# US 2022/0286162 A1 · Method for directional transmission of energy in the form of at least one energy packet

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0286162 A1
- Document ID: US20220286162A1
- Status: Published application
- Kind: A1
- Application number: 17/539,961
- Filed: 2021-12-01
- Published: 2022-09-08
- Earliest claimed: 2020-12-08
- Assignees: Gip AG
- Applicants: Gip AG
- Inventors: Bernd REIFENHÄUSER
- Web page: https://patentgrep.com/patent/US20220286162A1

## CPC

- H02J 3/004: {Generation forecast, e.g. methods or systems for forecasting future energy generation}
- H02J 13/1311: {using the power network as support for the transmission}
- H02J 3/003: {Load forecast, e.g. methods or systems for forecasting future load demand}
- H02J 3/0075: {for providing alternative feeding paths between load and source according to economic or energy efficiency considerations, e.g. economic dispatch}
- H02J 3/02: using a single network for simultaneous distribution of AC power at different frequencies
- H02J 3/381: {Dispersed generators}
- G05B 2219/2639: Energy management, use maximum of cheap power, keep peak load low
- H02J 2103/30: Simulating, planning, modelling, reliability check or computer assisted design [CAD] of electric power networks
- H02J 2103/35: Grid-level management of power transmission or distribution systems, e.g. load flow analysis or active network management
- Y02D 30/70: in wireless communication networks
- Y02E 40/70: Smart grids as climate change mitigation technology in the energy generation sector
- Y02E 60/00: Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y04S 10/12: Monitoring or controlling equipment for energy generation units, e.g. distributed energy generation [DER] or load-side generation
- Y04S 40/121: using the power network as support for the transmission
- H02J: ELECTRIC POWER NETWORKS; CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02E: REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y04S: SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS

## Abstract

The present invention relates to a method for directional transmission of power in the form of energy packets via a transmission network. The power to be transmitted is adjustable at at least one location x of each edge by a data and computer network. A data packet is biuniquely assigned to each energy packet, which data packet is formed in a control instance of the data and computer network by means of predictions. This data packet describes an optimized transport path via which optimized transport path a source transmits the physical power in a fixed transmission period T to a load for partial demand coverage. Furthermore, the data packet describes the power class of the energy packet, wherein this power class is defined by the temporal course of a nominal power Pnom(t) determined by predictions and by a remainder R(t) in the transmission period determined by a function of an uncertainty of the predictions. For the transmission of the energy packet, the transmitted power equal to Pnom(t) plus a fraction of R(t) is set for each point in time t within T at at least one location x of each edge of the transport path.

## Claims

### Claim 1

A method for directional transmission of energy in the form of at least one energy packet from a plurality of sources Q via at least two nodes, one of which is a supply node KQ connected via a supply edge to one of the plurality of sources Q and one of which is a demand node KS connected via a demand edge to a load S, and via a plurality of edges to a plurality of loads S in a transmission network, the transmission network being controllable by means of a data and computer network in such a way that for at least one location x on each of the plurality of edges an actually flowing physical power pφ(x,t) is controllable by a control instance of the data and computer network, which method comprises the steps of:
A) forming a data packet for each energy packet, wherein the data packet is biuniquely assigned to exactly one energy packet, wherein the data packet defines the respective energy packet, and wherein the data packet describes
i. a transmission period T of the energy packet with a duration DT and an execution time t0, which execution time t0 identifies the start of the transmission period T,
ii. a predetermined transport path of a transport graph of the transmission network for the directed transmission of the energy packet, which transport path connects at least one of the plurality of sources Q with exactly one of the plurality of loads, wherein the exactly one of the plurality of sources Q is connected to a supply node KQ of the predetermined transport path via exactly one supply edge and the exactly one load S is connected to the demand node KS via a demand edge of the predetermined transport path, and
iii. an equivalence class {pnom(t), R(t),T} of the energy packet,
wherein the equivalence class is given by
a nominal power pnom(t) of the energy packet as a function of time, wherein the nominal power pnom(t) is determined beforehand by at least one prediction, and a remainder R(t) as a function of time and as a function of a prediction uncertainty of the at least one prediction,
wherein for each point in time t within the transmission period T of the energy packet there is a μ(t), with −1≤μ(t)<1 or with −1<μ(t)≤1 for with −1≤μ(t)≤1, such that the physical power pφ(x,t) during the transmission of the energy packet at each point in time t and at at least one location x on each edge of the transport path is fixed as the sum of the nominal power pnom(t) and the product of μ(t) and the remainder R(t),
wherein the equivalence class is defined by an equivalence relation, according to which equivalence relation, for any point in time t of the transmission period T, a first physical power t) at any location {circumflex over (x)} of an edge of the transport path and a second physical power pφ(x,t) at any location x of an edge of the transport path are equivalent only if there is a predetermined remainder R(t) greater than or equal to zero and less than or equal to a limit value Rmax and a μ(t), with −1≤μ(t)<1 or with −1<μ(t)≤1 or with −1≤μ(t)≤1, such that the first physical power t) is equal to the sum of the second physical power pφ(x,t) and the product of and the remainder R(t),
B) prior to the execution time t, transmitting the data packet to all control instances of at least all nodes or all edges on the transport path,
C) beginning with the execution time t0 transmitting the energy packet biuniquely assigned to the data packet,
wherein for all points in time t within the transmission period T of the energy packet, the physical power pφ(x,t) flowing on the transport path between the supply node KQ and the demand node KS set at at least one location x of each edge of the transport path in such a way that the physical power fed into the demand nodes KS is an element of the equivalence class {pnom(t),R(t),T} described by the data packet.

### Claim 2 (depends on 1)

The method according to claim 1, wherein
the data packet furthermore describes
iv) a predetermined compensation path of the transport graph of the electrical transmission network connecting a predetermined compensation node KSAS node to the demand node via a plurality of edges, wherein the compensation node KSAS at any point in time t of the transmission period T of the energy packet either can receive a physical power up to a predetermined maximum compensation power C(t), wherein the predetermined maximum compensation power C(t) as a function of the remainder R(t), or can deliver a physical power up to the maximum compensation power C(t), optionally increased by a loss drop determined by a prediction, via the compensation path, and
wherein, in step C), the physical power flowing between the compensation node KSAS and at least the demand node K, and is controlled for at least one location x on the plurality of edges of the compensation path such that the physical power at the at least one location is at any point in time t within the transmission period T an element of an equivalence class {pnom(t)=0,R(t)=C(t),T} with a nominal power pnom(t) of zero and a remainder having the value of the previously determined maximum compensation power C(t) and such that at at least one location x of the demand edge the physical power pφ(x,t) flowing through the demand edge between the demand node KS and the load S connected to this demand node is equal to the nominal power pnom(t) of the energy packet plus a compensation power δ′(t) as a function of the prediction,
and/or the data packet describes
v) a previously determined balancing path of the transport graph of the transmission network, which via a plurality of edges connects a previously determined balancing node KAus and the supply node KQ, wherein the balancing node KAus at any point in time t of the transmission period T of the energy packet can receive a physical power up to a predetermined maximum balancing power, wherein the predetermined maximum balancing power D(t) is determined as a function of the remainder R(t), or provide a physical power up to the maximum balancing power D(t), optionally increased by the loss drop determined by a prediction, via the balancing path, and
wherein, in step C), the physical power flowing between the balancing node KAus and the supply node KQ for at least one location x on the plurality of edges of the balancing path is controlled such that at this at least one location x the physical power at any point in time t within the transmission period T is an element of an equivalence class {pnom(t)=0, R(t)=D(t),T} with a nominal power pnom(t) of zero and a remainder with the value of the predetermined maximum balancing power D(t) and such that the physical power pφ(x,t) flowing between the supply node KQ and the demand node KS for at least one location x on the plurality of edges of the transport path is an element of the equivalence class {pnom(t),R(t),T} of the energy packet.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the method for an arbitrary prediction period Tprog with a prediction start t0prog and a prediction duration DTprog further comprising the steps of, at a prediction point in time t0, which is prior to the prediction start t0prog,
D) generating of a supply prediction for a supply power pprogEin(t) to be supplied at any point in time t in the prediction period Tprog from the respective source, wherein the supply prediction has a supply prediction uncertainty, and
E) generating a demand prediction for a power demand (t) predicted to be required by the respective load at any point in time t in the prediction period Tprog, wherein the demand prediction has a demand prediction uncertainty,
wherein the temporal course of the nominal power pnom(t) of the energy packet is determined at least from the supply prediction and the demand prediction, and wherein the temporal course of the remainder R(t) is a function of at least the supply prediction uncertainty and the demand prediction uncertainty.

### Claim 4 (depends on 3)

The method according to claim 3, the method comprising the steps of
F) forming a demand data packet {pprogBed(t),R(t),Tprog} of the load, the demand data packet describing the predicted temporal course of the nominal power required by the load pnomBed(t) at any point in time t within the prediction period Tprog and the temporal course of a remainder R(t) as a function of the demand prediction uncertainty of the demand prediction at any point in time t within the prediction period Tprog,
G) forming a supply data packet of the source, the supply data packet describing the predicted temporal course of the nominal power pnomEin(t) to be supplied by the source at any point in time t within the prediction period Tprog and the temporal course of a remainder R(t) as a function of the supply prediction uncertainty of the supply prediction at any point in time t within the prediction period Tprog and
wherein forming of the equivalence class of an energy packet for transmission from a source to a load comprises at least one allocation of a sub-packet of the supply data packet to a sub-packet of the demand data packet at least to meet the demand of the load.

### Claim 5 (depends on 1)

The method according to claim 1, wherein, when a plurality of energy packets is transmitted simultaneously via exactly one edge of the transmission network, the following steps are carried out
H) superpositioning the simultaneously transmitted energy packets to exactly one simultaneous energy packet with a superpositioned simultaneous data packet,
I) adjusting the physical power flowing via the exactly one edge, so that the physical power at at least one location of the exactly one edge is an element of the equivalence class {pnom(t),R(t),T} of the superpositioned simultaneous data packet of the exactly one simultaneous power packet formed by the superpositioning in step H), wherein the nominal power pnom(t) of this equivalence class {pnom(t),R(t),T} is the sum of the nominal powers of the simultaneously transmitted energy packets, wherein in the summation for the nominal power pnom(t) the flow directions of the power is taken into account, and wherein the remainder R of the equivalence class {pnom(t),R(t),T} is the sum of the remainders of the simultaneously transmitted energy packets and wherein for the summation of the remainder the flow direction is disregarded.

### Claim 6 (depends on 5)

The method according to claim 5, the method further comprising the steps of
J) at each node on the transmission network determining the fractions of energy packets pending at each edge of the respective node in an arbitrary test period for simultaneous transmission to the next node based on the data packets,
K) for the duration of the test period, adjusting the power supply to each edge of the respective node so that the physical power t at at least one location x of the respective edge is an element of an equivalence class {pnom(t),R (t),T} whose nominal power pnom(t) is the sum of the nominal powers of the simultaneously transmitted energy packets, the flow directions of the electrical power being taken into account in the summation for the nominal power, and the remainder of which is the sum of the remainders of the simultaneously transmitted energy packets, the flow direction being disregarded for the summation of the remainder, wherein the flow direction for an electrical power fed out via an edge of the node being counted with the opposite sign compared to the flow direction of an electrical power supplied via an edge.

### Claim 7 (depends on 3)

The method according to claim 3, wherein the method at a reservation point in time t00 and for a reservation period Tres, wherein the time t00 is prior to the start of the reservation period Tres, which reservation period has a start time t0res and a reservation duration DTres, and wherein the reservation period Tres is within the prediction period Tprog, further comprises the steps of,
L) for each load, determining the amount of all transport graphs, each of which transport graphs is connected to the load, such that for each of these transport graphs, each source connected to the transport graph is connected to the load via a transport path of the transport graph and, for the reservation period Tres via a sub-packet E′Ein of a supply packet defined by the supply data packet for the reservation period, wherein this available sub-packet E′Ein is formed by a coverage packet defined by a coverage data packet ED and a loss packet Eν, the coverage packets of the sources Q of the transport graph together cover a maximized portion of a demand packet of the load S defined by the demand data packet and wherein the respective loss data packet describes the loss determined by a prediction during the transmission of the respective coverage packet defined by the coverage data packet via the corresponding transport path,
M) associating a loss data packet Eν to each of the coverage data packets, wherein each of the coverage packets defined by the coverage data packets is transmitted via a transport path assigned thereto as a subgraph of a transport graph out of the amount of transport graphs formed according to the preceding step L), wherein a nominal power of the equivalence class of a loss packet defined by the loss data packet is determined by a prediction of the loss and the remainder of the equivalence class of the loss packet defined by the loss data packet is determined as a function of a prediction uncertainty of a loss prediction, wherein the nominal power of the loss is the power predicted to be lost in the transmission of the cover packet via the transport path from the source Q to the load S by the loss prediction, and wherein for the loss prediction all energy packets to be transmitted via the considered transport path during the reservation period are taken into account, and the transport path has an available transmission capacity to transmit the sub-packet E′Ein which is composed of the coverage packet defined by the coverage data packet ED and the loss packet defined by the loss data packet Eν, wherein the coverage packet ED defined by the coverage data packet and the loss packet Eν defined by the loss data packet have the same transmission direction,
N) wherein the equivalence class of the sub-packet E′Ein is formed by adding the nominal powers of the coverage packet and the loss packet and the remainders, and wherein the loss packet has the same flow direction as the coverage packet and the transmission period T is set equal to the reservation period Tres,
O) for each load, determining the optimized transport paths as subgraphs of the transport graph optimizedly determined with respect to a selected metric, wherein each source of each transport path so optimizedly determined, connecting that source Q with the load S of the transport graph for the transmission of a sub-packet of the supply packet, at any point in time t within the reservation period Tres provides an available power which is an element of the equivalence class of the sub-packet determined in accordance with step N) above,
P) forming in each case a reservation data packet for defining a reservation packet of the transport path from step O), wherein the reservation data packet of the reservation packet describes the equivalence class of the respective sub-packet of the supply packet as an equivalence class of the reservation packet, the supply edge of the respective sub-packet of the supply packet as the supply edge of the reservation packet, the transport path and the reservation period Tres,
Q) for each source of the transport graph determining a sub-packet of the supply packet of each source that is still available, so that the supply packet is formed by superposition of all reservation packets and this available sub-packet for the reservation period,
R) for each load for the reservation period Tres determining the not yet covered sub-packet of the demand packet, so that the demand packet is formed from the coverage packets according to step N), the reservation packets being assigned to the load and the not covered sub-packet,
S) for each time t in the reservation period Tres determining the remaining available transmission capacity of the transmission network, such that the given transmission capacity of the transmission network is the sum of the remaining available transmission capacity and the sum of the nominal powers plus the remainders of the reservation packets,
T) for each supply prediction, demand prediction and loss prediction by which at least one of the values determined by them is changed, repeating the preceding steps,
U) for each reservation packet formed in steps L) to O) at an arbitrary booking time tBuch, which is prior to the start time t0res of the reservation packet, bindingly booking the reservation data packet of the reservation packet as the data packet of the energy packet, wherein the transmission period of the energy packet is the reservation period, wherein the equivalence class {pnom(t),R(t),T=Tres} is the equivalence class of the reservation packet, wherein the supply edge of the reservation packet is the supply edge of the energy packet, and wherein the transport path of the reservation packet is the transport path of the energy packet, and
V) at the execution time t0 transmitting of the respective energy packet according to the data packet of the energy packet.

### Claim 8 (depends on 2)

The method according to claim 2, the method, for a reservation period Tres, further comprising the steps of, at the reservation point in time t00,
W) for each load, determining the set of transport graphs each connected to the load such that, for each one of this set of transport graphs, each source connected to the transport graph is connected to the load via a transport path of the transport graph and has an available sub-packet of the remaining available supply packet of the source, the available sub-packet being formed from a coverage packet defined by a coverage data packet and a loss packet defined by a loss data packet, wherein the coverage packets of the sources Qi of the transport graph together cover a maximized portion of the demand data packet of the load S and wherein the loss packet describes the loss in the transmission of the respective coverage packet over the corresponding path, which loss is determined by a prediction,
X) for each load and each transport graph determined according to step W), extending the respective transport graph to an extended transport graph by determining the compensation nodes of the respective extended transport graph such that for each source of the extended transport graph the compensation nodes are determined, wherein each of these compensation nodes is connected via at least one compensation path, as a subgraph of the extended transport graph, to the load S and the compensation nodes together provide a maximum total compensation power C(t) determined beforehand in order to receive a physical power up to the maximum total compensation power C(t) at each point in time t within the reservation period Tres, wherein the maximum total compensation power C(t) is a function of the remainders of the coverage packets of the respective source Q and in order to supply via the compensation paths in sum a physical power up to this maximum total compensation power C(t), which is increased by the power losses determined by prediction, which power losses drop during the transmission of the sub-packets from the compensation nodes via the compensation paths into the demand node KS,
Y) for each load determining exactly one extended transport graph optimized according to a given metric from the set of all extended transport graphs determined according to steps W) and X), each of the extended transport graphs being connected to the load via a demand edge,
Z) for the optimized extended transport graph of step Y) forming the reservation data packets for defining the reservation packets of the energy packets to be transmitted over the optimized extended transport graph, wherein the reservation data packet of each reservation packet includes the equivalence class of the respective available supply packet as the equivalence class of the reservation packet, the supply edge of the respective supply packet as the supply edge of the corresponding source node KQ of the reservation packet, the corresponding transport path connecting the source to the load, the reservation period Tres, as well as the compensation nodes with the corresponding compensation potentials and the compensation paths connecting the compensation nodes with the load node KS to which the load is connected via the demand edge,
AA) determining the still available sub-packets of the supply packets and the not yet covered portions of the demand packet of the still available transmission capacities according to step O), reducing the still available capacity of the reserved compensation nodes by the reserved maximum powers for receiving and for delivering the corresponding powers, reducing the still available transmission capacities of the compensation paths by the reserved potentials for delivering power of the reserved compensation nodes increased by the power loss,
BB) for each supply prediction, demand prediction and loss prediction by which at least one of the values determined by the respective prediction is changed, repeat the previous steps,
CC) for each reservation packet of an optimized extended transport graph formed in the preceding steps, at an arbitrary booking point in time tBuch, which booking point in time tBuch is prior to the start time t0res of the reservation packet, bindingly booking the reservation data packets of the reservation packets as the data packets of the energy packets, wherein for each of these data packets the transmission period of the energy packet is the reservation period, wherein the equivalence class is {pnom(t),R(t),T} is the equivalence class of the reservation packet and the supply edge of the reservation packet is the supply edge of the energy packet, and bindingly booking the reserved transmission capacities on the optimal extended transport graph, and
at the execution time t0 transmission of the respective energy packet according to the data packet of the energy packet via the booked capacities of the booked extended transport graph, and/or
DD) for each load, extending the possible transport graphs and/or possible extended transport graphs previously determined in steps W) and X) by the possible balancing nodes and the possible balancing paths, such that each of these balancing nodes is connected via a balancing path to one of the sources of the transport graph and/or extended transport graph, for which the associated supply data packet requires a balancing, and these balancing nodes can individually or collectively provide a balancing for the corresponding sources and can balance the losses determined by at least one loss prediction occurring during the transmission of the balancing packet from the balancing node to the source,
EE) according to a predetermined metric, determining an optimized transport graph with balancing or an extended transport graph with compensation with a balancing out of the set of possible transport graphs with a balancing determined according to step DD) or a transport graph with compensation extended by the compensation, and
FF) for the optimized transport graph with balancing determined in this way or a transport graph with balancing extended by the compensation carrying out steps Z) to CC).

### Claim 9 (depends on 3)

The method according to claim 3, wherein the method for a real-time prediction period TRT with a real-time prediction start t0RT and a real-time prediction duration DTRT, wherein the realtime prediction period TRT is a sub-period of the reservation period Tres of the reservation packet, further comprises the steps of, at a real-time prediction point in time in time t000 after the reservation point in time in time t00 of the reservation period Tres and prior to the real-time prediction start t0RT,
GG) measuring at least one parameter,
HH) creating a real-time prediction pRT(x,t) of the physical power pφ(x,t) at any point in time t within the real-time prediction period TRT at a location x of the transmission network, the real-time prediction pRT(x,t) having negligible real-time prediction uncertainty and being created with consideration of the at least one parameter, wherein the physical power pφ(x,t) is a function of the at least one parameter, and
II) forming a real time energy packet for transmission over a subgraph of the transmission network by using the real time data packet associated with the real time energy packet, which real time data packet defines the real time energy packet, which real time data packet describes the time history of the nominal power pnomRT(x,t)=pRT(x,t) at at least one location x of each edge of the subgraph as a function of the predicted time history of the physical power supplied into that subgraphs pφ(x,t) at the location x, wherein the remainder R(t) is set equal to constant zero and the transmission period T is set equal to the real-time prediction period T.

### Claim 10 (depends on 5, 9)

The method according to claim 5, wherein transmitting the energy packet associated with the data packet over a directed transport path of the booked transport graph connecting the source for the energy packet to the load, for each real-time prediction time period TRT, which is a part of an overlap of the transmission period by real-time prediction periods and for each point in time t within the respective real-time prediction period TRT comprises the following steps
JJ) creating a real-time prediction for the temporal course of the physical power pφ(x,t) for at least one location x on at least each edge of the transport path according to step ii. of claim 9, including at least one measurement of the physical power fed into the respective transport graph at the real-time prediction point in time t000, supplied into the transport graph,
KK) forming a real-time energy packet for the supply via the supply edge into this transport path according to step II), in that in the associated real-time data packet the temporal course of the nominal power over the real-time prediction period TRT at at least one location x on each edge of the transport paths, over which transport paths the transfer of the power packet from the source supply edge to the corresponding load demand edge occurs is set equal to the corresponding predictions from steps HH) and II), and the remainder for all points in time t within the real-time prediction period Tres and all subgraphs is constantly set equal to zero,
LL) transmitting the real-time data packet to the control instances of the transport graph; and
MM) setting the physical power pφ(x,t) for at least one location x on each edge of the transport path, such that the physical power pφ(x,t) is a function of the nominal power pnormRT(x,t) described in the real-time data packet for that location x, and the physical power pφ(x,t) at any point in time t within the transmission period T satisfies the equivalence relation described in the data packet of the energy packet, and
NN) for simultaneous transmission of energy packets via a single edge of a node for each transmission period T setting the physical power pφ(x,t) for at least one location A on this edge, so that the physical power pφ(x,t) is a function of the sum of the nominal powers of the real-time packets associated with the simultaneously transmitted energy packets at that location x taking into account the direction of flow, so that the physical power pφ(x,t) at any point in time t within the transmission period T is an element of the equivalence class formed for the simultaneous energy packet according to claim 5.

### Claim 11 (depends on 8)

The method according to claim 8, wherein the method for the real-time prediction period TRT further comprising the steps of, for each real-time prediction period TRT, which is a part of an overlap of the transmission period by real-time prediction periods, and for each point in time t within the respective real-time prediction period TRT
OO) for each supply node KQi to which a source is connected via a booked transport path of the booked transport graph extended by compensation, whereby the source Qi is booked to cover the demand of the load S, determining a power supplied by the source via the edge of the transport path, via which edged the load node KS of the load S is connected, at at least one location on x of that edge by a real-time supply prediction,
PP) determining the difference δ(t) between the sum of the physical power supplied into all booked transport paths which connect all sources Q with the load S from the real-time supply predictions, and the physical power taken by the load S, which load S is connected to the demand node K, via the demand edge, at a location x of the demand edge from the real-time prediction,
QQ) for each point in time t within the real-time prediction period TRT forming a maximum total compensation power C(t) as the sum of all maximum compensation powers Ci(t) of all booked sources Qi and the load S and for each booked compensation node KSASi determining of a quotient γi(t) of the maximum compensation power Ci(t) of the respective compensation node KSASi and the maximum total compensation power C(t),
RR) determining real-time compensation packets for transmission on the compensation graph as a subgraph of the booked extended transport graph, which includes the compensation nodes KSASi and the load S, by adding, for each compensation path of the booked compensation graph connecting a compensation node KSASi with the load, in the associated real-time compensation data packets, the nominal power pnomKomRT(x,t) for at least one location x of each edge of the respective compensation path a function of the product of the difference δ(t) determined in step PP) with the quotient γi(t) but only up to a maximum of the value of the maximum compensation power Ci(t) and the remainder R(t) is constantly set equal to zero, wherein if the product γi(t)·δ(t) is negative, the product γi(t)·δ(t) is increased by an amount determined by a real-time prediction for a location x of the edge of the compensation path via which edge the demand node KS is connected to the compensation node to compensate for the losses incurred in the transmission from the compensation node to the corresponding load 5,
SS) transmitting the real-time compensation data packets to the control instances, and
TT) transmitting the real-time compensation packets by, at any point in time t within the transmission period, at at least one location x on the edges of the compensation paths, setting the physical power pφ(x,t) such that the physical power pφ(x,t) is equal to the nominal power pnomKomRT(x,t) of the real-time compensation packets for this location x, and by simultaneously at any point in time within the transmission period T setting the physical powers pφ(x,t) to be equal to the nominal powers of the real-time energy packets pnomRT(x,t) at the locations x on the transport graph previously selected by the real-time predictions, so that the physical power pφ(x,t) at a location x selected by the real-time prediction for the demand on the demand edge is equal to the sum of the nominal powers of the energy packets sent from the supply nodes KQi to the demand node KS plus the compensation power δ′(t), wherein the compensation power δ′(t) is equal to the difference between the physical power pφ(x,t) of the power taken by the load according to the real-time demand prediction and the sum of the nominal powers of the energy packets transmitted to the load S.

### Claim 12 (depends on 8)

The method according to claim 8, the method comprising, for each real-time prediction period TRT, which is part of the overlap of the transmission period by real-time prediction periods with the real-time prediction point in time t000 further comprising the steps of, for each point in time t within the respective real-time prediction period TRT
UU) for each supply node KQi to which a source Qi is connected via a supply edge wherein the source Qi is to cover the demand of the load S for each point in time t within the real-time prediction period, determining a fraction ρ(t) of the supplied physical power pφ(x,t) from the real-time supply predictions at at least one location x of the respective supply edges, which fraction ρ(t) exceeds or falls below the limits defined by the equivalence relation of the power to be transmitted from the source node KQi to the load S,
VV) for each point in time r within the real-time prediction period TRT forming a maximum total balancing power D(t) as the sum of all maximum balancing powers Di(t) of the balancing nodes booked for the sources Qi and for each balancing node KAusi determining a quotient gi(t) from the maximum balancing power Di(t) of the respective balancing node KAusi and the maximum total balancing power D(t),
WW) determining real-time balancing packets for transmission on all booked balancing paths that include the balancing nodes KAusi and the source Q1, wherein in the associated real-time balancing data packets the nominal power pnomAusRT(x,t) for at least one location x of each edge of the respective balancing path is a function of the product of the deviation μ(t) determined in step UU) with the quotient gi(t) but only up to a maximum of the value of the maximum compensation power Di(t), and the remainder Ri(t) is constantly set equal to zero, wherein if the product gi(t)·ρ(t) is negative, this product gi(t)·ρ(t) is increased by the amount determined by a real time prediction for a location x of the edge of the balancing path via which edge the supply node KQi is connected to the balancing node KAusi to compensate for losses incurred during transmission from the balancing node to the corresponding source,
XX) transmitting the real-time balancing data packets to the control instances associated with the nodes on the balancing graphs,
YY) transmitting the real-time balancing packets by at any point time t within the transmission period T for at least one location x on each edge of the balancing graph setting the physical power pφ(x,t) to be equal to the nominal power pnomAusRT(x,t) of the real-time compensation packets for the location x, and
ZZ) at any point in time t within the transmission period T setting the physical power pφ(x,t) for at least one location x of each edge of the transport path, so that the physical power pφ(x,t) is equal to the nominal power pnomRT(x,t) of the real-time energy data packets for the location x and the physical power pφ(x,t) is an element of the power class of the energy packet to be transmitted.

### Claim 13

A system for directional transmission of energy in the form of at least one energy packet with a transmission network and a data and computer network,
wherein the transmission network comprises
at least two nodes, one of which is a supply node KQ and one of which is a demand node KS,
at least one edge, wherein each edge connects exactly two nodes,
at least one power controller, wherein the at least one power controller is arranged and located such that, by means of the at least one power controller for at least one location x on each edge, a physical power pφ(x,t) actually flowing via the respective edge is settable,
a plurality of sources Q, wherein each of the sources Q is connected to a supply node via a supply edge KQ, and
a plurality of loads S, wherein each of the loads S is connected to a demand node KS via a demand edge,
wherein the data and computer network comprises
at least one computer node and
at least one control instance executed on the at least one computer node, wherein the data and computer network is configured and operatively connected to the at least one power controller such that, during operation of the system, the physical power pφ(x,t) is controllable by the at least one power controller for at least one location on each edge by the control instance,
wherein the at least one control instance of the at least one computer node is configured such that, during operation of the system, the control instance executes a method comprising the steps of:
A) forming a data packet, wherein the data packet is biuniquely assigned to exactly one energy packet, wherein the data packet defines the respective energy packet and wherein the data packet describes
i) a transmission period T of the energy packet with a duration DT and an execution time t0, which marks the beginning of the transmission period T,
ii) a predetermined transport path of a transport graph of the transmission network for the directed transmission of the energy packet, which transport path connects at least one of the plurality of sources Q with exactly one of the plurality of loads S, wherein the exactly one of the plurality of sources Q is connected via exactly one supply edge to a supply node KQ of the previously determined transport path and the exactly one load S is connected via a demand edge of the predetermined transport path to the demand node KS, and
iii) an equivalence class {pnom(t),R(t),T} of the energy packet,
wherein the equivalence class {pnom(t),R,T} is given by
a nominal power pnom(t) of the energy packet as a function of time, wherein the nominal power pnom(t) is determined beforehand by at least one prediction, and a remainder R(t) as a function of time and as a function of a prediction uncertainty of the at least one prediction,
wherein for each point in time t within the transmission period T of the energy packet there is a μ(t), with −1≤μ(t)<1 or with −1<μ(t)≤1 or with −1≤μ(t)≤1 so that the physical power pφ(x,t) during the transmission of the energy packet at any point in time t and at at least one location on each edge of the transport path is fixed as the sum of the nominal power pnom(t) and the product of μ(t) and the remainder R(t),
wherein the equivalence class is defined by an equivalence relation, according to which equivalence relation, for each point in time t of the transmission period T, a first physical power p′φ({circumflex over (x)},t) at any location {circumflex over (x)} of an edge of the transport path and a second physical power pφ(x,t) at any location x of an edge of the transport path are equivalent only if there is a predetermined remainder R(t) greater than or equal to zero and less than or equal to a limit value Rmax and a μ(t), with −1≤μ(t)<1 or with −1<μ(t)≤1 or with −1≤μ(t)≤1 so that the first physical power p′φ(x,t) is equal to the sum of the second physical power pφ(x,t) and the product of μ(t) and the remainder R(t),
B) prior to the execution time t0 transmitting the data packet to all control instances of the data and computer network to which the power controllers are connected for controlling the physical power, so that all data necessary for transmitting an energy packet are coherent on the control instances, and
C) beginning with the execution time t0 transmitting the energy packet biuniquely assigned to the data packet,
wherein for all points in time t within the transmission period T of the energy packet, the physical power pφ(x,t) flowing on the transport path between the supply node KQ and the demand node KS is set at at least one location x of each edge of the transport path in such a way that the physical power fed into the demand nodes KS is an element of the equivalence class {pnom(t),R(t),T} described by the data packet.

### Claim 14

A method for directional transmission of energy in the form of at least one energy packet from a plurality of sources Q via at least two nodes, one of which is a supply node KQ connected via a supply edge to one of the plurality of sources Q and one of which is a demand node KS connected via a demand edge to a load S, and via a plurality of edges to a plurality of loads S in a transmission network, the transmission network being controllable by means of a data and computer network in such a way that for at least one location x on each of the plurality of edges an actually flowing physical power pφ(x,t) is controllable by a control instance of the data and computer network, which method comprises the steps of:
A) forming a data packet for each energy packet, wherein the data packet is biuniquely assigned to exactly one energy packet, wherein the data packet defines the respective energy packet, and wherein the data packet describes
i. a transmission period T of the energy packet with a duration DT and an execution time t0, which execution time t0 identifies the start of the transmission period T,
ii. a predetermined transport path of a transport graph of the transmission network for the directed transmission of the energy packet, which transport path connects at least one of the plurality of sources Q with exactly one of the plurality of loads, wherein the exactly one of the plurality of sources Q is connected to a supply node KQ of the predetermined transport path via exactly one supply edge and the exactly one load S is connected to the demand node KS via a demand edge of the predetermined transport path,
iii. an equivalence class {pnom(t),R(t),T} of the energy packet,
wherein the equivalence class is given by
a nominal power pnom(t) of the energy packet as a function of time, wherein the nominal power pnom(t) is determined beforehand by at least one prediction, and a remainder R(t) as a function of time and as a function of a prediction uncertainty of the at least one prediction, and
iv. a predetermined compensation path of the transport graph of the electrical transmission network connecting a predetermined compensation node KSAS node to the demand node via a plurality of edges, wherein the compensation node KSAS at any point in time t of the transmission period T of the energy packet either can receive a physical power up to a predetermined maximum compensation power C(t), wherein the predetermined maximum compensation power C(t) as a function of the remainder R(t), or can deliver a physical power up to the maximum compensation power C(t), optionally increased by a loss drop determined by a prediction, via the compensation path,
wherein for each point in time t within the transmission period T of the energy packet there is a μ(t), with −1≤μ(t)<1 or with −1<μ(t)≤1 or with −1≤μ(t)≤1, such that the physical power p t) during the transmission of the energy packet at each point in time t and at at least one location x on each edge of the transport path is fixed as the sum of the nominal power pnom(t) and the product of μ(t) and the remainder R(t),
wherein the equivalence class is defined by an equivalence relation,
according to which equivalence relation, for any point in time t of the transmission period T, a first physical power p′φ({circumflex over (x)},t) at any location of an edge of the transport path and a second physical power pφ(x,t) at any location x of an edge of the transport path are equivalent only if there is a predetermined remainder R(t) greater than or equal to zero and less than or equal to a limit value Rmax and a μ(t), with −1≤μ(t)<1 or with −1<μ(t)≤1 or with −1≤μ(t)≤1, such that the first physical power p′φ({circumflex over (x)},t) is equal to the sum of the second physical power pφ(x,t) and the product of μ(t) and the remainder R(t),
B) prior to the execution time t0 transmitting the data packet to all control instances of at least all nodes or all edges on the transport path,
C) beginning with the execution time t0 transmitting the energy packet biuniquely assigned to the data packet,
wherein for all points in time t within the transmission period T of the energy packet, the physical power pφ(x,t) flowing on the transport path between the supply node KQ and the demand node KS is set at at least one location x of each edge of the transport path in such a way that the physical power fed into the demand nodes K, is an element of the equivalence class {pnom(t),R(t),T} described by the data packet,
wherein, in step C), the physical power flowing between the compensation node KSAS and at least the demand node KS and is controlled for at least one location on the plurality of edges of the compensation path such that the physical power at the at least one location x is at any point in time t within the transmission period T an element of an equivalence class {pnom(t)=0,R(t)=C(t),T} with a nominal power pnom(t) of zero and a remainder having the value of the previously determined maximum compensation power C(t) and such that at at least one location x of the demand edge the physical power pnom(x,t) flowing through the demand edge between the demand node KS and the load S connected to this demand node is equal to the nominal power pnom(t) of the energy packet plus a compensation power δ′(t) as a function of the prediction,
the method for directional transmission of energy for an arbitrary prediction period Tprog with a prediction start t0prog and a prediction duration DTprog further comprising the steps of, at a prediction point in time which is prior to the prediction start t0prog,
D) generating of a supply prediction for a supply power pprogEin(t) to be supplied at any point in time t in the prediction period Tprog from the respective source, wherein the supply prediction has a supply prediction uncertainty, and
E) generating a demand prediction for a power demand pprogBed(t) predicted to be required by the respective load at any point in time t in the prediction period Tprog, wherein the demand prediction has a demand prediction uncertainty,
wherein the temporal course of the nominal power pnom(t) of the energy packet is determined at least from the supply prediction and the demand prediction, and wherein the temporal course of the remainder R(t) is a function of at least the supply prediction uncertainty and the demand prediction uncertainty.