DK179419B9

DK179419B9 - Self-regulated distribution of power from energy sources

Info

Publication number: DK179419B9
Application number: DKPA201670958A
Authority: DK
Inventors: Anker Skaarup Lübker Poul
Original assignee: Udviklingsselskabet GH A/S
Priority date: 2016-12-05
Filing date: 2016-12-05
Publication date: 2018-08-15
Also published as: DK201670958A1; DK179419B1

Abstract

The present invention relates to a method of allocating electrical power from one or more energy sources to one or more energy sinks comprising acts of receiving electrical power from one or more energy sources to a distribution unit, distributing electrical power to one or more energy sinks from a distribution unit, and balancing the distributed electrical power to the received electrical power.

Description

DANMARK (10)

(12)

PATENTSKRIFT

Omtryk publiceret 2018-08-15

Patent- og Varemærkestyrelsen

Int.CI.: H02J 3/00 (2006.01)

Ansøgningsnummer: PA 2016 70958

Indleveringsdato: 2016-12-05

Løbedag: 2016-12-05

Aim. tilgængelig: 2018-06-07

Patentets meddelelse bkg. den: 2018-06-18

Korrektionen bkg. den: 2018-08-15

Patenthaver:

Udviklingsselskabet GH A/S, Ndr. Kajgade 7, 8500 Grenaa, Danmark

Opfinder:

Poul Anker Skaarup Liibker, HammerGut 111 6330 Cham, Schweiz

Fuldmægtig:

Patrade A/S, Ceresbyen 75, 8000 Århus C, Danmark

Titel: Self-regulated distribution of power from energy sources

Fremdragne publikationer:

US 2013/0043725 A1

EP 3002848 A1

US 2005/0165511 A1

WO 2012/078433 A2

US 2011/0082598 A1

Sammendrag:

Fortsættes...

100 k

Fig. 1 i

[Self-regulated distribution of power from energy sources]

Field of the Invention

Background of the Invention

In countries like Denmark and Germany renewable energy sources now provide more than 30% of the power on an average basis. In some cases renewable energy sources based on solar and wind power supplied more than 100% of peak power demand at specific times of the day. So far these countries have managed to integrate and balance these high shares of renewable energy with very modest changes to its public electricity network system. The major reasons are: The existing strength of its public electricity network system with a high security of power supply, and the flexible operation of coal and nuclear plants to dynamically increase power production to meet peak power demand or reduce power production when the power consumption is low. The modest changes to its public electricity network system include implementations of better system control software and day-ahead weather forecasting.

In the future, a variety of additional measures will be required on public electricity network systems to improve handle energy storage, power demand response, smart inverters, two-way energy flows, virtual power plants combining generation of power with flexible loads, integration of heat supply and heat storage, and other developments yet to be encountered on public electricity network systems.

US 2013/0043725 Al discloses a method and system for distributing energy from one or more energy producing devices to one or more energy consuming devices. The energy is transformed into electricity for distribution. The energy is supplied from at least one high voltage producing unit and possibly low voltage producing devices, which are connected through a supply network to energy consuming devices. The supply network comprises at least one transformer station that converts from high voltage to low voltage. The system monitors and controls energy production and consumption by establishing multiple grids with different sizes for measuring the energy supply and energy demand, by performing more frequent measurements or by performing frequent measurements on multiple parameters. These measurements are performed for achieving optimized prediction of the power distribution and thus the prediction is based solely on measured energy supply and energy consumption.

Energy storage has played almost no role in integrating and balancing renewable energy production so far. Innovations and changes specifically in this field will be necessary in the future to improve the ability to integrate and balance renewable energy production.

Object of the Invention

It is an objective of this invention to overcome one or more of the aforementioned shortcomings of the prior art.

Description of the Invention

An object of the invention may be achieved by a method of allocating electrical power from one or more energy sources to one or more energy sinks comprising acts of receiving electrical power from one or more energy sources to a distribution unit, distributing electrical power to one or more energy sinks from a distribution unit, and balancing the distributed electrical power to the received electrical power. A method where the distribution unit may be connected to the one or more energy sources and the one or more energy sinks, where the distribution unit may be configured with controlling means to control the acts of receiving and distributing electrical power, and where the distribution unit may be configured to execute the act of balancing according to one or more algorithms (Al_Bal) executable in the distribution unit.

The one or more algorithms for balancing electrical power, Al_Bal may each include several algorithms working dependent on each other or comprise a cluster of algorithms.

The controlling means to control the acts of receiving and distributing electrical power may operate as a switch, a regulator or to divert the electrical power. The controller means may also operate by adjusting the individual energy sinks to regulate the power output.

The energy sources may include renewable energy sources such as wind turbines, solar panel, systems for thermal energy and systems for wave energy amongst others.

An effect of this method is that fluctuations in produced and demanded energy are levelled out or balanced thereby achieving reduced energy loss due to surplus production of energy.

An effect of this method is that a local network electrical power centre may be established where the local energy production may be adjusted to the local electrical power demand thus avoiding surplus energy production. The energy production may be regulated by adjusting the output from local conventional energy sources with the advantage of minimizing the use of fossil fuels.

The energy production may further be regulated by receiving electrical power from renewable energy sources located at a distance which necessitates using the public electricity power system to transport this electricity power with the advantage of minimizing the use of fossil fuels.

The energy production may further be regulated by adjusting the output from local renewable energy sources with the advantage of minimizing the tear and wear of these. This could be the case for wind turbines, where these may be adjusted to a lower output than maximum to balance the received power of the distributing centre to the demanded electrical power and thereby reduce the load on the wind turbines. For solar panels a similar situation may be the case, where the solar irradiation input is reduced by regulating the inflow angle of solar radiation for the benefit of reducing the heat induced in the panels. Thus, environmental benefits due to reduced use of fossil fuels may be achieved and increased life span of the equipment for renewable energy may be achieved.

An object of the invention may be achieved by a method of allocating electrical power comprising the further acts of collecting and storing one or more data sets comprising actual electrical power production of each energy source, and providing executing means configured for executing one or more algorithms, which algorithm(s) (Al_Rec) regulates the level of received electricity power from each energy source to the distribution unit.

In general, when referring to one or more algorithms in the described method the algorithm may each include several algorithms working dependent on each other or the multiple algorithms may comprise a cluster of algorithms which may be referred to by a single reference, for example: Al-Bal, Al_Rec, algorithm for electrical power output to mention a few.

An object of the invention may be achieved by a method of allocating electrical power comprising the further acts of collecting and storing one or more data sets comprising actual electrical power demand for each energy sink, and providing one or more controllers configured for executing one or more algorithms, which algorithm(s) (Al_Dist) regulates the level of distributed electricity power to each energy sink from the distribution unit.

One effect of the above mentioned methods are in line with the previous mentioned effect of regulating the output from local conventional energy sources and local renewable energy sources with the advantage of minimizing the use of fossil fuels and minimizing the tear and wear of these. One effect may also be to regulate the demand for electrical power of one or more individual energy sinks to a later timer, where the general demand are lower or where the energy output from renewable energy sources are high without exploiting the energy sources to a maximum output. And thus as previously mentioned achieving environmental benefits due to reduced use of fossil fuels may be achieved and increased life span of the equipment for renewable energy.

An object of the invention may be achieved by a method of allocating electrical power comprising the further act of storing electrical power for balancing the distributed electrical power to the received electrical power from the one or more energy sources wherein the distributed electrical power comprises an actual electrical power demand.

One effect of this embodiment is that the fluctuations in received energy and distributed energy may be more easily accommodated without causing rapid changes in the energy production or energy demand. An advantage of this may be that a sudden increase in energy demand may be met without using fossil fuel. A sudden increase in energy demand in energy supply systems are often met by increasing the conventional energy production as this is often the fastest way of regulating energy production. In this method, due to energy storage facilities, the regulation may be achieved by a slower increase and thus over a longer time period and thus, the increase may be achieved by regulating the power output from the renewable energy sources.

A further object of the invention may be achieved by a method of allocating electrical power wherein the act of receiving electrical power from one or more energy sources to the distribution unit comprise the further acts of collecting and storing data sets comprising time stamped data sets of received electrical power from each energy source, and synchronizing and processing the said time stamped data sets of received electrical power with time stamped data sets of predicted electrical power output from each energy source. A method wherein the act of receiving electrical power from the one or more energy sources to the distribution unit may be controlled by an algorithm (Al_Rec).

Yet a further object of the invention may be achieved by a method of allocating electrical power wherein the act of distributing electrical power from the distribution unit to one or more energy sinks comprise the further acts of collecting and storing data sets comprising time stamped data sets of distributed electrical power to each energy sink, and synchronizing and processing the said time stamped data sets of predicted electrical power demand from each power sink. A method wherein the act of distributing electrical power from the distribution centre to the one or more energy sinks may be controlled by an algorithm (Al_Dist).

For this invention by synchronizing is meant that the data is time synchronized based on a time stamp on the individual data. This means that the data may be extrapolated in time to achieve synchronization.

Thus, one effect of time stamping the data is that synchronization of the data may be performed. Synchronisation ensures synchronized data prior to processing and thus advantageously allow for implementation of calibration of parameters.

An object of the invention may be achieved by a method of allocating electrical power wherein the act of collecting data synchronized is performed using at least one synchronization system, and wherein the synchronization system transmits a trigger signal activating synchronized measurement of data from one or more sensors.

The effect of this embodiment is that the data is synchronization due to that data are measured in parallel on a trigger signal and thus is synchronized in real time. In this case the data may still be time stamped for further use in case of synchronization of other measurements not performed in parallel.

An object of the invention may be achieved by a method of allocating electrical power comprising the further act of calibrating at least one algorithm from the one or more time stamped and synchronized data sets.

The calibrating process may involve an automatic self-calibrating process. The calibrating process may have the effect of improving the accuracy of the algorithm based on measured data. Calibration of one or more algorithms based on local measurements may be advantageous in regard to achieving an adjusted and optimised algorithm for the individual allocating system with all the energy sinks, sources and storage facilities which may be different for each individual system. Furthermore, the individual energy sources and sinks may perform differently under the same meteorology conditions if placed in different terrains.

An object of the invention may be achieved by a method of allocating electrical power comprising the further acts of collecting and storing actual meteorology data, collecting and storing weather forecast data, and collecting and storing predicted meteorolo gy data, wherein the actual meteorology data and the predicted meteorology data are time stamped and synchronized.

One effect of this embodiment is that that meteorology data may be used to predict electrical power output from the renewable energy sources and to predict electrical power demand, as the demand may be dependent on environmental conditions such as temperature, wind speed or others. For example in the case of power demand for heating is dependent on the ambient temperature. Thus this is advantageous in regard to predicting the amount of received electrical power and the demand for electrical energy. Furthermore, it may be advantageous in regard to manage the energy storage facilities.

An object of the invention may be achieved by a method of allocating electrical power wherein the electrical power measurements on one, more or all energy sources and/or energy sinks are performed by frequent power sampling in the time scale of milliseconds to seconds.

The frequent power sampling may also be performed as power sampling using a fast sampling rate.

A fast sampling rate and real-time data recording of current, voltage, frequency, power, power factor, phase angle, grid harmonics and/or the like enables use for operational measurement as well as for condition monitoring of the overall electrical power applications.

The measuring intervals on several devices have to be time synchronized in order to evaluate the collected energy data consistently. This also simplifies and speeds up the engineering, enabling comprehensive diagnostics and fault analysis.

An object of the invention may be achieved by a method of allocating electrical power wherein the one or more algorithms perform an act of calibrating by comparing actual meteorology data to predicted meteorology data, wherein the predicted meteorology data is extrapolated and estimated from short-term and/or long-term local weather forecast, by comparing actual electrical power production to predicted electrical pow er production, by comparing actual electrical power demand to predicted electrical power demand, and/or by comparing actual electrical power storage capacity to predicted electrical power storage capacity. A method wherein the predicted data may be extrapolated and estimated from predicted meteorology data, and databases of collected and stored data.

An object of the invention may be achieved by a method of allocating electrical power wherein the one or more algorithms (Al_Bal) for balancing the distributed electrical power to the received electrical power evaluates the state of operation of the individual energy sources, energy sinks and/or meteorology sensors.

An object of the invention may be achieved by a method of allocating electrical power wherein machine learning is used for calibrating the one or more algorithms.

Machine learning centres on the development and use of algorithms that can learn to make predictions based on past data. The recent advances in machine learning lets people take very large data sets and analyse them using large-scale computing systems or neural networks.

Thus, using machine learning may govern for a continuous and rapid calibration of the algorithms wherein the amount of historical data to be handle and used for calibrating the individual algorithms may be enormous. Furthermore, using machine learning may enable calibration of inter-dependent algorithms in a manageable way, which is basically impossible in practice without employing machine learning.

The calibration of the algorithm(s) may be based on historical data of weather, electrical power production from individual energy sources, electrical power demand of the individual energy sink(s) and capacity of any storage facilities and may take into account a weather forecast for prediction of basically the same parameters but present and future values of the electrical power production, electrical power demand and capacity of any storage facilities

A person skilled in the field of artificial intelligence and machine learning will know how to apply and implement machine learning in a beneficial way. A skilled person in the art may be inspired from the approach and implementation presented by Microsoft using machine learning for predicting weather patterns in the coming 24 hours.

Unlike more typical weather forecasting approaches, which have traditionally relied on physical simulations, Microsoft’s research took a data-centric approach: They just looked at the data and didn’t try to make restrictive assumptions about how nature tends to act. The researchers used historical data for several weather variables — atmospheric pressure, temperature, dew point and winds — to train their systems to make predictions about future weather patterns based on past data.

This invention may be using smart-grid technologies, new monitoring and data acquisition systems and better systems for day-ahead and/or week-ahead weather forecasting to forecast output from local renewables and local demand to self-regulate the distribution of electrical power from different energy sources on a local electricity network system. This local electricity network system will automatically perform balancing and peak-shaving by storing excess power and distributing it as needed by managing the electrical power supply to short term energy storage technology systems like for example local combined-heat-and-power plants and their heat storage facilities and to long term energy storage technology systems like for example production of hydrogen facilities.

Example of a method (which including renewables) for predicting and balancing incoming and outgoing electricity, and a method (which includes renewables) using energy storage for balancing predicted supply and predicted demand for short- and long term fluctuations. The method described herein may encompass a system which includes renewable energy sources.

The input to the method may encompass manual data from operators and measured data. The data may be time stamped and synchronize. The data may be stored in a number of databases.

The method of allocating electrical power may provide a regulated electrical power output which is based on predicting and balancing electrical power output from energy ίο sources and storage facilities, wherein the electrical power output from renewables may be predicted in consideration of short- and long term fluctuations. The regulated electrical power output may further be regulated in consideration of short- and long term fluctuations in predicted electrical power demand of the consumers.

The allocating of electrical power may require multiple algorithms for performing the relevant analysis on actual and predicted data. The algorithms may be multidimensional self-learning and self-calibrating algorithms.

The example is illustrated schematically, wherein the main algorithm is the algorithm for balancing electrical power in the system, Al_Bal. Al_Bal communicates with the algorithms Al_Rec, the algorithm for controlling the receiving of electricity power and Al_Dist, the algorithm for controlling the actual distribution of all available electrical power. Al_Bal further receives and transmit data regarding meteorology data and data from all clients collected in a database: Database for all clients. Here clients are energy sources, energy sinks and energy storage facilities.

Al_Bal	Database for all clients
Al_Rec
Meteorology data
Al_Dist (Actual)

Database for all clients'. Database for all clients delivering and receiving electricity directly to/from the distribution centre and for all clients connected to a heating pump. The database may comprise information on priority code of the client, agreed prices, minimum and maximum sale/purchase, special expectations to variations, and other relevant information.

Al_Rec: This algorithm communicates yet with another algorithm Algorithm for electrical power output and receives data regarding predicted electrical power demand from internal conventional energy source and predicted electrical power demand from public electricity network.

π

The predicted electrical power demand from internal conventional energy source and from public electricity network may be long-term and/or short-term forecasts.

Al_rec	Algorithm for electrical power output	Algorithm predicting renewable energy source output.
Algorithm checking if actual output from each renewable source is as expected and predicted
Algorithm predicting output from renewable based on short-range weather forecast.
Algorithm predicting output from renewable based on long-range weather forecast.
Predicted electrical power demand from internal conventional energy source.
Predicted electrical power demand from public electricity network.

Algorithm for electrical power output·. Self-calibrating algorithm checking actual elec5 trical power output and predicting electrical power output from renewable energy sources on a short-range time scale and on a long-range time scale.

The short-range time scale may be measured in hours.

The long-range time scale may be measured in days.

The long-range time scale may define the planning period.

The predicted and actual measured data is constantly evaluated using an alarm and a level indicator. The level indicator (L) and alarm (A) is used for calibrating the algorithm(s). The level indicator rates the level by which the predicted and measured data 15 agree and the impact of the data in the calibration algorithm. The alarm indicator is used when the predicted and measured data is inconsistent which extends beyond a given threshold value.

Algorithm predicting renewable energy source output.	Database “Actual meteorology”.	A	L
Collect, time stamp, synchronize and store dataset of produced electrical power from each renewable energy source in the database: Database “produced electrical

power”.

Algorithm predicting renewable energy source output may predict the output from each renewable source in the planning period.

Database “Actual meteorology” may comprise time-stamped and synchronized data from a meteorology station.

Database “produced electrical power” may comprise high frequency power measurements from each renewable source.

High frequency power measurements may include current, voltage and power output characteristics.

Algorithm checking if actual output from each renewable source is as predicted.	Database “Actual meteorology”
Actual production output data from high frequency power inflow measurements from each renewable source.
Data base with time stamp, time synchronize and store data about production output from high frequency power measurements from each renewable source.

Algorithm predicting output from renewable based on short-range weather forecast.	Data base: Short-range weather forecast	A	L
Algorithm predicting expected output from each renewable source in the planning period

The short-range weather forecast may be a weather forecast on a short-range time scale. The long-range weather forecast may be a weather forecast on a long-range time scale. The weather forecasts may be updated on an hourly basis.

Algorithm predicting

Data base: Long-range weather forecast

A

L

output from renewable based on long-range weather forecast.

Algorithm predicting expected output from each renewable source in the planning period

Predicted electrical power demand from internal conventional energy source.	Database: Price (Intern produced)	A	L
Maximum electricity output possible from internal conventional energy source
Data base with time stamped and synchronized data from high frequency

Database: Price (Intern produced) - with price in planning period for electricity produced on actual internal conventional energy source.

Predicted electrical power demand from public electricity network.	Database with price in planning period for electricity purchased from electricity grid (also considering if prices differs in this period)	A	L
Maximum electricity inflow possible from the public electricity grid
Data base with time stamped and time synchronized data from high frequency power measurements from public electricity grid source.

Meteorology data	Actual weather data from meteorology station are collected, time stamped, time synchronized and stored in data base	Time stamped wind speed data	A	L
Time stamped temperature data
Time stamped humidity data
Time stamped air pressure data
Time stamped wind direction data
Time stamped UV-index data
Time stamped solar irradiation data
Weather forecast	Short range	Time stamped wind speed exp.
Time stamped temperature exp.

	are stored in data base	Time stamped humidity exp.
Time stamped air pressure exp.
Time stamped wind direction exp.
Time stamped UV-index exp.
Time stamped solar irradiation exp.
Long range	Time stamped wind speed exp.
Time stamped temperature exp.
Time stamped humidity exp.
Time stamped air pressure exp.
Time stamped wind direction exp.
Time stamped UV-index exp.
Time stamped solar irradiation exp.

In the above the work expectations are abbreviated to exp.

Al_Dist (Actual)	Algorithm for allocation for electricity to internal clients connected directly to the distribution centre	Algorithm constantly updating expected need for electricity to internal clients for direct use in their operation.
Algorithm for balancing a renewable energy storage system
Allocation of electricity to external clients where the public utility network is used to transport the electricity
Allocation of electricity to the local utility network

Algorithm constantly updating expected need for electricity to internal clients for direct use in their operation - (in this example large consumers / large companies) based on data bases with time stamped and time synchronized data and manual input related to special expectations / variations from normal and specific requirements

Algorithm for balancing a renewable energy storage system (in this case a hot water storage system) - Algorithm constantly updating expected electricity need distributed to (in this case) to a large heat pump system producing hot water to INTERNAL storage system in the planning period - based on algorithms and analysis of data in data 5 bases with time stamped and time synchronized data and manual input related to expectations / requirements from operators.

Algorithm constantly updating expected need for electricity to internal clients for direct use in their operation.	Data base with time stamped and time synchronized data from high frequency power outflow measurements to each internal client/ large users (current, voltage and power output characteristics)	A	L
Data base with time stamped and time synchronized data from meteorology station
Data base: short-range weather forecast
Data base: long-range weather forecast
Algorithm for balancing a renewable energy storage system	Algorithm constantly updating expected remaining hot water storage capacity in INTERNAL hot water storage tank in planning period
Algorithm constantly updating expected efficiency of heating pump system output of hot water to INTERNAL hot water storage tank in planning period based on analysis of data bases with time stamped and time synchronized data and Manuel input of expectations and requirements from operational staff
Algorithm constantly updating expected need in planning period for distribution of hot water from INTERNAL hot water storage tank to central district heating network (Assuming that hot water generation with conven-	Algorithm constantly updating expected need in planning period for hot water consumption from end users connected to central district heating network
Algorithm constantly updating expected need in

	tional sources at the central district heating is used entirely as backup)	planning period for hot water from INTERNAL hot water storage tank to be distributed to other hot water storage tanks connected to central district heating network
Algorithm constantly updating expected need in planning period for distribution of hot water to large clients connected directly to INTERNAL hot water storage tank - (Assuming that hot water generation with conventional sources at those large clients is used entirely as backup)	Algorithm constantly updating expected need in planning period for hot water consumption from large clients/users connected directly to INTERNAL hot water storage tank
Algorithm constantly updating expected need in planning period for hot water from INTERNAL hot water storage tank to be distributed to other hot water storage tanks owned by large clients/users connected directly to INTERNAL hot water storage tank
Algorithm constantly updating expected need for electricity used by water cooker system for increasing water temperature in INTERNAL water storage tank - based on time stamped and synchronized data and manual input related to special expectations / variations from normal and specific requirements

Algorithm constantly updating expected remaining hot water storage capacity in INTERNAL hot water storage tank in planning period	Data base with time stamped, time synchronized and stored measurements of actual temperature in INTERNAL hot water storage tank	A	L
Algorithm constantly updating expected electricity need distributed to be used by large heat pump producing hot water to INTERNAL storage system in the planning period
Algorithm constantly updating expected heating pump system output efficiency of hot water to INTERNAL hot water storage tank in planning period
Algorithm constantly updating expected need in planning period for distribution of hot water from INTERNAL hot water storage tank to central district heating network
Algorithm constantly updating expected need in planning period for distribution of hot water to large clients connected directly to INTERNAL hot water storage tank

Algorithm constantly updating expected efficiency of heating pump system output of hot water to INTERNAL hot water storage tank in planning period based on	Data base with time stamped, time synchronized and stored measurements of high frequency power outflow measurements to heating pump system (current, voltage and power output characteristics)	A	L
Data base with time stamped, time synchronized and stored measurements of Actual flow of hot water to INTERNAL accumulation tank from heating pump system
Data base with time stamped, time synchronized and stored measurements of Actual temperature of hot water to INTERNAL accumulation tank from heating pump system
Algorithm constantly optimizing actual	Data base with time stamped, time synchronized and stored measurements of high frequency power outflow measure-

analysis of data bases with time stamped and time synchronized data and Manuel input of expectations and requirements from operational staff	flow of sea water system in planning period based on analysis of data bases with time stamped and time synchronized data and Manuel input of expectations and requirements from operational staff	ments to INTERNAL sea water pump system (current, voltage and power output characteristics)
Data base with time stamped, time synchronized and stored measurements of Actual flow of sea water to INTERNAL sea water storage
Data base with time stamped, time synchronized and stored measurements of Actual temperature of sea water to INTERNAL sea water storage
Data base with time stamped, time synchronized and stored measurements of Actual flow of sea water from INTERNAL sea water storage
Data base with time stamped, time synchronized and stored measurements of Actual temperature of sea water from INTERNAL sea water storage

Algorithm constantly updating expected need in planning period for hot water consumption from end users connected to central district heating network	Data base with time stamped, time synchronized and stored measurements of weather data from meteorology station	A	L
Data base: short-range weather forecast
Data base: long-range weather forecast
Data base with time stamped, time synchronized and stored measurements of hot water consumption from end users connected to central district heating network (including losses)
Database with new additional users connected to central district heating system in planning period

Algorithm constantly updating expected need in planning period for hot water from INTERNAL hot water storage tank to be distributed to other hot water storage tanks connected to central district heating network	Data base with time stamped, time synchronized and stored measurements of Actual temperature in each external hot water storage tank connected to central district heating system	A	L
Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of incoming cold water and outgoing hot water to INTERNAL hot water storage tank directly from end user network
Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of incoming hot water from central district heating conventional hot water source to end user network
Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of incoming cold water and outgoing hot water from INTERNAL hot water storage tank to storage tanks connected to end user network
Algorithm constantly updating expected hot water supply from renewables in planning period connected to central district heating network - for example solar	Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of incoming cold water from central district heating network to renewable system
Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of outgoing hot water from renewable to central district heating network
Database with collected, time stamped, time synchronized and stored actual weather data from meteorology station
Data base: short-range weather forecast
Data base: long-range weather forecast

Algorithm constantly updating expected need in planning period for hot water consumption from large clients/users connected directly to INTERNAL hot water storage tank	Database with collected, time stamped, time synchronized and stored actual weather data from meteorology station	A	L
Data base: short-range weather forecast
Data base: long-range weather forecast
Data base with time stamped hot water consumption from large clients/users connected directly to INTERNAL hot water storage tank (including losses)
Database with extraordinary fluctuations in expected consumption of hot water in planning period

Algorithm constantly updating expected need in	Data base with time stamped, time synchronized and stored measurements of Actual temperature in each external hot water storage tank connected to large clients/users heating system	A	L
planning period for hot water from INTERNAL hot water storage	Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of incoming cold water and outgoing hot water to INTERNAL hot water storage tank from large clients/users network
tank to be distributed to other hot water storage tanks	Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of incoming hot water from large clients/users conventional hot water source to large clients/users network
owned by large clients/users connected directly	Algorithm constantly updating expected hot wa-	Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of incoming
to INTERNAL hot water storage tank	ter supply from renewables in planning period	Data base with time stamped, time synchronized and stored measurements of Actual temperature on and flow of outgoing

	connected to large clients/users heating network - for example solar	Database with collected, time stamped, time synchronized and stored actual weather data from meteorology station
Data base: short-range weather forecast
Data base: long-range weather forecast

Algorithm constantly updating expected need for electricity used by water cooker system for increasing water temperature in INTERNAL water storage tank based on time stamped and synchronized data and manual input related to special expectations / variations from normal and specific requirements	Algorithm constantly updating expected remaining hot water storage capacity in INTERNAL hot water storage tank in planning period	A	L
Algorithm constantly updating expected heating pump system output efficiency of hot water to INTERNAL hot water storage tank in planning period based on analysis of time stamped and synchronized data and Manuel input of expectations and requirements from operational staff
Algorithm constantly updating expected need in planning period for distribution of hot water from INTERNAL hot water storage tank to be distributed to central district heating network - (Assuming that hot water generation with conventional sources is used entirely as backup)
Algorithm constantly updating expected need in planning period for distribution of hot water to large clients connected directly to INTERNAL hot water storage tank
Data base with time stamp, time synchronize and store data about electricity consumption from water cooker system (current, voltage and power output characteristics)

Specifications on water cooker system defining electricity input related to heating capacity of water

Allocation of electricity to external clients where the public utility network is used to transport the electricity

Data base with time stamped and time synchronized data from high frequency power outflow measurements to each external client/ large users (current, voltage and power output characteristics)

A

E

Allocation of electricity to the local utility network

Data base with time stamped and time synchronized data from high frequency power outflow measurements to local utility network (current, voltage and power output characteristics)

A

L

Description of the Drawings

Figure 1 illustrates an electrical power allocation system.

Figure 2 illustrates a method for balancing received and distributed electrical power in a connected system.

Figure 3 shows a top-level hardware and implementation view of one embedded com10 puter system.

Figure 4 illustrates an embodiment of a system for performing the method of allocating electrical energy.

Figure 5 illustrates an embodiment of the method of allocating electrical energy.

In figure 6 is illustrated an embodiment of electrical power measurements on energy sources and energy sinks by frequent sampling.

Figure 7 illustrates an embodiment in which the electrical power is allocated to drive pumps to drive a water heating system.

Figure 8 illustrates another embodiment in which the electrical power is allocated to drive pumps to drive a water heating system and a cooling water system.

Figure 9 illustrates an embodiment in which the electrical power is allocated to drive a cooker system.

Figure 10 illustrates a cooling water buffer system and storage tanks.

Figure 11 illustrates an embodiment of a meteorology station.

Figure 12 illustrates another embodiment of a system for performing the method of allocating electrical energy.

Figure 13 illustrates the legends used in figure 12.

Detailed Description of the Invention

No.	Item
10	Distribution unit
20	Energy source
22	Renewable energy source
30	Energy sink
40	Electrical power
42	Received electrical power
44	Distributed electrical power
50	Public electricity network
60	Energy storage facility
70	Large consumers
90	Meteorology station

100	System
110	Connected
112	Configured
120	Controlling means
122	Executing means
124	Controller
130	Time stamped
140	Synchronized
150	Processing
160	Sampling using a fast sampling rate
200	Data set
210	Actual electrical power production
212	Predicted electrical power production
220	Actual electrical power demand
222	Predicted electrical power demand
230	Actual electrical power storage capacity
232	Predicted electrical power storage capacity
240	Actual meteorology data
242	Predicted meteorology data
242A	Short-term local weather forecast
242B	Long-term local weather forecast
244	Local weather forecast
310	Embedded computer system
320	Sensors
330	Database
340	Interfaces
350	Input
360	Synchronization system
400	Method
410	Allocating
412	Receiving
414	Distributing
416	Balancing

418	Executing
420	Collecting
422	Storing
424	Providing
430	Time stamping
432	Synchronizing
434	Processing
450	Algorithm
452	Al_Bal
454	Al_Rec
456	Al_Dist

Figure 1 illustrates a system 100 in which one or more energy sources 20 are connected 110 to a distribution unit 10. The energy sources 20 may consist of conventional energy sources and/or renewable energy sources. The system may also be connected to a public electricity network 50 (not illustrated) which in that case may comprise one energy source 20. The distribution unit 10 is then further connected 110 to one or more energy sinks 30. The energy sinks 30 may be internal consumers, external consumers. Again, in the case of energy sinks 30 the system 100 may be connected to a public electricity network 50 (not illustrated) which in that case may comprise one energy sink 30. The distribution centre 10 allocates electrical power 40 from the energy source(s) 20 to the energy sink(s) 30. Where the electrical power 40 from the energy source(s) 20 are received electrical power 42, and the electrical power 40 to the energy sink(s) 30 are distributed electrical power 44.

The renewable energy source(s) may comprise wind turbines, solar cells, thermal solar panels, units for harvesting wave energy or thermal energy from the underground, any other existing or future renewable energy source, or a combination of these. The energy produced by the energy source(s) 30 are converted to electrical power 40 before reaching the distribution centre which then distributes the received electrical power 42 to the one or more energy sinks 30. In case of thermal energy this is used through heat exchange systems driven by the electrical power allocated by the distribution centre.

Thus, the harvested thermal energy is not converted to electrical power and back and thus, energy losses of such conversion are avoided.

The system may also comprise energy storage facilities. The storage facilities may act as both an energy source 20 and energy sink 30.

Figure 2 illustrates an embodiment of the method 400 of allocating 410 electrical power 40 from one or more energy sources 20 to one or more energy sinks 30. The distribution center receives 412 electrical power from one or more energy sources 20 and then allocates 410 the received electrical power by distributing 414 the electrical power to the energy sink(s) 30 which has an actual demand for electrical power 40. The allocating 410 of electrical energy may be balanced 440 using one or more algorithms 450 executable in the distribution unit 10. The algorithm(s) (Al_Bal 452) for balancing the allocated electrical power may be further dependent on other algorithms which regulates the level of received electrical power 42 (Al_Rec 454) from each energy source 20 to the distribution unit (10) and regulates the level of distributed electrical power 42 (Al_Dist 456) from the distribution unit 10 to each energy sink 30.

Figure 3 shows one embodiment of a hardware implementation of one embedded computer system 310 used for condition monitoring. On the left side is shown the interfaces relating to input 350 from sensors 320 and databases 330.

In this embodiment a range of sensors 320 measuring meteorology data are illustrated comprising wind speed sensors, temperature sensors, rain sensors, humidity sensors, air pressure sensors, wind direction sensors, UV sensors, solar irradiation sensors.

Furthermore, sensors measuring electrical power are illustrated along with sensors measuring on liquid parameters, which may comprise flow, temperature, density or other relevant parameters. These sensors may measure on energy sinks, sources and storage facilities to monitor the produced, consumed and stored energy.

All data collected by these sensors or a selected number of the sensors may be time stamped (430) and synchronized (432). Alternatively the data collected from the sensors may be time stamped (430) and subsequently synchronized (432).

Additionally, optional input is illustrated which may consist of additional relevant condition monitoring and measurement instruments which can be added to extend the invention to support an improved condition monitoring. This may include fault detection and instant alarm system for other key components in the electrical power allocation system.

Furthermore, the input 350 to the embedded computer system 310 may comprise data sets 200 from various databases 330. These databases 330 may comprise meteorology data sets, data sets on electrical power from energy sources, sinks and storage facility. The data sets may both be predicted values and actual measured values.

The embedded computer system 310 in figure 3 may comprise at least one processor for processing said input. There is at least one storage means for the storing of collected measured and calculated values etc. to be used in the algorithms comprised in the method.

The output interfaces 340 are also illustrated in figure 3. These may comprise at least one power backup with sufficient capacity to safely shut down all software in the embedded computer system 310 and attached systems in case of sudden loss of permanent power supply, at least one power supply to the embedded computer system 310 and attached systems, at least one terminal interface and one USB interface option, and at least one communication interface providing option to transfer larger data amounts - could be WAN interface, GPRS / 3G / 4G / 5G interface, or any other communication interface that may become relevant in the future to be able to transfer larger data amounts.

Thus the output from the embedded computer system controls the distribution of the available electrical power based on the received input.

Figure 4 illustrates an embodiment of a system 100 for performing the method 400 of allocating electrical energy 40. In this embodiment wind turbines 22 acts as energy sources 20 and a hardware setup, acting as the distribution centre 10 receives the electrical power from the wind turbines and distributes electrical power 42 to large consumers 70 and energy storage facilities 60. The distribution centre 10 either receives or distributes electrical power to the public electricity network 50 depending on if there is a surplus or deficit of received electrical power compared to the distributed electrical power. Thus a balancing of received and distributed electrical power 40 is performed.

Figure 5 illustrates an embodiment of the method 400 of allocating electrical energy 40. In this embodiment is illustrated how software, algorithms and artificial intelligence are used to balance the received and distributed electrical power in a balancing algorithm. The balancing algorithm receives data input of actual and predicted electrical power production and demand, where the predicted data is based on weather forecasts and historical data on electrical power demand and production. Figure 5 further more illustrates how the hardware and software in the distribution unit are arranged such that the allocating of electrical power may be performed in a balanced way.

In figure 6 is illustrated an embodiment of collecting data on energy 20 sources and energy sinks 30 by sampling using a fast sampling rate on the connections transferring electrical energy to and from the distribution centre 10. The data may be collected synchronized by collecting data from all connections on a trigger signal transmitted from the distribution centre 10. By sampling using a fast sampling rate it is possible to retrieve a broad range of data of the individual energy sources 20 and/or energy sinks 30.

Figure 7 illustrates an embodiment in which the electrical power is allocated to drive pumps to drive a water heating system wherein heat from warm seawater is extracted to district heating or other methods of distributing heated water to consumers. In this embodiment sampling using a fast sampling rate is used for measuring on the pumps and retrieve data on their performance.

In figure 8 another embodiment is illustrated wherein the electrical power also is allocated to drive pumps to drive a water heating system. The embodiment in figure 8 is an extended version of the embodiment illustrated in figure 7. In this embodiment an additional cooling exchanger is added to provide cooling water to consumers.

Figure 9 illustrates an embodiment in which the electrical power is allocated to drive a cooker system for heating water to distributed heating water for consumers. In this embodiment sampling using a fast sampling rate is used for measuring on the cooker system and retrieve data on the performance of the electrical components.

In figure 10 a buffer system and storage tanks comprising cold water is illustrated. Figure 9 and figure 10 both illustrates water storage systems which may be connected to the distribution centre 10 through electrical components working as energy sinks of electrical power but storage facilities of thermal or mechanical energy. Furthermore, the water storage systems may be connected to the distribution centre 10 through sensors. The sensors may be temperature sensors, flow sensors, level sensors or the like, which may provide the method 100 with data regarding the state of the storage facilities.

In figure 11 an embodiment of a meteorology station (90) is illustrated. The meteorology station may be localized on the premises or on the close vicinity of the system 100 thereby providing the method 400 with accurate data of the actual meteorology state where the renewable energy source(s) 20 and the energy sink(s) 30 are positioned.

Figure 12 illustrates an embodiment of a system 100 for performing the method 400 of allocating electrical energy 40. The arrows in the figure illustrate how the energy is allocated from the energy sources to energy sinks. The types of energy sinks, energy sources and storage facilities illustrate in figure 12 are illustrated and explained in figure 13.

Figure 13 contains the legends of figure 12

A: Renewable energy source 22, wind turbine

B: Meteorology station, 90

C: Energy sink 30, pump based on sea water

D: Energy storage facility 60, storage tank for hot water

E: Energy storage facility 60, storage tank for cold water

F: Energy storage facility 60, hydrogen production plant

G: Energy storage facility 60, hydrogen filling station

H: Energy sink 30, electrically powered ferry

I: Public electricity network 50

J: Large consumer 70, heating plant

K: Energy sink 30, city

L: Distribution unit 10

M: Energy sink 30, hydrogen powered car

N: Energy sink 30, industrial plant A

O: Energy sink 30 industrial plant B

P: Renewable energy source 22, solar panels.

Claims

KRAVREQUIREMENTS

1/131/13

100100

1. Fremgangsmåde (400) til fordeling af (410) elektrisk effekt (40) fra en eller flere energikilder (20) til en eller flere energiforbrugende enheder (30) omfattende fremgangsmådeskridt af, at:A method (400) for distributing (410) electrical power (40) from one or more energy sources (20) to one or more energy consuming units (30) comprising the method steps of:

modtage (412) elektrisk effekt (40) fra en eller flere energikilder (20) til en fordelingsenhed (10);receiving (412) electrical power (40) from one or more energy sources (20) to a distribution unit (10);

fordele (414) elektrisk effekt (40) til en eller flere energiforbrugende enheder (30) fra en fordelingsenhed (10); og balancere (416) den fordelte elektriske effekt (44) med den modtagne elektriske effekt (42), og hvori fordelingsenheden (10) er forbundet (110) til den ene eller flerheden af energikilder (20) og den ene eller flerheden af energiforbrugende enheder (30);distributing (414) electrical power (40) to one or more energy consuming units (30) from a distribution unit (10); and balancing (416) the distributed electrical power (44) with the received electrical power (42), and wherein the distribution unit (10) is connected (110) to the one or the plurality of energy sources (20) and the one or the plurality of energy consuming units. (30);

konfigureret (112) med styreorganer (120) til at styre fremgangsmådeskridtene af, at modtage (420) og fordele (414) elektrisk effekt; og konfigureret (112) til at udføre (418) en eller flere algoritmer konfigureret til at udføre fremgangsmådeskridtet af at balancere (416) elektrisk effekt i systemet, fremgangsmådeskridtet af at modtage (412) elektrisk effekt (40) og fremgangsmådeskridtet af at fordele (414)) elektrisk effekt (40), kendetegnet ved, at det førnævnte fremgangsmådeskridt af at modtage (412) yderligere omfatter fremgangsmådeskridtene af, at:configured (112) with control means (120) for controlling the method steps of receiving (420) and distributing (414) electrical power; and configured (112) to perform (418) one or more algorithms configured to perform the method step of balancing (416) electrical power in the system, the method step of receiving (412) electrical power (40) and the method step of distributing (414) ) electric power (40), characterized in that the aforementioned method step of receiving (412) further comprises the method steps of:

• indsamle (420) og lagre (422) datasæt (200), hvilke omfatter tidsstemplede datasæt (130) for modtaget elektrisk effekt (42) fra hver enkelt energikilde (20); og • synkronisere (432) og behandle (434) de tidsstemplede datasæt (130) for modtaget elektrisk effekt (42) med tidsstemplede datasæt (130) for forudsagt energi produktion (212) fra hver energikilde (20), og det førnævnte fremgangsmådeskridt (414) yderligere omfatter fremgangsmådeskridt af, at:Collecting (420) and storing (422) data sets (200), which comprise time-stamped data sets (130) for received electrical power (42) from each energy source (20); and • synchronizing (432) and processing (434) the time-stamped data sets (130) for received electrical power (42) with time-stamped data sets (130) for predicted energy production (212) from each energy source (20), and the aforementioned method step (414 ) further comprises procedural steps of:

• indsamle (420) og lagre (422) datasæt (200), hvilke omfatter tidsstemplede datasæt (130) for distribueret elektrisk effekt (44) til hver enkelt energiforbrugende enhed (30); og • synkronisere (432) og behandle (434) de tidsstemplede datasæt (130) for forudsagt efterspørgsel (222) på elektrisk effekt fra hver enkelt energiforbrugende enhed (30).Collecting (420) and storing (422) data sets (200), which comprise time-stamped data sets (130) for distributed electrical power (44) for each individual energy consuming unit (30); and • synchronizing (432) and processing (434) the time-stamped data sets (130) for predicted demand (222) for electrical power from each individual energy consuming unit (30).

2. Fremgangsmåde (400) ifølge krav 1 omfattende de yderligere fremgangsmådeskridt af, at:The method (400) of claim 1 comprising the further method steps of:

indsamle (420) og lagre (422) et eller flere datasæt (200), hvilke omfatter den faktiske produktion (210) af elektriske effekt fra hver enkelt energikilde (20); og tilvejebringe (424) eksekveringsorganer (122) konfigureret til at udføre (418) en eller flere algoritmer konfigureret til at udføre fremgangsmådeskridtet med at regulere niveauet for modtaget elektrisk effekt (42) fra hver enkelt energikilde (20) til distributionsenheden (10).collecting (420) and storing (422) one or more data sets (200), which comprise the actual production (210) of electrical power from each energy source (20); and providing (424) execution means (122) configured to perform (418) one or more algorithms configured to perform the method step of controlling the level of received electrical power (42) from each energy source (20) to the distribution unit (10).

3. Fremgangsmåde (400) ifølge krav 1 eller 2 omfattende de yderligere fremgangsmådeskridt af, at:A method (400) according to claim 1 or 2 comprising the further method steps of:

indsamle (420) og lagre (422) et eller flere datasæt (200), hvilke omfatter den faktiske efterspørgsel (220) på elektrisk effekt fra hver eneklte energiforbrugende enhed (30); og tilvejebringe (424) en eller flere styreenheder (124) konfigureret til at udføre en eller flere algoritmer konfigureret til at regulere niveauet af distribueret elektrisk effekt (44) til hver eneklte energiforbrugende enhed (30) fra fordelingsenheden (10).collecting (420) and storing (422) one or more data sets (200), which include the actual demand (220) for electrical power from each individual energy consuming unit (30); and providing (424) one or more controllers (124) configured to perform one or more algorithms configured to control the level of distributed electrical power (44) for each individual energy consuming unit (30) from the distribution unit (10).

4. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 3, omfattende et yderligere fremgangsmådeskridt af, at lagre (422) elektrisk effekt (40) til at balancere (416) den fordelte elektriske effekt (44) til den modtagne elektriske effekt (42) fra en eller flere energikilder (20), hvori den fordelte elektriske effekt (44) omfatter en faktisk efterspørgsel (220) på elektrisk effekt.A method (400) according to any one of claims 1 to 3, comprising a further method step of storing (422) electrical power (40) to balance (416) the distributed electrical power (44) to the received electrical power. power (42) from one or more energy sources (20), wherein the distributed electrical power (44) comprises an actual demand (220) for electrical power.

5. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 4, hvori fremgansmådeskridtet af, at indsamle data udført synkroniseret, udføres under anvendelse af mindst et synkroniserings system (360), hvori synkroniseringssystemet (360) transmitterer et trigger-signal, der aktiverer synkroniseret måling af data fra en eller flere sensorer.A method (400) according to any one of claims 1 to 4, wherein the step of collecting data performed synchronized is performed using at least one synchronization system (360), wherein the synchronization system (360) transmits a trigger signal, enabling synchronized measurement of data from one or more sensors.

6. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 5 omfattende et yderligere fremgangsmådeskridt af, at kalibrere mindst en algoritme ved at sammenligne faktiske data med forudsagte data fra det ene eller flereheden af tidsstemplede (130) og synkroniserede (140) datasæt (200).A method (400) according to any one of claims 1 to 5 comprising a further method step of calibrating at least one algorithm by comparing actual data with predicted data from one or more of timestamped (130) and synchronized (140) data set (200).

7. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 6 omfattende de yderligere fremgangsmådeskridt af, at:A method (400) according to any one of claims 1 to 6 comprising the further method steps of:

indsamle (420) og lagre (422) faktiske meteorologidata (240);collecting (420) and storing (422) actual meteorological data (240);

indsamle (420) og lagre (422) vejrudsigt data (244), og indsamle (420) og lagre (422) forudsagte meteorologidata (242), hvori de faktiske meteorologidata (240) og de forudsagte meteorologidata (242) er tidsstemplet (130) og synkroniseret (140).collecting (420) and storing (422) weather forecast data (244), and collecting (420) and storing (422) predicted meteorological data (242), wherein the actual meteorological data (240) and the predicted meteorological data (242) are time-stamped (130) and synchronized (140).

8. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 7, hvori måling af elektrisk effekt på en, flere eller alle energikilder (20) og/eller energiforbrungende enheder (30) udføres ved hyppig prøvetagning af elektrisk effekt på tidsskalaen fra millisekunder til sekunder.A method (400) according to any one of claims 1 to 7, wherein measuring electrical power on one, several or all energy sources (20) and / or energy consuming units (30) is performed by frequent sampling of electrical power on the time scale from milliseconds to seconds.

9. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 8, hvori den ene eller flere algoritmer udfører et fremgangsmådeskridt af at kalibrere ved at sammenligne faktiske meteorologidata (240) med forudsagte meteorologidata (242), hvor de forudsagte meteorologidata (242) ekstrapoleres og estimeres udfra en lokal korttids- og/eller langtidsvejrudsigt (242A, 242B), faktiske produktion (210) af elektriske effekt med forudsagt produktion (212) af elektriske effekt, faktisk efterspørgsel (220) på elektrisk effekt med forudsagt efterspørgsel (222) på elektrisk effekt, og/eller faktisk lageringskapacitet (230) af elektrisk effekt med forudsagte lageringskapacitet (232) af elektrisk effekt, hvori de forudsagte data ekstrapoleres og estimeres fra forudsagte meteorologidata (242) og fra databaser af indsamlede og lagrede data.A method (400) according to any one of claims 1 to 8, wherein the one or more algorithms perform a method step of calibrating by comparing actual meteorology data (240) with predicted meteorology data (242), wherein the predicted meteorology data (242 ) extrapolated and estimated from a local short-term and / or long-term forecast (242A, 242B), actual production (210) of electrical power with predicted production (212) of electrical power, actual demand (220) for electrical power with predicted demand (222 ) on electrical power, and / or actual storage capacity (230) of electrical power with predicted storage capacity (232) of electrical power, in which the predicted data are extrapolated and estimated from predicted meteorological data (242) and from databases of collected and stored data.

10. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 9, hvori den ene 5 eller flerheden af algoritmer til at balancere (416) den fordelte elektriske effekt (44) til den modtagne elektriske effekt (42) er konfigureret til at evaluere driftstilstanden af de individuelle energikilder (20), energiforbrugende enheder (30) og/eller meteorologiske sensorer.A method (400) according to any one of claims 1 to 9, wherein the one or more of the algorithms for balancing (416) the distributed electrical power (44) to the received electrical power (42) is configured to evaluate the operating status of the individual energy sources (20), energy consuming units (30) and / or meteorological sensors.

10 11. Fremgangsmåde (400) ifølge et hvilket som helst af kravene 1 til 10, hvori maskinlæring anvendes til kalibrering af den ene eller flereheden af algoritmer.A method (400) according to any one of claims 1 to 10, wherein machine learning is used to calibrate the one or more plurality of algorithms.

2/132/13

400400