WO2015130289A1

WO2015130289A1 - Method and apparatus for coordinated reactive power control in a wind park

Info

Publication number: WO2015130289A1
Application number: PCT/US2014/019014
Authority: WO
Inventors: Debrup DAS; Reynaldo Nuqui; Zhenyuan Wang
Original assignee: Abb Technology Ag
Priority date: 2014-02-27
Filing date: 2014-02-27
Publication date: 2015-09-03

Abstract

According to one aspect of the teachings herein, one or more converters, such as Variable Speed Drives or other AC-to-DC or AC-to-DC-to-AC converters in an electric power distribution network, are synchronized with respect to a central controller that is configured to control the one or more converters with respect to one or more network objectives. Example objectives includes reactive power consumption targets, Total Harmonic Distortion, THD, limits, reduced power consumption via Conservative Voltage Reduction, CVR, control, and overload mitigation.

Description

METHOD AND APPARATUS FOR COORDINATED REACTIVE POWER

CONTROL IN A WIND PARK

TECHNICAL FIELD

The present invention generally relates to electric power distribution systems and particularly relates to coordinated control of power-electronic converters in such systems.

BACKGROUND

Power-electronic converters find increasing use in electric power distribution networks. While converter architectures vary, a general class of converters may be regarded as having a "network side" facing the distribution network or grid and a "load side" facing the load(s). The network side of the converter may consist of various semi-conductor based switches. Where the network side of the converter consists of one or more controllable switches, the converter is said to have an "active front end" or AFE. Examples of controllable semi-conductor switches include IGBT, MOSFET, etc., while examples of uncontrollable semi-conductor switches include diodes.

Hence a converter can be either have a "diode front end" or an AFE The power-electronic switches of the AFE enable control of the network-side voltage or reactive power of the converter at the network node, where the converter is connected to the network. Example converters that may be implemented with AFEs include AC-to-DC converters and AC-to-DC-to- AC converters, such as may be used for powering and controlling industrial drives within a factory or other industrial distribution network.

SUMMARY

According to one aspect of the teachings herein, one or more converters, such as Variable Speed Drives or other AC-to-DC or AC-to-DC-to-AC converters in an electric power distribution network, are synchronized with respect to a central controller that is configured to control the one or more converters with respect to one or more network objectives. Example objectives include reactive power consumption targets, Total Harmonic Distortion, THD, limits, reduced power consumption via Conservative Voltage Reduction, CVR, control, and overload mitigation.

In one example, a central controller is configured for controlling one or more converters that are in an electric power distribution network. Each converter has an active front end, AFE, that is controllable by the converter responsive to commands from the central controller, which thereby allows the central controller to, among other things, effectuate desired reactive power control settings for each converter. The example central controller includes a signaling interface that is configured to receive network feedback comprising indications of bus voltages and line currents for the distribution network, and to receive converter feedback comprising indications of operational status and active and reactive power consumption. The signaling interface is further configured for sending converter control commands to each converter operating under control of the central controller. In at least one embodiment, and particularly where the central controller controls more than one converter, the converters are synchronized with respect to the central controller by way of timing based on the Global Positioning System, i.e., GPS-based timing, or other common timing reference.

Additionally, the central controller includes a processing circuit that is configured to monitor bus voltages and lines currents for the distribution network, based on the network feedback received through the signaling interface, and to monitor active and reactive power consumption at each converter, based on the converter feedback received through the signaling interface. The processing circuit is further configured to compute an actual reactive power consumption of the distribution network at a point of common coupling, PCC, between the distribution network and a supply network that supplies electric power to the distribution network.

Still further, the processing circuit is configured to use an electrical model of the distribution network, including model constraints on bus voltages and line currents, to compute reactive power control settings for the one or more converters that drive the actual reactive power consumption of the distribution network towards a target reactive power consumption. Additionally, the processing circuit is configured to generate converter control commands to effectuate the computed reactive power control settings at the one or more converters and transmit the generated converter control commands to the one or more converters via the signaling interface.

In another example, a method of controlling one or more converters that are in an electric power distribution network includes monitoring bus voltages and lines currents for the distribution network, based on receiving network feedback from one or more monitoring devices configured to determine the bus voltages and line currents, and monitoring active and reactive power consumption at each of the converters, based on receiving converter feedback for each converter. Here, each converter at issue includes an AFE as before, and the method includes computing an actual reactive power consumption of the distribution network at a point of common coupling between the distribution network and a supply network that supplies electric power to the distribution network.

The method further includes computing reactive power control settings for the one or more converters that drive the actual reactive power consumption of the distribution network towards a target reactive power consumption. These settings comprise, for example, a target voltage that each converter is to maintain at its point of connection to the distribution network. Equivalently, the reactive power control settings are expressed in terms of the targeted reactive power consumption of each converter— i.e., the targeted reactive power injection of each converter. In either case, it will be understood that an example converter has local intelligence or operational control, such as is known for conventional converters. Thus, each converter performs its ongoing converter operations, including supplying its attached load(s) as needed, but performs such operations according to the control settings represented by the converter commands it receives from the central controller.

The computation of the reactive power control settings for each converter is determined by the central controller using an electrical model of the distribution network. The model includes or associates with model constraints that stipulate the permissible limits or ranges for bus voltages and line currents in the distribution network. Thus, the method includes constraining the reactive power control settings in accordance with the model constraints, and further includes generating and sending converter control commands to the one or more converters, to effectuate the computed reactive power control settings at the one or more converters.

In another example embodiment, a reactive power control system includes two or more converters coupled to different nodes in an electric power distribution network. Each converter has an AFE that can be configured to control the reactive power consumption of the converter— i.e., to determine the amount the reactive power injected by the converter at its point of connection to the distribution network. The reactive power control system further includes a central controller that is configured to receive network feedback indicating an operating state of the distribution network and converter feedback indicating active and reactive power consumption at each of the two or more converters in the distribution network. The central controller is further configured to control an actual reactive power consumption of the distribution network at a point of common coupling between the distribution network and a supply network supplying electric power to the distribution network. This control is based on the central controller being configured to jointly coordinate the reactive power consumption at each of the two or more converters.

Notably, one or more of the converters may be an "asymmetric converter," having excess power handling capacity on its input side. By virtue of its excess input-side capacity, an asymmetric converter offers additional reactive power resources to the central controller, in the context of the central controller controlling the overall reactive power consumption of the distribution network. Of course, the present invention is not limited to the above features and advantages. Those of ordinary skill in the art will recognize additional features and advantages upon reading the following detailed description, and upon viewing the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a block diagram of one embodiment of a central controller, shown in context with an example electric power distribution network that includes a number of converters, such as AC-to-DC and/or AC-to-DC-to-AC converters.

Fig. 2 is a block diagram of further example details for the central controller introduced in Fig. 1.

Figs. 3 and 4 are block diagrams of example embodiments for a converter having an Active Front- End, AFE, which is directly or indirectly controllable by a central controller, e.g., for controlling the reactive power consumption of the converter.

Fig. 5 is a logic flow diagram of one embodiment of a method of centrally controlling one or more converters in a distribution network, to drive an actual reactive power consumption of the distribution network towards a target reactive power consumption.

Fig. 6 is a logic flow diagram of one embodiment of a method of centrally controlling two or more converters, to reduce an overall Total Harmonic Distortion, THD, in a distribution network.

Figs. 7 and 8 are block diagrams of a distribution network, at least in relevant part, illustrating one approach for determining the equivalent upstream impedance of any given converter in the distribution network, and for projecting the THD caused by a given converter upstream to a point of common coupling, PCC, for example.

Fig. 9 is a logic flow diagram illustrating example details for controlling two or more converters in a distribution network, to effectuate a reduction in the overall net THD at a PCC between the distribution network and a supply network.

Figs. 10 and 11 are a block diagram and a logic flow diagram, respectively, and provide further example details for jointly controlling the reactive power consumptions of two converters in a distribution network, to effectuate a desired reactive power consumption of the distribution network at a PCC between the distribution network and a supply network.

Fig. 12 is a plot of example voltage profiles for a distribution network, with and without Conservative Voltage Reduction, CVR.

Fig. 13 is a logic flow diagram of determining control settings for one or more converters in a distribution network, for implementing CVR control via the converters. DETAILED DESCRIPTION

Fig. 1 illustrates one embodiment of a central controller 10 that is configured to control one or more converters 12 in an electric power distribution network 14, for one or more of reactive power consumption control, Total Harmonic Distortion, THD, control, Conservative Voltage Reduction, CVR, control, and overload mitigation control. As a non-limiting example, the diagram depicts converters 12-1 through 12-6, which include a mix of AC-DC-AC and AC- DC converters.

When not needed for clarity, the reference number "12" is used to refer to converters in the plural and singular senses, e.g., "converters 12" or "converter 12." Each converter 12 of interest with respect to such control has an active front end or AFE, comprising power electronic switches that are actively controllable by the central controller 10, e.g., by way of commanded settings, in terms of conduction or phase angle, to effectuate any of the foregoing control objectives.

The electric power distribution network 14— hereafter "distribution network 14"— couples to a supply network 16 at Point of Common Coupling or PCC 18, which may be associated with a variable transformer 20 on the distribution side of the PCC 18. While the illustrated topology of the distribution network 14 shall be understood as a non-limiting example, in general, the distribution network 14 includes one or more buses 22 and distribution lines 24. Fig. 1 depicts buses 22- 1 through 22-N and provides example identification of distribution lines 24-1 through 24-5. Note that the distribution network 14 may include many distribution lines 24 and buses 22. Further, note that when suffixes are not needed for clarity, the reference number "22" is used to refer to any given bus 22 or buses 22, and the same approach is used for the distribution lines 24.

The various converters 12 are associated with loads 30, e.g., the converter 12-1 supplies a load 30-1, the converter 12-2 supplies a load 30-2, and so on. In this regard, it will be understood that each converter 12 includes a "network side" 26 and a "load side" 28. The network side 26 faces the distribution network 14 at a given point of connection with one of the distribution network 14. The network side 26 of each converter 12 includes an AFE, which can be commanded by the central controller 10, for controlling switching of the AFE, thereby allowing the central controller 10 to effectuate, for example, control of the reactive power consumption of each converter 12.

In this regard, and with respect to any other control objectives implemented by the central controller 10, the central controller 10 is communicatively coupled to the one or more converters 12, e.g., by way of signaling connections not explicitly shown in the diagram. Although the label "CMD" is generically used in the diagram, it will be understood that to the extent that the distribution network 14 includes two or more converters 12, the central controller 10 may generate individualized commands targeted to individual converters 12. Depending on the signaling connections involved, the central controller 10 may have separate communications with each converter 12 and/or the converters 12 may be addressable, thereby allowing a given converter 12 to recognize communications directed to it, while ignoring communications directed to other converters 12.

In the same vein, the central controller 10 can recognize converter feedback from individual ones of the one or more converters 12. This converter feedback is denoted as "CFB" in the diagram and the CFB from one converter 12 may be distinguished from the CFB by another converter 12 based on any one or more mechanisms— such as the order or timing of CFB receipt, the address or ID included in the CFB, the signaling connection over which the CFB is received, etc. The CFB from a given converter 12 indicates any one or more of: the converter status, e.g., online versus offline, active power consumption, reactive power consumption, network-side voltage, converter capacity or other operational limits, current capacity headroom, etc.

As will be later detailed, the central controller 10 uses the CFB as one input— or set of inputs— into its computation of reactive power control settings for the one or more converters 12. The central controller 10 also uses network feedback, denoted as NFB in the diagram, to determine its control of the one or more converters 12. In this respect, it will be appreciated that any number of measuring devices or sensors 32 are strategically located in the distribution network 14, as is known in the art, for monitoring distribution line currents and bus voltages. Such sensors and/or equipment associated with them provide the central controller 10 with at least distribution line current and bus voltage measurements for the distribution network 10.

In an example implementation, the network controller 10 includes a signaling interface 40, which may include one or more physical interface circuits and/or protocol processing circuits. The signaling interface 40 is configured to output the CMD signaling, to receive the CFB and NFB signaling, to provide any number of additional output signals, such as a synchronization clock, CLK, signal, and one or additional control signals, such as for controlling the variable transformer 20 in one or more embodiments.

Further in this example implementation, the network controller 10 includes a processing circuit 42, which for example comprises one or more microprocessor circuits or other digital processing circuitry. In this regard, it will be understood that the network controller 10 in one or more embodiments is a computer system that is specially adapted according to the teachings herein, and includes the necessary signaling and control circuitry to perform converter control as described herein. In at least one such embodiment, the central controller 10 includes storage 44, which may comprise one or more types of memory circuits and/or storage devices, such as electromagnetic disk storage, solid state disk storage, etc.

The storage 44 provides a computer-readable medium and stores configuration information 46, which includes a network model 48. The network model 48 comprises, for example, a data structure, file or collection of records, which together represent the topology and electrical characteristics of the distribution network 14. The network model 48 includes, for example, all line impedances and other characteristics that are known for the distribution network 14, and further includes or is associated with a number of model constraints. The model constraints reflect, for example, real- world limitations on the distribution network 14, such as maximum line currents, maximum/minimum bus voltages or voltage balances, etc.

Further, in at least one such embodiment, the computer-readable medium further stores a computer program 50 that, when executed by the processing circuit 42, configures the processing circuit 42 according to the teachings herein— i.e., configures the processing circuit 42 to carry out the any one or more embodiments of converter control disclosed herein.

With the above examples and explanations in mind, in one or more embodiments contemplated herein, the central controller 10 is configured for controlling one or more converters 12 that are in a distribution network 14, where each of the one or more converters 12 has an AFE and the central controller 10 comprises a signaling interface 40 and an associated processing circuit 42. The signaling interface 40 is configured to: receive network feedback, NFB, comprising indications of bus voltages and line currents for the distribution network 14; receive converter feedback, CFB, comprising indications of operational status and active and reactive power consumption, for each converter 12 to be controlled; and send converter control commands, CMD, to each such converter 12.

The processing circuit 42 is configured to: monitor bus voltages and lines currents for the distribution network, based on the network feedback received through the signaling interface 40; monitor active and reactive power consumption at each converter 12, based on the converter feedback received through the signaling interface 40; compute an actual reactive power consumption of the distribution network 14 at a point of common coupling, PCC 18 between the distribution network 14 and a supply network 16 that supplies electric power to the distribution network 14; use a network model 48 of the distribution network 14, including model constraints on bus voltages and line currents, to compute reactive power control settings for the one or more converters 12 that drive the actual reactive power consumption of the distribution network 14 towards a target reactive power consumption; and generate converter control commands to effectuate the computed reactive power control settings at the one or more converters 12 and transmit the generated converter control commands, CMDs, to the one or more converters 12 via the signaling interface 40.

Note that the target reactive power consumption may be fixed or may be a dynamically changeable value, and in either case it may be held in the storage 44 as part of the configuration information 46. The configuration information 46 further includes, for example, specifications or parameters for the one or more converters 46, e.g., the central controller 10 may be loaded or otherwise configured with information regarding the number, type, and operational

specifications, for the one or more converters 12 subject to its control. Alternatively, the central controller 10 may obtain such information using discovery and handshaking protocols, once it is communicatively coupled to the one or more converters 12 to be controlled.

The central controller 10 in some embodiments includes a synchronization circuit 52 that is configured to generate a synchronized clock signal as a function of a reference-timing signal. In turn, the processing circuit 42 is configured to output the synchronized clock signal to the one or more converters 12 via the signaling interface 40, to establish a common time reference for each converter 12 with respect to at least one of: measurements of the real and reactive power consumption of the converter 12, for use in generating the converter feedback from the converter 12; and control timings used for AFE switching at the converter 12 with respect to the converter control commands received by the converter 12 from the central controller 10.

More broadly, the processing circuit 42 in one or more embodiments is configured to synchronize the converter control commands to a common timing reference that is used by each of the one or more converters 12. Here, the converter feedback received from each converter 12 is synchronized to the common timing reference.

As for controlling the one or more converters 12, in one or more embodiments, the central controller 10 is configured to use an equivalent upstream impedance for each converter 12 to calculate an effective reactive power consumption of each converter 12 at the PCC 18, e.g., on the downstream side of the variable transformer 20 if one is used. Correspondingly, the central controller 10 is configured to compute the actual reactive power consumption of the distribution network 14 as a function of the effective reactive power consumptions, so that the actual reactive power consumption of the distribution network 14 is controlled at the PCC 18. It will be understood that the equivalent upstream impedance of any given converter 12 is based on knowledge of the impedances between the projection point— e.g., the PCC 18— and the actual network connection point of the converter 12.

However, whether the reactive power consumption is controlled at the PCC 18, or at the connection point of each of the one or more converters 12, the processing circuit 42 in one or more embodiments is configured is configured to constrain the reactive power control settings, according to the model constraints on bus voltages and line currents. That is, the processing circuit 42 uses the network model 48 and its included or associated constraints on distribution line current and bus voltages to constrain the reactive power control settings computed by it for any of the converters 12 being controlled. The processing circuit 42 further observes applicable operating limits on each converter 12— that is, the processing circuit 42 does not command any individual converter 12 to operate outside of its known or determined operational limits.

The central controller 10 can achieve at least some aspects of the disclosed control objectives, such as reactive power control, using a single converter 12. However, other control objectives, such as THD reduction, require two or more converters 12 to be available for control. Further, if two or more converters 12 are available for control, the central controller 10 enjoys greater control freedom. For cases where the one or more converters 12 comprise two or more converters 12, the processing circuit 42 is configured to jointly control reactive power consumption across the two or more converters 12, according to operating limits and operating margins of each converter 12, the target reactive power consumption, and the model constraints on bus voltages and line currents.

Here, "joint control" denotes an interdependence between the control settings computed for one converter 12 versus one or more other converters 12. For example, when jointly computing the control settings for two converters, the processing circuit 42 may coordinate the control settings computed for respective ones of the converters 12, to balance the reactive power consumptions among the two or more converters 12, so that each converter 12 contributes a portion of the overall reactive power consumption and so that no converter 12 is pushed to its operating limits. In particular, the processing circuit 42 is configured to coordinate the computation of the reactive power control settings across the two or more converters 12, to avoid operating limit violations at any given one of the converters 12.

In embodiments where the processing circuit 42 provides THD reduction by controlling two or more converters 12, the processing circuit 42 is configured to coordinate AFE switching across the two or more converters 12, to reduce an aggregate or net THD produced at the PCC 18 by the two or more converters 12. For example, the central controller 10 coordinates the AFE switching across the two or more converters 12 by staggering switching phases across the AFEs of the two or more converters 12, so that harmonic distortion produced by AFE switching in one or more given ones of the two or more converters 12 tends to cancel harmonic distortion produced by AFE switching in one or more other given ones of the two or more converters 12.

In the same or other embodiments, the central controller 10 is configured to generate converter control commands for one or more converters 12, to effectuate Conservative Voltage Reduction, CVR, targets at one or more buses 22 of the distribution network 14. For example, the processing circuit 42 computes the reactive power control settings for a given converter 12, to lower the voltage to a target level at the network side 26 of the converter 12. This lowering at the network connection pulls the associated bus voltage down.

In more detail, the central controller 10 considers the loadings of each converter 12 subject to use in CVR control. Additionally, the central controller 10 considers the overall power at the PCC 18. In an example CVR control implementation, the central controller 10 calculates the optimal voltages at the PCC 18 and at the input side 26 of each of the involved converters 12. The objective is to set these voltages such that the energy consumption of the distribution network 14 is minimized, while ensuring that all the given loads 30 are satisfied. The calculations are carried out by the central controller 10 uses the network model 48, as that model provides the requisite topology and impedance information inter-relating the controllable voltages at the various converter connection points.

Note that for CVR and/or for basic reactive power consumption control, the processing circuit 42 may generate the converter commands for any given converter 12 in the form of targeted network- side voltage commands and/or in the form of targeted reactive power consumption values, which are mapped by the converter 12 to targeted network-side voltages for the converter 12. It is assumed here that the converters 12 are intelligent enough to monitor their own operational parameters, including network-side voltages, active and reactive power consumption, etc.

The teachings herein contemplate an overall "reactive power control system" for a distribution network 14, which makes use of an embodiment of the central controller 14, and which relies on the distribution network 14 including two or more converters 12 having AFE enabling the central controller 10 to use the converters 12 for controlling the actual reactive power consumption of the distribution network 14. In particular, the contemplated reactive power control system in one or more embodiments comprises: two or more converters 12 coupled to different nodes in the distribution network 14. Each converter 12 has an AFE that can be controlled to control the reactive power consumption of the converter 12. The reactive power control system further includes a central controller 10 as taught herein.

The central controller 10 is configured to receive network feedback indicating an operating state of the distribution network 14 and converter feedback indicating active and reactive power consumption at each of the two or more converters 12 in the distribution network 14. Correspondingly, the central controller 10 is configured to control an actual reactive power consumption of the distribution network 14 at the PCC 18 between the distribution network 14 and a supply network 16 supplying electric power to the distribution network 14. The control in this respect is based on jointly coordinating the reactive power consumption at each of the two or more converters 12.

At least one of the two or more converters 12 comprises an "asymmetrical" converter 12. As defined herein, an asymmetrical converter 12 has excess power handling capacity on its input side facing the distribution network 14, as compared to the power handling capacity on its output side facing the load 30 powered by the asymmetrical converter 12. The excess capacity on the network side 26 of the asymmetrical converter 12 provides a greater range of reactive power consumption control for the asymmetrical converter 12 than would be provided if the input and output sides— the network side 26 and the load side 28— had substantially the same power handling capacity. In one non- limiting example, the network side 26 of the asymmetrical converter 12 is oversized by a margin of five to thirty percent, as compared to the load side 28 of the asymmetrical converter 12. The excess network-side capacity of the asymmetrical converter 12 provides the central controller 10 with additional "resources" that, for example, can be used for controlling the overall reactive power consumption of the distribution network 14.

Fig. 2 illustrates additional example details for the central controller 10 in one or more embodiments. Further example details include the possible inclusion of a master clock 54 and/or the association of a timing reference 56, for use with the synchronization circuit 52. In one example, the timing reference 56 is included in or associated with the central controller 10 and comprises, for example, a GPS receiver-based timing reference, which provides the

synchronization circuit 52 with a GPS-based time reference. The synchronization circuit 52 is configured to save or otherwise reference its timing to the GPS-based timing and the central controller 10 may be configured to distribute a synchronization clock, CLK in Figs. 1 and 2, to each converter 12 being controlled by the central controller 10. Alternatively, each converter 12 has access to GPS-based timing, e.g., through a GPS receiver included in each converter 12.

In other embodiments, the synchronization circuit 52 synchronizes itself to a master clock circuit 54, which itself may represent an input clock from a higher layer in an associated distribution network control system, or which may be derived from such an input clock, or which may simply be a precision local time reference. In instances where the central controller 10 and the converters 12 do not all have access to GPS-based timing, it is advantageous for the central controller 10 to distribute its master clock signal, or a signal derived therefrom, as the CLK signal for synchronization of the converters 12 to the central controller 10.

Further details of interest include the processing circuit 42, which implements any one or more of the following computational blocks or computational processes: a reactive power control process 60, a THD reduction process 62, a CVR process 64, and an overload mitigation process 66. All such processing makes use of various configuration information 46, including the aforementioned network model 48 and its associated model constraints, along with one or more control objectives. The control objectives include, for example, a value representing the target reactive power consumption for the distribution network 14, and may also include target levels of THD for the distribution network 14, target CVR values or equivalent targets in power reduction, etc.

Fig. 3 provides relevant introductory details for an example converter 12, such as one that may be included in the distribution network 14 and controlled by the central controller 10 according to the teachings herein. The converter 12 may be an AC-DC- AC converter or an AC- DC converter, and it includes input-side circuitry 70 and output-side circuitry 72. The input-side circuitry includes an AFE 74 that comprises power electronic switches that are controlled to effectuate AC-DC power conversion, where the specific parameters— such as conduction angles and phasing can be adjusted via commands from the central controller 10.

Fig. 4 provides further example details for a converter 12, including the inclusion of a signaling interface 80 for receiving converter commands from the central controller 10, sending converter feedback to the central controller 10 and, optionally, receiving a synchronization clock signal, e.g., to synchronize it to the central controller 10.

The example converter 12 further includes a processing circuit 82, which comprises one or more microprocessor-based circuits or other digital processing circuitry. The processing circuit 82 implements a converter control and reporting process, whereby it reports converter status and other information to the central controller 10, either periodically, on a triggered basis, or as requested by the central controller 10, and whereby it controls its input-side and output-side circuitry 70 and 72 according to local demands of the load 30— not shown— and according to commands from the central controller 10.

The processing circuit 82 is associated with memory/storage 86, which comprises, for example, electromagnetic disk storage or solid state disk storage, or essentially any other type of storage device or memory circuit— e.g., FLASH, EEPROM, etc. In one or more embodiments, the storage/memory 86 provides non-transitory storage for configuration data 88 and for a computer program 90. In at least one such embodiment, the depicted converter 12 is specially adapted to operate according to the converter- side teachings herein, based on its execution of the computer program instructions comprising the computer program 90.

The illustration also depicts that the converter 12 may have its own timing reference 92, such as a GPS-based timing circuit. In general, it will be understood that the converter 12 in one or more embodiments is configured for stand-alone operation and thus is capable of performing its conversion duties, etc., absent active control the central controller 10, which can be understood as adding a layer of further sophistication to operation of the converter 12, so that the converter 12 performs its converter function in accordance with the one or more overall control objectives implemented at the central controller 10.

To this end, the example converter 12 includes one or more measurement circuits 94 that enable it to monitor its own local operating conditions— e.g., input/output voltages, load current, etc. Finally, the illustration depicts input power to the input-side circuitry 70 and output power to the load 30— not shown— from the output side circuitry 72.

With these example converter details in mind, Fig. 5 illustrates a method 500 at a central controller 10 of controlling one or more converters 12 in a distribution network 14. The method 500 includes monitoring (Block 502) bus voltages and line currents for the distribution network 14, based on receiving network feedback from one or more measuring or monitoring devices 32, which are shown by way of example in Fig. 1 and which are configured to determine the bus voltages and line currents for the distribution network 14.

The method 500 further includes monitoring (Block 504) active and reactive power consumption at each of the converters 12, based on receiving converter feedback for each converter 12. Further, the method 500 includes computing (Block 506) an actual reactive power consumption of the distribution network 14 at the PCC 18, which couples the distribution network 14 to a supply network 16 that supplies electric power to the distribution network 14.

Still further, the method 500 includes computing (Block 508) reactive power control settings for the one or more converters 12 that drive the actual reactive power consumption of the distribution network 14 towards a target reactive power consumption. Here, for computation of the reactive power control settings, the central controller 10 uses a network model 48 of the distribution network 14, which includes model constraints on bus voltages and line currents.

The method 500 further includes generating and sending (Block 510) converter control commands to the one or more converters 12, to effectuate the computed reactive power control settings at the one or more converters 12. Here, the central controller 10 can be understood as "mapping" the computed reactive power control settings for each converter into one or more generated commands for that converter, so as to cause the converter 12 to take on or otherwise adopt the control settings. The particular details of command generation are not germane inasmuch as the formats and protocols of the generated commands are implementation details that will vary with converter make and/or model.

Those of ordinary skill in the art will appreciate that one or more steps or processing operations represented in Fig. 5 may be performed in an order other than that suggested by the illustration. Moreover, one or more of the illustrated steps may be repeated or looped as needed, and/or performed in conjunction with other operations at the central controller 10. Broadly, the central controller 10 performs the method 500 with respect to one converter 12, or with respect to two or more converters 12. In this latter case, in one or more embodiments, the central controller 10 jointly computes the reactive power control settings of the two or more converters 12, based on considering the operating capacities and margins across the two or more converters 12, and considering the locations— e.g., in terms of impedance with respect to the PCC 18— of the converters 12.

Fig. 6 illustrates further method operations performed by the central controller 10 in one or more embodiments. In particular, the processing represented by Fig. 6 represents an example THD reduction strategy implemented by the central controller 10 in one or more embodiments. While Fig. 6 is illustrated as a continuation of the method 500, it should be understood that the THD reduction processing may be integrated into and performed in conjunction with the reactive power control processing at issue in Fig. 5.

In its example details, Fig. 6 includes the central controller 10 checking synchronization (Blocks 512, 514), e.g., checking whether the converters 12 are synchronized with respect to the central controller's synchronization clock signal or with respect to a common time reference used by the central controller 10 and the converters 12. If synchronization criteria— e.g., a defined tolerance or threshold value— are not met, processing continues with the central controller 10 determines (Block 516) time delays or offsets relative to the converter(s) 12 that are out of synchronization and sends the synchronization corrections to the targeted converter(s) 12 and receives return synchronization acknowledgements from the commanded converters 12 (Block 520).

Processing then returns to the synchronization check operation (Blocks 512, 514). Upon this return, or upon initial invocation of the illustrated method, if the synchronization check passes (YES from Block 514), processing continues with the central controller 10 measuring THD at the PCC 18 (Block 522) and checking (Block 524) whether the measured or actual THD at the PCC 18 is below a defined THD MAX value, which represents a permissible or targeted maximum level of THD.

If no (NO from Block 524), the central controller 10 takes action to reduce THD at the PCC 18, based on computing new staggering angles for two or more of the converters 12 in the distribution network 14 (Block 526) and sending the new staggering angles to the targeted converters 12 (Block 528). Here, the "staggering angles" will be understood to comprise, for example, the phasing or timing offsets between switching times in the AFEs of the converters 12.

On the other hand, if the actual THD at the PCC 18 is already at or below the maximum permissible or target value (YES from Block 524), then the central controller 10 waits for some defined time (Block 530) before rechecking synchronization and/or THD. Of course, the central controller 10 continues other processing— e.g., reactive power control, etc.— during this "waiting" period. Instead, the waiting period can be understood as some practical limit or interval controlling how frequently the central controller 10 measures and attempts to correct THD.

As part of its operations for THD control, the central controller 10 advantageously uses the network model 48. In particular, the central controller 10 uses its knowledge of the involved impedances— i.e., as between the PCC 18 and the network connection of each converter 12— to compute the THD at the PCC. The central controller 10 uses this information and knowledge of the loading conditions at the converters 12 to compute new staggering angles, to reduce the aggregate or overall TDH in the distribution network 14.

Figs. 7-9 provide further example details for THD control according to one or more embodiments of the central controller 10. With Fig. 1 being understood as a simplified line diagram or schematic of an example distribution network 14, Fig. 7 illustrates an equivalent model of at least a portion of the distribution network 14 for any harmonic "H." Three examples, buses 22-1, 22-2 and 23-3, are illustrated, along with two example converters 12-1 and 12-2. The harmonic currents produced by the converters 12-1 and 12-2 in this example are shown as IH2 and IH3. Note that both these currents are vectors, i.e., they have magnitude and phase angles. Notably, the diagram represents the involved distribution lines 24 according to their "PI" models, but it should be understood that other electrical models can be used to represent the distribution lines 24.

Fig. 8 illustrates how the upstream current, resulting from a given converter 12 as a downstream current source, can be calculated. This calculation can be applied to the section of the distribution network 14 between bus 22-1 and bus 22-2, or to the section between bus 22-2 and bus 22-3, and so on. If any downstream current source in the section of interest-e.g., a converter 12- is drawing II 0° , then the resulting upstream current may be determined as I^AU=((Zs+Zc))/Zc I^AD<5°.

The above equation shows the relationship between the upstream and downstream currents. Since the terms Zs and Zc are complex numbers, the upstream current will have a different phase angle compared to the downstream current. This aspect is important in the context of cancel harmonic currents, as the central controller 10 must account for the upstream phase angles when determining how to stagger the converters 12, to effectuate a cancelation or reduction in THD at the upstream point of interest.

With this approach, the central controller 10 finds the upstream current at bus 22-2, for example, arising from the current source(s) in bus 22-3. Let this current be denoted as IH32. So the total harmonic current at Bus 2 is IH32 + IH2. To minimize the harmonic, the currents IH2 and IH32 must have opposite phase angles, because of the properties of vector addition. Here, in this two-converter example context, the central controller 10 computes the phase angle of the harmonic current produced in the bus 22-3 by the converter 12-2, as projected upstream into the bus 22-2. For example, represent the phase angle of IH3 as a°, and represent the projected upstream phase angle as of IH32 as β°.

The best or "optimum" phase angle for the harmonic current IH2 produced by the converter 12-1 in bus 22-2 is - β°, i.e., the negative of phase angle of IH32. Of course, with the various converters 12 mainly producing harmonics at specific multiples, e.g., the fifth, seventh and eleventh harmonics, the central controller's overall objective would be to minimize the THD in a cumulative sense, with respect to all harmonics, or at least a subset of the dominant harmonics.

For the case where two converters 12 are involved, only one staggering angle is required to relate the operation of one converter 12 to the other converter 12. Either such converter can be selected as the "reference" for staggering angle computation. For the case where the central controller 10 controls N converters 12, where N is an integer number greater than two, the number of staggering angles would be "N-l". Hence, the number of unknown variables would be "N-l". Advantageously, the central controller 10 uses this approach, or variations of it, to coordinate the operation of more than one converter 12 across any number of different nodes or buses in the distribution network 14 to reduce THD.

Fig. 9 illustrates the above mathematical computations in more detail and it can be understood as representing an example of the computations performed in the context of the earlier-described Fig. 6. The calculation uses a starting or initial value of the staggering angles for the N converters 12 being controlled for THD reduction by the central controller 10. Here, N is at least two. In particular, for a particular set of initial or prior staggering angles, the harmonic current angles can be calculated (Block 902). Further, the central controller 10 obtains the magnitudes of the harmonic currents at issue in Block 904— e.g., based on loading information included in the converter feedback from each of the converters 12 at issue.

Using the magnitude and phase information, along with the impedance information from the network model 48, a "load flow" analysis of the distribution network 14 is performed by the central controller 10 (Block 906). The load flow analysis includes calculation of the effective or projected upstream harmonic currents at the PCC 18, for the involved converters 12. Such calculations are performed for all harmonics of interest— e.g., for the harmonics that are detected or known for the converters 12. Correspondingly, the overall or aggregate net THD is computed as a function of all of the harmonic components treated in the load flow analysis (Block 908). If the overall net THD is within acceptable limits (YES from Block 910), the central controller 10 dispatches the staggering angles to the converters 12, or otherwise maintains those staggering angles if they are already in force at the converters 12 (Block 912). On the other hand, if the overall net THD is above the target threshold (NO from Block 910), the central controller updates (Block 914) the staggering angles and the process repeats with new computations for the updated staggering angles. In any case, synchronization among all the converters 12 is required to ensure that the measurements passed to the central controller 10 are coherent— i.e., referenced to or resolvable to a known timing reference. For example, whether synchronized via individual GPS timing references or from a synchronization clock signal output by the central controller 10, each converter 12 may time stamp its measurements and the central controller 10 can therefore correctly perform its phase computations for establishing the inter-converter staggering angles.

As for performing the load flow analysis, it will be appreciated that the distribution network 14 may include any number of loads 30 having known active (real) and reactive power, along with one or more converter-controlled loads that, by virtual of the converters 12 with their AFEs, have fixed active power requirements but variable reactive powers. Of course, this scenario extends to buses 22 in the distribution network 14 that include loads 30 driven by converters 12 and those not driven by converters. The voltages at the PCC 18 and a reference bus 22 are known, e.g., from measurements provided to the central controller 10. The central controller 10 may form an initial guess on the amount of reactive power consumption at each converter 12. Here, "consumption" is a broad term and encompasses reactive power injection from the converter 12 into the distribution network 14.

Using this set of values as an initial "dispatch" and the network model 48, the central controller 10 runs the load flow analysis for the distribution network 14. By running this analysis, the central controller 10 obtains the bus voltages and line currents for the distribution network 14 and checks them against the corresponding model constraints— i.e., the defined limits or ranges permissible for the bus voltages and line currents. For example, the central controller 10 checks whether any of the bus voltages, in per unit scale, are off by more than five percent of the nominal. If no limits are violated, the current dispatch may be deemed acceptable. Otherwise, the central controller 10 may adopt a new initial dispatch solution for the distribution network and rerun the load flow analysis, check for violations, etc.

Consider Figs. 10 and 11, for example. Fig. 10 depicts a portion of the distribution network 14, including example buses 22-1, 22-2 and 22-3. There are two converters 12 subject to control by the central controller 10, i.e., converter 12-1 and 12-2. The converter 12-1 operates with a variable— and controllable— reactive power consumption Ql, and with an active power consumption PI, as required by its associated load 30-1. Similarly, the converter 12-2 operates with a variable and controllable reactive power consumption Q2, and with an active power consumption P2, as required by its associated load 30-2. One also sees another load 30-3 that has fixed active and reactive power consumptions P3 and Q3, respectively.

Thus, in terms of the load flow analysis and degrees of freedom, the central controller 10 can dynamically adjust the reactive power control settings for the converter 12-1 and/or the converter 12-2, to effectuate a desired value for Ql and/or Q2. Note that Q3 is not controllable, at least not in terms of converter-based manipulation. Preferably, the central controller 10 jointly computes the reactive power control settings for the converters 12-1 and 12-2, e.g., it treats the Q2 and Q3 values as a pair or otherwise considers them jointly. Joint consideration allows the central controller 10 to achieve changes in reactive power consumption for the overall distribution network 14 via manipulation of both Ql and Q2.

Joint consideration therefore allows the central controller 10 to "balance" the overall reactive power consumption in the sense that the controllable reactive power capacity of the available converters 12 can be used in part to effectuate the overall desired reactive power consumption of the distribution network 14 at the PCC 18 or, equivalently, at the associated substation. By using two or more converters 12 as a "pool" or collection of reactive power control resources, the central converter 10 avoids overloading the reactive power control capabilities of any one converter 12 and, likewise, avoids undesirable or impermissible voltages and currents within the distribution network 14. In all cases, the central controller 10 can use the equivalent upstream impedance of any given converter 12 at any given point of connection to the distribution network 14, to project to the effective reactive power consumption of that converter 12 back to the PCC 18.

Fig. 11 illustrates operations at the central controller 10 for reactive power control of the two example converters 12-1 and 12-2, in the context of the simplified network depiction given in Fig. 10. It will be appreciated that the processing of the illustrated method 1000 can be understood as example implementation details for the reactive power control method 500 more generally depicted in Fig. 5.

According to the method 1100, the central controller 10 assumes an initial dispatch— i.e., initial values for Ql and Q2 (Block 1102). The initial dispatch may be based on historical settings or default targets defined in the configuration information 46, for example, or may be based on the last-computed values for Ql and Q2— i.e., the current reactive power control settings for the converters 12-1 and 12-2. Processing continues with the central controller 10 running the load flow analysis using the initial dispatch and using the network mode 48 (Block 1104). The central controller 10 thereby determines the bus voltages and line currents for the distribution network 14 (Block 1106) and checks for the presence of any voltage or current violations (Block 1108). If there are violations (YES from Block 1108), the central controller 10 updates (Block 1110) the dispatch— i.e., updates the values of Ql and/or Q2— and runs the analysis again (repeats Blocks 1104, 1106 and 1108). If there are no violations (NO from Block 1108), the central controller 10 accepts the Ql and Q2 values as the dispatch solution. At least to the extent that the new dispatch solution is changed from the prior solution, the central controller 10 generates the corresponding converter commands and sends them to the converters 12-1 and/or 12-2.

Further, in some embodiments, the central controller 10 considers one or more added metrics in its load flow analyses. For example, the central controller 10 considers total losses in the distribution network 14 and seeks to minimize or at least reduce power losses. In such cases, in addition to checking that model constraints are not violated, the central controller 10 evaluates dispatch solutions for the distribution network 14 in terms of power losses and may iteratively adjust the network dispatch and rerun the load flow analysis, to obtain a dispatch offering lower losses. Here, the "dispatch" includes or maps to reactive power control settings for the one or more converters 12 subject to control by the central controller 10.

For example, the reduced-power solution for the distribution network 14 may be represented in terms of bus voltages or voltage levels at the connection points of the converters 12, and the central controller 10 translates those voltage levels treats those voltage levels as the corresponding reactive power control settings and sends corresponding commands to the converters 12, to effectuate the dispatch voltages. Alternatively, the central controller 10 translates or otherwise maps the actual voltage levels into corresponding targeted reactive power consumption (injection) values, and generates converter commands to effectuate the reactive power consumption target for each converter 12. Whether computed in terms of reactive power consumption values or equivalent connection-point voltage levels, it will be understood that the central controller 10 determines reactive power control settings for the converters 12, according to its one or more overall control objectives for the distribution network 14.

Examples of control objective includes but are not limited to CVR, THD reduction or minimization, feeder power line overload prevention, etc. As a further example of CVR-related control, consider Figs. 12 and 13. Fig. 12 provides an illustration of the results of example CVR control, while Fig. 13 illustrates one embodiment of a method 1300 of CVR control implemented by the central controller 10 in one or more embodiments.

To better appreciate these example details, consider that there are various types of electrical loads. The simplest types of loads are resistive loads. The power absorbed by a resistive load is given by

where V is the RMS voltage and R is the resistance of the load. Examples of such loads include incandescent lights.

As the power of these loads vary with squares of the voltage across them, the power can be increased by increasing the voltage. Likewise, the power can be decreased by decreasing the voltage. In most countries, including the United States, there are standards which dictate the band of admissible voltage. In a typical radial distribution system, voltage can only be controlled at the PCC. Such control is typically achieved by using a tap changing transformer, by using one or more shunt VAR compensators.

Fig. 12 illustrates example voltage curves without CVR and with CVR control by the central controller 10. In particular, the solid line depicts an example voltage curve without CVR and the dashed line depicts an example voltage curve with CVR. Absent CVR, to maintain the remotest bus 22 in the distribution network 14 within the permissible voltage limit, the voltage would have to be raised at the PCC 18 towards the higher end of the permissible voltage range. However, simply raising the voltage at the PCC 18 in that manner would result in higher power draws for loads 30 nearer to the PCC 18.

Through CVR, the central controller 10 maintains a relatively flat voltage profile in the distribution network 14, as seen from the dashed-line voltage curve in Fig. 12. The central controller 10 effectuates this flattened voltage curve by using various ones of the converters 12 in the distribution network 14 for VAR compensation. More particularly, rather than relying on expensive additional circuitry for VAR compensation, the central controller 10 uses the AFEs of individual converters 12 to control the reactive power injections at various buses 22 in the distribution network 14, so as to maintain a relatively flat profile. This control behavior advantageously lowers the power consumption of resistive loads throughout the distribution network 14 and generally does not affect the power consumption of loads 30 powered by the given converters 12.

Fig. 13 illustrates an example processing method 1300, as implemented by the central controller 10 in one or more embodiments, to effectuate CVR in the distribution network 14. Processing "begins" with collecting inputs (Block 1302), e.g., collecting measurements like bus voltages, line currents, loadings, load status, location and size of resistive loads. Note that at least some of this information is collected for other control processes run by the central controller 10 and the "collection" may be ongoing, triggered, periodic, etc.

Processing continues with the central controller 10 "guessing" (Block 1304) a value for Q_CONV^ASET, which represents establishing initial or default VAR settings for at least those converters 12 involved in CVR control. Note that the current reactive power control settings for all of the converters 12 may already be known to the central controller 10, as a consequence of its baseline reactive power control process. In any case, the central controller 10 uses the Q_CONV^ASET values to perform a load flow analysis (Block 1306) and from these calculations, the central controller 10 determines the real power consumptions of the resistive loads involved in the analysis, which may be all resistive loads known for the distribution network 14, or a selected subset of them.

The reactive power at the PCC 18 is also computed, which is denoted as Q_PCC in the diagram. Processing continues with computing a "cost" (Block 1308), which should be understood as corresponding to an overall objective function being minimized or otherwise optimized by the central controller 10. For example, consider the objective function

Cost=wl-P_R+w2-Q_PCC,

where P_R is the (overall) real power consumption of the resistive load(s), Q_PCC is, as noted, the reactive power for the distribution network 14 at the PCC 18, and wl and w2 are weighting factors. Thus, if the central controller 10 were configured only to minimize Q_PCC, then it would the following weighting values: wl=0,w2=l.

Using different values for wl and w2 varies the weighting of the terms being considered in the objective function. Of course, the central controller 10 can be configured to minimize costs with respect to a more complex objective function, which includes numerous terms, e.g., a term for reactive power at the PCC 18, a term for THD at the PCC 18, a term for real power consumption by the distribution network 14, etc. Further, the central controller 10 may use a non-linear objective function, e.g., Cost=P_R P_R+Q_PCC Q_PCC.

In any case, it will be appreciated that the central controller 10 optimizes the objective function based on computing or adjusting the control settings to be dispatched to the involved converters 12. After determining the converter settings via objective function optimization, processing continues with determining whether the contemplated settings would, if implemented at the converters 12, result in constraint violations (Block 1310). If one or more constraints would be violated as a result of the contemplated settings, then processing returns to Block 1302, and Block 1304 is repeated with Q_CONV^ASET now being adjusted. The adjustment may be intelligent, in view of the particular constraint violations. In any case, processing continues forward again with the adjusted Q_CONV^ASET.

If at any iteration of execution, no constraint violations are detected at Block 1310, processing continues with determining whether the cost is at or below a minimum cost threshold (Block 1312). If not, processing returns to Block 1302 and repeats. If so, processing continues with storing (Block 1314) the then-current value(s) of Q_CONV^ASET as a "best guess." Processing further involves checking (Block 1316) whether any controlling convergence criterion or criteria have been met. For example, the central controller 10 may evaluate any one or more of: whether Cost is below a defined cost threshold, whether an iteration limit has been reached, whether a minimum iteration count or timer has been satisfied, and/or whether an iteration limiting timer has expired. If the convergence criterion or criteria is/are met, then the central controller 10 dispatches the best-guess Q_CONV^ASET to the converters 12 (Block 1318).

If the criterion or criteria for iteration is/are not met, then processing repeats from Block

1302, and it will be appreciated that the method 1300 can be integrated into or performed along with any of the other processing methods disclosed herein for the central controller 10. The method 1300 may be repeated on a triggered basis, e.g., on an on-demand basis, and/or may be performed on a recurring basis.

Likewise, the other control methods disclosed herein for the central controller 10 may be performed on a recurring basis, or may be performed on a triggered or on-demand basis.

Consider overload mitigation, for example. If a feeder or substation associated with the distribution network 14 is overloaded, the central controller 10 in one or more embodiments commands one or more of the one or more converters 12 in the distribution network 14 to reduce their individual loadings on the distribution network 14. Where two or more converters 12 are available for commanded reductions, the aggregate reduction in loading may be uniformly allocated across the two or more converters 12, or unequally allocated according to priorities. For example, a converter 12 with a higher priority would have less load reduction applied to it, versus a converter 12 of lower load priority.

It is contemplated in one or more embodiments to define an "essential" or "top priority" designation, wherein any converter 12 so designated is not included in any commanded load reductions for overload mitigation. That is, the central controller 10 will command load reductions at lower priority converters 12, while allowing the converter(s) 12 associated with essential loads to continue operating at demand loading. In this way overloading of lines can still be mitigated with minimal or no inference with the essential loads.

For this and other control objectives, the central controller 10 will be understood as receiving one or more signals and measurements from the converters 12, as well as from measurement devices or sensors 32 at the PCC 18 and, possibly, elsewhere in the distribution network 14. These signals includes, but are not limited to converter status, converter loading, converter switching frequencies, THD level at the PCC 18 or at the substation, line/feeder loading at the PCC 18 or associated substation, voltage level at the PCC 18 or associated substation, etc. Broadly, the measurements of interest include but are not limited to voltages, currents, active and reactive power, frequency, etc.

Notably, modifications and other embodiments of the disclosed invention(s) will come to mind to one skilled in the art having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the invention(s) is/are not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of this disclosure. Although specific terms may be employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims

CLAIMS What is claimed is:

1. A central controller configured for controlling one or more converters that are in an electric power distribution network, each of the one or more converters having an active front end, AFE, and the central controller comprising:

a signaling interface configured to: receive network feedback comprising indications of bus voltages and line currents for the distribution network; receive converter feedback comprising indications of operational status and active and reactive power consumption, for each converter; and send converter control commands to each converter; and

a processing circuit configured to:

monitor bus voltages and lines currents for the distribution network, based on the network feedback received through the signaling interface;

monitor active and reactive power consumption at each converter, based on the converter feedback received through the signaling interface;

compute an actual reactive power consumption of the distribution network at a point of common coupling between the distribution network and a supply network that supplies electric power to the distribution network;

use an electrical model of the distribution network, including model constraints on bus voltages and line currents, to compute reactive power control settings for the one or more converters that drive the actual reactive power consumption of the distribution network towards a target reactive power consumption; and

generate converter control commands to effectuate the computed reactive power control settings at the one or more converters and transmit the generated converter control commands to the one or more converters via the signaling interface.

2. The central controller of claim 1 , wherein the central controller includes a

synchronization circuit configured to generate a synchronized clock signal as a function of a reference timing signal, and wherein the processing circuit is configured to output the synchronized clock signal to the one or more converters via the signaling interface, to establish a common time reference for each converter with respect to at least one of: measurements of the real and reactive power consumption of the converter, for use in generating the converter feedback from the converter; and

control timings used for AFE switching at the converter with respect to the converter control commands received by the converter from the central controller.

3. The central controller of claim 1, wherein the processing circuit is configured to synchronize the converter control commands to a common timing reference that is used by each of the one or more converters, and wherein the converter feedback received from each converter is synchronized to the common timing reference.

4. The central controller of claim 1, wherein the central controller is configured to:

use an equivalent upstream impedance for each converter to calculate an effective

reactive power consumption of each converter at the point of common coupling; and

compute the actual reactive power consumption of the distribution network as a function of the effective reactive power consumptions, so that the actual reactive power consumption of the distribution network is controlled at the point of common coupling.

5. The central controller of claim 1, wherein the processing circuit is configured to constrain the reactive power control settings, according to the model constraints on bus voltages and line currents.

6. The central controller of claim 1 , wherein, for cases where the one or more converters comprise two or more converters, the processing circuit is configured to jointly control reactive power consumption across the two or more converters, according to operating limits and operating margins of each converter, the target reactive power consumption, and the model constraints on bus voltages and line currents.

7. The central controller of claim 1, wherein, for cases where the one or more converters comprise two or more converters, the processing circuit is configured to coordinate AFE switching across the two or more converters, to reduce an aggregate or net Total Harmonic Distortion, THD, produced at the PCC by the two or more converters.

8. The central controller of claim 7, wherein the central controller coordinates the AFE switching across the two or more converters by staggering switching phases across the AFEs of the two or more converters, so that harmonic distortion produced by AFE switching in one or more given ones of the two or more converters tends to cancel harmonic distortion produced by AFE switching in one or more other given ones of the two or more converters.

9. The central controller of claim 1, wherein the central controller is further configured to generate the converter control commands to effectuate Conservative Voltage Reduction, CVR, targets at one or more buses of the distribution network.

10. The central controller of claim 1, wherein, for cases where the one or more converters comprise two or more converters, the processing circuit is configured to coordinate the computation of the reactive power control settings across the two or more converters, to avoid operating limit violations at any given one of the converters.

11. A reactive power control system comprising:

two or more converters coupled to different nodes in an electric power distribution

network, wherein each converter has an active front end, AFE, that can be controlled to control the reactive power consumption of the converter; and a central controller configured to:

receive network feedback indicating an operating state of the distribution network and converter feedback indicating active and reactive power consumption at each of the two or more converters in the distribution network; and control an actual reactive power consumption of the distribution network at a point of common coupling between the distribution network and a supply network supplying electric power to the distribution network, based on jointly coordinating the reactive power consumption at each of the two or more converters.

12. The reactive power control system of claim 11, wherein at least one of the two or more converters comprises an asymmetrical converter having excess power handling capacity on its input side facing the distribution network as compared to the power handling capacity on its output side facing the load powered by the asymmetrical converter, to provide a greater range of reactive power consumption control for the asymmetrical converter than would be provided if the input and output sides had substantially the same power handling capacity.

13. A method of controlling one or more active-front end, AFE, converters that are in an electric power distribution network, said method comprising:

monitoring bus voltages and lines currents for the distribution network, based on

receiving network feedback from one or more monitoring devices configured to determine the bus voltages and line currents;

monitoring active and reactive power consumption at each of the converters, based on receiving converter feedback for each converter;

computing an actual reactive power consumption of the distribution network at a point of common coupling between the distribution network and a supply network that supplies electric power to the distribution network; and

computing reactive power control settings for the one or more converters that drive the actual reactive power consumption of the distribution network towards a target reactive power consumption, wherein said computing uses an electrical model of the distribution network, including model constraints on bus voltages and line currents; and

generating and sending converter control commands to the one or more converters, to effectuate the computed reactive power control settings at the one or more converters.

14. The method of claim 13, wherein the method includes synchronizing the one or more converters to a common timing reference, for synchronizing at least one of:

measurements made by the converter of the real and reactive power consumption of the converter, for use in generating the converter feedback from the converter; and control timings used for AFE switching at the converter with respect to the converter control commands received by the converter.

15. The method of claim 13, wherein the method includes synchronizing the converter control commands to a common timing reference that is used by each of the one or more converters, and wherein the converter feedback received from each converter is synchronized to the common timing reference.

16. The method of claim 13, wherein the method includes:

using equivalent upstream impedance for each converter to calculate an effective reactive power consumption of each converter at the point of common coupling; and calculating the actual reactive power consumption of the distribution network as a function of the effective reactive power consumptions, so that the actual reactive power consumption of the distribution network is controlled at the point of common coupling.

17. The method of claim 13, wherein the method includes constraining the reactive power control settings, according to the model constraints on bus voltages and line currents.

18. The method of claim 13, wherein, for cases where the one or more converters comprise two or more converters, the method includes jointly controlling reactive power consumption across the two or more converters, according to operating limits and operating margins of each converter, the target reactive power consumption, and the model constraints on bus voltages and line currents.

19. The method of claim 13, wherein, for cases where the one or more converters comprise two or more converters, the method further includes coordinating AFE switching across the two or more converters, to reduce an aggregate or net Total Harmonic Distortion, THD, produced at the PCC by the two or more converters.

20. The method of claim 19, wherein reducing the aggregate or net THD comprises staggering switching phases across the AFEs of the two or more converters, so that harmonic distortion produced by AFE switching in one or more given ones of the two or more converters tends to cancel harmonic distortion produced by AFE switching in one or more other given ones of the two or more converters.

21. The method of claim 13, wherein the method includes generating the converter control commands further to effectuate Conservative Voltage Reduction, CVR, targets at one or more buses of the distribution network.

22. The method of claim 13, wherein, for cases where the one or more converters comprises two or more converters, the method includes coordinating the computation of the reactive power settings across the two or more converters, to avoid operating limit violations at any given one of the converters while simultaneously controlling the actual reactive power consumption of the distribution network at the point of common coupling, with respect to the target reactive power consumption.