TECHNICAL FIELD

The present invention relates to electrical energy supply, and in particular to an electrical power control apparatus and process for use in the electricity grid.

BACKGROUND

The Global Energy Problem

The ability to harness the stored energy of fossil fuels in the form of electricity has enabled humans to make amazing advances toward our wellbeing. However, as the global demand for electricity continues to increase, it is widely anticipated that the resulting impact on the environment will ultimately reach a point where it becomes a threat to our survival. Therefore, resolving the constraints to universal access to electricity without unsustainable environmental consequences is a social and economic priority of the highest order.

To meet the legitimate aspirations of the world's population and impact poverty, the US Energy Information Administration estimates an increase of over 50% in global energy requirements by 2040 (Reference: International Energy Outlook 2016). At the same time, significant reductions in the environmental impact and cost of energy is required. Our traditional power generation methods use fossil fuels that are all constrained resources. At present, our only proven clean generation options are hydro-electric, geothermal, biomass, solar and wind. Of these, solar and wind present the best opportunity to meet our future needs, as the others have very specific geographic inputs and requirements.

The global consensus of policymakers is that we need to dramatically increase our clean energy generation, as evidenced by the recent Paris Climate or COP21 agreement. However, with our current and forecast advances in clean generation technology, the design of our electricity networks is incapable of allowing us to achieve the mandated goals. In particular, integration is a well-known and unsolved issue. Electricity System Overview

The entire electricity supply chain can be generally grouped into three areas: generation, distribution, and consumption. In order to maintain a reliable power supply, energy generation is controlled to meet consumption through an end-to-end network.

Electricity distribution networks are the largest manmade objects ever created. To date, they have all been conceived and constructed for very specific operating conditions under a single design schematic. The design schematic is relatively simple : generation in the past was easily and adequately equilibrated with consumption. A small number of dispatchable generation sources supplied a series of tranched or grouped consumers. The topology of electrical networks tends to be a mixture of radial (a single large generator with consumers progressively tranched in a tree structure around it), and more complex and redundant path network topologies such as mesh and ring topologies. These topologies are designed to provide a good balance of reliability and cost, based on the current network demands of a few large stable and dispatchable generators and many consumers.

At present, the vast majority of electricity generation is directly controllable because it is produced from fossil fuels. The table below outlines the current makeup of electricity generation for the US and the UK.

Table 1 Percentage of Total Electricity Generation by Source (2015)

As the level of variable renewable energy generation increases, our ability to control the supply of electricity to match demand becomes increasingly difficult. If this balance of supply and demand is not maintained, the stability and availability of electricity is threatened.

As the makeup of generation is changing toward more renewable energy, the characteristics of the electricity supply are changing, including both the physical architecture of the network, and our ability to control generation levels. This causes increasing issues of stability and efficiency, with the threshold capability of current electricity network architecture falling far short of required levels to meet our climate change targets. Attempting to resolve the situation through control of generation and/or consumption alone has intolerable social, economic and environmental effects. The configuration of the electricity distribution network therefore needs to adapt if our electricity supply chain is to remain viable and deliver on our social and economic intentions.

Electricity System Balancing Requirement

Maintaining a reliable electricity supply requires the voltage and frequency of the electric power grid to be maintained within a narrow band of about +/-1%. Other than the limited storage options currently available, electricity must be consumed when it is generated, and consequently supply and demand must be balanced to maintain the required target voltage and frequency. Until now, this has been accomplished by monitoring the grid at a coarse high level, and then adjusting the output of controllable generation sources that are largely fossil fuel, nuclear or hydro-electric.

Balancing of the grid can be categorised into three response times:

• Long Term (days to weeks),

• Medium Term (hours), and

• Short Term (milliseconds to minutes).

The UN report 'Global Trends in Renewable Energy Investment 2016' states there are currently four potential balancing options, with an unacceptable fifth option currently also being utilised globally.

In the case where demand exceeds supply:

i. increasing the amount of faster responding conventional generation; i.e., gas, coal or diesel;

ii. interconnectors to transport electricity from one grid to another; and iii. demand response by paying larger industrial and commercial consumers to reduce usage when supply is falling short of demand.

In the case where supply exceeds demand :

iv. energy storage to store excess electricity when it is available and release it back into the grid when required; and

v. curtailment of renewable energy generation to directly reduce supply. A combination of these solutions in parallel would be plausible to attempt to manage the long and medium term balancing of electric power grids. Currently, each option either has prohibitive costs, unacceptable consequences, or both.

For the short term response (milliseconds to seconds and minutes) to balance our electric power grids, none of these options will successfully allow the increased penetration and consumption of renewable generation whilst maintaining a reliable power supply, for at least the reasons discussed below.

Short Term Grid Balancing Issues

The following outlines the issues that must be overcome in relation to renewable energy integration into the electric power grid.

System Frequency

All generators inject power into the grid as alternating current (AC), and are synchronised to operate at the same frequency and phase. The amount of power injected by each generator is balanced through the ratio of its power output rating compared to all other generators injecting power in order to evenly distribute the load. This occurs naturally unless modified by operator control.

Traditional fossil fuels, nuclear reactors, and even hydro-electric power are all synchronous generators that introduce inertia to help maintain this frequency, and are controllable, providing frequency response and stability. They remain synchronised due to the self-regulating properties of their interconnection. If one generator deviates from its synchronous speed, power is transferred from the other generators in the system in such a way as to reduce the speed deviation. The stored inertial energy of the generators provides a short-term counteraction to frequency change, with governors taking over after a few seconds.

In contrast, wind and photovoltaic solar power generation use significantly different technologies, producing DC power and injecting it into the AC grid through converters. This means that they are decoupled from the grid frequency, and results in asynchronous operation with no inertial energy to contribute. It is possible for converters equipped with governor-like controls to respond to frequency drops; however, this cannot occur fast enough to adequately compensate and maintain grid stability. It can also only occur when the generation source is operating in a curtailed condition.

Grid Architecture

Our electricity grids have been specifically designed to deliver a reliable electricity supply from power sources through a transmission network over long distances to load centres on a distribution network. However, the entire ontology of our grids is changing due to the local and dispersed nature of renewable energy generation. Our current grid hardware is incapable of adequately distributing these new power sources bi-directionally and both vertically and horizontally through the network, causing a myriad of power engineering problems, including a reduction in the capacity of the network.

Current methods for addressing these issues primarily involve additional hardware and software systems to mitigate undesired effects. However, these technologies are generally accepted as increasing network fragility and cost without addressing the root cause.

Control

Renewable energies such as wind and solar are not dispatchable like traditional fossil fuels, nuclear or hydro power. As we do not have control over the energy input (i.e. the wind or the sun), we are unable to ramp up or down as required to balance the system, or maintain a steady state of output. We can only actively manage the output to maintain the required power level through energy storage solutions or by curtailing the generation. However, curtailment is pure waste.

Variability

The rate at which the power output of renewable energies such as wind and solar can change is much faster than traditional generation technology. This occurs in two major forms:

• intermittence - renewable energy sources have long periods of unavailability due to input requirements outside of direct control (i.e., sunlight and wind); and

• volatility - at all times constant variation in the output from renewable energy generation is occurring. The two main constituents of this are the rapid rate of change of output generation, and the noise inherent to the output signal.

The law of averages helps in part to mitigate the instantaneous effects of volatility with the vast number of solar and wind generation sources. However, maintaining voltage and frequency in the short term (milliseconds to seconds) remains a significant unresolved challenge. Currently available responsive dispatchable generation technology is still significantly slower to react than the rate of change introduced by volatile renewable generation. There is no current solution to this issue.

Efficiency

Electricity grids are designed to work at a specific operating point, with a narrow band of operation due to consumption requirements. When the voltage or frequency deviates from the optimal point, the efficiency of the grid and its devices decreases, resulting in greater energy losses. Energy losses in developed grids are 5% - 10%, with up to half of this loss due to non-fixed inefficiency losses. When the voltage or frequency goes outside the set operating boundaries, system protection actions are automatically undertaken, which leads to both brownouts and blackouts for hardware protection and safety.

Network Hardware

Electric power networks around the world use predominantly AC (alternating current) transmission and distribution. DC (direct current) is typically only used for high capacity and long distance interconnectors between separate networks (e.g., undersea cable connections). Network voltages (and sometimes frequency) must be altered at different locations to minimise transmission and distribution losses, and deliver power to consumers at manageable levels. These voltages and the resulting power flows are managed with the following devices which are designed to operate most efficiently at a fixed maximum capacity with minimum variations in demand or supply:

Circuit breakers

Circuit breakers allow substations to be disconnected from the transmission network, or for distribution lines to be disconnected.

Transformers

Transformers are used for most voltage conversion duties in AC electrical power networks. Transformers are passive devices: they operate using simple electromagnetic principles without any active modulation or switching schemes. Simple tap-changers are used on some transformers to regulate network voltage in discrete steps within relatively narrow ranges according to varying demand. Transformers can also transfer power bi- directionally, depending upon the balance of generation and load on each side of the device. Rectifiers

Rectifiers perform direct conversion from AC to DC electrical power using semiconductor diodes or similar devices. These are also passive devices in the sense that there is no inherent switching or control capability built into their design. Large scale rectifiers are used for HV DC transmission.

Frequency Converters

These are more sophisticated devices which use active high-speed electronic switching of the mains supply to deliver frequency conversion between two different parts of the network. They range in size from small domestic network-connected solar panel inverters up to long distance HV DC to AC converters.

Power Correction Devices

There are a number of devices with the sole purpose of undertaking corrective actions on the power in the system to maintain a steady and clean power supply for use. This group of devices includes, but is not limited to, filters, capacitors, and inductors.

Network Limitation

Increasing penetration of variable and non-dispatchable electricity generation sources (/.e., renewable energy sources such as wind and solar) are changing both the physical architecture of the network, and our ability to control generation levels. This causes increasing issues of stability and efficiency, with the threshold capability of current electricity network architecture falling far short of required levels.

A New Approach to Existing Electricity Networks

It is unanimously agreed that our current electricity transmission and distribution networks are unable to provide a usable power supply above a certain threshold penetration of clean energy from wind and solar. This threshold point varies for each network based on physical architecture, generation and load profiles, as well as a myriad of other factors.

The design architecture and technology that our electricity networks use have significantly improved in cost and efficiency over the last 120 years. Yet they still utilise the same fundamental technology and design architecture that were established in the 19th Century. All currently proposed options to upgrade our networks rely on existing operating methodologies, technologies and systems. In the case of energy storage, the technology to make this economically feasible has not yet been invented. All these options add significant cost, complexity and fragility into the network, and reduce its efficiency.

It is desired to alleviate one or more difficulties of the prior art, or to at least provide a useful alternative.

SUMMARY

In accordance with some embodiments of the present invention, there is provided a n electrical power control apparatus, including :

a magnetic core including at least three interconnected limbs defining a plurality of magnetic circuits, the at least three interconnected limbs including a first limb and second limbs;

a primary winding to receive input electrical power having an input voltage, an input current, and an input power factor, the primary winding being wound around the first limb of the limbs of the magnetic core to generate a corresponding first magnetic flux in the first limb that is divided between the second limbs;

secondary windings to provide output electrical power having an output voltage and an output power factor to a load, the secondary windings being wound around the first and second limbs of the magnetic core to generate the output voltage across the secondary windings from magnetic flux in the first and second limbs of the magnetic core, wherein the secondary windings around the second limbs are wound in opposite polarities; and

control windings wound around the second limbs to modify the magnetic flux in the second limbs and thereby modify the output voltage and the input power factor, wherein the control windings around the second limbs are wound in opposite polarities.

In some embodiments, the apparatus includes a further winding around the first limb of the magnetic core to supply electric power to electronic components of the apparatus.

In some embodiments, a portion of the secondary coil wound around the first limb is wound over the primary winding so that the impedance of the apparatus is substantially equal to a predetermined value. In some embodiments, the apparatus includes a control component to measure the input voltage, the input current, and the output voltage, and to control the energising of the control windings on the basis of the measured input voltage, input current, and output voltage in order to control the output voltage and the input power factor.

In some embodiments, the control component controls the energising of the control windings to maintain the output voltage at a desired voltage substantially independently of fluctuations and/or deviations of the input voltage, and to maintain the input power factor substantially at unity and substantially independently of the output power factor.

In some embodiments, the control component includes:

a reactive power control loop to generate an output voltage phase signal representing a desired phase angle of the output voltage based on a phase angle of the input voltage and a comparison of the reactive power of the input power and reactive power of the output power; and

a voltage control loop to process the voltage phase signal, a voltage amplitude signal representing the measured output voltage, and a reference voltage amplitude signal representing a desired output voltage to generate a control winding signal representing a corresponding signal to be applied to the control windings so that the output voltage is substantially equal to the desired output voltage, and the input power factor is maintained at substantially unity.

In some embodiments, the control component includes:

a phase locked loop (PLL) to generate an input phase signal representing a phase angle of the input voltage, and an input frequency signal representing a frequency of the input voltage;

wherein the input phase signal and the input frequency signal are used by the reactive power control loop and the voltage control loop.

In some embodiments, the reactive power control loop includes:

a reactive power estimator that generates an input reactive power estimate signal representing reactive power of the input power based on the measured input voltage, the measured input current, and the input frequency signal; and

an input reactive power control component that processes the input phase signal, the input reactive power estimate signal and a reactive power demand signal representing reactive power of the output power to generate the voltage phase signal.

In some embodiments, the control component suppresses harmonics such that harmonics present in the input voltage are substantially reduced in the output voltage.

In some embodiments, the input electrical power and the output electrical power are three phase powers, and the apparatus includes three of said magnetic cores connected in parallel to the control component, wherein, for each of the three magnetic cores, the control component maintains the corresponding phase output voltage at the desired voltage substantially independently of the corresponding phase input voltage, and maintains the input power factor of the corresponding phase substantially at unity and substantially independently of the output power factor for each of the corresponding phases.

In some embodiments, power is transferred between the magnetic cores in order to balance power across all three phases.

In some embodiments, the primary windings of the respective cores are interconnected in a delta configuration, and the secondary windings of the respective cores are interconnected in a wye configuration, wherein power is transferred between the magnetic cores in order to balance power across all three phases.

In some embodiments, the apparatus includes a power electronics component to receive control coil signals generated by the control component and to generate corresponding control coil currents to power the control coils and thereby control the output voltage and the input power factor.

In some embodiments, the power electronics component receives electric power from an additional winding around the first limb of the core.

In some embodiments, the power electronics component includes a rectifier to rectify received AC power, and an inverter component to generate control coil currents to power the control coils and thereby control the output voltage and the input power factor.

In some embodiments, the power electronics component includes a low-pass filter to remove high frequency digital artefacts from the control coil currents. In some embodiments, the inverter component includes only two transistors, the transistors being connected in series between voltage rails, and each transistor being bypassed by a corresponding diode, wherein digital control signals applied to the gates of the transistors enable the generation of the control coil currents digital signals at the junction between the two transistors.

In some embodiments, the rectifier component includes transistors that are controlled by the control component in order to control the DC output voltage generated by the rectifier component.

In some embodiments, output of the rectifier component is coupled to the inverter component via a pair of series-connected capacitors in parallel with the rectifier component, wherein a neutral rail is connected to the common terminals of the capacitors, and the control component compares voltages across the capacitors and adjusts control signals sent to the rectifier component and the inverter component in order to balance the voltage rails about the neutral rail.

In some embodiments, the apparatus is configured to lift a sagging input voltage by generating the control coil signals so as to over-correct the input power factor.

In accordance with some embodiments of the present invention, there is provided a n electrical power control process, including :

receiving, at a control component of an electrical power control apparatus, signals representing input voltage, input current, and output voltage of a magnetic core of the electrical power control apparatus, the magnetic core including :

at least three interconnected limbs defining a plurality of magnetic circuits, the at least three interconnected limbs including a first limb and second limbs; a primary winding to receive input electrical power having an input voltage, an input current, and an input power factor, the primary winding being wound around the first limb of the limbs of the magnetic core to generate a corresponding first magnetic flux in the first limb that is divided between the second limbs;

secondary windings to provide output electrical power having an output voltage and an output power factor to a load, the secondary windings being wound around the first and second limbs of the magnetic core to generate the output voltage across the secondary windings from magnetic flux in the first and second limbs of the magnetic core, wherein the secondary windings around the second limbs are wound in opposite polarities; and

control windings wound around the second limbs to modify the magnetic flux in the second limbs and thereby modify the output voltage, wherein the control windings around the second limbs are wound in opposite polarities;

processing the received signals to control the energising of the control windings on the basis of the measured input voltage, input current, and output voltage in order to control the output voltage and the input power factor.

In some embodiments, the energising of the control windings filters harmonics present in an input signal received at the primary winding and corresponding to the input voltage so that the harmonics are absent or at least substantially reduced in an output signal generated at the secondary winding and corresponding to the output voltage.

In accordance with some embodiments of the present invention, there is provided a n electrical power control apparatus for use with an electromagnetic device having a magnetic core with limbs defining a plurality of magnetic circuits and windings around the limbs of the magnetic core, and that receives input electrical power having an input voltage, an input current, and an input power factor, and switches magnetic flux between the magnetic circuits under control of electric current in control windings of the windings to control the output voltage and input power factor, the apparatus including :

a control component to measure the input voltage, the input current, and the output voltage, and to generate control signals to control the energising of the control windings on the basis of the measured input voltage, input current, and output voltage in order to control the output voltage and the input power factor; and

a power electronics component to receive the control signals generated by the control component, and to generate corresponding control winding currents to power the control coils and thereby control the output voltage and the input power factor.

In some embodiments, the power electronics component receives electric power from an additional winding around the first limb of the core. In some embodiments, the power electronics component includes a rectifier to rectify received AC power, and an inverter component to generate control coil currents to power the control coils and thereby control the output voltage and the input power factor.

In some embodiments, the power electronics component includes a low-pass filter to remove high frequency digital artefacts from the control coil currents.

In some embodiments, the inverter component includes only two transistors, the transistors being connected in series between voltage rails, and each transistor being bypassed by a corresponding diode, wherein digital control signals applied to the gates of the transistors enable the generation of the control coil currents digital signals at the junction between the two transistors.

In some embodiments, the rectifier component includes transistors that are controlled by the control component in order to control the DC output voltage generated by the rectifier component.

In some embodiments, output of the rectifier component is coupled to the inverter component via a pair of series-connected capacitors in parallel with the rectifier component, wherein a neutral rail is connected to the common terminals of the capacitors, and the control component compares voltages across the capacitors and adjusts control signals sent to the rectifier component and the inverter component in order to balance the voltage rails about the neutral rail.

In some embodiments, the apparatus is configured to lift a sagging input voltage by generating the control coil signals so as to over-correct the input power factor.

BRIEF DESCRIPTION OF THE DRAWINGS

Some embodiments of the present invention will now be described, by way of example only, with reference to the accompanying figures, wherein :

Figure 1 is a functional block diagram of an electrical power control

apparatus/process in accordance with some embodiments of the present invention;

Figure 2 is a block diagram of an electrical power control apparatus in accordance with some embodiments of the present invention, including a magnetic core having multiple magnetic circuits, and a control system to control the modulation of magnetic flux in those magnetic circuits;

Figure 3 is a schematic diagram of a magnetic core of an embodiment of the electrical power control apparatus, showing the configurations and connections of the primary, secondary, and control windings;

Figure 4 is a graph showing the B-H curve of electrical steel used in the described embodiments of the electrical power control apparatus;

Figure 5 is a magnetic circuit diagram of an embodiment of the electrical power control apparatus, showing the configuration of primary, secondary, control, and rectifier windings/coils;

Figure 6 is a schematic diagram showing the configuration of the magnetic core and windings/coils of an embodiment of the electrical power control apparatus;

Figure 7 is a cross-sectional side view of the magnetic core of an embodiment of the electrical power control apparatus as shown in Figure 26 showing the arrangement of the copper turns in the coils, and the locations of thermocouple sensors;

Figure 8 is a three-dimensional wire frame view showing the three-dimensional spatial arrangement of the windings/coils shown in Figure 26, together with a structural frame to support and move the apparatus;

Figures 9 to 12 are three-dimensional surface plots of respectively the output to input voltage ratio (Figure 18), the output power (Figure 19), the control power (Figure 20), and the power factor (Figure 21), each of these being plotted as a function of input voltage amplitude and phase for different control voltages;

Figure 13 is a functional block diagram of an embodiment of the control system of the electrical power control apparatus;

Figure 14 is a block diagram of a phase lock loop (PLL) component of the control system;

Figure 15 is a block diagram of a vector signal generator component of the control system;

Figure 16 is a block diagram of a half park transform component of the control system;

Figure 17 is a block diagram of a reference voltage generator component of the control system; Figure 18 is a block diagram of a voltage controller component of the control system;

Figure 19 is a block diagram of a reactive power estimator component of the control system;

Figure 20 is a block diagram of a reactive power regulator component of the control system;

Figure 21 is a Nichols plot of the response of the phase lock loop component of Figure 14;

Figure 22 is a Nichols plot of the response of the voltage controller component of Figure 18 for different power factors;

Figure 23 is a Nichols plot of the response of the reactive power components of Figures 13 for different power factors;

Figure 24 to 29 are screen capture images showing actual measured waveforms of input signals provided to the apparatus and corresponding output signals generated by the apparatus, including the input voltage waveform, RMS voltage, and power factor, and the output voltage waveform, RMS voltage and power factor;

Figure 30 is a graph of a measured input voltage waveform, and the corresponding control signal voltage and output voltage waveforms, demonstrating removal of 3 ^rd , 5 ^th and 7 ^th harmonics present on the input waveform from the output generated by the electrical power control apparatus;

Figure 31 is a plot of a physical voltage dip test completed to IEC standard 610000-4-11, showing the reduction in voltage dip in the output generated by the electrical power control apparatus;

Figure 2 is a circuit diagram with a rectifier circuit, a DC link and two bridge circuits to provide pulse width modulation of a control winding of the electrical power control apparatus;

Figure 33 is a block diagram of power electronics used to drive the control windings and rectifier winding of the apparatus of Figure 25;

Figure 34 is a circuit diagram of the power electronics 204 in accordance with one embodiment of the present invention, including a rectifier circuit, a DC link and two bridge circuits to provide pulse width modulation (PWM) of the control windings and the rectifier; Figure 35 is a block diagram of the power electronics component 204 of the apparatus 100 in accordance with an alternative embodiment of the present invention;

Figure 36 is a circuit diagram of a low pass filter of the power electronics component;

Figure 37 is a circuit diagram of an inverter of the power electronics component;

Figure 38 is a cross-sectional side view of a portion of the central limb of the magnetic core, showing the placement of an additional secondary coil around the primary coil to provide a specified impedance so that the device has the characteristics of a standard transformer when the device is inactive;

Figures 39 and 40 show the control laws used by the control system to control the inverter and the rectfier, respectively, of the power electronics component of the apparatus.

DETAILED DESCRIPTION

Embodiments of the present invention include an electrical power control apparatus and process that involve receiving input electrical energy in the form of an input signal having some voltage waveform and root-mean-square (RMS) voltage, and converting that input electrical energy to output electrical energy in the form of an output signal having a desired or 'target' voltage waveform and a desired or 'target' output RMS voltage. The input electrical energy will typically vary over time (that is, its voltage waveform and/or RMS voltage is time-dependent), and thus the apparatus and process operate to dynamically control the conversion so that the output electrical energy has the desired target voltage waveform and target RMS voltage, which themselves may also vary over time.

Additionally and simultaneously, the output electrical energy of the apparatus will have a power factor determined by the downstream load drawing power from the apparatus. The apparatus determines that power factor on its output, and provides a unity power factor on its input, such that (the input of) the apparatus appears as an ideal (purely resistive) load.

Embodiments of the present invention are thus able to provide voltage waveform and RMS voltage conversion while simultaneously providing power factor correction, utilising high-speed electromagnetic path modulation instead of electronic circuit switching (used in power electronics devices) to deliver improved efficiency and performance (while also electrically isolating the two systems).

Although embodiments of the present invention are primarily described herein in the context of mains power distribution, it will be apparent to those skilled in the art that other embodiments may be configured for use in essentially any electrical system application that requires control of output power factor and/or voltage waveform and/or RMS voltage, including transformers, power factor compensation, and harmonic filters, for example. Many other applications of the electrical power supply system and process described herein will be apparent to those skilled in the art in light of this disclosure.

In this specification, unless the context indicates otherwise, the term "signal" is used for convenience of reference, and is to be construed broadly as referring to a form of electrical energy characterised by a voltage, current, and at least one fundamental frequency (which would be zero in the case of a DC voltage), and does not necessarily require that any form of information is represented by or conveyed by the signal, notwithstanding that some embodiments of the invention may involve the communication of information encoded in the signal.

As shown in the functional block diagram of Figure 1, an electrical power control apparatus 100 includes a voltage control component 104, a power factor control component 108, and a controller 106. As described above, the voltage control component 104 receives an input signal of some characteristic input waveform and RMS voltage, and provides a corresponding output signal having a selected or desired output waveform and RMS voltage, regardless of fluctuations or deviations in the input voltage.

Similarly, the power factor control component 108 receives an input signal of some output power factor, and provides a corresponding signal having a selected or desired input power factor, regardless of fluctuations or deviations in the input signal power factor. In combination, the voltage control component 104 and the power factor control component 108 act in concert so that the apparatus 100 receives an input signal having some waveform and (typically time-varying) characteristic RMS voltage Vm and power factor PFin, and generates a corresponding output signal having a desired or selected waveform and RMS voltage Vout and a desired or selected power factor PF _0Ut , where the output RMS voltage Vout is maintained substantially at a desired value and the input power factor PFin is maintained substantially at unity. For example, where the electrical power control apparatus 100 is used as part of a mains electricity distribution system or grid, the apparatus 100 can be configured so that the output voltage V _out is maintained at the required RMS mains (or distribution) voltage, independent of Vm, and the output waveform is maintained as a clean (or at least improved) sine wave independent of noise and/or distortions of the input waveform, and the power factor on the input side is maintained at 1, independent of the output power factor (which is determined by the load on the apparatus 100). Waveform harmonics and voltage deviations are a major source of losses and instability within an electricity grid, and consequently the operation of the apparatus 100 reduces losses and improves stability of the mains electricity supply network).

The controller 106 uses step down devices ( e.g ., buck converters in some embodiments) to monitor the input voltage Vm, and generates corresponding voltage control and power factor control signals that are respectively used to control the operation of the voltage control component 104, and the power factor control component 108.

It will be apparent that the same result can be achieved using a current control component instead of the voltage control component, measuring electric current instead of voltage, at the same location(s), but using current sensors instead of voltage sensors.

Similarly, the same result can be achieved using a magnetic flux control component instead of a voltage (or current) control component, using flux sensors instead of voltage (or current) sensors, at the same location(s), to measure magnetic flux instead of voltage (or current).

Figure 2 is a block diagram illustrating the architecture or high-level structure of the electrical power control apparatus 100 in accordance with the described embodiments of the present invention, and in the context of mains electricity supply. The apparatus 100 includes an electromagnetic device 202, a power electronics component 204, and a control system 206. Typically, the operation of the apparatus 100 will be monitored and managed at a high level by a SCADA (Supervisory Control and Data Acquisition) system of an electricity supply organisation.

Most fundamentally, the conversion of electrical power by the apparatus 100 is effected by the electromagnetic device 202 (also referred to hereinafter as "the device" for convenience). The device 202 is similar to a transformer, having a magnetic core with primary and secondary windings, but includes multiple magnetic circuits controlled by additional control and secondary windings. The current in the control windings is supplied by the power electronics component 204, under control of the control system 206.

As known by those skilled in the art, a transformer is an electromagnetic device that transfers electric energy from one circuit to another circuit via mutual inductance, and is typically made up of a magnetic core, a primary winding wound around a first part of the core, and a secondary winding wound around a second part of the core. When an alternating voltage is applied to the primary winding, an alternating current flows through the primary winding. This magnetizing current produces an alternating magnetic flux. The flux is mostly constrained within the magnetic core, and induces a corresponding voltage in the linked secondary winding, which, if connected to an electrical load, produces an alternating current. This secondary load current produces its own alternating magnetic flux which links back to the primary winding.

The secondary voltage is determined by the product of the primary voltage and the ratio of the number of turns in the secondary winding and the number of turns in the primary winding. Transformers are commonly used to convert between high and low voltages, but they are bulky by necessity at mains power distribution frequencies. They offer high efficiency, simplicity of design, and bidirectional power transfer. However, their passive nature affords limited regulation of the power transferred, requiring the addition of inefficient power factor control and voltage regulation components. The physics of operation at mains supply frequencies also makes transformers comparatively large for a given power rating, increasing costs of materials, fabrication and insulation management.

Electromagnetic Device

As shown in Figure 3, in the described embodiment, the device 202 includes a magnetic core in the form of an El core (of single phase shell configuration) where primary and secondary coils or windings (these terms being equivalent in this specification) are wound around the centre limb of the El core; however, unlike a conventional transformer, the secondary windings around the central limb are connected in series to an additional pair of coils on the two outer limbs, one wound with the same polarity or sense as the centre secondary coil, and the other wound with the opposite polarity or sense. In this specification, these series-connected windings or coils are all referred to as 'secondary' coils, notwithstanding that they are wound around different limbs of the magnetic core. A corresponding pair of control coils are also wound around the outer pair of limbs, with the two control coils wound in opposite polarities or senses on each limb. This geometry, combined with selection of the magnetic core material (being an M400- 50A non-grain-oriented soft electrical steel in the described embodiments), results in good magnetic coupling between the primary coils and secondary coils, good magnetic coupling between the secondary coils and control coils, but poor magnetic coupling between the primary coils and the control coils, this general arrangement providing the best performance.

The primary winding on the centre arm has Ni turns, and the secondary winding is split across all 3 limbs, with the centre limb having I turns, and the outer limbs having xI h and -xl\l2 turns, meaning that they are wound in opposite directions. The control windings have an equal number of turns wound in opposite directions.

When the control windings are turned off (have no power flowing through them), the flux flowing in the core is evenly split between the two outer flux paths (/.e., defined by the outer limbs) because the geometry is symmetrical. As the two outer secondary windings are wound in opposite directions, they effectively cancel each other out in the magnetic circuit, and do not affect the magnetic field in the central limb, and consequently the voltage transformation between the primary and secondary windings is Ni to I h, as would be the case for a normal transformer. A consequence of this configuration is that if there is no control of the device (for example if the control component fails for any reason), the device operates as though it is a normal transformer.

By energising the control windings (putting AC or DC power through them) the balance of the flux between the two magnetic pathways is altered. This results in the two outer secondary windings each having a different amount of flux flowing through them. As they are wound with opposite polarities this results in the effect of adding or subtracting turns from the secondary winding via a proportion of N2 + xN2 or a proportion N2 + - xN2, depending on the relative proportions of flux in each limb. This changes the voltage transformation ratio NI to N2 of the device, with no moving parts or power electronics in the power flow path.

Furthermore, as they do not interact with the primary coil, a phase shift in the control coil voltage will produce a phase shift between the primary and secondary coils. This allows the input and output power factors to be adjusted independently. For the device 202 to operate optimally, it is advantageous to have good magnetic coupling between the primary and the secondary windings, and between the secondary and the control windings, but little or (ideally) no coupling between the primary and the control windings. When in operation, uniquely, flux circulates around the outer shell of the core, not passing through the centre limb. This flux is stored magnetic energy that allows the apparatus 100 to provide power factor control by absorbing or injecting energy The amount of magnetic energy that can be stored within the core is dependent on the physical size of the core, because more magnetic material is required to store a larger amount of magnetic energy without the material becoming magnetically saturated. As the core becomes larger, the efficiency of the voltage transform will reduce due to the increased iron losses associated with a larger core. There is therefore a design trade off between efficiency and magnetic energy storage which can be optimised for the specific design requirements for each device and application.

In some embodiments, an additional coil is added to the centre arm of the core, as shown in Figure 26, and is used to supply power to the control coils (and in some embodiments also to the control system 206) via the power electronics 204 and under control of the control system 206, as described below.

In the described embodiments, the device operates in a different section of the core material B-H curve to a normal transformer. A material with linear properties in the operating area of the B-H curve is desirable to facilitate controlling the device to deliver voltage waveform, RMS voltage and power factor control. Similarly, the core material should have a permeability curve that has a soft increase (/.e., does not rapidly change from one value to another) in the magnetic field up to material saturation to enable the device to be controlled to deliver voltage waveform, RMS voltage and power factor control.

The core of the device can be composed of any electromagnetic material that will allow magnetic flux to be transferred between windings. In the described device, this is a non- grain-oriented electrical grade steel. The windings of the device can be made with any electrical conductor, in the described device this is high conductivity copper.

The specific configuration of a device (including winding and core materials and configurations, laminations, dimensions, core size and shape, number of phases and face configurations, et cetera) can be determined using standard electromagnetic design methods known to those skilled in the art, and manufactured using standard core cutting and stacking techniques (see, for example, http://sites.ieee.org/qms- pes/files/20 4/ l/Tr3nsformer-Manuf3Ctu rinq-Frocesses. pdf).

In the described embodiments, the core material that is used is M400-50A electrical steel. The relevant properties of this material are given in the table below, and its flux density against field strength (B-H curve) is shown in Figure 4.

The sizing of the magnetic core is dependent on the amount of energy passing through the device, in the same way that the core of a standard transformer is dependent on its power rating.

The number of windings in the primary, secondary and control coils is dependent on the required performance of the device, including the standard voltage transformation and the range of voltage control performance. The configuration and dimensions of one embodiment of the device are shown in Figure 5, with each coil having the following number of turns: A 113, B 54, C 13, D13, E 14, F 14, G 79. A cross-sectional view of these windings in the gaps between the central and outer limbs of the core is shown in Figure 6. Figure 8 is a three-dimensional wire frame view showing the three- dimensional spatial arrangement of the windings/coils, together with a structural frame to support the device and facilitate transport.

Device Performance Relationships

The performance of the device 202 is dependent on several parameters such as the mutual and self inductances of the primary, secondary and control coils, as well as a number of variables such as the input, output and control voltages, power throughput and load power factor.

Let L (H) and R (W) denote the inductance and resistance matrices between the device primary (1), secondary (2) and control (3) coils. Define w = 2pί (rad/s) as the angular frequency of the supply voltage, i = [ii , i ₂ , i ₃ ] (A) as the current through the coils and v = [vi , V2 , v ₃ ] (V) as the voltage across them. The relationship between current and voltage across the device is given by:

[R - jcoL] i = v

If the load impedance Z = R (1 + j (1 - p ² ) ^1/2 / p ) where R (W) is the load resistance and p the power factor then v ₂ = Z i ₂ and the above equation can be solved explicitly if vi and v ₃ are specified. In the following sections, input to output relationships are plotted as a function of the control voltage v ₃ for the parameter values listed in the table in order to demonstrate the characteristics of the apparatus 100.

With respect to the described embodiment, Figure 9 shows the voltage ratio VI / V2 x | v2 1 / | vl | x 100% as a function of the control voltage magnitude and phase relative to the primary voltage signal. The voltage ratio clearly varies with the control voltage and in fact the relationship is almost linear when the control is in phase with the primary voltage. This implies that ratio across the device can be altered and, since the variation is monotonic, it can be controlled.

With respect to the described embodiment, Figure 10 shows the output power as a function of the control voltage magnitude and phase relative to the primary voltage signal. The output power varies as the output voltage squared hence the profile of output power is very similar to that of output voltage. In practice the output power will be determined by the behaviour of the electrical load. With respect to the described embodiment, Figure 11 shows the control power as a function of the control voltage magnitude and phase relative to the primary voltage signal. Control power can clearly be either positive or negative depending on the magnitude and phasing of the applied control voltage and the power electronics needed to drive the control coils must be able to inject and absorb power from the device.

With respect to the described embodiment, Figure 12 shows the input power factor as a function of the control voltage magnitude and phase relative to the primary voltage signal. Power factor is clearly influenced by the control voltage, particularly its phase. As the power factor relationship with respect to control voltage phase is monotonic, and that of the voltage ratio with respect control voltage amplitude is also monotonic, then it is possible to control both voltage ratio and input power factor simultaneously.

Control System

As shown in Figure 2, the control system 206 monitors the apparatus output voltage and input current and voltage, with the objective of ensuring that the output voltage waveform is a sinusoid of the correct amplitude and the input power factor is unity. Those skilled in the art will appreciate there are many ways to achieve the desired control, of various algorithms and configurations, using a variety of different hardware implementation types. One such implementation is described below, implemented as discrete processes executed by digital processors, as described below.

Control Architecture

The high level structure of the control system 206 and its control system process is shown in Figure 13, including three closed control loops: a phase lock loop (PLL), a Voltage Control Loop (VCL), and a Reactive Power Control Loop (RPL). The phase lock loop (PLL) monitors the input voltage and synchronizes the output frequency and phase angle of a reference voltage generator to the input waveform. In the VCL, a voltage controller monitors the error between this reference and the measured output voltage and generates the control demand needed to drive the voltage error to zero. The different loops are configured to operate at different frequencies spaced apart to reduce mutual interference. For example, the loop frequencies can be an order of magnitude apart, although in practice it is found that can be closer in frequency without apparent detriment to performance.

The phase of the reference voltage signal is augmented by the reactive power controller. This drives the estimated input reactive power to zero, and hence the input power factor to unity. Reactive power is estimated by monitoring input current and voltage over an entire cycle.

Phase lock loop (PLL)

The components of the phase lock loop are shown in Figures 14, 15, and 16. As shown in Figure 14, the input voltage is first scaled approximately to generate a signal that has a maximum value around unity. The scaling doesn't have to be precise, as the PLL is relatively insensitive to input amplitude. A vector consisting of the input signal phase shifted by 90 degrees is produced by an AB vector generator (shown in Figure 16). This is then Park transformed (by the half-Park transform shown in Figure 16) into the DQ frame, using the phase angle generated by the PLL. Only the Q (quadrature) component is needed.

The Q component is zero when the PLL phase is synchronised to the input signal. A PI control law (upper right of Figure 14) increases or lowers the estimated frequency (and hence its phase) in order to drive the Q error to zero. LP (low pass) filters (lower right of Figure 14) are included to remove any unwanted noise from the voltage signal. The feedforward reduces feedback requirements, allowing the bandwidth of the closed loop to be made as low as possible, which reduces the sensitivity to signal noise even further. Parameters for the PLL are defined in the table below.

In the described embodiment, the PLL provides the phase synchronisation needed by the other two control loops, but its operation is virtually independent. The loop is deliberately made as low bandwidth as possible to enhance suppression. The open loop characteristics are shown in the Nichols plot of Figure 21, and key control parameters are summarised in the table below.

Reference voltage generator

The purpose of the reference voltage generator is to produce a voltage demand for the voltage control loop to track. As shown in Figure 17, the amplitude input is set to the desired output voltage, and the phase is set by the PLL and reactive power regulator. As such, no parameters are needed.

High bandwidth voltage control

The voltage control law shown in Figure 18 is a high gain integral action control law equipped with low bandwidth feedback to wash out unwanted DC signal. Parameters for the control law are defined in the table below.

In the described embodiment, the voltage control loop is required to track the voltage reference signal precisely, which requires a high bandwidth to attenuate harmonics. A Nichols plot of the voltage open loop is shown in Figure 22, and the key control parameters are summarised in the table below. Note that the open loop transfer function (OLTF) is relatively independent of the load power factor due to careful selection of the core magnetic material to minimise core flux dispersion (leakage) and primary to control coil cross coupling inductance. The low frequency roll-off of the control system is required to avoid saturating the control amplifier input with a DC signal.

Reactive power estimator

An estimate of reactive power is needed to control the power factor. Reactive power is computed by multiplying the quadrature component of the input voltage by the current, and averaging this over a complete cycle. The reactive power estimator is shown in block diagram form in Figure 19. Note that ¹ / ₄ (quadrature) and full cycle delays are approximated to the nearest sample time using the z ^_d operator.

Reactive power regulator

The reactive power regulator is a low gain limited integrator, as shown in Figure 20. The limiter is used to curtail the degree of control action used, to avoid saturating the magnetic core with unrealistic power factor compensation demands. Parameters for the regulator are defined in the table below.

The reactive power regulator is required to hold the input power factor at unity. The control loop is inherently nonlinear by virtue of the product of voltage and the quadrature current used in the reactive power estimation process. Stability and performance should therefore be evaluated using a nonlinear method of analysis, such as phase plane. The Nichol's plot of Figure 23 shows the OLTF of the loop when linearised about a primary voltage. The key parameters are summarised in the table below.

Figures 24 to 29 are screenshot images showing the actual measured input and outputs of the described embodiment apparatus in operation. Figure 24 shows the primary AC voltage waveform, with significant random noise. As shown in Figure 25, the output waveform on the secondary is a relatively clean AC waveform. In the described embodiment, the apparatus is able to reduce the total harmonic distortion (calculated as the harmonics energy divided by the total energy) from 35% on the input to less than 2% on the output.

Figure 26 shows the RMS voltage, with random noise and a step function on the primary windings, and Figure 27 shows a stable output voltage on the secondary winding.

Figure 28 shows a load power factor varying on the secondary winding. However, this is not passed upstream to the primary winding, as can be seen by the stable unity power factor of the primary winding shown in Figure 29. The harmonic suppression is further shown in Figure 30. Figure 31 shows the apparatus performance for a voltage dip.

The power flow through the apparatus 100 is regulated by controlling its control windings shown in Figure 3. As the input power and output power draw can be constantly changing, the control windings are correspondingly adjusted by changing the DC or AC control current through the control windings to provide a selected level of reluctance in the outer limbs of the core. In various embodiments, the control current can be either a direct analog signal, where the current level is changed directly, or can take the form of a digital signal. For example, the described embodiments use pulse width modulation (PWM) to provide an equivalent average current. PWM uses a digital signal switched at a rate much higher than will affect the load to control the power supplied. Switching the voltage to a load with the appropriate duty cycle approximates the desired voltage level. In the same manner, the duty cycle can be varied to deliver an approximation of an analog waveform using digital sources. Modern semiconductors are able to provide this switching in microseconds, meaning that power loss is very low, but the imperfect waveform produce can cause significant harmonics and losses in some applications.

In some embodiments, the PWM signals are generated in the control system 206 by a high speed microprocessor (such as the 100MHz Texas Instruments device described at which has sensor inputs to receive signals representing the voltages of the input and output (the circuit directly before the primary and directly after the secondary windings). In the described embodiments, the control system 206 achieves PWM control of the current through the control windings using the power electronics circuit 204, consisting of a rectifier, a DC link, and an inverter. In one embodiment, the power electronics 204 consists of two bridges, as shown in Figure 32. The two bridges are made up of four diodes and four fast digital switches, being insulated-gate bipolar transistors (IGBTs) in the described embodiment. The four IGBTs are switched on and off by signals generated by the control system 206 in order to provide the correct current level in the control windings (and thus the correct flux in the outer limbs of the core) to deliver the target signal output. It will be apparent to those skilled in the art that this outcome can be achieved by a number of different circuit topologies. A control circuitry topology utilising step down converters (buck converters) allows a greatly reduced power usage to achieve the desired magnitude and range of flux control.

By controlling the PWM at a speed sufficiently higher than the target mains frequency (e.g., 7.5 kHz for a target mains frequency of 50 Hz), the waveform of the output signal is modulated using the control windings in order to smooth out harmonics, as shown in Figure 30. Faster control speeds better compensate and correct harmonics.

Additionally, in some embodiments the rectifier is connected to the magnetic core of the device through an additional winding on the centre limb, as shown in Figures 5, 33 and 34 to provide the input to the rectifier (instead of an independent power supply) . The rectifier is actively controlled by the control system 206 (via generating PWM signals to gate the transistors in a desired manner) in order to regulate the DC link voltage and the input reactive power. Similarly, the inverter is able to control output voltage and also input reactive power. The combined control of both the rectifier and the inverter by the control system 206 substantially improves the regulation of input power factor compared to controlling only one of these components. Balancing reactive power injection between the control coil and the rectifier to hold unity power factor reduces the size of both the rectifier and the inverter.

In another embodiment, the power electronics 204 is of the form shown in Figure 35, consisting of a rectifier portion coupled to an inverter portion via a via a DC link with a pair of capacitors in series. In the rectifier portion, AC power received from the rectifier coil wound around the central limb of the magnetic core is filtered by a low-pass filter, and the resulting filtered signal is then fed to a digital rectifier circuit. As shown in Figure 36, the low-pass filter is an LCLCR(L) analog filter, where the final (L) represents the inductance of the rectifier and rectifier coil. In the described embodiment, the output of the rectifier is further low-pass filtered by the pair of series-coupled capacitors in parallel with the output, and is then fed to the inverter portion of the circuit.

In the inverter portion of the circuit, first an inverter circuit generates an initial coil control signal under control of PWM signals generated by the control system 206. In the described embodiment, the inverter circuit is of the surprisingly simple form shown in Figure 37, operating at a frequency of 7.5 kHz, but in another embodiment the coil control signal generator circuit is a multi-modular converter (MMC) circuit known to those skilled in the art. As shown in Figure 37, the inventors developed the simplified inverter circuit for such active control, using only two transistors and two diodes (compared to the four transistors and six diodes of prior art inverter circuits typically used to control motors and the like.)

The high frequency switching components or artefacts resulting from the digital synthesis are filtered by a second filter, consisting of another LCLCR(L) analog filter, as shown in Figure 36, with the final (L) in this instance representing the inductance of the control coils themselves. This filter configuration is used to provide a frequency response that is quite flat at and above the mains frequency (of 60 Hz in the described embodiment) up to a frequency of about 15 kHz, after which higher frequencies are very steeply rolled off at -60 dB/decade. This is in contrast with standard filters used to filter PWM outputs, which generally have a resonance between the mains frequency and the high frequency rolloff.

The coupling of the two portions of the circuit is via a pair of capacitors (of equal capacitance) in series, with the connection between the capacitors being tied to the neutral supply line. A pair of voltage sensors measure the voltages between the neutral and the upper and lower rails (as VH and VL), and any imbalance between the magnitudes of these voltages causes the controller to modify the rectifier and inverter PWM signals in order to effectively introduce an offset so that the upper and lower rail voltages become equally spaced from the neutral rail potential. This is referred to herein as bridge voltage stabilisation.

Figure 38 shows the control law used to control the rectifier (transistors). The upper part of the Figure represents voltage control, whereby the desired reference voltage for the DC link is compared to the actual DC link voltage, and the difference is used to adjust the PWM signal so as to drive the voltage difference to zero. The lower part of Figure 38 is for power factor control, and the inputs are all quadrature components: for the desired reference QREF (in the context of mains supply, this is always zero, representing a desired power factor of unity), the measured input voltage Qm, and Qv, which is used for line voltage control. Specifically, by over driving the quadrature component (/.e., by correcting more than is required), the input line voltage can be lifted if required. This capability allows the apparatus to lift a sagging mains supply line voltage to its desired value, and thus the mains supply line voltage can be maintained over longer transmission distances than would otherwise be the case.

Figure 39 shows the control law used to control the inverter. The first part of the law is a power factor control component that is the same as described above for the rectifier. However, this is used with the PLL to generate a reference voltage VREF, which is compared to the measured load voltage to generate an error value that is used to adjust the PWM signals generated for the inverter circuit.

A significant feature of the electromagnetic device 202 is that it can operate as a standard transformer by disconnecting the power electronics 204 from the rectifier coil, and shorting the control coils. For example, the control system 206 may do this upon detecting a fault (e.g., voltage under or over specified limits) or in response to a severe power surge (e.g., as result of a lightning strike). The shorting of the control coils has the effect of balancing the magnetic flux in the outer limbs of the core. When used in an electricity grid, it may be an operational requirement for the device 202 to have a specified impedance equal to that of a standard mains transformer under these conditions. In the described embodiments, this is achieved by providing an additional secondary coil wound around the primary coil, as shown schematically in cross-section in Figure 40.

Three-phase power

The electrical power control apparatuses described above can be extended from controlling single phase power to control three-phase power by using three of the electromagnetic devices 202, one for each phase. This configuration is able to control voltage and power factor of each phase in the same manner as described herein for a single phase, with each phase having its own dedicated set of control loops and power electronics component 204 as described above. However, the control system 206 provides an additional capability to balance power across the 3 phases. In the three phase embodiments, the primary coils of the three electromagnetic control devices are interconnected in a delta configuration and the secondary coil in a wye configuration. This delta-wye configuration of the device provides automatic balancing between phases when no power is applied to the control coils, but as known by those skilled in the art, it also introduces a 30 degree phase shift that the control system 206 needs to correct as described above.

In some embodiments, an energy recovery circuit is used to store the excess energy that would be otherwise lost. This energy can be stored within a capacitor, inductor, battery or other energy storage device, and subsequently reinjected into the main power flow through the device directly, or directly from the capacitive, inductive or other storage.

Frequency Stability

As the frequency of the coupled electrical system is dependent on the frequency of the generated electricity and balanced with the load, as this balance between generation and load varies, the frequency of the system will vary. Therefore changing the control winding current, the reluctance of the circuit will change and energy will be injected into or taken from the power flow as described above. This can provide short term frequency stability for an electrical system.

Energy Storage

As described above, energy is stored in the magnetic core within an internal loop. With reference to Figure 3, energy injected into the core through the primary winding can circulate around the outer shell of the core. This effect allows out of phase reactive power (both capacitive and inductive reactive power) from the input to be phase shifted within the magnetic core, reducing reactive power on the output. This ability achieves the required power factor control, and is available for both capacitive and inductive power factors.

In some embodiments, that apparatus 100 includes an additional form of energy storage that can absorb and inject energy into the magnetic flux of the core. This can consist of capacitors or any other fast responding means of energy storage to increase the available stored energy. For instance this would increase the total available stored energy within the outer shell of Figure 3. In the described embodiments, Hall effect flux sensors are utilised located at the specific points where the control winding(s) interact and affect the flux in the magnetic core. These measurements are used in conjunction with the input and output voltage and frequency to determine the control signal required. However, it will be apparent to those skilled in the art that other measurement sensors and techniques can be used to monitor the flux; for example, a winding around the magnetic core that will have a current induced based on the flux.

The operational speed of the control system 206 is dependent on the frequency of the power flow being controlled by the system. Given this can be in the kHz range, high speed control can be utilised if required by the application requirements in use within a system. Within an electricity grid of 50Hz or 60Hz, microprocessors with relative low clock frequencies in the MHz range are sufficient.

The control system 206 constantly measures the voltage and frequency of the input signal and, in some embodiments, also the voltage and frequency of the output signal to directly control the control windings by way of the voltage and frequency and power factor control signals in order to maintain target setpoints.

In situations where the input is providing more power than the output requires at that instant, the controller reduces the power flow through the device by increasing the reluctance of the magnetic circuit. Increasing the reluctance of the device means that additional power is stored in the magnetic field of the device.

In situations where the input is providing less power than the output requires at that instant, the controller determines this by monitoring the input and output signals, and generates corresponding control signals to increase the power flow through the device by decreasing the control current in its control winding(s). This decreases the reluctance and increases the flux in the device, discharging stored magnetic energy within the device to the output, thereby delivering an essentially instantaneous power output that is greater than the power input to the device. If the device is already at a minimum level of reluctance (/.e., the control current is already zero and thus cannot be decreased further), in some embodiments the energy stored within additional energy storage devices such as capacitors can be used to compensate for this lack of power in the short term. An alternative control methodology for the electrical power supply system is a lagging control (rather than a leading control as described above). In this configuration, the apparatus acts in a synchronous manner, with the input and output injecting and sucking energy out of the magnetic field as required and based on the instantaneous input and output power levels. This then changes the reluctance of the magnetic circuit, and therefore the magnetic coupling between the primary and secondary windings allowing the control of the energy transformation through the device 202 as described above (including RMS voltage, harmonics, and power factor) . The control system 206 monitors the input and output, and then responds to the difference between the input and output measurements, changing the control setpoints to return the apparatus back to balance.

In the described embodiments, the control system 206 is implemented using a microprocessor for the PLL and a field programmable gate array (FPGA) for the other control processes, all of these being powered by the power flow through the device, and the electrical power supply processes are implemented as configuration data stored in non-volatile memory for the FPGA, and as executable instructions for the microprocessor. However, it will be apparent to those skilled in the art that in other embodiments the control system 206 could alternatively be implemented as an application-specific integrated circuit (ASIC), or as a microprocessor (for example, an Intel™ Architecture IA-64 Core i7 multi-core processor) programmed to execute instructions stored in non-volatile memory, or combinations of these. It will also be apparent to those skilled in the art that in other embodiments the controller may be powered by a separate local power supply where available, such as local control power from a distribution board.

Extending upon the PWM control implementation described above, the measured flux is used for a control feedback loop with the FPGA (or other controller device, where applicable) as shown in Figure 6. The reference signal for this control is provided by a phase lock loop (PLL) driven by the measured flux. The phase lock loop uses a phase detector, filter and voltage control oscillator with a feedback loop to lock the input and output frequency with each other as shown in Figure 7. In the described embodiments, the phase lock loop is controlled at a speed of 1kHz; however, it will be apparent to those skilled in the art that a different control speed can be used in other embodiments.

The same process applies for reverse power flow by interchanging the primary and secondary inputs, outputs and setpoints. The same process can be used for 3 phase power by applying to each phase. It will be apparent that the electrical power supply apparatus and processes described herein are particularly advantageous as they are able to dynamically and rapidly respond to changes in the input energy received by the apparatus in order to generate corresponding output energy having a target voltage and a target frequency. In particular, this ability allows the described apparatus and process to match the output energy to the energy required by the loads on the apparatus. Moreover, the system and process are bi-directional, meaning that they are able to do this for energy supplied from an energy grid and flowing in one direction, for example, and also for energy supplied from renewable energy sources, which may be flowing through the system in the opposite direction. For example, changes in local energy generation arising from changes in wind and/or changes in available sunlight are able to be mitigated by the system and process to provide a relatively constant output for a fixed load. Similarly, changes in the load on the apparatus can be compensated for within the ability of the corresponding device.

Many modifications will be apparent to those skilled in the art without departing from the scope of the present invention.