MULTI-LEVEL MEDIUM VOLTAGE DATA CENTER STATIC SYNCHRONOUS COMPENSATOR (DCSTATCOM) FOR ACTIVE AND REACTIVE POWER CONTROL OF DATA CENTERS CONNECTED WITH GRID ENERGY STORAGE AND SMART GREEN DISTRIBUTED ENERGY SOURCES

Abstract

Systems and methods for supplying power (both active and reactive) at a medium voltage from a DCSTATCOM to an IT load without using a transformer are disclosed. The DCSTATCOM includes an energy storage device, a two-stage DC-DC converter, and a multi-level inverter, each of which are electrically coupled to a common negative bus. The DC-DC converter may include two stages in a bidirectional configuration. One stage of the DC-DC converter uses a flying capacitor topology. The voltages across the capacitors of the flying capacitor topology are balanced and switching losses are minimized by fixed duty cycle operation. The DC-DC converter generates a high DC voltage from a low or high voltage energy storage device such as batteries and/or ultra-capacitors. The multi-level, neutral point, diode-clamped inverter converts the high DC voltage into a medium AC voltage using a space vector pulse width modulation (SVPWM) technique.

Claims

1. A transformerless DCSTATCOM for an IT electrical load, comprising: an energy storage device; a two-stage bi-directional DC-DC converter; a multi-level inverter outputting a medium AC inverter voltage; and a controller configured to control the magnitude of the medium AC inverter voltage of the multi-level inverter to generate or absorb reactive power to achieve a power factor equal to or approximately equal to unity at PCC, the controller further configured to charge or discharge the energy storage device by adjusting the angle of the medium AC inverter voltage with respect to a medium AC grid voltage, wherein a negative terminal of the energy storage device, a negative terminal of the two-stage DC-DC converter, and a negative terminal of the multi-level inverter are electrically coupled to a common negative bus.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Various embodiments of the present disclosure are described with reference to the accompanying drawings wherein:

[0027] FIG. 1 is a schematic block diagram including ‘medium voltage UPS utilizing step-up transformer’ of a power supply system for a data center according to the prior art;

[0028] FIG. 2 is a schematic block diagram including ‘medium voltage UPS utilizing step-up transformer’ and STATCOM with reactive power control of a power supply system for a data center according to the prior art;

[0029] FIG. 3 is a schematic block diagram including a multi-level DCSTATCOM without any output transformer with both simultaneous active and reactive power control capability connected at upstream PCC of a data center according to embodiments of the present disclosure;

[0030] FIG. 4 is a schematic block diagram including a multi-level DCSTATCOM without any output transformer with both simultaneous active and reactive power control capability connected at downstream PCC of a data center according to embodiments of the present disclosure;

[0031] FIG. 5 is a circuit diagram of an embodiment of a multi-level two-stage bidirectional DC-DC converter of the DCSTATCOM of FIG. 3 or 4;

[0032] FIG. 6 is a circuit diagram of an embodiment of a five-level diode-clamped inverter of the DCSTATCOM of FIG. 3 or 4;

[0033] FIG. 7 is a circuit diagram of the multi-level two-stage bidirectional DC-DC converter of FIG. 5 coupled to the five-level diode-clamped inverter of FIG. 6;

[0034] FIG. 8 is a space-vector modulation diagram showing switching states for Sector A of the 5-level inverter of FIG. 6;

[0035] FIG. 9 shows the sequence of switching states and waveform of phase U for region 1 in Sector A (U.sub.A1) of the space-vector modulation diagram of FIG. 8;

[0036] FIG. 10 is a block diagram of an overall control system of DCSTATCOM;

[0037] FIGS. 11-13 are flowcharts illustrating methods for controlling the flow of power between a transformerless DCSTATCOM and an electrical load;

[0038] FIG. 14 is a table showing various operating conditions of DCSTATCOM; and

[0039] FIG. 15 is a table showing cost savings per year and corresponding freeing up of data center capacity for four different cases.

DETAILED DESCRIPTION

[0040] Embodiments of the present disclosure are described in detail with reference to the drawing figures wherein like reference numerals identify similar or identical elements.

[0041] The present disclosure relates to multi-level, transformer-less DCSTATCOM system that includes a multi-level DC-DC converter and a multi-level inverter coupled together. The efficiency of a conventional STATCOM using a transformer is about 96%. In contrast, the transformer-less DCSTATCOM according to present disclosure can achieve efficiencies of about 97%.

[0042] STATCOM (Static Synchronous Compensator), which is a family of FACTS (Flexible AC Transmission System) controllers, is a shunt connected voltage source inverter and is connected to the grid through a smoothing reactor, as shown in FIG. 3.

[0043] Existing STATCOM generates low voltage AC output through a two-level inverter. Therefore, it requires a transformer at its output to match the utility voltage value (for example, 13.8 kV).

[0044] In this application, active energy storage is not possible for the STATCOM 60 (FIG. 2). The STATCOM 60 is not suitable to function as a UPS or to store energy from the utility power feed 20′ supplied from the utility grid (not shown). Only reactive power compensation is possible. The step-up transformer 68 that is required at the output of the smoothing reactor 66 is a large, bulky device which requires extra space and imposes additional weight and cost in addition to incurring power losses.

[0045] FIG. 3 illustrates a DCSTATCOM applied to a utility power feed of a data center to control both active (to provide uninterruptible power supply (UPS), grid energy storage, peak demand supply operations) and reactive power (to provide PF corrector, grid voltage stiffness, grid voltage transient stabilizer operations) according to one embodiment of the present disclosure. More particularly, new or existing data center 110 has a utility-data center boundary interface 105 defining a utility side 151 and a data center side 152. On the utility side 151, a utility power feed 120 supplies, on data center side 152, power to double conversion AC-DC/DC-DC server power supplies 45a′-45n′ with an AC input and a DC output. AC output from the step-down transformer 135 is supplied via AC feed line 124 to IT loads 55 and mechanical loads 60 via 132. A generator 115 on the data center side 152, which connects to transfer switch 130, starts to operate once a disturbance in the utility power feed 120, e.g., a loss of all or a portion of the electricity provided by the utility power feed 120, is more than approximately two seconds. During a disturbance in the utility feed 120, a surge protector 25 dampens the disturbance. For disturbances beyond a pre-determined acceptable level that are beyond the dampening capabilities of the surge protector 25 or during a utility black out condition, the IT loads 55 and mechanical loads 60 are now powered by a medium voltage DCSTATCOM 260 located on the utility side 151 of the utility-data center interface boundary 105.

[0046] Medium voltage DCSTATCOM 260 is electrically coupled to the utility power feed 120 at a point of common coupling 172 on the utility side 151. In contrast to STATCOM 60 of FIG. 2, DCSTATCOM 260 includes a DC-DC converter 262 whose DC output 265 is electrically coupled as input to a multi-level inverter 268. The DC-DC converter 262 is supplied power via an energy storage device 264. The AC output of the multi-level inverter is electrically coupled to the PCC 172 through a smoothing reactor or inductor 266 in feed line 170 that electrically couples the smoothing reactor 266 to the utility power feed line 120 at PCC 172, which, in turn, supplies AC power to the transfer switch 130. When an interruption or disturbance in the power supplied by the utility supply 120 is detected, the STS 130 opens and the DCSTATCOM system 260 starts supplying about 100% of the total power to the loads 55 and 60. The DCSTATCOM system 260 can supply power to the loads 55 and 60 for a short period, e.g., approximately five minutes, but generally the generator 115 or on-site green power starts generating power if the interruption is more than a few seconds. It reduces data center CAPEX/OPEX cost as there is no UPS in the data center and there are no corresponding UPS losses that reduce PUE of the data center.

[0047] DCSTATCOM also provides reactive power at PCC to maintain a unity power factor of the devices upstream from the PCC. This reactive power compensation avoids a penalty bill from the utility and reduces utility component (transformer, cable (not shown)) heat loss by 1.4% at 0.85 PF. It also frees up 19.25% capacity of utility components (transformer, cable) at 0.85 PF. The IT server loads 55 are supplied power via double conversion AC-DC/DC-DC power supplies 45a′-45n′.

[0048] FIG. 4 illustrates a DCSTATCOM connected at downstream of data center (up to and including the static transfer switch) to control both active (to provide uninterruptible power supply (UPS), grid energy storage, peak demand supply operations) and reactive power (to provide PF corrector, grid voltage stiffness, grid voltage transient stabilizer operations).

[0049] DCSTATCOM also provides reactive power at PCC to maintain a unity power factor of the devices at the utility upstream (FIG. 3) and downstream (FIG. 4) of the data center. This reactive power compensation avoids a penalty bill from the utility and may reduce heat losses in utility components (transformer, cable (not shown)), generator, transfer switch, and static transfer switch by 1.4% at 0.85 PF. It also frees up about 20% capacity of the utility components (transformer, cable), generator, transfer switch, and static transfer switch of the data center at 0.85 PF.

[0050] FIG. 5 depicts DC-DC converter 262, described above with respect to FIG. 3 (262), which is a bi-directional two-stage DC-DC converter 224, 226. The first DC-DC stage 224 converts the voltage from the energy storage device 264 into voltage V1. Voltage V1 is a DC voltage higher than the voltage of the energy storage device 264. The second DC-DC stage 226 converts the voltage V1 into voltage V2, which is higher than voltage V1. The voltage boost from the first and second stages 224, 226 can range from about 1:5 to about 1:10. The voltage boost of the DC-DC converter 262 can be adjusted by changing the size of the switches at each level, the number of stages, and/or the number of levels in each stage. The optimum boost voltage requirement is based on the given voltage of the energy storage device 264 and the required voltage output from the inverter 268. For lower voltage outputs from the inverter 264 the boost voltage ratio can be lower. For higher voltage outputs from the inverter 264 the boost voltage ratio can be higher. The efficiency of the DC-DC converter 224 is reduced when the boost ratio is greater than about 7.

[0051] In FIG. 5, output capacitor C2 and inductor L2 connect the first stage 224 to the second stage 226. More particularly, inductor L2 is connected from a positive junction 2245 to a positive junction 2260, which forms a common positive junction for the second stage 226.

[0052] The first stage 224 of the DC-DC converter 262 is shown as a bidirectional, two-level DC-DC converter having one insulated gate bipolar transistor (IGBT) switch S1 connected in series with another IGBT switch S2. The switches S1 and S2 are connected to the energy storage device 264 through an LC filter, which includes capacitor C1 and inductor L1. Capacitor C1 is connected in parallel across the terminals of energy storage device 264 from junction 2241 on the negative terminal to junction 2242 on the positive terminal. Inductor L1 is connected from the positive junction 2242 to the collector terminal of switch S1 at junction 2243.

[0053] The switch S1 is connected from the positive junction 2243 to junction 2244 on the negative terminal side of energy storage device 264, which is at an equipotential with junction 2241. Capacitor C2 is connected from positive junction 2245 to negative junction 2246 with is at an equipotential with junctions 2241 and 2244. Voltage V1 is the potential difference between junction 2245 and junction 2246 across capacitor C2. Thus, switch S2 and capacitor C2 are connected in series with respect to the energy storage device 264.

[0054] If the switch S1 is formed into a boost converter, the first stage 224 may provide a range of duty or boost ratios. For example, as shown in Table 1 below, the boost ratio may range from 0 to 0.9. Thus, if the input voltage (VS) to the first stage 224 is about 1 kV, the output voltage (V1) ranges from 1 kV to 10 kV depending on the value of the boost ratio, as shown in Table 1. The voltage V1 varies depending upon the inductance of L1 multiplied by the rate of change of current di/dt. As used herein, voltage V1 refers to the voltage output of the first stage of a DC-DC converter. Also, as used herein, voltage V2 refers to the output voltage of the final stage of a DC-DC converter.

TABLE-US-00001 TABLE 1 VS (~1 kV) Duty (Boost) ratio V1 1 kV 0 1 kV 1 kV 0.2 1.25 kV 1 kV 0.4 1.66 kV 1 kV 0.6 2.5 kV 1 kV 0.7 3.3 kV 1 kV 0.8 5 kV 1 kV 0.9 10 kV

[0055] The IGBT in switch S1 may be configured in such a way as to handle a lower voltage and a higher current. Furthermore, because the IGBT of switch S1 is handling a lower voltage, the overall size of the IGBT may be smaller.

[0056] FIG. 5 shows an embodiment of the DC-DC converter 262 of FIG. 3, which is a two-stage, bidirectional DC-DC converter. The two-stage bidirectional DC-DC converter 262 can be used to supply power from the energy storage device 264 to the load 152 when power from the generator 115 or utility supply 120 is interrupted or to charge the energy storage device 264 with power from the utility supply 120 when the utility supply 120 is supplying power to the data center load 152.

[0057] The two-stage bidirectional DC-DC converter 262 is a bi-directional converter 262 of FIG. 3. Switches S1 and S20-S23 are used to supply power (discharging of energy storage 264) to the load 155 and switches S2 and S24-S27 are used to charge the energy storage device 264. In particular, switch S1 is configured as a boost converter that converts the voltage Vs of the energy storage device 264 to a higher voltage and switch S2 is configured as a buck converter that converts voltage from the utility supply 120 to a lower voltage appropriate for charging the energy storage device 264, e.g., a voltage slightly more than Vs.

[0058] Voltage V201 is measured across switches S20 and S27 and capacitor C3 from junction 2261 to junction 2271. Voltage V202 is measured across switches S21 and S26 and capacitors C4 and C5 from junction 2262 to junction 2272. Voltage V203 is measured across switches S22 and S25 and capacitors C6, C7, and C8 from junction 2263 to junction 2273. Voltage V2 is then measured across switches S23 and S24 and capacitors C9, C10, C11, and C12 from junction 2264 to junction 2274.

[0059] Each of the switches S20-S27 outputs a voltage equal to the input voltage V1. Thus, the capacitance of capacitor C9 equals the capacitance of capacitor C2, the capacitance of capacitor C10 equals the capacitance of capacitor C2, the capacitance of capacitor C11 equals the capacitance of capacitor C2, and the capacitance of capacitor C12 equals the capacitance of capacitor C2. Since the switches S20-S27 are connected in series, the output voltage V2 is equal to the sum of the voltages output from each of the switches S20-S27. Thus, the boost ratio is 4:1 and V2 equals 4×V1.

[0060] The capacitors C3-C12 are relatively small capacitors, e.g., capacitors rated for about 5 kV with a capacitance value that is about ten times less than a capacitor for a conventional DC-DC converter. For example, if a conventional two-level DC-DC converter needs a capacitor having a value of about 2000 μF, then the multi-level flying capacitor arrangement (e.g., C3-C12) needs a capacitor having a value of about 200 μF. In a five-level arrangement, each switch S20-S23 operates at a fixed duty cycle of 25% and a fixed switching frequency without pulse width modulation. The voltages V201, V202, V203, and V2 across the capacitors C3-C12 may be balanced in every switching cycle due to fixed duty cycle operation. Additionally, the voltage across each switch S20-S23 maintains 25% of the high voltage V2.

[0061] For a conventional one-stage DC-DC converter, the boost ratio is about 1:18 to about 1:24 for lower energy storage voltages, e.g., 1 kV. The efficiency of a DC-DC converter is reduced when the high boost conversion ratio is greater than about 7. For the two-stage DC-DC converter 220, 230, or 234, the boost ratio of each stage is about 1:4 to about 1:6. In the case of the DC-DC converter 262 of FIG. 5, the voltage of the energy storage device is high (e.g., about 4 kV to about 6 kV), which reduces the boost conversion ratio to around 5 to 7. This improves the efficiency of the DC-DC converter 262.

[0062] As shown in FIG. 5, the number of capacitors coupled in series between the collectors of switches arranged in the upper portion of a stage and the emitters of the switches arranged in the lower portion of the stage depends on the level of the switch to which the capacitors are coupled. The DC-DC converter 262, however, may include any number of capacitors coupled in series between the collectors and emitters of appropriate switches to achieve a desired result. The DC-DC converter 262 of FIG. 5 is a five-level converter in flying capacitor configuration.

[0063] FIG. 6 shows inverter 268 which may be used to convert the DC voltage output V2 from the converter 262 to three-phase AC voltage V3. FIG. 6 shows a five-level diode-clamped inverter 268. The five-level inverter 268 includes three groupings of switches and diodes 242, 244, and 246 to generate the three phases V3a, V3b, and V3c of the AC voltage V3, which is the output voltage of the inverter 240. Each grouping of diodes D30-D35, D40-D45, and D50-D55 and corresponding switches S30-S37, S40-S47, and S50-S57 are connected together in a diode-clamped configuration.

[0064] Switches S30-S37, S40-S47, and S50-S57 may be IGBTs. IGBTs allow for higher voltages/currents and higher switching frequencies. The five-level inverter 240 illustrated in FIG. 6 allows for sharing of the high voltage among the switches S30-S37, S40-S47, and S50-S57, and reduces harmonic distortion.

[0065] The switches S30-S37, S40-S47, and S50-S57 are controlled by a microprocessor (not shown) such as a digital signal processor (DSP) (not shown). The DSP may use a space vector pulse width modulation (SVPWM) technique for operating the switches S30-S37, S40-S47, and S50-S57 in such a way that the neutral-point voltage remains balanced in open-loop operation. The SVPWM technique is an inverter modulation technique for synthesizing a voltage space vector V* (described below with respect to FIG. 8) over a modulation sampling period T.sub.s (see FIG. 9 discussed below).

[0066] The SVPWM technique provides the advantages of superior harmonic quality and large under-modulation range that extends the modulation factor from 78.5% to 90.7%. Alternatively or in addition to the SVPWM, an artificial neural network (ANN) control technique can be used to reduce harmonics outputted from the inverter 268.

[0067] FIG. 7 shows the combination of the converter 262 of FIG. 5 and the inverter of FIG. 6. The capacitors C9-C12 may be the same as capacitors C13-C16, respectively. Additionally, the converter/inverter combination includes smoothing reactors (266 a-c) at the output of the inverter.

[0068] Space vector pulse width modulation of three-level inverters with respect to neural networks is described in “A Neural-Network-Based Space-Vector PWM Controller for a Three-Level Voltage-Fed Inverter Induction Motor Drive”, by Subrata K. Mondal, Joao O. P. Pinto and Bimal K. Bose, published in IEEE Transactions on Industry Applications, Vol. 38, No. 3, May/June 2002, Paper IPCSD 02-005, presented at the 2001 Industry Applications Society Annual Meeting, Chicago, Ill., Sep. 30-Oct. 5, 0093-9994 ©2002 IEEE, and in “Neural-Network-Based Space-Vector PWM of a Three-Level Inverter Covering Overmodulation Region and Performance Evaluation in Induction Motor Drive”, by Cong Wang, Bimal K. Bose, Valentin Oleschuk, Subrata Mondal, and Joao O. P. Pinto, 0-7803-7906-3/03 ©2003 IEEE, the entire contents of both of which are hereby incorporated by reference herein.

[0069] Additionally, space vector pulse width modulation of three-level inverters is described in “Space Vector Pulse Width Modulation of Three-Level Inverter Extending Operation Into Overmodulation Region,” by Subrata K. Mondal, Bimal K. Bose, Valentin Oleschuk and Joao O. P. Pinto, published in IEEE Transactions on Power Electronics, Vol. 18, No. 2, March 2003, 0885-8993 ©2003 IEEE, the entire contents of which is hereby incorporated by reference herein.

[0070] FIG. 8 is a space-vector modulation diagram 300 showing switching states for Sector A of the five-level inverter 268 of FIG. 6 according to embodiments of the present disclosure.

[0071] FIG. 9 shows the sequence of switching states of phase U for region 1 in Sector A (Um) of the space-vector modulation diagram 300 in FIG. 8. The switching states for the space-vector modulation diagram 300 are such that the sequence of switching causes balancing of the voltages across the capacitors C13, C14, C15, and C16 of the five-level inverter 240 of FIG. 6 in open loop operation.

[0072] Referring to FIGS. 8 and 9, space-vector modulation diagram 300 is formed by a hexagon 302. The hexagon 302 for the five-level design has six sectors, i.e., sectors A, B, C, D, E, and F, each of which has sixteen regions (1-16), giving altogether 96 regions of operations, i.e., 16 regions×6 sectors=96 regions of operations.

[0073] There are 125 switching states in five-level inverters, such as the five-level diode-clamped inverter 268 of FIG. 6. There are 120 active states and the remaining five states are zero states occurring at the center point V.sub.0 of the diagram 300. U, V, and W (designated as V3a, V3b, and V3c, respectively, in FIG. 7) are the phases, and P.sub.2, P.sub.1, O, N.sub.1, and N.sub.2 are DC-bus points. The number of switching states is determined by raising the number of levels, e.g., five, to the power of the number of phases, e.g., three for phases U, V, and W. Therefore, the number of switching states is 125 (5.sup.3). Since there are six sectors, i.e., Sectors A, B, C, D, E, and F, with 20 active states per sector, the total number of active states is 120 (6×20). The active states are those states extending beyond the center point V.sub.0.

[0074] As described above, the SVPWM technique is an inverter modulation technique for synthesizing a voltage space vector V*. In FIG. 8, voltage space vector V* originates at the center point V.sub.0. In the example shown, the voltage space vector V* is characterized by a constant voltage value represented by a first circle 310 so that the voltage space vector V* may rotate around the center point V.sub.0. Therefore, all switching states at the circumference of the first circle 310 are at the same voltage V*. Similarly, if the voltage space vector V* is characterized by a constant voltage value represented by a second circle 312 that is concentric with first circle 310, the voltage space vector V* then assumes a constant voltage represented by the second circle 312.

[0075] In the example shown, since the second circle 312 has a radius that is greater than the radius of the first circle 310, the constant voltage represented by the second circle 312 is greater than the constant voltage represented by the first circle 310. Alternatively, if the second circle 312 were to have a radius that is less than the radius of the first circle 310, then the constant voltage represented by the second circle 312 would be less than the constant voltage represented by the first circle 310.

[0076] Table 2 below illustrates the switching states for switches SX0-SX7 of the inverter 260, where X is 3, 4, or 5. Operation of each set of switches SX0-SX7 of FIG. 7 produces a phase of the three-phase AC output. In Table 2, the closing of a switch is represented by the numeral “1” and the opening of a switch is represented by the numeral “0.”

TABLE-US-00002 TABLE 2 (where X = 3, 4, or 5) Switching State SX0 SX1 SX2 SX3 SX4 SX5 SX6 SX7 P2 1 1 1 1 0 0 0 0 P1 0 1 1 1 1 0 0 0 O 0 0 1 1 1 1 0 0 N1 0 0 0 1 1 1 1 0 N2 0 0 0 0 1 1 1 1

[0077] In FIG. 6, junction 2400 represents state O, which represents neutral point balancing so that the average current injected at O should be zero. Voltage V2 is measured at junction 2413 and 2414 located between capacitors C13 and C16. States P1 and P2 represent positive bus voltages. States N1 and N2 represent negative bus voltages.

[0078] State P1 is represented by a voltage at junction 2401 between capacitors C14 and C13. State P2 corresponds to a voltage at junction 2413 on common positive bus 2411 that electrically couples junction 2400, capacitor C14, junction 2401, and capacitor C13 to junction 2420 for phase V3a or U. Switches S30, S31, S32, and S33 are electrically coupled to common positive bus 2411 at junction 2402 via the collector side of switch S30. Similarly, state N1 corresponds to a voltage at junction 2401′ between capacitors C15 and C16. State N2 corresponds to a voltage at junction 2414 on common negative bus 2412 that electrically couples junction 2400, capacitor C15, junction 2401′, and capacitor C16 to junction 2420 for phase V3a or U. Switches S34, S35, S36, and S37 are electrically coupled to common negative bus 2412 at junction 2402′ via the emitter side of switch S37.

[0079] Referring again to Table 2, in conjunction with FIG. 7, phase W (V3c in FIG. 7), for example, is in state P2 (positive bus voltage) when the switches S30, S31, S32, and S33 are closed or “ON”, and switches S34, S35, S36, and S37 are open or “OFF”. The phase W is in state P1 (positive bus voltage that is less than P2) when switches S30, S35, S36, and S37 are open or OFF, and switches S31, S32, S33, and S34 are closed or ON. The phase W is in state O when switches S30, S31, S36, and S37 are open or OFF and switches S32, S33, S34, and S35 are closed or ON.

[0080] The phase W is in state N1, which corresponds to a negative bus voltage that is greater than a negative bus voltage that corresponds to state N2, when switches S30, S31, S32, and S37 are turned off (i.e., open) and switches S33, S34, S35, and S36 are turned on (i.e., closed). The phase W is in state N2, which corresponds to a negative bus voltage that is less than the negative bus voltage that corresponds to state N1, when switches S30, S31, S32, and S33 are turned off (i.e., open) and switches S34, S35, S36, and S37 are turned on (i.e., closed).

[0081] In FIG. 9, the states P2, P1, O, N1, and N2, and waveform 320 are plotted versus a sampling period Ts or symmetrically over half a sampling period Ts/2 for the phase U.sub.A in Sector A. The top portion of FIG. 9 also shows the switching states P2, P1, O, N1, and N2 of all three phases UA, VA, and WA. The modulation strategy illustrated in FIG. 7 is a DSP-based SVPWM modulation strategy for a 5-level DCSTATCOM system, e.g., the 5-level inverter 268 of FIG. 6.

[0082] It should be noted that in addition to the space-vector pulse width modulation method, those skilled in the art will recognize and understand that, as described in the publications referenced above, artificial neural network (ANN) control may be applied for the modulation of the switching states of the 5-level inverter 268 of FIG. 6.

[0083] DCSTATCOM, without an output transformer, has a small footprint as it replaces the output transformer by a compact power electronic voltage converter. It significantly improves transient stability and regulates dynamic voltage at PCC (Point of common coupling). It also regulates both lag and lead reactive power. Therefore, DCSTATCOM provides stable voltage for a weak grid along with continuous reactive power regulation.

[0084] DCSTATCOM can also be controlled to manage active power. DCSTATCOM acts as active power sources like UPS if a DC energy source is available at the input of the DCSTATCOM inverter. So, UPS is not needed in the data center if DCSTATCOM is connected at the input supply of the data center, which significantly reduces CAPEX cost of the data center. In embodiments, batteries and ultra-capacitors may be used as energy sources to supply active power. Also, in embodiments, the generator may be taken out from the data center design if on-site distributed power like solar, fuel cells, wind, etc. is available to be used as back-up power. This also reduces CAPEX costs of a proposed data center.

[0085] Available battery voltage of energy storage is a low value (approximately 700 to 1000 V DC). So two stages of DC-DC boost converters are needed to boost voltage efficiently as shown in FIG. 5. A multi-level inverter is needed at the output of the two stages of the DC-DC boost converters to generate 13.8 KV AC directly without any step-up transformer as shown in FIG. 6. Therefore, the overall footprint of the proposed DCSTATCOM without an output transformer would be a smaller footprint, as shown in FIG. 7.

[0086] The DCSTATCOM at the input supply of the data center takes out significant CAPEX costs from the data center as the DCSTATCOM is considered to be part of the utility side. The DCSTATCOM of the present disclosure improves voltage stability of both grid and on-site distributed sources at PCC near the data center. It also improves overall PF of the data center with a reactive component load.

[0087] In comparison to traditional STATCOM, which compensates for reactive power, DCSTATCOM compensates for both active and reactive power. DCSTATCOM needs a smaller capacitor with respect to traditional STATCOM to smooth battery current due to the integration of battery and capacitor in parallel mode. Depending on the data center back-up time requirement, a number of battery cells can be connected in parallel.

[0088] FIG. 10 is a block diagram of the overall DCSTATCOM control system including the DCSTATCOM system controller 1006 and three main sub-system controllers: the Battery Management System (BMS) controller 1000, the DC-DC converter controller 1002, and the inverter controller 1004. Charging and discharging of the battery 264, which may alternatively be an ultracapacitor or a combination a battery and an ultracapacitor, is controlled by BMS controller 1000 based on cell temperature and State of Charge (SOC) conditions of the battery 264 and availability of power at PCC. In embodiments, if the temperature of one or more cells of the energy storage device exceeds a predetermined threshold, the BMS controller 1000 may determine that the energy storage device cannot be charged.

[0089] DC-DC converter 262 is controlled by DC-DC converter controller 1002. An example embodiment of the DC-DC converter 262 is illustrated in FIG. 5. The multi-level inverter 268 is controlled by inverter controller 1004. An example embodiment of the multi-level voltage source inverter 268 is illustrated in FIG. 7. An example embodiment of a control methodology implemented by the inverter controller 1004 is illustrated in FIGS. 8 and 9.

[0090] The three sub-system controllers 1000, 1002, and 1004 are controlled by the DCSTATCOM system controller 1006. The DCSTATCOM system controller 1006 communicates with the grid and receives grid voltage and current information. The DCSTATCOM system controller 1006 also communicates with other energy sources, e.g., a solar energy source, connected at PCC using SCADA of the Smart Grid controller to decide control function at any particular instant.

[0091] DCSTATCOM has dual independent simultaneous control capability of both active and reactive power due to the integration of a battery storage system with traditional STATCOM. The following equations indicate that 0 (the angle between V.sub.GRID and V.sub.INV) and V.sub.INV can be controlled to control active (P) and reactive (Q) power of DCSTATCOM respectively:

P=3*V.sub.GRID*V.sub.INV*sin ∂/ω*L, and

Q=3*V.sub.GRID*(V.sub.INV−V.sub.GRID)/ω*L,

where V.sub.GRID=grid/utility voltage, V.sub.INV=inverter voltage, ω=line frequency, and L=smoothing reactor.

[0092] Due to implementation of fast power electronics, IGBT-based devices, the controller has the ability to effect changes (+Q capacitive VAR compensation to −Q inductive VAR compensation) within a quarter of a cycle. By controlling output voltage DCSTATCOM (V.sub.INV) magnitude higher or lower than the PCC grid system voltage (V.sub.GRID), the system is able to regulate line voltage by absorbing or generating VAR. The ability of operation of DCSTATCOM with energy storage in four quadrant operation (both reactive plus active power operation) benefits/reduces data center CAPEX and OPEX costs.

[0093] Flow of reactive power can be controlled by adjusting the inverter voltage (VINO. The following conditions happen: [0094] 1. Inverter of DCSTATCOM acts as a generator of reactive power (Q>0) when ‘Magnitude of V.sub.INV’ is greater (>) than ‘Magnitude of V.sub.GRID’. The grid sees DCSTATCOM as a capacitance connected to its terminals, i.e. DCSTATCOM is seen to provide capacitive VARS to the system. [0095] 2. Inverter of DCSTATCOM acts as an absorber of reactive power (Q<0) when ‘Magnitude of V.sub.INV’ is less (<) than ‘Magnitude of V.sub.GRID’. The grid sees DCSTATCOM as an inductance connected to its terminals. [0096] 3. Reactive power exchange is zero (Q=0) when |V.sub.INV|=|V.sub.GRID|.

[0097] The amount of reactive power depends on the magnitude of the voltage V.sub.INV and on the drop across smoothing reactor L. The fundamental component of the voltage V.sub.INV is controlled by varying the DC capacitor voltage (V.sub.DCCAP). The inverter voltage of DCSATCOM (V.sub.INV) is normally kept in phase (i.e., ∂=0) with the grid voltage (V.sub.GRID), but the voltage angle is temporarily phase shifted in order to vary the V.sub.DCCAP.

[0098] Flow of active power can be controlled by adjusting the phase angle (∂) of the inverter voltage of DCSTATCOM (V.sub.INV) and grid voltage (V.sub.GRID). The following conditions happen: [0099] 1. DCSTATCOM absorbs real power (P<0) from Grid and battery is charged (rectification mode) if V.sub.INV lags V.sub.GRID by angle ∂. [0100] 2. DCSTATCOM supplies real power (P>0) to grid and data center and battery is discharged (inversion mode) if V.sub.INV leads V.sub.GRID by angle ∂.
FIG. 14 shows the various operating conditions of DCSTATCOM. The active and reactive power control may be implemented by a digital signal processor (DSP) or other suitable processor of the DCSTATCOM system controller 1006 of FIG. 10.

[0101] FIG. 11 is a flowchart of a method for exchanging power between a transformerless DCSTATCOM and an electrical load. After starting in step 1101, a first DC voltage is supplied from a low voltage energy storage device to a DC-DC converter in step 1102. In step 1104, the first DC voltage is converted into a second DC voltage and, in step 1106, the second DC voltage is provided to a multi-level inverter. In step 1108, an AC inverter voltage is generated from the second DC voltage. The AC inverter voltage is a medium voltage less than the second DC voltage. In step 1110, the AC inverter voltage of the multi-level inverter is controlled to generate or absorb reactive power to achieve a power factor at PCC equal to or approximately equal to unity. Before ending in step 1113, the energy storage device is charged or discharged by adjusting the phase angle of the AC inverter voltage with respect to the AC grid voltage.

[0102] FIG. 12 is a flowchart of a method for compensating for reactive power according to embodiments of the present disclosure. After starting in step 1201, the AC grid voltage and the AC grid current are measured in step 1202. In step 1204, a phase angle (Θ) between the measured AC grid voltage and the measured AC grid current is calculated. In step 1206, reactive power is calculated based on the phase angle, the measured AC grid voltage, and the measured AC grid current. In step 1207, it is determined whether the measured AC grid current is leading or lagging with respect to the measured AC grid voltage.

[0103] If it is determined that the measured AC grid current is leading with respect to the measured AC grid voltage, the calculated reactive power is generated by adjusting the magnitude of the AC inverter voltage so it is greater than the magnitude of the AC grid voltage in step 1208. If it is determined that the measured AC grid current is leading with respect to the measured AC grid voltage, the calculated reactive power is absorbed by adjusting the magnitude of the AC inverter voltage so it is less than the magnitude of the AC grid voltage in step 1210. Then, the method ends in step 1211.

[0104] FIG. 13 is a flowchart of a method for charging and discharging an energy storage device according to embodiments of the present disclosure. After starting in step 1301, the AC grid voltage, the AC grid current, and the phase angle between the AC grid voltage and the AC grid current is measured in step 1302. In step 1304, the power factor is calculated based on the measured phase angle. In step 1306, the active power demand at the grid PCC is calculated based on the measured AC grid voltage, the measured AC grid current, and the calculated power factor. Then, in step 1307, it is determined whether to charge or discharge the energy storage device based on the calculated active power demand.

[0105] If it is determined that the energy storage device should be charged, the angle of the AC inverter voltage is adjusted so that it lags with respect to the AC grid voltage in step 1308. If it is determined that the energy storage device should be charged, the AC inverter voltage is adjusted so that it leads with respect to the AC grid voltage in step 1310. Then, in step 1311, the method ends.

[0106] The methods according to the present disclosure may be performed by any combination of the DCSTATCOM system controller 1006, the Battery Management System (BMS) controller 1000, the DC-DC converter controller 1002, and the inverter controller 1004. One or more of these controllers may include sensors for measuring voltage, current, phase angle, etc., a suitable processor and memory for performing calculations and other functions based on these measurements, and any other hardware, firmware, and/or software components necessary for carrying out the method steps of the present disclosure.

[0107] FIG. 15 shows cost savings per year and corresponding freeing up data center capacity for four different cases: (1) at 0.85 PF, (2) at unity PF with existing reactive power STATCOM, (3) at unity PF with DCSTATCOM at upstream of data center, and (4) at unity PF with DCSTATCOM at downstream of data center). It is assumed that IT load is 11\4 W and PUE is 1.5.

[0108] In one example, DCSTATCOM, which includes both STATCOM and UPS functions, not only may save the above demand penalty of $16,733.10 per year, but also may reduce data center energy cost of $78,840/year for a 1 MW IT load at 1.5 PUE due to the elimination of MVUPS loss and cable copper loss (5%) (1.5 PUE×1000 kW×0.05 (losses)×24 hours×365 days×$0.12=$78,840/year). DCSTATCOM may also free up 19.25% of data center capacity (e.g., Transfer switch, Static Transfer Switch, Generator) if it is placed downstream (after Static Transfer Switch) from the data center source.

[0109] DCSTATCOM integrates multiple functions (like STACOM VAR compensator, active power support, active filtering, voltage/frequency support, power quality/reliability) into one system controller. DCSTATCOM leverages the same capital investment ($/kW of DCSTATCOM and MVUPS are in same range) and generates better cost-benefit ratio. DCSTATCOM also eliminates UPS requirements in data centers and, in turn, reduces CAPEX of data centers. DCSTATCOM integrates grid and data center functionality into one solution.

[0110] While several embodiments of the disclosure have been shown in the drawings and/or discussed herein, it is not intended that the disclosure be limited thereto, as it is intended that the disclosure be as broad in scope as the art will allow and that the specification be read likewise. Therefore, the above description should not be construed as limiting, but merely as exemplifications of particular embodiments. Those skilled in the art will envision other modifications within the scope and spirit of the claims appended hereto.