Power conditioning for high speed mechanical generators

Abstract

An exemplary method and system for power conditioning using a combined diode bridge rectifier and series transformer power converter is disclosed, where the power converter is configured to inject VARs into a high speed rotating machine to compensate for its mechanical reactance. By using a power converter to inject VARs into the machine via a series transformer, the power converter can be rated slightly lower in terms of power rating than would otherwise be required to be rated higher due to reactance requirements from high frequency inputs. Furthermore, due to the lower power rating, the transformer can be beneficially rated for the corresponding machine power level. The operation of the diode bridge rectifier and transformer does not add any complexity to the system control or power converter control.

Description

[Technical field]

[0001] Related Applications This international PCT application claims priority to and the benefit of U.S. Provisional Patent Application No. 63 / 323,741, entitled "Power Conditioning for High-Speed ​​Permanent Magnet Generators," filed March 25, 2022, which is incorporated by reference in its entirety. [Background technology]

[0002] High speed turbine systems driving permanent magnet generators may result in the generators operating at high frequencies in the kilohertz range. Power converters are used to convert the high frequency AC output of the generators to DC output. These machines are finding use in a wide range of applications, from aerospace to distributed power generation.

[0003] Inverters are typically designed to switch at multiples of the fundamental frequency of the machine. For higher operating frequencies, one approach is to provide an inverter using wide bandgap materials such as silicon carbide to provide a silicon carbide-based inverter, which can be costly. Another approach is to use conventional power switching devices and rate the converters of such machines at higher power ratings. While it is possible to lower the inductance of the machine, at such high operating frequencies the resulting impedance can still be high, leading to reactive power losses, reduced operating power rate of the machine, reduced power delivered at a given current level, and limitations on the maximum power that can be transmitted. The higher power rating addresses reactive power losses.

[0004] There is strong interest in improving the operation of high speed turbines capable of driving permanent magnet generators and other high speed mechanical generators. Summary of the Invention

[0005] An exemplary power conditioning method and system is disclosed using a combined diode bridge rectifier and series transformer power converter, where the power converter is configured to inject VARs into a high speed rotating machine to compensate for machine reactance. One example of a high speed rotating machine is a high speed turbine driving a permanent magnet generator, among other machines described herein. By using a power converter to inject VARs into the machine through a series transformer, the power converter can be rated slightly lower in terms of power rating than would otherwise be required to be rated higher due to reactance requirements from high frequency inputs. Furthermore, due to the lower power rating, the transformer can be beneficially rated for the corresponding machine power level. The operation of the diode bridge rectifier and transformer does not add any complexity to the system control or power converter control.

[0006] In some embodiments, the exemplary power conditioning unit comprises a single- or double-wound coaxial winding transformer for injecting leading or lagging VARs. The exemplary power conditioning unit may be used with a power converter (e.g., inverter) that can use ¼ the rated power of conventional solutions, providing size, cost, and efficiency advantages.

[0007] In one aspect, a system is disclosed that includes a rectifier coupled to each phase output of a high speed mechanical generator and configured to convert AC power output from the machine to a DC output; and a series compensation power regulator coupled to a diode bridge and the high speed mechanical generator, the series compensation power regulator including a power converter coupled in series to the high speed mechanical generator via a series transformer to inject reactive power (VAR) into each phase output of the high speed mechanical generator.

[0008] In some embodiments, a series compensation power regulator provides leading reactive power to the high speed machine to reduce reactive power losses.

[0009] In some embodiments, the system further includes a controller configured to direct the power converter to inject a leading voltage into a mechanical reactance for each phase output of the high speed mechanical generator.

[0010] In some embodiments, the series transformer comprises a coaxially wound transformer (CWT), which includes a compound winding surrounded by a single-turn primary winding structure.

[0011] In some embodiments, each phase output of the power converter is connected in series with a compound winding, and each phase output of the high speed mechanical generator is connected in series with a rectifier via a single turn primary winding structure.

[0012] In some embodiments, the rectifier comprises a diode bridge rectifier.

[0013] In some embodiments, the rectifier further comprises an LC filter.

[0014] In some embodiments, the controller is configured to generate a notched pulse width modulated (PWM) output to reduce harmonic distortion of predefined harmonics.

[0015] In some embodiments, the high speed mechanical generator is selected from the group consisting of a wound field synchronous machine, a switched reluctance machine, a permanent magnet machine, and a permanent magnet synchronous motor machine.

[0016] In some embodiments, the high speed mechanical generator is a single phase output machine.

[0017] In some embodiments, the high speed mechanical generator is a three phase output machine.

[0018] In another aspect, a method is disclosed that includes converting AC power output from each phase output of a high speed mechanical generator to a DC output via a rectifier, and injecting reactive power (VAR) via a series transformer into each phase output of the high speed mechanical generator via a series compensation power regulator coupled to the rectifier and the high speed mechanical generator, the series compensation power regulator comprising a power converter coupled in series to the high speed mechanical generator via the series transformer.

[0019] In some embodiments, a series compensation power regulator provides leading reactive power to the high speed machine to reduce reactive power losses.

[0020] In some embodiments, reactive power is injected as a leading voltage into a mechanical reactance for each phase output of the high speed mechanical generator.

[0021] In some embodiments, a series transformer (eg, a CWT) includes a compound secondary winding having N windings proximate to a primary winding structure having fewer than N windings.

[0022] In some embodiments, each phase output of the power converter is connected in series with a secondary compound winding, and each phase output of the high speed mechanical generator is connected in series with a rectifier via a primary winding structure.

[0023] In some embodiments, the rectifier comprises a diode bridge rectifier and an LC filter.

[0024] In some embodiments, the leading reactive power is injected as a pulse width modulated (PWM) output or a notched PWM output.

[0025] In some embodiments, the high speed mechanical generator is selected from the group consisting of a wound field synchronous machine, a switched reluctance machine, a permanent magnet machine, and a permanent magnet synchronous motor machine.

[0026] In some embodiments, the high speed mechanical generator is a single phase output machine or a three phase output machine.

[0027] The following detailed description of certain embodiments of the present disclosure will be better understood when read in conjunction with the accompanying drawings. For the purpose of illustrating the present disclosure, certain embodiments are shown in the drawings. It should be understood, however, that the disclosure is not limited to the precise arrangements and instrumentalities of the embodiments shown in the drawings. [Brief description of the drawings]

[0028] [Figure 1] 1 illustrates an example power conversion and conditioning system for a high speed machine-generator configured to inject a set of VARs into the high speed machine-generator to compensate for machine reactance, according to an example embodiment. [Figure 2A] 2 illustrates an example of a high speed machine-generator and the example power conversion and conditioning system of FIG. 1 for injecting a series of VARs into the high speed machine-generator in accordance with an example embodiment. [Figure 2B] 1 illustrates an example of injection voltage operation in relation to the operation of other high speed mechanical generators. [Figure 2C] 2 shows the total harmonic distortion performance of the power conversion system of FIG. 1. [Figure 2D] 2 shows the rectifier terminal voltage of the power conversion and conditioning system. [Figure 2E] Indicates the maximum obtainable power of the system. [Diagram 3] 1 illustrates an exemplary series transformer that may be used to couple a power conversion and conditioning system to a high speed machine. [Figure 4A] 1 illustrates an example of a notched PWM output that may be generated by an example power conversion and regulation system for injecting a VAR into a high speed machine-generator, according to an example embodiment. [Figure 4B] An example of injection voltage operation via notched PWM is given in the context of operation of other high speed mechanical generators. [Figure 4C] 2 illustrates the total harmonic distortion performance of the power conversion and conditioning system of FIG. 1 using notched PWM. [Figure 4D] 2 shows the rectifier terminal voltage of the power conversion and conditioning system. [Figure 4E]Indicates the maximum obtainable power of the system. [Figure 5A] 3 illustrates an example of the high speed mechanical generator of FIG. 2 coupled with a conventional power conversion system. [Figure 5B] 1 illustrates the operation of a conventional power conversion system without reactive power injection. [Figure 5C] The total harmonic distortion performance of the power conversion system is shown. [Figure 5D] 2 shows the rectifier terminal voltage of the power conversion system. [Figure 5E] Indicates the maximum obtainable power of the system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0029] To facilitate an understanding of the principles and features of various embodiments of the present invention, these are described below with reference to their implementation in exemplary embodiments.

[0030] Several references, which may include various patents, patent applications, and publications, are cited in the reference list and discussed in the disclosure provided herein. Citation and / or discussion of such references is provided merely to clarify the description of the present disclosure and is not an admission that any such reference is "prior art" to any aspect of the present disclosure described herein. For purposes of notation, "[n]" corresponds to the nth reference in the list. All references cited and discussed herein are incorporated herein by reference in their entirety and to the same extent as if each reference was incorporated by reference separately.

[0031] Exemplary System FIG. 1 illustrates an exemplary power conversion and conditioning system 100 for a high-speed machine-generator 102 (shown as motor 102a and generator 102b) configured to inject a series of VARs into the high-speed rotating machine to compensate for the machine reactance, according to an exemplary embodiment. In the example illustrated in FIG. 1, the power conversion and conditioning system 100 includes a power converter 106 (shown as inverter 106) as a series compensation power regulator and an AC / DC converter 108 (shown as diode bridge rectifier 108) coupled to a generator output 109 (shown as 109a, 109b, 109c) of the high-speed machine-generator 102 via a set of transformers 110. The diode bridge 108 and the transformer 110 collectively form a series compensator as a series compensation power conditioning element configured to increase the power available from the high-speed machine-generator and thus drive its rated value. In other words, the series compensator is capable of injecting a series of reactive powers (VARs) into the high-speed rotating machine 102 to compensate for the machine reactance.

[0032] An example of a high speed mechanical generator 102 includes a permanent magnet synchronous motor machine (PMSM). The generator may include a permanent magnet rotor and three-phase stator windings, which may be terminated at the generator terminals. Other examples of high speed mechanical generators include wound field synchronous machines, switched reluctance machines, permanent magnet machines, or other types of machines.

[0033] 1, the power converter 106 is coupled between the generator terminals 109 and the DC link 112 and configured as a three-phase bridge inverter. The generator terminals 109 are connected to the inverters connected in series for each phase. Although this example is described in the context of three-phase AC power, the machine / generator may alternatively be configured to output one-phase, two-phase AC power, six-phase AC power, etc. for a corresponding power conditioning system.

[0034] In the example shown in FIG. 1, the power converter 106 includes six switches 114, including a set of upper switches 114 (shown as 114a, 114b, 114c) and a set of lower switches 116 (shown as 116a, 116b, 116c) connected across a DC link 120. Each switch 114 and 116 of the power converter 106 includes a controllable solid-state device, such as an IGBT or a MOSFET, and an anti-parallel diode 117 across the solid-state device. A DC link capacitor 118 is additionally coupled across the DC link 112 to create a low source impedance for the power converter and reduce voltage ripple on the DC link 112. The switches 114 and 116 are controlled by a controller 119. The controller 119 may be implemented in software, hardware, or any combination thereof.

[0035] The generator 120 of the machine 102 may generate a voltage as a back EMF lagging reactance 122. The voltage generated by the generator may be referred to as the generator voltage or generator EMF, and the voltage at the terminals of the power converter 106 may be referred to as the power converter terminal voltage.

[0036] As described above, the series compensator includes a diode bridge 108 and a set of series transformers 110. In this example, the diode bridge 108 includes six diodes 124 (shown as 124a, 124b, 124c, 124d, 124e, 124f) in a second DC link 126 that outputs to a bridge output 134 through an LC filter 128 including an inductor 130 and a capacitor 132. The set of transformers 110 is configured with three primary windings 136 of the generator 102 and a set of series transformers that couple to a three-phase power converter (e.g., having multiple windings) through a diode bridge (e.g., having at least one winding) of the primary winding 136 (shown as 136a, 136b, 136c) and a secondary winding 138 (shown as 138a, 138b). In some embodiments, the series transformer 110 is configured as a coaxial winding transformer.

[0037] Example of VAR injection operation using a power conditioning unit 2A illustrates an example of a high speed motor / generator 202 having a high speed turbine driving a permanent magnet generator. The high speed turbine permanent magnet generator 202 can be configured to operate with the example power conditioning system and power converter of FIG.

[0038] An exemplary specification for a high speed turbine driven permanent magnet generator may have a turbine operating at a rotational speed of 75,000 rpm (75k RPM) to generate hundreds of kilowatts of power. Turbines include 10k RPM, 11k RPM, 12k RPM, 13k RPM, 14k RPM, 15k RPM, 16k RPM, 17k RPM, 18k RPM, 19k RPM, 20k RPM, 21k RPM, 22k RPM, 23k RPM, 24k RPM, 25k RPM, 26k RPM, 27k RPM, 28k RPM, 29k RPM, 30k RPM, 31k RPM, 32k RPM, 33k RPM, 34k RPM, 35k RPM, 36k RPM, 37k RPM, 38k RPM, 39k RPM, 40k RPM, 41k RPM, 42k RPM, 43k RPM, 44k RPM, 45k RPM, 46k RPM, 47k RPM, 48k RPM, 49k The turbine may be operated or rated at 50 k RPM, 50 k RPM, hi some embodiments, the turbine is operated at greater than 50 k RPM, hi some embodiments, the turbine is operated at greater than 100 k RPM, This type of machine can be used in a wide variety of applications ranging from aerospace to distributed power generation.

[0039] In one example, the generator (e.g., 202) may be a four-pole permanent magnet (PM) machine (e.g., 102) that produces a nominal output at an electrical frequency of 2500 Hertz. The maximum power such a machine (e.g., 202) can supply, say, a resistive load, may be limited by the voltage drop across the machine reactance (e.g., shown as 140 in FIG. 1). This voltage drop can reduce (i) the machine's operating power rate and (ii) the power delivered at a given current level, limiting the maximum power that can be transferred. An example is a PM generator with a line-to-neutral back emf of 183.7 volts at 2500 Hertz, with a machine inductance of 14 microhenries and windings rated to carry 660 amps. At these operating points, the generator can only supply approximately 220 kW to the resistive load, as opposed to the 363 kW that could be supplied if the inductance did not limit the flow of power. This is a significant penalty (40% reduction) to the capability of such a generator. A similar penalty would apply if the generator output was rectified to a DC voltage to power a DC load.

[0040] The exemplary method and power conditioning system can extract higher power from the machine using power converters to compensate the machine VAR, for example by providing a leading VAR at 2500 Hz in the above example and injecting a VAR in series to compensate the machine reactance. In the example shown in Figures 1 and 2, the power conditioning unit uses series-connected power converters per phase (e.g., series-connected inverters per phase). The power converters 106 may be configured to float at the terminal voltage and inject the desired leading (or lagging) voltage to increase or decrease the generator current. Such power converters must carry very high currents (660 amps in the above example), which can lead to high switch losses. A single-phase configuration can also lead to high ripple currents in the DC bus capacitance, increasing the requirements on the rating.

[0041] An alternative approach is to use a series transformer with a three-phase power converter (e.g., inverter) connected to three primary windings, which results in low losses and low leakage inductance at high frequencies and high currents. An exemplary power conditioning unit may use a coaxial winding transformer (CWT) with a single-winding primary [4], which may then be connected to a fractional-rated power converter that is configured to switch at a relatively low frequency, e.g., about 2.5 kHz to 7.5 kHz. This CWT may have an N:1 turns ratio to handle high currents (e.g., 660 amps) in one winding and convert it to lower currents in the second winding. The CWT design also results in extremely low leakage inductance. Furthermore, at a frequency of 2.5 kHz, the CWT can be 1 / 40 the size of an equivalent 60 Hertz transformer, making it very efficient and cost-effective.

[0042] CWTs are configured to insert voltages into line-to-neutral or line-to-line voltages, although it is speculated that the line-to-line configuration has less core and copper losses. The design examples presented herein may use high frequency core materials such as nanocrystalline cores, amorphous iron cores, or thin silicon steel laminates suitable for operation at 2.5 kHz, and most often configure the high current winding as a single winding made from continuous copper tubing. This single winding carries the high machine currents and inserts the voltages required for series VAR compensation. The internal windings are configured with multiple turns and are generally high voltage, low current windings. The coaxial design is capable of providing extremely low leakage inductance between the primary and secondary windings. CWTs are well suited to operation at high current levels, and previous CWTs have been built to handle thousands of amps at both low and high frequencies.

[0043] The power converter as an inverter may be a voltage source inverter (or current source inverter, CSI) and typically switches at the electrical frequency of the generator output, in this case 2500 Hertz. The lower switching frequency reduces switching losses compared to inverter switching at 25-50 kHz. In scenarios where inverter switching causes high levels of harmonics in the motor current, additional pulses can be introduced into the inverter switching to eliminate the lowest order harmonics. Alternative advanced PWM control methods can also be used to eliminate harmonics. The inverter may be of standard system design, but may also be configured with customized programmed PWM to eliminate low order harmonics. This may be an iterative process and may benefit from the use of advanced techniques such as neural nets or gradient methods to find the lowest loss point, as well as various AI / ML techniques described herein.

[0044] The exemplary system also allows re-optimization at the machine level to increase the back EMF, for example by up to 25%, further reducing the winding current and reducing losses. An optimal integrated solution needs to be developed that can provide a DC voltage in the range of 600-700 volts from no load to full load, and adjust the DC voltage in smaller ranges as required. This will significantly reduce the overall implementation complexity and cost, and increase the versatility of the turbine generator.

[0045] In contrast, a conventional approach may use a high frequency switching inverter (e.g., 106) to connect to the machine. The inverter 106 is capable, at least in principle, of generating any voltage and phase angle at 2500 Hz. This allows independent control of P and Q, allowing higher power to be extracted from the generator. The challenge is often the low inductance of the machine. To keep the current ripple at a reasonable level, the system may use a very high switching frequency. For example, as in this example, with a machine with an inductance of 14 μH, and a 500 volt DC bus (e.g., 112) at the inverter, the system may use a switching frequency of 50-100 kHz. If this were realized with a wide bandgap device (e.g., SiC), the system would be subject to dv / dt levels in excess of 50 kV / μs, with associated EMI and / or insulation failure issues, and potentially even more losses. And to extract 350 kW of power from this generator configuration, the system may use an inverter of about 500 kVA switching at about 50 kHz.

[0046] In addition to being technically challenging, this approach can have other problems such as cost and system efficiency. For example, a high frequency inverter rated at 500 kVA can be costly and can be inefficient due to device switching losses.

[0047] Exemplary Coaxial Winding Transformer FIG. 3 shows a coaxial winding transformer (CWT) 302 having a compound winding 304 surrounded by a single turn structure in an assembled view 300a and a cross-sectional view 300b. In this example, the turns of the secondary winding 304 are arranged coaxially with the main (or primary) conductor 301 in a core 305. The core 305 may be a nanocrystalline core, an amorphous iron core, a thin silicon steel laminate, a silicon steel core, or other cores equivalent or suitable for the application. A U-shaped or elliptical tube 306 may provide the primary winding (and a portion of the main conductor 301 to the high speed mechanical generator 102) through a connection at the opposite end. The U-shaped or elliptical tube 306 may be copper or other suitable conductive material. The turns of the secondary winding 304 may pass inside the copper tube 306 (e.g., with a turns ratio of 1:50) and connect to the power converter 106. The coaxial configuration reduces the level of leakage flux in the CWT 302.

[0048] In diagram 300a, an oval tube 306 includes two straight sections 308 between rounded ends 310 (shown as 310 and 310') and is split along the length of the tube to facilitate insertion of the secondary winding 304 between both halves 306a and 306b of the oval tube 306. At one of the rounded ends 310', the oval split tube 306 is cut to form a single turn primary winding for the CWT 302. A connecting plate 312 is attached to both halves of the split tubes 306a and 306b on the opposite side of the cut in the rounded end 310. As can be seen, the two halves of the oval tubes 306a and 306b are mirror images of each other that can be positioned to form the completed oval split tube 306.

[0049] For assembly, a compound winding may be formed for the secondary winding 304 of the CWT 302. The secondary winding 304 is disposed in one half of the elliptical split tube 306a. The lead wire for the secondary winding may be passed through an opening in one of the rounded ends 310' of the elliptical split tube 306a half. For example, an opening may be formed opposite the cut end of the half tube 306a, with the connection plate 312 being between the two openings. The secondary winding 304 may be passed through an opening for connection to a switching circuit, such as the switching circuit of the inverter 106. A protective sleeve may be provided around the lead wire passing through the opening to provide additional insulation and wear protection. With the winding 304 positioned in the half tube 306a, the other half tube 306b may be aligned over the tube half 306a such that the secondary winding 304 is surrounded by the elliptical split tube 306. The connecting plates 312 of each tube half can be configured to attach together to hold the halves of the oval split tube 306 in alignment around the secondary winding 304. For example, corresponding connecting plates 312 can be secured together using fasteners (e.g., bolts or screws).

[0050] A mounting brace 314 may also be included to hold the oval split tube 306 in alignment and secure the core 305 around the straight section 310 of the oval split tube 306. The mounting brace 314 may be made from an insulating material having sufficient strength to hold the core 305 in place on the oval split tube 306. The mounting brace 314 may be secured in place at both ends of the core 305 using fasteners (e.g., nuts and bolts) that run through the center and both outside sides of the oval split tube 306 to secure the two halves together. The structure of the CWT 302 may be oil bathed for cooling during operation. Holes or openings may be provided in the tube 306 to allow oil to flow within the tube 306 and around the secondary winding 304.

[0051] The machine (202) (with the series CWT 302) may be connected (e.g., using silicon carbide (SiC) diodes) to a three-phase diode rectifier bridge with a capacitor filter 118. The inverter 108 and CWT 302 may be used to inject a controllable leading voltage that cancels a portion of the voltage drop due to the machine reactance (e.g., 140). For example, to cancel 40% of the voltage drop across the machine reactance, the inverter 106 is rated at about 150 kVA (30% of the alternative) and switched at 7.5 kHz with a simple control strategy and significantly lower losses. Such an approach allows for conversion of the turbine to DC output. While the inverter 106 injects a leading VAR to increase the power delivered by the generator under normal conditions, it can also inject a lagging VAR to significantly limit the current that may flow under fault conditions and simplify protection requirements.

[0052] Referring back to FIG. 1, the power converter 106 may use a coaxial winding transformer 302 to inject a square wave 202 (shown as injected voltage 202) in series with the winding at the machine frequency (2500 Hz). The CWT 302 includes a single turn copper tube 306 for the high current winding with a compound winding 304 inside. CWTs rated at 2500 amps are manufactured and commercially available. As an example, a 6:1 turns ratio may be used. As previously explained, the (machine (e.g., 102, 202) and the CWT 302) can go directly into a diode capacitor rectifier. The resistor 204 models the load on the system.

[0053] Experimental Results and Additional Examples Two cases are shown for illustrative purposes. The "base case" shows the performance of a PM generator with zero injection and a simple diode bridge supplying a DC resistive load. Figures 4-5 show various waveforms of the system, including the generator current, DC output voltage, back EMF, and rectifier terminal voltage. At operating conditions, the DC voltage is 265 volts DC, the maximum output power available is approximately 210 kW, and the machine current is 850 amps peak and 600 amps RMS. The machine current output has a total harmonic distortion (THD) of 4.4%, which is considered acceptable.

[0054] Base case: 0V injection (=0%) (power converter bus) (i.e. power converter bus voltage OV at rectifier power of 210KW) Simulation studies have been performed on the circuit of FIG. 1 and corresponding results are shown. FIGS. 5A-5E show the results for a reference case of a power converter (e.g., inverter) without VAR injection, i.e., 0V injection. FIG. 5A shows a power converter model 502 corresponding to the inverter 106. Plot 504 in FIG. 5B shows the scenario injection voltage is 0V. Plot 506 shows the machine current, plot 508 shows the back EMF, and plot 510 shows the rectifier DC bus voltage. FIG. 5C shows the THD of the current at 4.72%. FIG. 5D shows the rectifier terminal voltage. FIG. 5E shows the maximum power available at the rectifier load from the PMSM drive. As shown in FIG. 5E, the maximum available power without compensation is limited to about 210KW.

[0055] Case "1": 180V injection with rectifier power of 350kW In order to increase the power available from the PMSM drive to its rated value, a series compensator is disclosed herein and simulation studies based on MATLAB / Simulink have been conducted to validate the approach. Figures 2A-2E show the results for a 180V VAR injection with a rectifier power of 350kW. Figure 2A shows a circuit schematic of the series compensation architecture of a series compensated PMSM. Plot 206 in Figure 2B shows the injected voltage at + / - 180V. Plot 508 shows the machine current, plot 510 shows the back EMF, and plot 512 shows the rectifier DC bus voltage. It can be observed that the injected positive voltage is added to the falling waveforms of the machine current and back EMF, and the injected negative voltage is added to the rising waveforms of the machine current and back EMF.

[0056] In plot 512, it can be observed that the DC bus voltage is about 400V, which is much higher than the bus voltage of 270V output by the power converter of FIG. 5. FIG. 2C shows the THD of the current at 4.72%. FIG. 2D shows the rectifier terminal voltage. FIG. 2E shows the maximum power available at the rectifier load from the PMSM drive. As shown in FIG. 2E, the maximum available power without compensation is limited to about 350kW (compared to 210kW without compensation shown in FIG. 5E).

[0057] In fact, with the newly injected compensation, the rectifier side power could be delivered at 350kW, significantly higher than the 210kW possible without zero compensation. This is achieved with a power converter rated at 150kVA instead of 500kVA and switching at about 1 / 10th the frequency of the conventional solution. This significantly reduces losses in the exemplary method and system as opposed to a full rated power converter. In a 350kW system, losses can be reduced from 30kW (9% loss) to about 7kW (2% loss), which is 75% of the converter losses.

[0058] Case "2": 200V injection with 350kW rectifier power by modifying notch PWM waveform To reduce harmonic content in the machine current, the controller 119 may generate a modified square wave injection with notches that can be voltage injected to reduce the overall THHO and specific harmonics. Figure 4A shows an example of a modified square wave injection scheme. The modified square wave includes an "on" portion 402 and an "off" portion 404, each with an "off" notch 406 or an "on" notch 408, respectively, therein to address specific harmonics.

[0059] Figures 4B-4E show the results when a 200V VAR is injected with 350kW rectifier power. Plot 402 in Figure 4B shows the injected voltage of + / -200V. Plot 410 shows the injected voltage with a notch (used in the simulation), plot 412 shows the machine current, plot 414 shows the back EMF, and plot 416 shows the rectifier DC bus voltage. Figure 4C shows the THD of the current at 3.79%. Figure 4D shows the rectifier terminal voltage. Figure 4E shows the maximum power available at the rectifier load from the PMSM drive.

[0060] It can be observed that the exemplary series compensated PMSM with modified injection voltage has a reduced overall THD along with a significant reduction in the 5th and 7th harmonics (FIG. 4C). By adding a notch to the waveform, the current THD is reduced to 3.4%. It can also be observed that the power available after adding the notch remains rated at 350 kW for the machine, as shown in FIG. 4E.

[0061] The modified PWM with notches may be generated using selective harmonic elimination (SHE) PWM, space vector PWM, instantaneous current control PWM, hysteresis band current control PWM, sigma-delta modulation, etc. Another exemplary method of generating a PWM with a notched output is described in U.S. Pat. No. 4,245,290.

[0062] Exemplary Computing System It should be understood that the logical operations described above may be implemented (1) as a sequence of computer-executed operations or program modules operating on a computing system, and / or (2) as interconnected machine logic circuits or circuit modules within a computing system. The implementation is a matter of choice dependent on the performance and other requirements of the computing system. Thus, the logical operations described herein are variously referred to as state operations, operations, or modules. These operations, operations, and / or modules may be implemented in software, firmware, special purpose digital logic, hardware, and any combination thereof. It should also be understood that more or fewer operations may be performed than those shown in the figures and described herein. These operations may also occur in orders different from those described herein.

[0063] In its most basic configuration, the controller (e.g., 119) may include at least one processing unit and system memory configured to execute in a real-time control loop. Depending on the exact configuration and type of computing device, the system memory may be volatile (such as random access memory (RAM)), non-volatile (such as read-only memory (ROM), flash memory, etc.), or some combination thereof. The processing unit may be a standard programmable processor that performs arithmetic and logical operations necessary for the operation of the computing device. As used herein, processing units and processors refer to physical hardware devices that execute coded instructions to perform functions on inputs and create outputs, including, but not limited to, microprocessors (MCUs), microcontrollers, graphic processing units (GPUs), and application specific integrated circuits (ASICs). Thus, while instructions may be discussed as being executed by a processor, the instructions may be executed simultaneously, sequentially, or otherwise by one or more processors. A computing device may also include a bus or other communication mechanism for communicating information between various components of the computing device.

[0064] A computing device may have additional features / functionality. For example, a computing device may include additional storage, such as removable and non-removable storage. A computing device may also include network connection(s) that allow the device to communicate with other devices, such as via the communications paths described herein.

[0065] The processing unit may be configured to execute program code encoded on a tangible computer-readable medium. A tangible computer-readable medium refers to any medium that can provide data that causes a computing device (i.e., a machine) to operate in a specific manner. A variety of computer-readable media may be utilized to provide instructions to a processing unit for execution. Examples of tangible computer-readable media may include, but are not limited to, volatile, non-volatile, removable, and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. System memory, removable storage, and non-removable storage are all examples of tangible computer storage media.

[0066] In an exemplary implementation, the processing unit may execute program code stored in a system memory. For example, a bus may transmit data to the system memory, from which the processing unit receives and executes instructions. Data received by the system memory may optionally be stored in a removable or non-removable storage device before or after execution by the processing unit.

[0067] Artificial Intelligence and Machine Learning Examples As noted above, advanced techniques such as neural nets, gradient methods, or other AI techniques may be used to find the lowest loss point, as well as various AI / ML techniques described herein. The term "artificial intelligence" is described herein to include any technique that enables one or more computing devices or computing systems (i.e., machines) to mimic human intelligence. Artificial intelligence (AI) includes, but is not limited to, knowledge-based, machine learning, representation learning, and deep learning. The term "machine learning" is defined herein to be a subset of AI that enables machines to acquire knowledge by extracting patterns from raw data. Machine learning techniques include, but are not limited to, logistic regression, support vector machines (SVMs), decision trees, naive Bayes classifiers, and artificial neural networks. The term "representation learning" is defined herein to be a subset of machine learning that enables machines to automatically discover representations required for feature detection, prediction, or classification from raw data. Representation learning techniques include, but are not limited to, autoencoders. The term "deep learning" is defined herein to be a subset of machine learning that enables machines to automatically discover representations required for feature detection, prediction, classification, etc., using layers of processing. Deep learning techniques include, but are not limited to, artificial neural networks or multi-layer perceptrons (MLPs).

[0068] Machine learning models include supervised, semi-supervised, and unsupervised learning models. In a supervised learning model, the model learns a function that maps inputs (also known as feature(s)) to outputs (also known as targets or targets) during training with a labeled data set (or data sets). In an unsupervised learning model, the model learns a function that maps inputs (also known as feature(s)) to outputs (also known as targets or targets) during training with an unlabeled data set. In a semi-supervised model, the model learns a function that maps inputs (also known as feature(s)) to outputs (also known as targets or targets) during training with both labeled and unlabeled data.

[0069] Neural Networks An artificial neural network (ANN) is a computing system that includes multiple interconnected neurons (e.g., also referred to as "nodes"). The present disclosure contemplates that the nodes may be implemented using a computing device (e.g., a processing unit and memory as described herein). The nodes may be arranged in multiple layers, such as an input layer, an output layer, and optionally one or more hidden layers. An ANN with hidden layers may be referred to as a deep neural network or a multi-layer perceptron (MLP). Each node is connected to one or more other nodes in the ANN. For example, each layer is composed of multiple nodes, and each node is connected to all nodes in the previous layer. The nodes in a given layer are not interconnected with each other, i.e., the nodes in a given layer function independently of each other. As used herein, the nodes in the input layer receive data from outside the ANN, the nodes in the hidden layer modify the data between the input layer and the output layer, and the nodes in the output layer provide the results. Each node is configured to receive an input, implement an activation function (e.g., a binary step, linear, sigmoidal, hyperbolic tangent, or rectified linear unit (ReLU) function), and provide an output according to the activation function. In addition, each node is associated with a respective weight. The ANN is trained with a data set to maximize or minimize an objective function. In some implementations, the objective function is a cost function that is a measure of the performance of the ANN during training (e.g., an error such as L1 or L2 loss), and the training algorithm adjusts the weights and / or biases of the nodes to minimize the cost function. The present disclosure contemplates that any algorithm that finds a maximum or minimum of an objective function may be used to train the ANN. Training algorithms for ANNs include, but are not limited to, backpropagation. It should be understood that the artificial neural network is provided only as an exemplary machine learning model. The present disclosure contemplates that the machine learning model can be any supervised, semi-supervised, or unsupervised learning model. Optionally, the machine learning model is a deep learning model.Machine learning models are known in the art and therefore will not be described in further detail herein.

[0070] A convolutional neural network (CNN) is a type of deep neural network that is applied, for example, to image analysis applications. Unlike traditional neural networks, each layer of a CNN has multiple nodes arranged in three dimensions (width, height, and depth). A CNN can include different types of layers, for example, convolutional layers, pooling layers, and fully connected (also referred to herein as "dense") layers. Convolutional layers include a set of filters and perform the majority of the computations. Pooling layers are optionally inserted between convolutional layers to reduce computational power and / or control overfitting (e.g., by downsampling). Fully connected layers include neurons, each neuron connected to all neurons in the previous layer. Layers are stacked similarly to traditional neural networks. GCNN is a CNN adapted to operate on structured datasets such as graphs.

[0071] Other supervised learning models A logistic regression (LR) classifier is a supervised classification model that uses a logistic function to predict the probability of a target, which can be used for classification. During training, the LR classifier is trained on a data set (also referred to herein as a "data set") to maximize or minimize an objective function, e.g., a measure of the performance of the LR classifier (e.g., an error such as an L1 or L2 loss). The present disclosure contemplates that any algorithm that finds a minimum of a cost function can be used. LR classifiers are known in the art and therefore will not be described in further detail herein.

[0072] The Naive Bayes (NB) classifier is a supervised classification model based on Bayes' theorem, which assumes independence between features (i.e., the presence of one feature in a class is independent of the presence of any other features). NB classifiers are trained on a dataset by calculating the conditional probability distribution of each feature given and applying Bayes' theorem to calculate the conditional probability distribution of the label given which an observation was made. NB classifiers are known in the art and therefore will not be described in further detail herein.

[0073] A K-NN classifier is a supervised classification model that classifies new data points based on a similarity measure (e.g., a distance function). During training, the K-NN classifier is trained on a data set (also referred to herein as a "data set") to maximize or minimize an objective function, e.g., a measure of the performance of the K-NN classifier. The present disclosure contemplates that any algorithm that finds a maximum or minimum of an objective function may be used. K-NN classifiers are known in the art and therefore will not be described in further detail herein.

[0074] A majority voting ensemble is a meta-classifier that combines multiple machine learning classifiers for classification via majority voting. In other words, the final prediction (e.g., class label) of the majority voting ensemble is the one that is most frequently predicted by the member classification models. The majority voting ensemble is known in the art and therefore will not be described in further detail herein.

[0075] It should be understood that the various techniques described herein may be implemented in connection with hardware or software, or, where appropriate, with a combination thereof. Thus, the methods and apparatus of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in a tangible medium, such as a hard drive or any other machine-readable storage medium, where the program code, when loaded and executed on a machine, such as a computing device, causes the machine to become an apparatus for practicing the presently disclosed subject matter. When executing the program code on a programmable computer, the computing device generally includes a processor, a processor-readable storage medium (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. One or more programs may implement or utilize the processes described in connection with the presently disclosed subject matter, for example, through the use of application programming interfaces (APIs), reusable controls, or the like. Such programs may be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, the program(s) may be implemented in assembly or machine language, as appropriate. In any case, the language may be a compiled or interpreted language, and combined with hardware implementations.

[0076] Although exemplary embodiments of the present disclosure have been described in detail herein in certain instances, it should be understood that other embodiments are contemplated. Thus, the present disclosure is not intended to be limited in scope to the details of construction and the arrangement of components set forth in the following description or illustrated in the drawings. The present disclosure is capable of other embodiments and of being practiced or carried out in various ways.

[0077] It should also be noted that, as used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. Ranges may be expressed herein as from "about" or "approximately 5" one particular value, and / or to "about" or "approximately" another particular value. When such a range is expressed, other exemplary embodiments include from the one particular value and / or to the other particular value.

[0078] "Comprising" or "containing" or "including" means that at least the named compound, element, particle, or method step is present in a composition or article or method, but does not exclude the presence of other compounds, materials, particles, or method steps, even if other such compounds, materials, particles, or method steps have the same function as the named one.

[0079] In describing the exemplary embodiments, technical terms will be used for clarity. Each term is intended to assume the broadest meaning of that term as understood by those skilled in the art and to include all technical equivalents that operate in a similar manner to achieve a similar purpose. It should also be understood that the reference to one or more steps of a method does not preclude the presence of additional or intervening method steps between those steps explicitly identified. The steps of the method may be performed in a different order than described herein without departing from the scope of the present disclosure. Similarly, it should also be understood that the reference to one or more components in a device or system does not preclude the presence of additional or intervening components between those components explicitly identified.

[0080] The term "about" as used herein means approximately, in the region of, roughly, or around. When the term "about" is used in conjunction with a numerical range, it modifies that range by extending the boundaries above and below the stated numerical values. In general, the term "about" is used herein to modify a numerical value above and below the stated value by a variance of 10%. In one embodiment, the term "about" means ±10% of the numerical value of the number with which the term is used. Thus, about 50% means within a range of 45% to 55%. Numerical ranges recited herein by endpoints include all numbers and fractions subsumed within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.90, 4, 4.24, and 5).

[0081] Similarly, numerical ranges recited herein by endpoints include the subranges subsumed within that range (e.g., 1 to 5 includes 1 to 1.5, 1.5 to 2, 2 to 2.75, 2.75 to 3, 3 to 3.90, 3.90 to 4, 4 to 4.24, 4.24 to 5, 2 to 5, 3 to 5, 1 to 4, and 2 to 4). It is also to be understood that all numbers and fractions thereof are presumed to be modified by the term "about."

[0082] The following patents, applications, and publications, mentioned below and throughout the specification, are hereby incorporated by reference in their entireties. [1] Monica Chinchilla, Santiago Arnaltes and Juan Carlos Burgos. “Control of Permanent-Magnet Generators Applied to Variable-Speed ​​Wind-Energy Systems Connected to the Grid,” IEEE Transactions on Energy Conversion, Vol. 21, o. 1, March 2006, pp. 130-135. [2]Charles Maria Jenisha,Nanjappagounder Ammasaigounden,Natarajan Kumaresan,Kadi BhagyaSri,“Power electronic interface with de-coupled control for wind-driven PMSG feeding utility grid and DC load,”Power Electronics!ET,vol.11,no.2,pp.329-338,2018. [3]F.Morel,L.Xuefang,J.-M Retif,B.Allard,C.Buttay,“A Comparative Study of Predictive Current Control Schemes for a Permanent-Magnet Synchronous Machine Drive,”IEEE Transactions on Industrial Electronics,vol.56,No.7,pp.2715-2728,2009. [4]L.A.C.Lopes,G.Joos,“Pulse width modulated capacitor for series compensation,”Power Electronics IEEE Transactions on,vol.I 6,no.2,pp.167-174,200 I. [5]J.Kuang and B.T.Ooi,“Series Connected Voltage-Source Converter Modules for Force-Commutated SVC and DC-Transmission,”IEEE Trans.Power Delivery,Vol.9,no.2,pp.977-981,1994. [6]L.Gyugyi,“Dynamic Compensation of Ac Transmission Lines by Solid State Synchronous Voltage Sources,”IEEE Trans.Power Delivery,no.9,pp.904-911,1994. [7]L.Gyugyi,C.D.Schauder,and Kelyan K.Sen,“Static Synchronous Series Compensator:A solid-State Approach to the Series Compensation of Transmission Lines,”IEEE Trans.on Power Delivery,Vol.12,no.1,pp.406-417,1997.

Claims

1. It is a system, A rectifier configured to couple to each phase output of a high-speed mechanical generator and convert the AC power output from the high-speed mechanical generator into a DC output, A system comprising: a series compensating power regulator coupled to the rectifier and the high-speed mechanical generator, wherein the series compensating power regulator includes a power converter coupled in series to the high-speed mechanical generator via a series transformer for injecting reactive power (VAR) into each phase output of the high-speed mechanical generator.

2. The system according to claim 1, wherein the series compensating power regulator provides leading reactive power to the high-speed mechanical generator to reduce reactive power loss.

3. The system according to claim 1 or 2, further comprising a controller configured to instruct the power converter to inject a leading voltage into the mechanical reactance for each phase output of the high-speed mechanical generator.

4. The system according to claim 1 or 2, wherein the series transformer comprises a coaxial winding transformer (CWT), and the coaxial winding transformer includes a double winding surrounded by a single primary winding structure.

5. The system according to claim 4, wherein each phase output of the power converter is connected in series with the compound winding, and each phase output of the high-speed mechanical generator is connected in series with the rectifier via the single-winding primary winding structure.

6. The system according to claim 1 or 2, wherein the rectifier comprises a diode bridge rectifier.

7. The system according to claim 6, wherein the rectifier further comprises an LC filter.

8. The system according to claim 3, wherein the controller is configured to generate a pulse-width modulation (PWM) output with a notch to reduce harmonic distortion of predefined harmonics.

9. The system according to claim 1 or 2, wherein the high-speed mechanical generator is selected from the group consisting of a wound-field synchronous machine, a switched magnetoresistive machine, a permanent magnet machine, and a permanent magnet synchronous motor machine.

10. The system according to claim 1 or 2, wherein the high-speed mechanical generator is a single-phase output machine.

11. The system according to claim 1 or 2, wherein the high-speed mechanical generator is a three-phase output machine.

12. It is a method, The process involves converting the AC power output from each phase output of a high-speed mechanical generator into DC power via a rectifier, A method comprising injecting reactive power (VAR) into each phase output of the high-speed mechanical generator via a series transformer through a series compensating power regulator coupled to the rectifier and the high-speed mechanical generator, wherein the series compensating power regulator comprises a power converter coupled in series to the high-speed mechanical generator via the series transformer.

13. The method according to claim 12, wherein the series compensating power regulator provides leading reactive power to the high-speed mechanical generator to reduce reactive power loss.

14. The method according to claim 12 or 13, wherein the reactive power is injected as a leading voltage to the mechanical reactance for each phase output of the high-speed mechanical generator.

15. The method according to claim 12 or 13, wherein the series transformer includes a double-wound secondary winding having N windings, located proximal to a primary winding structure having less than N windings.

16. The method according to claim 15, wherein each phase output of the power converter is connected in series with the secondary winding of the compound winding, and each phase output of the high-speed mechanical generator is connected in series with the rectifier via the primary winding structure.

17. The method according to claim 12 or 13, wherein the rectifier comprises a diode bridge rectifier and an LC filter.

18. The method according to claim 13, wherein the leading reactive power is injected as a pulse-width modulated (PWM) output or a notched PWM output.

19. The method according to claim 12 or 13, wherein the high-speed mechanical generator is selected from the group consisting of a wound-field synchronous machine, a switched magnetoresistive machine, a permanent magnet machine, and a permanent magnet synchronous motor machine.

20. The method according to claim 12 or 13, wherein the high-speed mechanical generator is a single-phase output machine or a three-phase output machine.

Images