102 and the second power conversion module 104 are coupled in parallel between an input voltage (Vin) and an output voltage (Vo). When the apparatus 100 is operating at a pre-determined minimum voltage gain (Mmin), the firstpower conversion module 102 is configured to produce a nominal gain (M≈1). When theapparatus 100 is operating at a pre-determined maximum voltage gain (Mmax), the second power conversion module 104 is configured to produce the nominal gain (M≈1). The disclosed converter configuration and control scheme can efficiently deliver a flatter full-load efficiency over wide range variation of input and output voltages without the use of extra components. Since both power conversion modules 102, 104 are operated under close loop control, theconverter apparatus 100 is able to achieve a fast-transient response.Generally, for optimal and efficient performance, a DC-to-DC power conversion module shouldbe able to efficiently deliver full-load power over wide gain variations. Switching losses shouldbe kept low such as by maintaining zero-voltage-switching, and conduction losses should bekept low such as by maintaining a gain close to unity over wide range variation in input andoutput voltage. Ideally, these efficiencies should be achieved without employing additionalcomponents. In the exemplary apparatus 100 two phases are formed by coupling a first power conversionmodule 102 and a second power conversion module 104 in parallel between an input voltageVinand an output voltage Vo. As will be described in more detail below, each power conversionmodule 102, 104 is configured as an SRDAB converter topology.The first SRDAB conversion module 102 includes a series resonant tank Z1formed by a series connected resonant inductor Lr1, resonant capacitor Cr1, and primary winding 120 of atransformer T1. Alternatively, the series connected resonant tank components Z1, may beconnected in any order as desired. A primary full bridge switching network 106 is coupled to the series resonant tank Z1andconfigured to receive the input voltage Vin, and a secondary full bridge switching network 108is coupled to a secondary winding 122 of the transformer and is configured to produce theoutput voltage Vo. When desired, a link capacitor C1 may be included in the first powerconversion module and coupled in parallel with the output voltage Vo. The second SRDAB conversion module 104 includes a series resonant tank Z2 formed by a series connected resonant inductor Lr2, resonant capacitor Cr2, and primary winding 124 of a transformer T2. Alternatively, the series connected resonant tank components Z2, may be connected in any order as desired.A primary full bridge switching network 110 is coupled to the series resonant tank Z2 andconfigured to receive the input voltage Vin. A secondary full bridge switching network 112 is coupled to a secondary winding 126 of the transformer and is configured to produce the output voltage Vo. When desired, a link capacitor C2 may be included in the second power conversion module 104 and coupled in parallel with the output voltage Vo. When desired, a first link capacitor Cin may be coupled in parallel with the input voltage Vin. Similarly, an additional link capacitor Co may be coupled in parallel with the output voltage Vo.Including link capacitors may be desirable for example to remove higher frequency componentsand provide smoothing of the DC input voltage Vin and DC output voltage Vo.It should be understood that the terms input voltage, and output voltage, are used herein as an aid to understanding only and should not be construed as limiting operation of the converter apparatus 100. Those skilled in the art will readily recognize that the power conversion apparatus 100 is capable of bi-directional operation and may be operated to transfer power in either direction (input voltage Vi to output voltage Vo, or output voltage Vo to input voltage Vo)by varying the phase relation θ between the primary full bridge switching circuits 106, 110 andthe secondary full bridge switching circuits 108, 112.In the first power conversion module 102, four switching devices S1, S2, S3, S4 are connected in a full bridge configuration to form the primary full bridge switching circuit 106 and an additional four switching devices S5, S6, S7, S8are connected in a full bridge configuration toform the secondary full bridge switching circuit 108. Similarly, in the second conversionmodule 104, four switching devices Q1, Q2, Q3, Q4 connected in a full bridge configurationform the primary full bridge switching circuit 110 and an additional four switching devices Q5,Q6, Q7, Q8 connected in a full bridge configuration form the secondary full bridge switching circuit 112. The switching devices S1, S2, S3, S4, S5, S6, S7, S8, Q1, Q2, Q3, Q4, Q5, Q6, Q7, Q8 are illustrated as metal oxide semiconductor field effect transistors (MOSFET) having a parasitic capacitance Cossand substrate diode D. However, any suitable switching device, capable of switching the desired power at the desired frequencies, may be advantageously employed as the switching devices S1, S2, S3, S4, S5, S6, S7, S8without straying from the spirit and scope of the present disclosure. As used herein, a switch or switching device is referred to as ‘on’ or ‘turned on’ when it is conducting current, and is referred to as ‘off’ or ‘turned off’ when it is not conducting current.An appropriately efficient power flow is facilitated by the exemplary apparatus 100 whenoperated by a controller 128 configured to receive the input voltage Vin, the output voltage Vo, and an output current Io2 of the second power conversion module 104. The controller 128 is configured to produce a first switch control signals 130 and a second switch control signals 132 configure to operate the first power conversion module 102 and the second power conversion module 104 respectively. The first switch control signals 130 are configured to operate the first power conversion module 102 under a voltage mode control based on a reference voltage Vrefand the output voltage Vo, and the second switch control signals 132 are configured to operatethe second power conversion module 104 under a current mode control based on a referencecurrent Io2refand the output current Io2of the second power conversion module 104. The reference voltage Vrefand reference current Io2refdetermine the output power of the apparatus 100, and may, when desired, be used to manage power sharing between the two power conversion modules 102, 104. As used herein the term nominal gain refers to a near unity gain, or a gain that is close to one M1≈1. As used herein the term nominal gain refers to a near unity gain which is a gain that is close to one, M1≈1. In certain embodiments the nominal gain may be a gain between zero point nine five (0.95) and one point zero five (1.05).To obtain a flatter full-load efficiency over the wide range voltage gain variation, the firstconversion module 102 is optimized for operation around a minimum gain Mmin, and the second power conversion module 104 is optimized for operation around a maximum converter gainMmax. Optimization of the first converter module 102 may be achieved by properly choosingthe first transformer T1 turns ratio N1 such that the gain of the first power converter module 102produces a nominal gain at a pre-determined minimum converter gain Mmin. Optimization ofthe second converter module 104 is achieved by properly choosing the second transformer T2 turns ratio N2such that the gain of the second power converter module 104 produces thenominal gain at a pre-determined maximum converter gain Mmax. When operating around theminimum converter gain Mmin, a majority of the power will be processed by the first power conversion module 102, and when operating around the maximum converter gain Mmax, a majority of the power will be processed by the second power conversion module 104. Figure 2 illustrates a block diagram showing an exemplary control scheme 200 incorporating aspects of the disclosed embodiments. The exemplary control scheme 200 is appropriate foroperating an asymmetric SRDAB DC-DC power converter such as the exemplary apparatus100 described above and with respect to Figure 1. The exemplary control scheme 200 includes two control loops configured to produce a flatter full-load efficiency over wide range gainvariations. A voltage mode control loop 218 receives the output voltage Vo, the input voltageVin, and a reference voltage Vref, and generates 208 the first set of control signals 130, wherethe first set of control signals 130 is configured to operate the first power conversion module102. A current mode control loop 220 receives the output voltage Vo, the input voltage Vin, theoutput current Io2, of the second power conversion module 104, and a reference current Io2ref,and generates a second set of control signals 132 configured to operate the second powerconversion module 104.In operation, the voltage control loop 218 determines 202 a first set of phase shift parameters^1, αp1, αs1 based on the input voltage Vin and the output voltage Vo. Determination of theoptimal phase shift parameters ^, ^^, ^^ for soft switching operation and minimum rmscirculating current will be described in more detail below. A voltage error signal ev is produced by comparing 204 the reference voltage Vref, with the output voltage Vo. This comparison 204creates a feedback control loop where a reference voltage Vref, which represents the desiredoutput voltage, is subtracted from the actual output voltage Vo to produce a voltage error signalev.A first loop compensation, or control algorithm, is applied 206 to the voltage error signal ev toproduce a first switching frequency fs1. In the illustrated embodiment, the compensationincludes a proportional plus integral loop compensation algorithm (PI). Variation of the first switching frequency fs1changes the gain introduced by the series resonant tank Z1, thereby facilitating control of the output voltage. The first set of switch control signals 130 is generated 208 based on the first switching frequency fs1and the determined first set of phase shiftparameters ^1, αp1, αs1.The current control loop 220 operates similarly to the voltage control loop 218 described abovewhere a second set of phase shift parameters ^2, αp2, αs2 are derived 210 based on the inputvoltage Vin and the output voltage Vo. A current error signal ei is produced by comparing 212the reference current Io2ref with the output current Io2, and a second switching frequency fs2 isproduced by applying 214 a second compensation to the current error ei. Then the second set of switch control signals 132 is generated 216 based on the second switching frequency fs2 andthe second phase shift parameters ^2, αp2, αs2.Any suitable control algorithm or compensation may be advantageously applied 206, 214, such as a proportional plus integral loop compensation algorithm (PI), or other appropriate loop compensation algorithm as desired. As used herein the term series resonant dual-active-bridge DC-DC power converter topology, also referred to herein as a SRDAB DC-DC power converter topology, refers to a bi-directional power converter topology having a transformer and series resonant tank coupled between two switching networks where the switching networks are configured to convert between a DCvoltage and an AC voltage. As an aid to understanding an exemplary SRDAB DC-DC powerconverter topology is analysed below. Figure 3 illustrates an exemplary series resonant dual active bridge (SRDAB) DC-DC power converter topology 300 incorporating aspects of the disclosed embodiments. The SRDAB topology 300 is appropriate for use as either or both the power conversion modules 102, 104 employed in the exemplary apparatus 100 described above and with respect to Figure 1. Similar to the first and second converter modules 102104 incorporated in the apparatus 100described above, the SRDAB DC-DC power converter topology 300 includes a series resonanttank having a resonant capacitor Cr, a resonant inductor Lr, and a transformer Tr, coupled between a primary full bridge switching circuit 306 and a secondary full bridge switching circuit 308. The primary full bridge switching circuit 306 includes four switches S1, S2, S3, S4arrangedin an H-bridge configuration and is adapted to receive an input DC voltage Vin and produce aprimary AC voltage vp. The secondary full bridge switching circuit 308 also includes four switches S5, S6, S7, S8arranged in an H-bridge configuration and adapted to receive a secondary AC voltage vsand produce a DC output voltage Vo. As described above, due to the bi- directional power handling capability of the SRDAB DC-DC power converter topology 300,the designation of input and output is arbitrary and assigned herein only for illustrativepurposes. Figure 4 illustrates graphs 400 showing exemplary operating waveforms of the SRDAB DC- DC power converter topology 300 for converter gains less than one incorporating aspects of thedisclosed embodiments. In the graphs 400 time is depicted along a horizontal axis ωt increasingto the right and voltage is depicted along a vertical axis V increasing upwards. The time intervalfrom t0 to t4 depicts one full cycle of the AC signals, with the primary side internal phase shiftbeing labelled as αp and the external phase shift labelled as ^.It should be understood that in the following discussion of operating waveforms of theexemplary SRDAB DC-DC power converter topology, switches not listed as being on during a time interval are in their off state during that interval.During the first interval depicted in the graphs 400 between time t0 to t1, switches S1, S3, S6,and S7 are on, forcing the primary voltage vp to zero, and the secondary voltage vs to a negativeof the output voltage Vo. During the next interval between t1 to t2, switches S1, S4, S5, and S8are on, forcing the primary voltage vp to the input voltage Vi, and the secondary voltage vs tothe output voltage Vo. During the next interval between t2 to t3, switches S2, S4, S5, and S8 areon, forcing the primary voltage vp to zero, and leaving the secondary voltage vs at the outputvoltage Vo. During the final interval in the cycle between t3 to t4, switches S2, S3, S6, and S7 areon, forcing the primary voltage vp to zero, and leaving the secondary voltage vs at a negative ofthe output voltage Vo. After time t4the cycle repeats. Figure 5 illustrates graphs 500 showing exemplary operating waveforms of the SRDAB DC- DC power converter topology 300 for converter gains greater than one incorporating aspects of the disclosed embodiments. In the graphs 500 time is depicted along a horizontal axis ωtincreasing to the right and voltage is depicted along a vertical axis V increasing upwards. Thetime interval from t0 to t4 depicts one full cycle of the AC signals, with the secondary side internal phase shift being labelled as αs and the external phase shift labelled as ^. During the first interval in the graphs 500 between t0 to t1, switches S1, S4, S5, and S7 are on,forcing the primary voltage vp to the input voltage Vi, and the secondary voltage vs to zero.During the next interval between t1 to t2, switches S1, S4, S5, and S8 are on, leaving the primaryvoltage vpat the input voltage Vi, and forcing the secondary voltage vsto the output voltage Vo.During the next interval between t2 to t3, switches S2, S3, S6, and S8 are on, forcing the primaryvoltage vp to a negative of the input voltage Vi, and forcing the secondary voltage vs to zero.During the final interval in the cycle between t3 to t4, switches S2, S3, S6, and S7 are on, forcingthe primary voltage vp to a negative of the input voltage Vi, and the secondary voltage vs to anegative of the output voltage Vo. After time t4 the cycle repeats.A fuller appreciation of the benefits and advantages provided by the exemplary apparatus (100)may be obtained through circuit analysis of the SRDAB DC-DC power converter topology 300 incorporated in the two phases 102,104 of the apparatus 100. Fundamental circuit analysis ofthe SRDAB DC-DC power converter topology 300 yields the output power as shown inequation (1): where ^^and ^^are the internal primary side and secondary side phase shift respectively, andtheta ^ is the external phase shift which controls power flow through the converter. The gainM, also referred to as a voltage gain, is the effective gain across the impedance Z: ^ =Equation (2) provides the rms primary current ^^^^^, also referred to as the rms tankcurrent: = Based on equation (1), the input current ^^^ and output current ^^ may be derived as shown inequations (3) and (4): ^ ^^^ ^^The optimal phase shift parameters ^, ^^, ^^ for soft switching operation and minimum rmscirculating current may be derived as shown in equation (5): ^ The rms primary current ^^^^^,^ in terms of input current ^^^ may be derived from equations(1), (3), and (5) as sown in equation (6): ^ Similarly, the rms primary current ^^^^^,^in terms of output current ^^may be derived from equations (1), (4), and (5) as shown in equation (7): Normalizing equation (6) in with respect to the input current Iin and equation (7) with respect to the output current Io, yields normalized per unit values that may be plotted together on the same graph as given by equations (8) and (9): ^ ^^^^^,^,^.^=^^^^,^^^^ ,^^^^ ^^^,^^^^^,^,^.^= ^^ . Figure 6 illustrates graphs 600 showing the per unit normalized rms current iprms,p.u with respectto the gain M of an SRDAB power converter topology 300 incorporating aspects of thedisclosed embodiments. In the graphs 600 voltage gain is depicted along a horizontal axis Mincreasing to the right, while current is depicted along a vertical axis iprms,p.u increasing upwards.The graphs 600 plot equations (8) and (9) where the solid line 604 shows the normalized primary rms current with respect to the input current ^^^^^,^,^.^as given by equation (8), and the dotted line 602 shows the normalized primary rms current with respect to the output current ^^^^^,^,^.^as given by equation (9).It can be observed from the graphs 600, that for gains M close to unity, the rms current is at itsminimum. However, when the gain M deviates from unity, the rms current increasessignificantly. As shown in the graphs 600, when the converter is operating at its minimum voltage gain Mmin or its maximum voltage gain Mmax the rms current is approximately 1.6 timesthe input or output current. This increase in rms current will increase conduction losses,resulting in reduced converter efficiency. This reduced efficiency typically occurs at theextreme operating points where the gain is at its minimum value Mmin or its maximum valueMmax. Increased power dissipation at the extreme operating points can complicate thermaldesign of the converter. In conventional multiphase SRDAB converters, all phases are optimized for a single operatingpoint (typically around unity gain M≈1), and produce their best efficiency when operating nearthis operating point. However, as the above analysis of the SRDAB DC-DC power convertertopology 300 shows, when the voltage gain deviates from unity, the rms current increasessignificantly, resulting in increased conduction and switching losses and reduction in the converter efficiency.In contrast, the exemplary apparatus 100 includes two modules or phases 102, 104 where a firstpower conversion module 102 is optimized for a pre-determined minimum gain Mmin and will take the majority of power when the apparatus 100 is operating at this minimum gain Mmin, and a second power conversion module 104 is optimized for a pre-determined maximum gain Mmaxand will take the majority of power when the apparatus 100 is operating at this maximum gainMmax. Optimization may be achieved by selecting the transformer turn ratio N such that thevoltage gain of the power conversion module is near one M≈1 at the desired operating point.Figure 7 illustrates graphs 700 showing the per unit normalized rms current iprms,p.u with respectto the voltage gain M of the an exemplary optimized asymmetric DC-DC converter, such as theapparatus 100, incorporating aspects of the disclosed embodiments. In the graphs 700, voltagegain is depicted along a horizontal axis M increasing to the right, while current is depicted alonga vertical axis iprms,p.u increasing upwards. In the graphs 700, the normalized primary rmscurrent with respect to the input current ^^^^^,^,^.^is shown by the dashed line 704, and the normalized primary rms current with respect to the output current ^^^^^,^,^.^is shown by the dash-dotted line 702. In the example depicted in graphs 700 the ratio of power sharing between the two power conversion modules 102, 104 at each optimization point, Mminand Mmax, is set to seventypercent to thirty percent (70:30). This means that when operating at either of the optimizationpoints, Mmin and Mmax, seventy percent of the power will be delivered to the output voltage Vothrough one power conversion module and thirty percent of the power will be delivered throughthe other power conversion module. The optimal ratio of power sharing between the two power conversion modules 102, 104 is application dependent and a may be selected as a matter of design choice. The graphs 700 show a reduction in primary current at the minimum and maximum gain and an increased current near unity gain leading to a flatter full-load efficiency. To aid comparison the graphs 600 showing primary current in the SRDAB converter topology 300 are superimposed on the graphs 700 showing primary current in the exemplary apparatus 100.Figure 8 illustrates graphs 800 comparing performance of an SRDAB DC-DC power convertertopology with performance of an optimized asymmetric DC-DC power converter apparatusincorporating aspects of the disclosed embodiments. In the graphs 800, voltage gain is depictedalong a horizontal axis M increasing to the right, while current is depicted along a vertical axisiprms,p.u increasing upwards. To facilitate comparison Figure 8 shows the graphs 600 superimposed on the graphs 700. Graphs 602 and 604 depict the primary currents for the SRDAB DC-DC power converter topology 300 described above and with respect to Figure 6. Graphs 702 and 704 depict the primary currents for the exemplary apparatus 100 as described above and with respect to Figure 7. The graphs 800 show that the normalized primary rms current with respect to the input current ^^^^^,^,^.^is reduced by thirteen percent (13%) at the minimum voltage gain Mminand reduced by four percent (4%) at the maximum voltage gain Mmaxby the exemplary apparatus 100 as compared to the SRDAB DC-DC power converter topology 300, while the normalized primaryrms current with respect to the output current ^^^^^,^,^.^ is reduced by eighteen percent (18%)at maximum voltage gain Mmax. However, the rms primary currents are increased by thirteen percent (13%) at unity gain. The reduction in rms currents at the extreme points may, in many applications, result in reduced conduction losses and increased efficiency over a wide range of voltage gains. Due to the reduction in the rms current around the extreme operating points, i.e., around the minimum and maximum gains, the high current turn-off losses will be significantlyreduced under full load conditions. A flatter full-load efficiency also aids thermal design.Referring now to Figure 9 there can be seen a flow chart of an exemplary method 900 foroperating an asymmetric two-phase SRDAB DC-DC power converter apparatus incorporatingaspects of the disclosed embodiments. The exemplary method 900 may be advantageously employed to operate any appropriate asymmetric two-phase SRDAB DC-DC power converterapparatus such as the exemplary converter apparatus 100 described above and with respect toFigure 1. During operation, the exemplary method 900 provides improved efficiency at theextreme operating points and a flatter full-load efficiency over a wide gain range without addingcomponents to the SRDAB power conversion modules.The exemplary method 900 begins at design time by coupling (902) two power conversionmodules in parallel between an input voltage and an output voltage, where each power conversion module includes a SRDAB DC-DC power conversion topology, such as the exemplary SRDAB DC-DC power conversion topology 300 described above. Any appropriate SRDAB DC-DC power converter topology capable of providing electrical isolation, bi- directional power flow, and gain modulation may be advantageously employed as the SRDAB DC-DC power conversion modules without straying from the spirit and scope of the disclosed embodiments. To minimize circulating currents and improve efficiency, a SRDAB DC-DC power converter should be operated with an effective gain across the impedance network near unity. As discussed above, when a SRDAB DC-DC power converter is operated away from a unity gain,converter efficiency deteriorates resulting in increased conduction losses near the extremeoperating points. Maintaining a flatter full-load efficiency aids thermal design of the converter.Improved full-load efficiency may be obtained by configuring (904) the first power conversionmodule to have a nominal gain when the power conversion is operating at a pre-determinedminimum gain and configuring (906) the second power conversion module to have the nominalgain when the power conversion is operating at a pre-determined maximum gain. In oneembodiment the nominal may be a near unity gain with a value between zero point nine five(0.95) and one point zero five (1.05).In operation the first power conversion module is operated 908 in a voltage control mode andthe second power conversion module is operated 910 in a current control mode. In this way, the first power conversion module regulates the output voltage and the second power conversion module regulates output power. The control scheme also allows power sharing to be influenced by the two control loops and ensure the first power conversion module takes a majority of the power when operating near the minimum converter gain and the second power conversion module takes a majority of the power when operating near the maximum converter gain. As described above and with reference to Figure 2, operating 908 the first power conversion module under voltage mode control includes forming a feedback control loop around the first power conversion module configured to maintain the converter output voltage at a desired setpoint as indicated by a reference voltage. The voltage mode control loop 218 determines 202a first phase shift parameters ^1, αp1, αs1 based on the input voltage Vin and the output voltage Vo. The optimal phase shift parameters ^1, αp1, αs1 for soft switching operation and minimum rms circulating current may be derived based on the input voltage Vin and the output voltage Vo as described in equation (5) above.A voltage error (ev) is then generated 204 by comparing, or differencing, a reference voltageVoref with the output voltage Vo, and a first switching frequency fs1 is generated by applying 206 a first compensation to the voltage error ev. In one embodiment the compensation is provided by a proportional plus integral (PI) control algorithm. Alternatively, any suitable controlalgorithm that is adapted to provide stable operation and acceptable transient performance maybe advantageously employed as the applied 206 loop compensation.A first set of switch control signals 130 is generated 208 based on the first switching frequencyfs1 and the first phase shift parameters ^1, αp1, αs1 where the first set of switch control signalsare configured to operate the first power conversion module under a voltage mode control (218). Operating 910 the second power conversion module under current mode control includesforming a feedback control loop around the second power conversion module configured tomaintain output current of the second power conversion module at a desired set point asindicated by a reference value Io2ref. The current mode control loop 220 determines 210 asecond phase shift parameters ^2, αp2, αs2 based on the input voltage Vin and the output voltage Vo. The optimal phase shift parameters ^2, αp2, αs2 for soft switching operation and minimum rms circulating current may be derived based on the input voltage Vinand the output voltage Voas described in equation (5) above.A current error (ei) is then generated 212 by comparing, or differencing, a reference currentIo2ref with an output current Io2 of the second power conversion module to produce a currenterror (ei). A second switching frequency fs2 is generated by applying 214 a secondcompensation to the current error ei. In one embodiment a proportional plus integral (PI) controlalgorithm is used as the loop compensation. Alternatively, any suitable control algorithmadapted to provide stable loop operation and acceptable transient performance may beadvantageously employed as the applied 214 loop compensation.A second set of switch control signals 132 is generated 216 based on the second switchingfrequency fs2 and the second phase shift parameters ^2, αp2, αs2 where the second set of switchcontrol signals are configured to operate the second power conversion module under a currentmode control 220.Thus, while there have been shown, described, and pointed out, fundamental novel features ofthe invention as applied to the exemplary embodiments thereof, it will be understood that various omissions, substitutions and changes in the form and details of devices and methods illustrated, and in their operation, may be made by those skilled in the art without departing from the spirit and scope of the presently disclosed invention. Further, it is expressly intended that all combinations of those elements, which perform substantially the same function in substantially the same way to achieve the same results, are within the scope of the invention. Moreover, it should be recognized that structures and / or elements shown and / or described in connection with any disclosed form or embodiment of the invention may be incorporated in any other disclosed or described or suggested form or embodiment as a general matter of design choice. It is the intention, therefore, to be limited only as indicated by the scope of the claims appended hereto.

Claims

CLAIMS 1. An apparatus (100) comprising: a first power conversion module (102) comprising a series resonant dual-active-bridge DC-DC power converter topology and a second power conversion module (104) comprising a series resonant dual-active-bridge DC-DC power converter topology, wherein the first power conversion module (102) and the second power conversion module (104) are coupled in parallel between an input voltage (Vin) and an output voltage (Vo), and wherein, when the apparatus (100) is operating at a pre-determined minimumvoltage gain (Mmin), the first power conversion module (102) is configured to produce a nominal gain (M≈1), and when the apparatus (100) is operating at a pre-determined maximum voltage gain (Mmax), the second power conversion module (104) is configured to produce the nominal gain (M≈1).

2. The apparatus (100) according to claim 1, wherein the nominal gain (M≈1) comprises a gain between zero point nine five (0.95) and one point zero five (1.05).

3. The apparatus according to any one of the preceding claims wherein the apparatus further comprises a controller (128) configured to receive the input voltage (Vin), the output voltage (Vo), and an output current (Io2) of the second power conversion module (104), and produce a first set of switch control signals (130) and a second set of switch control signals (132), wherein the first set of switch control signals (130) are configured to operate the first power conversion module (102) under a voltage mode control (218) and the second set of switch control signals (132) are configured to operate the second power conversion module (104) under a current mode control (220).

4. The apparatus (100) according to any one of the preceding claims wherein the series resonant dual-active-bridge DC-DC power converter topology (300) comprises: a transformer (T) comprising a primary winding (302) magnetically coupled to a secondary winding (304); aseries resonant circuit (Z) comprising an inductor (L), a capacitor (C), and the primarywinding (302) connected in series; a primary full bridge switching circuit (306) coupled to the series resonant circuit (Z) and configured to receive the input voltage (Vin); anda secondary full bridge switching circuit (308) coupled to the secondary winding (304) and configured to produce an output voltage (Vo).

5. The apparatus (100) according to any one of the preceding claims, wherein the first power conversion module (102) comprises a first transformer turn ratio (N1), and the first transformer turn ratio (N1) is configured to produce the nominal gain (M≈1) when the converter is operating at the pre-determined minimum voltage gain (Mmin), and wherein the second power conversion module (104) comprises a second transformer turn ratio (N2), and the second transformer turn ratio (N2) is configured to produce the nominal gain (M≈1) when the converter is operating at the pre-determined maximum voltage gain (Mmax).

6. The apparatus (100) according to any one of the preceding claims, wherein operating the first power conversion module (102) under the voltage mode control (218) comprises: determining a first phase shift parameters (^1, αp1, αs1) based on the input voltage (Vin) and the output voltage (Vo); comparing a reference voltage (Voref) with the output voltage (Vo) to produce a voltage error (ev); applying a first compensation to the voltage error (ev) to produce a first switching frequency (fs1); and generating the first set of switch control signals (130) based on the first switchingfrequency (fs1) and the first phase shift parameters (^1, αp1, αs1), andwherein operating the second power conversion module (104) under current mode control (220) comprises: determining a second phase shift parameters (^2, αp2, αs2) based on the input voltage (Vin) and the output voltage (Vo);comparing a reference current (Io2ref) with the output current (Io2) to produce a current error (ei); applying a second compensation to the current error (ei) to produce a second switching frequency (fs2); and generating the second set of switch control signals (132) based on the secondswitching frequency (fs2) and the second phase shift parameters (^2, αp2, αs2).

7. The apparatus (100) according to any one of the preceding claims wherein the firstcompensation (206) comprises a proportional plus integral (PI) control algorithm, and thesecond compensation (214) comprises a proportional plus integral (PI) control algorithm.

8. A method (900) for DC-DC power conversion, the method (900) comprising: coupling (902) a first power conversion module and a second power conversion module in parallel between an input voltage and an output voltage, wherein the first power conversion module comprises a series resonant dual-active-bridge DC-DC powerconverter topology and the second power conversion module comprises a series resonantdual-active-bridge DC-DC power converter topology; configuring (904) the first power conversion module to produce a nominal gain when the DC-DC power conversion is operating at a pre-determined minimum gain; configuring (906) the second power conversion module to produce the nominal gain when the DC-DC power conversion is operating at a pre-determined maximum gain; operating (908) the first power conversion module under voltage mode control; andoperating (910) the second power conversion module under current mode control.

9. The method (900) according to claim 8, wherein the nominal gain comprises a gain between zero point nine five (0.95) and one point zero five (1.05).

10. The method (900) according to any one of claims 8 or 9, wherein the first power conversion module comprises a first transformer turn ratio, and the first transformer turn ratio is configured to produce the nominal gain when the converter is operating at the pre-determined minimum voltage gain, and wherein the second power conversion module comprises a second transformer turn ratio, and the second transformer turn ratio is configured to produce the nominal gain when the converter is operating at the pre-determined maximum voltage gain.

11. The method (900) according to any one of 8 through 10, wherein operating the first power conversion module under the voltage mode control comprises: determining (202) a first phase shift parameters (^1, αp1, αs1) based on the input voltage (Vin) and the output voltage (Vo); generating (204) a voltage error (ev) by comparing a reference voltage (Voref) with the output voltage (Vo); generating a first switching frequency (fs1) by applying (206) a first compensation to the voltage error (ev); andgenerating (208) a first set of switch control signals (130) based on the first switchingfrequency (fs1) and the first phase shift parameters (^1, αp1, αs1) wherein the first set ofswitch control signals (130) are configured to operate the first power conversion module.

12. The method (900) according to any one of claims 8 through 11, wherein operating the second power conversion module under the current mode control comprises: determining (210) a second phase shift parameters (^2, αp2, αs2) based on the input voltage (Vin) and the output voltage (Vo); comparing (212) a reference current (Io2ref) with an output current (Io2) of the second power conversion module to produce a current error (ei); applying (214) a second compensation to the current error (ei) to produce a second switching frequency (fs2); and generating (216) the second set of switch control signals (132) based on the secondswitching frequency (fs2) and the second phase shift parameters (^2, αp2, αs2), wherein thesecond set of switch control signals (132) are configured to operate the first power conversion module.

Images