EP3238314A1

EP3238314A1 - Electrical power distributor and method for distributing electrical power

Info

Publication number: EP3238314A1
Application number: EP15813336.3A
Authority: EP
Inventors: Alexander Ebbes; Bernd REIFENHÄUSER
Original assignee: GIP AG
Current assignee: GIP AG
Priority date: 2014-12-22
Filing date: 2015-12-15
Publication date: 2017-11-01
Also published as: BR112017008706A2; WO2016102241A1; DE102014119431A9; US20170353032A1; MX2017006120A; CN107112754A; US10361564B2; DE102014119431A1

Abstract

The present invention relates to an electrical power distributor (2000) for an electrical grid, and to a method for distributing electrical power in an electrical grid. The method comprises: connecting at least three sources (3000) and sinks (1000) for electrical energy, each having a terminal of a distribution circuit, wherein the terminals of the distribution circuit are electrically connected to one another such that an electrical current can flow from each of the terminals to each of the other terminals; receiving data from the sources (3000) or sinks (1000); calculating the electrical power P(t) flowing via each of the terminals according to time t and according to the data received from the sources (3000) and sinks (1000); and controlling the electrical power P(t) flowing at a time t via each of the terminals by means of a respective power controller connected to the terminal.

Description

Electric power distributor and method for distributing electric power

The present invention relates to an electric power distributor for a power network with an electrical distribution circuit having at least three terminals, wherein the terminals are connected sources and sinks for electrical energy and wherein the three terminals are electrically connected together so that an electric current from each of the ports to each of the other ports.

The present invention also relates to a method for distributing electrical power in a power grid.

A central element in the use of renewable energy sources in a power grid, but also in the supply of energy to mobile consumers, such as motor vehicles with electric drive, are buffer storage. These are used to compensate for the volatility of the supply of electric power by the sources and the consumption by sinks for the electric power. Without a buffer memory, only the amount of electrical power can be transported and / or provided to the consumers or sinks that are being generated at the given time at any given time. Due to the volatility of both sources and sinks, security of supply can not be guaranteed.

In particular, the buffer memories must be able to compensate for short-term and short-term fluctuations on the source and sink sides. An example of this is the "kick-down", ie the short-term passage of the acceleration regulator of an electrically driven motor vehicle, but also network fluctuations due to short-term weather changes in a supply network with a large proportion of renewable energy sources and short-term load fluctuations are a suitable example of this The need for balancing mechanisms means that the buffers must be able to both dynamically and variably output and record power, and different types of cache buffer request situations can be distinguished:

1. bridging longer-term periods to season memories, 2. Balance sheet memory to compensate for under- and over-coverage in a power grid,

3. Balancing short-term and temporary shortfalls and surpluses, such as those that occur due to grid fluctuations in a grid, in order to stabilize the grid,

4. buffer memory for a digital endpoint in a routed packet-based transmission-based utility network and gateways or interfaces of such a routed packet-based transmission based utility network to the classic ohmic power grid;

5. Requirements resulting from new power grid concepts, such as packet-based power transport and

6. Requirements scenarios, as they occur, for example, in the power supply of electrically driven means of transport such as cars, ships and aircraft.

There may in principle be two cases to be distinguished from one another, namely that the electrical power demanded by one or a plurality of sinks at a time exceeds the power provided by the sources at that time and, secondly, that the one from the sources to one Power Provided is greater than the power fetched by the sinks at the same time.

Therefore, it is an object of the present invention to provide an electric power distributor and a method for distributing electric power in a power network, which makes it possible to orchestrate a plurality of sources such that one or a plurality of electric energy sinks provide the required power to each one Receives time. To do this, it is necessary to distribute the power provided by the sources to the sinks in such a way that the services requested by the sinks are also provided to them and any performance surpluses or performance deficits due to consideration of further sources and / or sinks, preferably by taking into account an energy store, which acts either as a source or sink to cover.

To achieve this object, an electric power distributor for a power supply is provided with an electrical distribution circuit having at least three terminals, with the terminals sources and sinks for electrical energy are connectable, wherein the three terminals are electrically connected to each other such that an electric current from each of the terminals to each of the other terminals, and each of the terminals each having a power divider arranged such that, in operation of the power distributor, the electrical power P (t) flowing through the respective terminal varies with time t is adjustable, a communication device which is connectable to a data network and which is arranged so that they are in the operation of the power distributor data from the sources and Sinks, and a controller for controlling a distribution of a flow of electric power P (t) in dependence on the time t at the terminals, wherein the controller is connected to the communication device such that the data received from the communication device of the controller processable wherein the controller is connected to each of the power controllers, the controller being arranged to operate, during operation of the power distributor, electrical power P (t) flowing across each of the terminals in response to data received from the sources or sinks and wherein the controller is set up such that, in operation of the power distributor, it controls the electrical power P (t) flowing through the respective connection as a function of the time t.

The idea underlying this electric power distributor is to provide time dependent power flows (also referred to as power profiles) having different characteristics, i. with different time histories of the power P (t) at the terminals of the distribution circuit of the power distributor, generated by the fact that the electric power distributor controls power flows from the sources connected to it and to the sinks connected to them, that of the sinks required service can be provided. The control mechanisms are based on data which the power distributor or its controller receives from the sources or sinks via a communication device which can be connected to a data network.

For the purposes of the present application, sources of electrical energy are understood to be all electrical devices which, when connected to the power distributor, provide electrical energy or power thereto. This can be very concrete, for example, power plants of all kinds or rechargeable energy storage, which are discharged. In addition, whole subnets may appear as sources, which are connected to one of the connections of the distribution circuit.

Correspondingly, sinks for electrical energy within the meaning of the present application are all types or types of electrical consumers, for example a household connection, but also whole subnetworks which receive power from the power distributor. A sink within the meaning of the present application is thus each unit of a power network into which electric power flows from the power distributor.

A power network, of which the electric power distributor forms a built-in state, is, for example, an electrical supply network in which the source and the sink are connected to the power distributor via an overhead line or a ground line or an electrical vehicle electrical system Aircraft, a ship or other means of transport. An electrical distribution circuit according to the present invention may be a bus bar in a first, simple embodiment, to which all terminals of the electric power distributor are connected in parallel. In such an embodiment of the distributor circuit, the power flows within the power distributor are determined solely by the power dividers in the respective connections.

In another embodiment of the invention, at least two of the terminals, which are connectable to sources or connected to sources when viewing an electrical power network, are connected via a so-called crossbar, i. a switching matrix, switchable with each other and with a bus bar, which in turn connected to the other terminals, with which further sources or preferably depressions are connected or connected. A switching network, also referred to in the past as a crossbar distributor, in the context of the present application serves for switching through the power provided by each source connected to the power distributor to at least one sink connected to the power distributor. Coupling fields for weak currents are known from the communication technology and belong there to the so-called space multiplexing. A switching matrix refers to a matrix connected together (so-called coupling multiple) of incoming and outgoing lines. While a switch fabric in communication technology is completely transparent, i. For the purposes of the present application, the term "switching network" in the context of the present application also includes switching networks which combine several power flows or divide line flows.

An embodiment of the distribution circuit as a switching matrix has the advantage over an embodiment in which all the connections are connected in parallel to a busbar, that the sources can be orchestrated irrespective of the voltages which they provide at the power distributor, and in this way each any voltage level or any temporal course of the power at the terminals, which are connected to sinks can be provided.

In a further embodiment of the power distributor according to the invention, all the sources and sinks are connected to a distribution circuit which is an electrical coupling field which selectively enables any connection of the connections with each other. Such a configuration has the highest complexity of the circuit, but also the greatest possible flexibility of the switching states. In particular, in such an embodiment it is not necessary to pay attention to whether a source or a drain is connected to a connection of the power distributor. In particular, with each of the ports in this Case, an energy storage are connected, which is both discharged and charged via the power distributor.

In one embodiment of the invention, all nodes of the switching matrix are formed by a controlled power divider according to embodiments of the present invention. In an alternative embodiment, all nodes of the switching matrix are formed by on / off switches.

In an embodiment of the power distributor according to the invention, the distributor circuit has a first and a second section, wherein the first section comprises a coupling field that is configured such that all connections of the first coupling field that can be connected to a source or a drain are parallel or in Series connected to the second portion of the distribution circuit are connectable. For this purpose, the nodes of the switching matrix of the first section of the distributor circuit are formed in an embodiment of regulated power controllers according to embodiments of the present invention. In one embodiment, the second section of the distributor circuit can be designed as a passive busbar or as a switching network, which has simple on / off switches at the node. The first section of such a distribution circuit is referred to in the context of the present application as a Physical Abstraction Layer. The advantage of an embodiment of a distribution circuit with a physical abstraction layer is that it has a high flexibility with simultaneously reduced complexity of the circuit.

The central element of the electric power distributor is a power divider in each of the terminals, wherein the power divider is set up such that, during operation of the power distributor, the electrical power P (t) flowing through the respective connection can be set as a function of time. In this case, a power divider is understood to mean any electronic device with which the electrical power P (t), which flows via the respective connection, can be set as a function of time t.

In one embodiment of the invention, the power divider is a combination of a boost converter and a buck converter whose voltage level can be controlled. For this purpose, the power divider in one embodiment has a control signal input which is connected to the control of the power distributor. In one embodiment, the control signal is in the form of the pulse width of a modulated voltage signal, the pulse width controlling the voltage level of the power controller.

In one embodiment, the boost converter and the buck converter of the power controller in each of the terminals are preferably both bidirectionally configured. In this way, a port of the power distributor can be connected to both a source and a sink become. This is particularly important if the source or sink is an energy store in the broadest sense, which can be both discharged and charged via the power distributor. While embodiments of the present invention are contemplated in which only a power divider is provided in the terminals of the distribution circuit, embodiments are preferred in which the power dividers are part of a device referred to in the present application as the Digital Flow Controller (DFC) in each of Connections of the distribution circuit is. Such a power flow controller in the sense of the present application can also be described as a regulated power divider whose power level is regulated by means of a control signal predetermined by the control of the power distributor.

Such a DFC therefore has, in addition to the power divider or the power controllers, a control which is connected, on the one hand, to the (central) control of the power distributor via a data network and, on the other hand, is connected to the one or more power controllers of the DFC, so that they are in operation Power distributor controls the state of the power divider or regulates it in one embodiment thereof. In one embodiment of the invention, the control of the DTC is arranged to convert each value for the power flow P (t) of the corresponding port which the control of the DTC receives from the (central) control of the power distributor into a control signal for driving the or the power divider implements. Such a control signal for driving the power controller is, in one embodiment, the pulse width of the voltage signals with which the gates of the controllable switches, e.g. Thyristors, the power controller are driven.

In one embodiment of the invention, each of the terminals of the power distributor has, in addition to the power divider, a measuring device for detecting an actual electrical power flowing across the terminal.

This meter is in one embodiment part of the DTC and connected to the controller of the DTC, the controller of the DTC being arranged to control the actual power through the power distributor terminal to be equal to one from the central power distributor controller predetermined power P (t), which represents the target power in this sense. Only a measuring device in each of the terminals allows control of the current flowing through the terminal power. In contrast, in one embodiment of the power distributor according to the invention, each of the terminals additionally has a voltage converter. This voltage converter is expediently part of the DFC. In one embodiment, the voltage converter is disposed between the measuring device and the power divider in the DFC.

Such voltage transformers are capable of converting an input voltage to a higher or lower output voltage. Voltage transformers which convert an input voltage to a higher voltage are also referred to as boost converters. Voltage transformers that convert an input voltage into a lower voltage are also referred to as a step-down converter. Step-up converters and step-down converters are typically present as direct current elements, so that when using boost converters and step-down converters in an alternating current system, preferably a DC or an alternating direction takes place in front of and behind such a voltage converter. The power distributor also provides certain functions essential to the invention when it receives data from either the sources or the sinks to calculate the electrical power P (t) flowing through each of the terminals at a time t. The power flows P (t) for the future times t for those sources and / or sinks for which there is no data must be estimated in this case.

Nevertheless, an embodiment is preferred in which at least data is received from the sources, and preferably also from the sinks.

A power flow control in an electrical power distributor, in one embodiment, requires that the power distributor at each time t have information about which sources at that time t can provide what maximum power P _ma x (t). This requires an information flow from the sources to the control of the power distributor. Therefore, the electrical power distributor according to the invention has a communication device which can be connected to a data network and which is set up in such a way that it receives data from the sources during the operation of the power distributor.

Such a communication device is in one embodiment an interface for connecting to a data network, it being immaterial for the present invention over which physical transmission path the data network transmits the data to the communication device of the power distributor. The data network may be, for example, a wired data network or a wireless network. The power distributor is arranged to receive data from the sources via the communication device. This assumes that the sources have the appropriate technology to generate data and transmit it to the communication facility of the distribution board. However, in one embodiment, in addition, sources may also be connected to the power distributor that do not provide a data connection to the power distributor. For this purpose, in one embodiment of the invention, the control of the power distributor is arranged so that for a source connected to the power distributor, which does not transmit state information in the form of data to the power distributor, an estimate of at a future time t maximum of this Source provides power P _ma x (t) in order to integrate this source in the power grid can. Such an estimate may be based, for example, on information about the type of source and / or measurement of the power provided by that source over a past period of time. In a further embodiment of the invention, the power distributor has at each time t information about which sinks connected to the power distributor decrease which power at precisely this point in time t. This requires an additional flow of information from the sinks to the control of the power distributor. Therefore, in such an embodiment, the electrical power distributor according to the invention has a communication device which can be connected to a data network and which is set up to additionally receive data from the sinks during operation of the power distributor.

The controller for controlling a distribution of the flow of electric power P (t) within the electric power distributor from the terminals connected to the sources to the terminals connected to the sinks is, for example, a microprocessor or a computer in general.

In one embodiment of the invention, the controller is configured and configured to control the power dividers in the operation of the power distributor such that at each time t, the electrical power P (t) provided at a port connected to a drain is equal to that power required by the sink at this time is Pdem (t).

The controller is connected to the communication device of the power distributor so that it can receive and process the data received from the communication device. In order to effect the distribution of the flow of electric power P (t) within the power distributor, the controller is also connected to each of the power dividers in the terminals of the power distributor. In one embodiment of the invention, the controller gives each of the power splitters at least one desired value for the power P (t) as a function of the time t, which is supposed to flow over the respective connection. While in one embodiment of the invention, the controller also performs a regulation of the power at the respective port and to this end receives a measured value for the actual power from the port, in other embodiments the power splitters themselves may have their own control circuit which is set up in this way in that it adjusts the actual power of the target power specified by the control. The combination of a power divider with the associated regulation of control and measuring device is referred to in the present application as DFC.

In one embodiment, the controller is informed for each time t about which of the terminals which power P (t) is fed into the power distributor and via which of the terminals which power P (t) is delivered at the time t.

For this purpose, in one embodiment of the invention, the controller is set up to calculate the electrical power P (t) flowing through each of the terminals during operation of the power distributor for each instant t

the maximum power available from each source at the time t

the electric power Pdem (t) required at the time t of each drain,

and that it controls the power dividers such that the calculated electrical power

P (t) is set at the time t at the respective terminal.

It is understood that the power P (t) fed into the power distributor from the sources at any time t is at most as large as the maximum electric power P _ma x (t) that can be provided by this source at that time. Also, in one embodiment, ideally, at the time t of each of the sinks connected to a terminal of the electric power distributor, an electric power equal to the electric power Pdem (t) required at that time t by the respective sink is provided. In other words, for each of the terminals of the electric power distributor, the controller determines a power profile, that is, a profile of the electric power P (t) flowing through the respective terminal for each time t.

In one embodiment of the invention, the controller is set up and configured to approximate the power P (t) as an integer multiple of an elementary power dP as a function of the time t at each of the terminals, where dP via a time interval dt is constant. Such an approximation can also be understood and referred to as digitization of the power profiles. This digitization of Performance Profiles with Discrete Elementary Performance Levels dP enables the algorithmic task of efficiently distributing the power of port-connected ports to ports connected to sinks with algorithms. Such algorithms will be described in detail below.

As an alternative to an approximation or digitization of the power profiles P (t) by an elementary power dP, where dP is constant over a time interval dt, it makes sense that the controller is set up and configured such that it performs in the operation of the power distributor P (t) is approximated as a function of the time t at each of the terminals as a formula P (t) = ^ _{= Q} 2 ^k dP. This approximation is known in the data processing as Power of 2 representation.

To accomplish the distribution of power from the sources to the sinks, two tasks have to be solved. On the one hand, the amounts of the power flows of all sources in the power distributor must be divided into the amounts of the individual power flows from the power distributor into the sinks. This task can be solved in particular for digitized power flows P (t) as described above. On the other hand, the physical paths required for the distribution within the distribution circuit must be determined and switched so that all requested power flows for the connected sinks are met.

Combined, i. the power distributor adds power flows from different sources so that the desired power flows are created towards the sinks. In one embodiment, this is achieved by adjusting the voltages at the terminals of the sources in such a way that the power flows occur there. Either in the case of parallel circuits by voltage leveling the currents are added or in the case of a series connection, the voltages are added, so that the individual power flows add.

The steps required in one embodiment can be summarized as follows:

1 . Determining the requested power profiles Pdem (t) of each sink associated with the power distributor for each future time t

2. Determine the power profiles for maximum provided power P _ma x (t) of each source for each future time t

2. Digitizing the power profiles of sources and sinks

3. Solution of the optimization task, how the power profiles of the sources can be distributed to the sinks, with an orchestration algorithm

4. Interconnecting the sources and sinks by means of the distribution circuit

5. Adjustment of the individual power flows at the connections of the sources and sinks by the power dividers 6. Control of the process by appropriate control and control circuits

The step listed in point 3 is based on the following task: Which sources must be switched in which time sequence over which routes of the distribution circuit and DFCs, so that at the terminals of the power distribution, which are connected to sinks, the power profiles requested by the sinks can be provided , This can be solved for more complicated tasks by means of suitable algorithms. Examples of suitable algorithms are the simplex method and genetic algorithms. Also suitable are algorithms according to embodiments of the present invention as described in detail below

In one embodiment of the invention, the controller is arranged to split digitized electrical power P (t) flowing through each of the terminals connected to a drain during operation of the power distribution unit digitized electrical power P (t). To do this, in one embodiment, the following steps are performed:

1 . Calculate for each time slice t and for all sources j the difference

2. A _{1 t} = A _lt - z _jt ,

2. if true even for a single source j A _LJ7 <0, the requirement can be met from a single source, end the calculation, and the residual value of the resource to the extracted value z j: = zj - _lt A and preferably more , from

Type of source dependent correction term time slices updated,

3. if a single source j does not satisfy the requirement, check if there are two sources j and i such that A _lj7 = A _lt - (z _{7 f} + z _it holds A _lj7 <0 and if this condition is met, update the residual values of both sources, so that for both sources z ₇ : = z ₇ - A _lt ,

4. if the request can not be fulfilled even with two sources, repeating step 2 with three and more sources,

where A _{kt is} the value of the k th request _profile and z _{7t is} the value of the maximum power provided by the y ^' th source during the t th time slice.

The paths through the distributor circuit from the sources to the sinks are also determined using known algorithms or read from corresponding linkage tables. In one embodiment of the invention, the control of the electric power distributor is set up and configured such that it controls the current flow under the assumption of elementary energy packets with an energy dp × dt during operation of the power distributor. In one embodiment of the electric power distributor, one of the terminals of the power distributor is electrically connected to an electrical energy store, wherein the energy store is set up so that it can receive, store and / or dispense electrical energy during operation, wherein the energy store is further configured such that on the basis of information about a current state of the energy store, the maximum electric power P _ma x (t) which can be supplied by the energy store at a future point in time t and a maximum power consumption P _cap (t) that is calculable at a point in time t is calculable, and wherein the control is such is set up and configured in the operation of the power distributor in calculating the current at each time t via each of the terminals flowing electrical power P (t) at this time t provided by the energy storage maximum electric power Pmax (t) or to this Z At the time t possible power consumption P _ca (t) of the energy storage considered. Such an embodiment has the advantage that "shortfalls" between the electrical power flowing into the power distributor at a time t and the electrical power flowing out of the power distributor at this time can be compensated for by such an energy store Charged with excess electrical energy or lacking electrical energy is provided by the energy store and thus the terminals connected to the sinks A prerequisite for the power distributor can work is that the energy store is configured such that from a state information about the state the energy storage at any time ti at any time t = ti + At in the future, both the maximum power consumption P _cap (t) of the energy storage and its at this time t maximum deployable electrical power P _ma x (t) is calculable. This is based on the assumption that all power flows are recorded from time ti and are available for calculating the powers P _ma x (t) and P _cap (t). In one embodiment of the invention, such an energy store is, for example, a super capacitor, a chemical energy store, a mechanical / kinematic energy store, a position energy store or a thermodynamic energy store.

While in one embodiment at least one such electrical energy store, which is electrically connected to one of the terminals of the power distributor, forms an integrated system with the electrical power distributor, the requirement for the electrical energy store is in relation to its maximum electrical power that can be supplied at any time t Power P _ma x (t) also to all other sources of electrical energy to be electrically connected to one of the terminals of the power distributor.

Therefore, at least one of the aforementioned objects is also achieved by an electric power network having an electric power distributor as described above in embodiments of the invention and having a data network connected to the communication device of the power distributor, at least one source of electrical energy wherein the source is electrically connected to one of the terminals of the power distributor, the source being configured such that based on information about a current state of the power source, the maximum power of the power source P _ma x (t ), and wherein the source has a communication device connected to the data network, which is set up in such a way that, during operation of the power network, it receives data with information about a current state of the energy source and / or with information about the e at a future time t transmits maximum electric power P _ma x (t) to the communication device of the power distributor, and at least one sink for electrical energy, wherein the sink is electrically connected to one of the terminals of the power distributor, and wherein the sink connected to the data network communication device which is adapted to transmit data in the operation of the power network with information about a required at a future time t from the sink electrical power Pdem (t) to the communication device of the power distributor.

An essential feature of the energy store is that the maximum power P _ma x (t) that can be provided by it at any future time t or the power P _ca (t) that can be received by the energy store at any future time is calculable, if there is information about the current state of the energy store at the time of the calculation. Such information includes in particular the state of charge of the energy store and its charging or discharging behavior with respect to time. Certain properties are already defined by the type of energy storage. For example, a supercapacitor has a different discharge curve than a conventional lithium ion secondary battery. Another information about the state is, for example, the temperature at which the energy store is located. Particularly in the case of accumulators, the discharge behavior depends, for example, on the temperature at which the accumulator is operated and on the number of charging cycles which the accumulator has already undergone.

An energy store, its maximum electrical power that can be provided at a future time t, and a maximum power consumption at a future time t can be used can be calculated on the basis of information about a current or current state of the energy store, denote as deterministic energy storage.

For the functionality of the power distributor, it does not matter whether the energy store transmits information about its current state via the data network and the communication device to the controller of the power distributor and this performs the calculation of Pmax (t) and Pcap (t) or whether P _ma x (t) and P _ca (t) are calculated on the side of the energy store itself, ie in a computing device provided there and then P _ma x (t) and P _cap (t) via the data network and the communication device to the control of Be transferred to the service distributor. Embodiments are also conceivable in which the energy store has only measuring devices which are read out by the control of the power distributor in order to determine P _ma x (t) and P _ca (t). In one embodiment of the invention P _ma x (t) and P _cap (t) based on information on the storage type and the type of energy storage based on measurements of current and voltage in a measuring device of a DFC in the terminal of the power distributor, with which the Energy storage is connected determined. The energy storage itself does not require any communication interface or measuring devices in such an embodiment.

Similarly, as previously described in detail in one embodiment of the electrical energy distributor with an electrical energy store, it is a prerequisite for the power grid according to the invention that the sources of electrical energy connected to the power distributor via both the power grid and a data network are deterministic sources which, based on information about a current state of the energy source, makes it possible to calculate the maximum electric power P _ma x (t) that can be provided by the energy source at a future point in time t. These sources of electrical energy can also be called deterministic sources. As before, it is unimportant whether the source transmits, as information over the data network, the maximum amount of electrical power P _ma x (t) that can be provided at any time t in the future to the control of the power distributor or state information which enables the control P _ma x (t) to calculate. Easy to calculate is the maximum deliverable power P _ma x (t), for example, for diesel generators with a constant power output over time.

In addition, in one embodiment, from the perspective of the power distributor, the sinks connected to the terminals of the power distributor must also be deterministic in the sense that the controller of the power distributor at each instant t has knowledge of what power Pdem (t) at that time t from the respective sink is needed. For this purpose, in one embodiment, the sink transmits over the data network and the communication device of the Distributor this information so that it is available to the control of the power distributor. Alternatively, in one embodiment, the required power of a sink may also be estimated or otherwise calculated by the controller of the power distributor. In a further embodiment of the invention, the source (s) and the sink (s) are connected to the power distributor via an overhead line or a ground line or underwater line. That is, in such a case, the electrical grid is a distribution network for connecting households, industrial companies or other consumers to energy producers, such as conventional power plants, renewable energy plants or energy storage facilities.

In an alternative embodiment of the invention, the electric power grid is an electrical system of a vehicle, an aircraft or a ship. In one embodiment of the invention, the energy store and the source of electrical energy, which are connected to terminals of the power distributor, have mutually different power profiles P (t). This makes it possible to comply with the combination of the energy storage and the source of electrical energy very different performance profiles on the side of the sink or sinks.

At least one of the aforementioned objects is also achieved by a method for distributing electric power in a power network comprising the steps of: connecting at least three sources and electrical energy sinks to one terminal of a distribution circuit, the terminals of the distribution circuit being electrically connected to one another in that an electric current can flow from each of the terminals to each of the other terminals, and receiving data from the sources or sinks, calculating the electric power P (t) flowing through each of the terminals as a function of time t and in response to the data received from the sources or sinks, and controlling the electrical power P (t) flowing through each of the terminals at a time t, by means of a respective power controller connected to the terminal.

As far as aspects of the invention with respect to the electric power distributor and the electric power grid have been described with this power distributor, they also apply to the corresponding method for distributing electrical power in a power grid. As far as the method is performed with the power distributor or the power grid according to this invention, the method has the corresponding steps for this. In particular, however, embodiments of the electric power distributor and of the electric power network are suitable for carrying out embodiments of the method. The power distributor according to the invention and the method according to the invention for distributing electrical power in a power network can be used advantageously in a number of applications. In electrical power grids, for example a supply network or a vehicle electrical system of a motor vehicle, the power distributor can serve in one embodiment for network stabilization, in particular for buffering short-term fluctuations in the range of services or in the demand for power. In another embodiment, the power distributor may serve as an interface between conventional networks and digitally controlled networks, so-called smart grids, or packet-based power transmission networks. In a further embodiment of the invention, the power distributor is used for the management of energy stores, preferably of accumulators. He controls in particular Umladeprozesse and a charge management of energy storage.

Further advantages, features and applications of the present invention will become apparent from the following description of embodiments thereof and the accompanying figures.

Figure 1 shows schematically an embodiment of a digitization of a Leistungspro- a capacitor during discharge.

FIG. 2 schematically shows a further embodiment of a digitization of a power profile of a capacitor during the discharge.

Figure 3 shows schematically the structure of an embodiment of an inventive

Power grid.

FIG. 4 schematically shows a detailed representation of the power network from FIG. 3.

FIG. 5 shows a block diagram of an embodiment of a power flow controller according to the present invention.

FIG. 6 shows a schematic circuit diagram of a bidirectional boost / buck converter according to an embodiment of the present invention.

Figure 7 shows a schematic diagram of a distribution circuit with a passive

Busbar for the power distributor according to the invention. FIG. 8 shows a schematic representation of a cascaded arrangement of passive ones

Busbars from FIG. 7. FIG. 9 shows a schematic representation of a completely reconfigurable switching matrix as a distributor circuit according to an embodiment of the present invention.

FIG. 10 shows a schematic representation of a variant of the switching matrix from FIG. 9.

Figure 1 1 a) shows an exemplary representation of a 4 x 1 switching matrix.

Figure 1 1 b) shows a schematic representation of the switching states of the Kopppelfeldes from Figure 1 1 a).

Figure 12 a) shows a schematic representation of a distribution circuit according to an embodiment of the present invention with a passive busbar and a so-called. Physical Abstraction Layer. FIG. 12 b) shows a schematic circuit diagram of a DFC from FIG. 12 a).

FIG. 13 shows a schematic circuit diagram of a distributor circuit according to another

Embodiment of the invention with a coupling field and a so-called. Physical Abstraction Layer.

Figure 14 shows an exemplary schematic representation of the distribution of elementary power units from three sources to one sink.

FIG. 15 shows a block diagram of an electrical system of a motor vehicle according to an embodiment of the invention.

FIG. 16 schematically shows the exemplary power flow in the vehicle electrical system from FIG. 15 in an excellent load situation. In the figures, identical elements are designated by identical reference numerals.

In the following description of concrete embodiments, first the power characteristic of energy storage and a concept for approximating the power profiles, ie the History of the output of the energy storage electrical power as a function of time, discussed. The following is a section on the structure and operation of an embodiment of the power control according to the invention and the connection of energy storage devices to this power control. The discussion ends with the description of several examples for electric power grids in which such a power distributor is used.

For the storage of electrical energy, there are a variety of methods, such as electrochemical storage, storage energy storage or capacitive storage. All of these memories have different performance characteristics, ie the maximum of the memory provided by the electric power P _ma x (t) and the time t absorbable electrical power P _cap (t) are different from each other. This not only between different storage principles, but also between the different concrete technologies within a storage principle. The different performance characteristics of the different types, technologies and shapes of electrical energy storage devices, when combined together, can be used to meet very different requirements within a power network that make up the power system sinks. In the field of electromobility, for example when considering an electrically driven motor vehicle, depending on the driving situation and environmental conditions, changing demands are placed on the power supply of the different consumers or sinks. Examples of such consumers are heating, headlamps, electroviscous shock absorbers and wheel hub motors. Each consumer has different characteristics, ie a different power consumption as a function of time. This results in a complex dynamics of the requirements for the electrical power to be provided.

Also, the provision of electrical energy within a power grid for the supply of households, industry and other consumers, in particular when the network has a large number of weather-dependent sources, e.g. Wind turbines or photovoltaic systems, has a high level of dynamics.

To ensure stable grid operation in an electrical grid, it is necessary that the sources produce as much power as the sinks consume.

In the language of the present application are referred to as sources all those elements in a power network that deliver electrical power. As sinks are considered all those elements that absorb electrical power. An energy store is in this sense both source and sink for electrical energy. A subnetwork of an electric power network having both sources and sinks may appear both as a source and as a sink from the point of view of a network node, ie a power distributor according to the invention, which connects this subnetwork to other subnetworks, depending on whether the power Distributor to consider, from the subnet receives electrical power or outputs electrical power to this.

In the event that there is no data for the sources and sinks, according to one embodiment of the invention, it is attempted to predict both the power consumption of the sinks and the power output of the sources for each time t by means of model-based methods, artificial intelligence-based methods or classical schedules ,

Short-term and small fluctuations in the power output of the wells and the power consumption of the sinks are buffered in conventional networks by the rotational energy of the generators. This bridging over the support of the rotational frequency of the generators takes place until the control mechanisms of the network have adapted the generation. In networks that do not have or only have a small number of conventional rotating electromechanical generators, such buffering must be managed by a power distributor according to the present invention. Typically, situations arise in which a difference in electrical power between sink demand and source supply must be temporarily trapped by a memory buffer until the power flow from the sources associated with the power distributor is so large again is how the power flow into the sinks connected to the power distributor.

Resulting under- or overlaps in the electrical power in the electric power distributor must therefore be compensated in an embodiment of the invention by appropriate energy storage or additionally switchable sources. In particular, to compensate for network fluctuations, the methods of controlling the power distributor described below are used to generate power profiles that provide the necessary power flows in the short term from energy storage devices connected to the power distributor to support the network. These power profiles are selected to support the network so that the predetermined power flow or the voltage level or the mains frequency are within the specified tolerances. In order to meet the power distribution requirements, in an electric power grid according to an embodiment of the present invention, the power profiles of both the sources and the sinks are approximated by elementary power profiles with constant power units dP over a period dt. The approximation of the actual Power profile P (t) by dP as digitizing the power profile P (t) are understood. This approximation simplifies the algorithms for distributing the electric power between the powers provided by the sources at a time t and the powers required by the sinks at that time.

Two different approximation methods are available for the approximation of the power profiles.

On the one hand, the power P in a given interval dt can be approximated as P #, where where P- P # <δ. That is, the power P in the interval dt is approximated as an integer multiple of the elementary power profile dP.

FIG. 1 shows an approximation of the power profile P (t) of a discharging capacitor carried out in this way. Such a capacitor could be used, for example, as an energy store, which is connected to a connection of the power distributor. For the power output of such a capacitor is general

P (t = P ₀ * exp (- ^), where R is the load resistance and C is the capacitance of the capacitor P ₀ is the power delivered in the fully charged state at time to.

Alternatively, the approximation can be performed using a Power of 2 education law, as known from digital technology. Such an approximated course of the power curve P (t) of the discharged capacitor is shown by way of example in FIG. The digitized approximation P # can then be described as

# P = dp £ £ _k = 2 ^k * dp, wherein DSSK = 2 ^k * dp.

It makes sense to realize capacitors whose performance profile can be approximated by a power of 2 education law by means of capacitor banks. In order to meet the power requirements Pdem (t) at each instant t of the individual sinks connected to the power distributor terminals, the power flows of the various sources (including energy storage such as the previously considered capacitor) must be combined, ie in the power divider be interconnected, that the requested power flows of each individual sink connected to the power distributor can be provided. For this purpose, in the illustrated embodiment it is necessary to provide at least one energy store which is able to supplement the power provided by the other sources connected to the power distributor either in the case of underfunding or in the case of overlapping which is not required at a time t To absorb power. In order to be able to solve the distribution task, the current state of all sources including a source known as energy storage and at least the power requirement Pdem (t) of the sinks must be known at each instant t. For each time slice dt belonging to a time t, via which the elementary power unit dP used to approximate the power flows P (t) is constant, it is determined in a state model how the maximum power P _ma x (t ) of the sources and the requested power Pdem (t) of the sinks and how the power of the sources can be distributed to the sinks. This is done as previously done by approximating the power curves of the sources and sinks. Alternatively to a consideration of the powers, the control of the power flow in one embodiment could also be based on the voltage U (t), since P (t) = where R is the resistance of the source or sink.

FIG. 3 schematically shows a power network consisting of a plurality of sinks or consumers 1000, a plurality of sources or generators 3000 and the power distributor 2000 according to the invention.

All of the terminals of the power distributor 2000 are bidirectionally configured, so that it makes no difference to the power distributor 2000 in the described embodiment whether a source or sink is connected to one of its terminals. Typically, individual elements connected to the power distributor 2000 may be both source and sink. An example of this is an energy store. This is true even if one of the with the Power Distributor 2000 connected elements, for example, one of the sinks 1000, a subnetwork of a power network, which is connected via the power distributor 2000 with other subnets. The consideration of whether an element 1000, 3000 connected to the power distributor 2000 is a source or sink depends only on whether this element provides electrical power to or receives electrical power from the power distributor 2000 at a given time t. For ease of consideration, it is always assumed in the following description that at a given time t, all elements labeled 1000 are sinks and all elements labeled 3000 are sources. Together, elements 1000, 3000 are also referred to as nodes.

FIG. 4 shows a detailed representation of the power network from FIG. 3, it being apparent in this representation that the power network according to the invention comprises two grids both logically and physically. On the one hand, this is a data network which enables an exchange of information between individual communication devices of the components of the power network. The data network connects each of the sink controllers 1200 of the wells 1000 to a communication device 2200 of the power distributor 2000, and the source controllers 3200 of the sources 3000 for electrical energy to the communication device 2200 of the power distributor 2000. In contrast, the electrical connections 1 are 100 and 3100 of the wells 1000, respectively and the sources 3000 are electrically connected to a connection system 2100, this connection system 2100 forming an electrical distribution circuit in the sense of the present application.

Via the distribution circuit 2100, the electrical connections 1 100, 3100 and thus the actual consumers 1300 and generators 3300 are interconnected. Components 1 100, 2100 and 3100 thus form the power section of the power grid shown. As used herein, the term generator 3300 includes any type of electrical energy source, such as a turbine, wind turbine, photovoltaic system or accumulator. A controller 2300 of the power distributor 2000 serves to control the electric power flows in the power unit 1100, 2100, 3100 required at a time t and the power flows required at this time from status information stored in the data network between the communication devices 1200 connected thereto , 2200, 3200 are to be exchanged.

The task of the controllers 1 100 of the sinks 1000 is to generate digitized requirement profiles for the required power Pdem (t) as a function of the time t on the basis of the current and expected power requirements of the actual consumers 1300 and these via the Data network to the communication device 2200 of the power distribution to communicate.

On the other hand, it is the task of the controllers 3200 of the sources 3000 to determine and update state information about the generators 3300 and to calculate from this state information the maximum achievable electrical power from the source 3000 at any future time t and this information to the communication device 2200 the power control 2000 to submit. Alternatively, in embodiments not described in detail herein, the controller 3200 could merely determine and update state information of the generator 3300 and communicate that state information over the data network to the communication device 2200 of the power distributor 2000, in which case the central controller 2300 will do the computing of the from the source 3000 at a time t maximum available electrical power P _ma x (t) takes over.

The structure of the distribution circuit 2100, which forms the central element of the power distributor 2000 controlled by the controller 2300, will now be explained in detail below. A necessary prerequisite for a power flow control in the electric power distributor is that each of the terminals has a power divider which makes it possible to set the electrical power P (t) flowing through the respective terminal as a function of the time t. An embodiment of such a power controller will be described below. The power divider is part of a power flow controller DFC, which has additional components in addition to the power controller. The power flow controller with the power divider may also be used as a variable switch in the electrical distribution circuit itself beyond its use in each of the terminals of the distribution circuit, as will be explained below. FIG. 5 shows a block diagram of the power flow controller DFC with its individual components. The power flow controller DFC has a computer unit 4 (as control of the DFC in the sense of the present application) for control, regulation, management and communication. The power flow controller also has a connection to a data network, in this case an IP network 8, a power divider 1 based on a bidirectional step-up / step-down divider, a DC / DC converter 2 and a measuring device 3 for detecting the actual electrical power a measurement of current and voltage. The computer unit 4 is connected via control lines 5, 6 to the boost / buck converter 1 and the DC / DC converter 2 and via a measuring line 7 to the measuring device 3. The DC / DC converter 2 serves to adjust the voltage level required by the network. The power flow controller DFC of FIG. 5 is a DC (DC) component. Even if the actual power flow controller DFC presupposes a DC operation, this power flow controller DFC can nevertheless also be used to implement a power distribution in an AC power grid. For this purpose, it is necessary to provide a bidirectional AC / DC converter between the power divider 1 of the DFC and the respective AC source or AC sink, which converts the coming from a source AC voltage into DC voltage for distribution in the power distributor or the Power Distributor converts incoming DC voltage into AC voltage to then provide the power to a sink.

Figure 6 shows a schematic circuit diagram of the bidirectional step-up / step down converter 2 of Figure 5. This is capable of producing a power flow in both directions, i. both into the power distributor and out of the power distributor out to sinks or consumers out.

The electrical distribution circuit 2100 of the electric power distributor 2000 may be implemented in a number of embodiments. FIG. 7 shows a first very simple embodiment of the distribution circuit 2100, wherein the respective power flow controllers DFC are depicted in the connections of the distribution circuit. The distributor circuit 2100 from FIG. 7 comprises a simple bus bar 9, to which all connections and thus the power controllers DFC are connected in parallel. It is also indicated in FIG. 7 that the individual power flow controllers DFC are connected to one another via a data network 8. The data network 8 in turn connects the power flow controllers DFC to the controller 2300 of the power distributor 2000.

This embodiment of the distributor circuit has the advantage that it is easy to implement and the power distributor as a whole manages with a number of power controllers DFC which is equal to the number of terminals of the distributor circuit. The disadvantage of this simple embodiment of the distribution circuit is that it has limitations in the configurability of power distribution from the sources to the sinks. When connecting the sources to the passive busbar 9 of Figure 7, all sources are connected in parallel to the rail. That is, all sources are operated on the rail with the same voltage, the currents add up. The power divider DFC will ensure that that the corresponding power is fed. So that all sources can be connected at the same voltage level, a DC / DC voltage converter is provided in each power controller. 2. A serial connection of the sources is not possible with the passive busbar 9. This requires a cascade of passive busbars that form a tree. Such an arrangement is shown schematically in FIG. In this circuit, individual connections of a busbar can be connected to one terminal of a busbar in a next higher busbar.

Then only one port of a power rail can be connected in series via a connection of the next higher power rail. If one wants to connect a connection Z1 on the passive busbar PPB1 1 in FIG. 8 to a connection Z2 on the busbar PPB12 serially, then all other connections except Z1 and Z2 must be switched off on PPB1 1.PPB12. Then Z1 and Z2 can be connected in series via PPB21. This means that all other connections on PPB1 1 must be switched off. This cascading allows a simple circuit design, but in addition to the reduced flexibility also leads to a shutdown of many connections. In contrast, Figure 9 shows schematically the structure of the power distributor 2100 in the manner of a fully reconfigurable crossbar, i. a six-by-four switching matrix. This makes it possible to connect each of the sources 3000 (shown in FIG. 9 only the power section 3100) to each of the loads 1000 (shown in FIG. 9 only the power section 1100). In addition, all sources 1000 can optionally be interconnected in parallel or in series with each other. For this purpose, the distribution circuit 2100 of FIG. 9 has a power flow controller DFC at each node of the switching network.

FIG. 10 shows a variant of the switching matrix from FIG. 9, the switches s being designed as simple on / off switches at the nodes of the switching matrix. In addition, however, 1000 DFCs are provided in the connections of the sources 3000 and the load to provide the required power flow control.

FIG. 1 a) shows, by way of example, a simplified crossbar with four sources 10 and only one single connection 1 1 for one sink, whereby it is taken into account that each of the DC sources 10 is connected via two lines to the connection 1 1 for the sink have to be. With 12 in Figure 1 1 a) in each case a part of a power flow controller, namely for a wire of a line, referred to. Figure 1 1 b) shows a schematic representation that all possible variants of series and parallel circuits can be realized with this arrangement. This form of switching matrix is scalable with respect to the number of connections for sources and sinks, with the associated complexity increasing substantially linearly as the number of sources and sinks increases.

Between the embodiment of the power distributor of FIG. 7 and the embodiment of FIG. 10, with regard to the complexity of the interconnection, there is an embodiment as described in addition in the following with reference to FIGS. 12 and 13. If, for example, several sources are to be connected in series on a power rail, then the power rail must be actively switched, i. a switch fabric as previously described. To reduce the complexity, however, the sources can be combined into a separate network, a so-called physical abstraction layer. As shown in FIG. 12 a), this physical abstraction layer 13 is connected to the bus bar 9 via many connections A1 to A8.

The physical abstraction layer 13 is used to connect the sources Z1 to Z4 in parallel and in series. Each individual source can be connected to the busbar 9. Sources Z1 to Z4 following each other can be connected in parallel and in series with the bus bar. Also, sources Z1 to Z4 can be interconnected by omitting other sources Z1 to Z4. However, this means that the skipped sources can no longer be used. By appropriate circuit of the DFCs any series and series circuits of the sources Z1 to Z4 can be realized. With reference to Figure 12 a) will now be described how sources Zi and Z2 and the sources Z3 and Z _{4 are} connected in series. The series-connected sources Z1 and Z2 are connected in parallel via the terminals A1 and A4 and via the terminals A5 and A8, where the serially connected sources z3 and z4 are connected, to the corresponding current conductors of a bus bar 9. The power flow controllers DFC1 to DFC15 each have three connections in the schematic view from FIG. 12 a). This is an equivalent circuit diagram, the exact configuration of which is shown in FIG. 12 b). On the left side of FIG. 12 b), the equivalent circuit diagram of the 2 × 1 DFC from FIG. 12 a) is shown once more, on the right side the actual switching circuit. On closer inspection, the 3-port DTC is implemented by a DTC as shown to the right in Figure 12 b, the output of which is connected to two switches S1, S2 connected in parallel, with both the DTC and the two switches S1, S2 connected to the controller 2300 of the power distributor 2000 are connected and controlled by this. FIG. 13 shows schematically the connection of a physical abstraction layer 13 to an active, that is to say switched, switching network. In this case, the switching network is a switching network with simple on / off switches S, as they have already been described for the switching network of Figure 10.

The wiring of the distribution circuit 2100 is performed by the controller 2300 of the power distributor. The controller 2300 and the controllers of the sources and sinks 1200, 3200 form a logical level, which is also referred to as a control plane. Tasks of this control plane are in particular:

Communication of the controllers with each other

Processing and storage of current power, voltage and current measurements

Storage of information about the current states of the individual sources Control and regulation of the DFCs so that the corresponding sources are interconnected with the corresponding sinks.

Administration and configuration of the switching networks

Determination of current load resistances of sources 1000

Determination of other parameters such as temperature, number of cycles, aging of sources or energy storage

system monitoring

Communication with higher-level systems

Receipt and processing of external timetables for plannable power flows

Requesting power to recharge energy stores connected to the power distributor

Handling spontaneous power delivery and recording

Determine appropriate sources to meet the requirement using the orchestration algorithm

Creation of the process control for the source and DFC interconnection as well as the control and regulation of the involved DFCs

The process control and source determination is iteratively optimized

Source Management: Requirement of external power for recharging as well as internal re-storage and selection of defective sources

Microaccounting and CDR creation for billing

Notification regarding the provision of positive and negative control energy

Bridging gaps in delivery in the event of short-term reordering of services. In other words, a sink during the consumption process determines that the ordered package too small, so she demands a reorder. Since this reordered power can usually be delivered with a time delay, but an interruption of the process is very damaging, this gap is bridged by the power distributor. For this, the sink receives the corresponding power requirement from the sink. · Communication with all elements of the electricity network

The distribution of the power flows P (t) of the sources 3000 to the power flows P (t) of the sinks 1000 through the power distributor 2000 will now be described with reference to FIG. There, the power profiles of three sources Z1 to Z3 and the power profile of a sink S1 for four time intervals dt are shown by way of example. Each of the power profiles P (t) is based on elementary power units dP, which have a constant power over a time interval dt dp, where the power at one time is equal to an integer multiple of the elementary power unit dP. In the illustration from FIG. 14, the time intervals dt are just selected such that they correspond to the time duration dt of an elementary power element dP.

The source controllers periodically or on request report the information about their current state of charge, but also the temperature, aging or the number of charge cycles already performed (in the case that the source is an accumulator) in the form of a state matrix M_i of its associated source Z_i the controller 2300 of the power distributor 2000. These status matrices are then stored in the controller 2300.

The state matrix is always a section through a high-dimensional map and is formed for fixed load resistance values, temperature values, etc.

The controllers 1200 of the sinks 1000 only transmit request profiles Pdem (t) to the controller 2300 with a time stamp. The load resistance of the sinks 1000 to be supplied is now estimated or determined by means of a measuring head in the power connections 1 100 of the sinks 1000. The controller knows the maps of the sources 3000 and thus the dependence of the states z. B. from the load resistance. Then the current state matrices are corrected according to the effect of the individual load resistances. In particular, for example, the discharge time of a storage capacitor used as a source changes as a function of the total load resistance associated therewith.

As a next step, the controller 2300 determines appropriate sources 3000 so that at any one time the sum of the power provided by the individual sources 3000 is equal to the sum of the power requested by the sinks. Figure shows this distribution task simplified insofar as only a single sink must be supplied whose requested power profile Pdem (t) is shown in Figure 14 below.

This requested power profile of the sink is now composed of the three power profiles shown in FIG. 14 above.

Two elementary power units dP are taken from the source Z1 in the first and second time intervals z_1 1 and z_12, and two elementary power units dP are also taken from the source Z2 for the second time interval, so that in the second time interval the control is effected by series connection of the two sources Z1 and Z2 can provide a total of four elementary power units dP for the duration dt. No power is needed for the third time interval. In the fourth time interval, source Z1 and source Z3 each provide two elementary power units dP. The algorithm as part of the orchestration algorithm for assembling the requested performance profile relies on a packaging problem. The requested performance profile forms the packaging space and the digitized performance profiles of the sources provide the packages. Added to this is the boundary condition that the packets have a given chronological sequence both on the source side and the lower side.

Such a packaging algorithm, as part of the orchestration algorithm, can look like this:

Let A _{kt be} the value of the k th requirement _profile and z _7t the value of the maximum power of the y ^' th source during the t th time slice.

Let's start with the case that only one simultaneous requirement k = 1 exists:

• During each time slice t compute the difference for all sources j

A _{1J t} = A _lt - z _Jt

• If true even of a source j A _LJ7 <0, the request could be met from a source that ends the calculation, and the residual value of the source is updated to the value corresponds taken: zj: = z j -A _lt plus any more, of the type

Source dependent correction terms for further time slices.

• If no single source j satisfies the requirement, it is checked whether there are two sources j and i such that A _lj7 = A _lt - (z _{7 f} + z _it ) holds A _lj7 <0. In this case, the

Residual values of both sources updated. • If the request can not be fulfilled even with two sources, it will be tried with three, with four, etc.

If two or more requested power profiles of the sinks are to be fulfilled with such an algorithm, two or more calculations are carried out in parallel. One invoice starts at the first source and the second from the last source. To save time, you can partition the sources and begin parallel calculations for the partitions. There must be a higher-level control and procurement mechanism for this. If the bills in their partitions are unsuccessful, they will perform this calculation iteratively in the next one. If an invoice in a partition has succeeded, the value for other invoices will be blocked. In the event that there is only one solution but two or more requested service packages, and there is no priority marking one of the requesting sinks, then the decision on which sink to serve may be decided randomly or the existing performance profiles of the sources will be uniform all sinks distributed.

It should be noted that the request for a service package must be answered by a sink in a defined time. In the example considered here, this means that if requests are to be processed in 10 milliseconds, ie at 100 Hz, and the CPU is clocked at 1 MHz, the above algorithm must convert the calculation to 10,000 clock cycles of the CPU. But this can not always be guaranteed. So that the response time is met, then the two best results are passed in response.

Possibilities to enable predictability in a given time frame include, for example, a Power of 2 approach to digitizing performance profiles, typing and classifying the requested performance profiles, parallelizing or transmitting typical features.

In particular, for the requirements on short time scales, e.g. Network stabilization, demand bridging and dynamic actuator requirements require requirement types and classes. That means for these scenarios there are only defined requirement profiles. Under certain circumstances these can be varied by parameters.

In a further embodiment, the sources are categorized into equivalence classes with regard to their performance properties, eg in capacitors, batteries, etc. With the aid of pattern recognition, the requested power profiles of the sinks are analyzed and then eliminated For the equivalence classes of the sources, those classes are selected whose performance profiles can be most suitably used to synthesize the respective requested performance profile. When selecting the sources, the algorithm must also address the question of whether the requested performance profile is realized by a series connection of sources (higher voltage and smaller currents) or by a parallel connection of sources (smaller voltage but higher current). When drawing up a power distribution roadmap for the sources' performance profiles at sinks or sink performance profiles, additional parameters may need to be taken into account, such as management requirements based on cyclization management or non-technical aspects such as maintenance contracts for batteries. The requirements are therefore to be provided with relevance. In the simplest case, the relevance is represented by a number. The higher the number, the higher the relevance. Each relevance is assigned a priority and the respective request is processed with the priority assigned to its relevance.

If the packaging task is not completely solvable, a solution with the smallest deviation or a predetermined deviation threshold is sought. This delta is then sent to the requesting sink. This can then change the request or trigger the immediate delivery.

Thus, e.g. Network stabilization requirements are more relevant compared to the relevance of hole bridging requirements between re-ordering and actual delivery. A request with higher relevance is processed with higher priority. In addition, in one embodiment, the sources, when it is energy storage, are assigned operational priorities for management purposes. For example, a certain type of high access frequency accumulator with only small power packets retrieved may be less suitable or one accumulator has already reached a higher number of charge cycles compared to another accumulator.

When prioritizing the spatial arrangement of the network should be considered. Thus, due to the line resistance, the relaxation time of a capacitor is changed, this affects the charging and discharging process. For Wegfind in the distribution circuit known algorithms are used. Deviating from the algorithms used in switching networks in communication networks, where the double use of paths is prohibited, the number of switching operations is for a switching network for the management of electrical power, which serve the supply of electrical consumers such as households or units of a motor vehicle to reduce. This means that in the context of the performance of the lines a multiple use may be advantageous.

A particularly instructive example of the use of a power distributor according to the invention in the electrical system of an electric car will be described below with reference to FIGS. 15 and 16.

Depending on the driving situation and environmental conditions, electric mobility places varying demands on the power and voltage supply of different consumers. Heater, headlamps, electroviscous shock absorbers, wheel scarred motors: each consumer has different characteristics, and depending on the combination and timing, a complex dynamic of the requirements of current and voltage results. Traction batteries are not suitable for on-board electronics, starter batteries not for heating. None of them are suitable for absorbing braking energy gained through recuperation, but storage capacitors would be better.

Figure 15 shows schematically the structure of a vehicle electrical system of an electric car. The motor M is an electric motor which is used both to drive the vehicle and to recuperate braking energy. The motor M is thus either source or sink for electrical energy depending on the viewing time t. The generator G is an electric generator, e.g. based on an internal combustion engine or a fuel cell whose primary energy sources are kept in a separate tank. Batteries B and capacitor C may also both accept or provide power, depending on the adjusted flow direction of the DFC.

The structure of the distribution circuit corresponds to the structure as shown in FIG. The essential feature of the DFCs is to generate a defined current flow and thus a defined power flow between the terminals in the vehicle electrical system by targeted generation of a potential difference between at least two terminals. By inserting these active elements, Kirchhoff's rules are only piecemeal and intermittent valid and lose their applicability for the global determination of the power flows in this network. The DTCs are linked through the power distributor control and are used to orchestrate the individual sources and sinks and the energy flows between them.

The control of the power flows between the DFCs involved and thus the distribution of the power profiles to the connections of the distribution circuit will now be illustrated by means of an acceleration process. As a starting point, a constant speed with constant power requirement is assumed. Additional power is needed for the acceleration process. This is to be regarded as additive to the existing constantly provided service. Additional power can only be provided at very short notice or, i. be provided in quasi-real-time, since the power requirement is not predictable by the nature of a car. FIG. 16 shows the composition of the power profile requested by the engine M and then provided to it with the onset of the sudden acceleration at the instant to. The service provided consists of three different shares, which are supplied by three different sources. The envelope of the power profile P (t) is equal to the power required for the abrupt increase in speed.

For t <t0, the generator G provides a power profile for the electric motor M. The power profile for the power requested by the electric motor as a sink connected to the power distributor is generated by the accelerator pedal of the vehicle and transmitted to the controller of the power distributor in the form of data. The electric motor works at one of its defined operating point which is characterized by a high efficiency. For t> t0 the acceleration process is initiated The motor requests the maximum total power At this stage, an additional power profile is provided by the capacitor bank C as this can provide the highest power gradient.

However, the energy content of the capacitor bank C is not sufficient for the entire acceleration process, therefore, a supplementary power flow is requested by the traction battery B. After reaching the target speed, a higher power from the generator G is requested, since the required base load has increased and about the maximum of the traction battery Provable power is too low. For this purpose, the generator changes to another operating point, for which it requires a certain amount of time, since it has a shallower gradient than, for example, the capacitor bank C.

LIST OF REFERENCE NUMBERS

DFC power flow controller

1 bidirectional boost / buck converter

2 DC / DC converters

3 measuring device

4 computer unit

5 control line

6 control line

7 measuring line

8 IP network

9 passive busbar

10 sources

1 1 connection

12 switches

13 Physical Abstraction Layer

14 distribution circuit

15 board network

1000 sink

1 100 electrical connection of the sink

1200 controllers

1300 consumers

2000 electrical power distributor

2100 distribution circuit

2200 communication device

2300 control

3000 source

3100 electrical connection of the source

3200 controller

3300 generator

10000 power grid

Claims

P a t e n t a n s r ü c h e

Electrical power distributor (2000) for a power grid

an electrical distribution circuit (2100) which has at least three connections, sources (3000) and sinks (1000) for electrical energy being connectable to the connections,

wherein the three terminals are electrically connected to one another such that an electrical current can flow from each of the terminals to each of the other terminals, and

wherein each of the connections has a power divider (1), which is set up in such a way that, during operation of the power distributor (2000), the electrical power P(t) flowing through the respective connection can be adjusted as a function of the time t,

a communication device (2200) which can be connected to a data network and which is set up so that it receives data from the sources (3000) or sinks (1000) during operation of the power distributor, and

a controller (2300) for controlling a distribution of a flow of electrical power P(t) as a function of time t at the connections,

wherein the controller (2300) is connected to the communication device (2200) in such a way that the data received from the communication device (2200) can be processed by the controller (2300),

wherein the controller (2300) is connected to each of the power dividers (1), the controller (2300) being set up in such a way that, during operation of the power distributor (2000), it controls the electrical power P(t) flowing through each of the connections as a function of the data received from the sources (3000) or sinks (1000), and

wherein the controller (2300) is set up in such a way that, during operation of the power distributor, it controls the electrical power P(t) flowing via the respective connection as a function of the time t.

Power distributor (2000) according to the preceding claim, characterized in that the controller (2300) is set up so that it calculates the electrical power P(t) flowing through each of the connections for each time t during operation of the power distributor

the maximum electrical power P _ma x(t) that can be provided by each source (3000) at time t and

the electrical power P _de m(t) required by each sink (1000) at time t, and that it controls the power divider (1) in such a way that the calculated electrical power P(t) is set at the respective connection at time t.

Power distributor (2000) according to one of the preceding claims, characterized in that the distribution circuit (2100) comprises a switching network, with nodes of the switching network preferably being formed by controllable power controllers (1).

Power distributor (2000) according to one of the preceding claims, characterized in that the distribution circuit (2100) has a first and a second section, the first section comprising a switching network which is designed such that all connections of the first switching network which are connected to a Source (3000) or a sink (1000) can be connected, connected in parallel or in series with the second section of the distribution circuit (2100).

Power distributor (2000) according to one of the preceding claims, characterized in that each of the connections has a voltage converter

(2) and/or each of the connections has a measuring device (3) for detecting an actual electrical power flowing through the connection and/or a controller (4), the controller (4) being designed in such a way that it measures the actual The power via the connection is controlled so that it is equal to the calculated power P(t).

Power distributor (2000) according to one of the preceding claims, characterized in that the controller (2300) is set up and designed in such a way that, during operation of the power distributor (2000), it controls the power P(t) as a function of the time t at each of the connections approximated as an integer multiple of an elementary power dP, where dP is constant over a time interval dt.

Power distributor (2000) according to one of claims 1 to 5, characterized in that the controller (2300) is set up and designed in such a way that, during operation of the power distributor (2000), it controls the power P (t) as a function of the time t at each of the connections is approximated as P (t) = Y _=Q 2 ^k dP.

Power distributor (2000) according to one of the preceding claims, characterized in that the controller (2300) is set up in such a way that, during operation of the power distributor (2000), it digitizes electrical power P(t) flowing via each of the connections connected to a source (3000). ) on those with a sink (1000) electrical power P(t) flowing through the connected connections, the following steps preferably being carried out:

3. Calculate the difference for each time slice t and for all sources j

4. A _{1 t} = A _lt - z _jt ,

5. if A _lj7 < 0 already applies to a single source j, then the requirement can be fulfilled from a single source, the calculation ends, and the residual value of the source is increased by the extracted value zj : = zj — A _lt and preferably others , from the

correction terms dependent on the source type, time slices updated,

6. if not a single source j meets the requirement, check whether there are two sources j and i such that if A _lj7 = A _lt - (z _{7 f} + z _it applies A _lj7 < 0 and if this condition is met, update the residual values of both sources, so that z ₇ : = z ₇ - A _lt applies to both sources,

7. if the requirement cannot be met with two sources, repeat step 2 with three or more sources,

where A _kt is the value of the k-th requirement profile and z _7t is the value of the maximum power provided by the y ^' th source during the t-th time slice.

Power distributor (2000) according to one of the preceding claims, characterized in that the controller (2300) is set up and designed in such a way that it controls the current flow during operation of the power distributor (2000) assuming elementary energy packets with an energy dP x dt.

0. Power distributor (2000) according to one of the preceding claims, characterized in that the controller (2300) is set up and designed in such a way that it controls the power divider (1) during operation of the power distributor (2000) in such a way that at every time t Electrical power P(t) provided at a connection that is connected to a sink (1000) is equal to the power Pdem(t) required by the sink (1000) at this point in time.

1 . Electrical power distributor (2000) according to one of the preceding claims with an electrical energy storage device which is set up so that it can absorb, store and/or release electrical energy during operation,

wherein the energy storage is electrically connected to one of the connections of the power distributor (2000), wherein the energy storage is designed such that, based on information about a current state of the energy storage, the maximum electrical power P _ma x (t) that can be provided by the energy storage at a future time t and a maximum power consumption P _cap (t) at a time t can be calculated, and wherein the controller (2300) is set up and designed in such a way that, during operation of the power distributor (2000), it calculates the electrical power P(t) flowing through each of the connections at a time t at this time t of The maximum electrical power P _ma x (t) that can be provided to the energy storage or the possible power consumption P _cap (t) of the energy storage at this time t is taken into account.

Electrical power network (10000) with an electrical power distributor according to claim 1 1,

a data network that is connected to the communication device of the power distributor,

at least one source (3000) for electrical energy,

wherein the source (3000) is electrically connected to one of the connections of the power distributor,

wherein the source (3000) is designed such that, based on information about a current state of the source (3000), the maximum electrical power P _ma x(t) that can be provided by the source (3000) at a future time t can be calculated, and

wherein the source (3000) has a communication device (3200) connected to the data network, which is set up so that, during operation of the power network, it transmits data with information about a current state of the source (3000) and / or with information about the transmits the maximum electrical power P _ma x(t) that can be provided at a future time t to the communication device (2200) of the power distributor (2000), and

with at least one sink (1000) for electrical energy,

wherein the sink (1000) is electrically connected to one of the connections of the power distributor (2000), and

wherein the sink (1000) has a communication device (1200) connected to the data network, which is set up so that during operation of the power network it receives data with information about an electrical power Pdem() required by the sink (1000) at a future time t t) transmits to the communication device (2200) of the power distributor (2000). Electrical power network (10000) according to the preceding claims, characterized in that the source (3000) and the sink (1000) are connected to the power distributor (2000) via an overhead line, an underground line or an underwater line or that the electrical power network is an on-board electrical system ( 15) a vehicle, an aircraft or a ship.

Electrical power network (10000) according to one of the preceding claims, characterized in that the energy storage and the source (3000) have a different power profile P(t).

Method for distributing electrical power in a power grid with the steps:

Connecting at least three sources and sinks for electrical energy, each with one connection of a distribution circuit,

wherein the terminals of the distribution circuit are electrically connected to one another such that an electrical current can flow from each of the terminals to each of the other terminals, and

Receiving data from the sources or sinks,

Calculating the electrical power P(t) flowing through each of the ports as a function of time t and as a function of the data received from the sources or sinks and

Controlling the electrical power P(t) flowing through each of the connections at a time t using a power controller connected to the connection.