EP3433828A1

EP3433828A1 - A method and a system for dynamic aggregation of a fleet of power units to provide frequency regulation of a power system

Info

Publication number: EP3433828A1
Application number: EP17714840.0A
Authority: EP
Inventors: Juhani RANTANIEMI; Janne HAPPONEN
Original assignee: Fortum Oyj
Current assignee: Fortum Oyj
Priority date: 2016-03-24
Filing date: 2017-03-03
Publication date: 2019-01-30
Also published as: WO2017162910A1; FI20165255A7; FI128279B

Abstract

The invention relates to a method and system for providing frequency regulation to a power system and for aggregating a fleet of power units by means of a central control unit. The power units are connected to the power system and are capable of consuming and/or producing electrical energy. At the beginning of a given frequency regulation period the central control unit reads a utility frequency target value for the power system and from a data source at least one of the following: the current status of the power units, the power rating of each power unit, and/or an estimated or measured energy demand for each power unit. The sizes of an up regulation reserve and a down regulation reserve required for the power system to reach said utility frequency target value are estimated and a priority count for each power unit is established based at least on the amount of time a power unit has been in its present status and/or its remaining power demand. The utility frequency is monitored and in response to a deviation there of the power units are transferred according to their priority from the down regulation reserve to the up regulation reserve if the utility frequency is rising, and power units from the up regulation reserve to the down regulation reserve, if the utility frequency is decreasing.

Description

A method and a system for dynamic aggregation of a fleet of power units to provide frequency regulation of a power system

Field of the Invention The invention relates to methods and systems for dynamic aggregation of fleets of loads and other power consuming and/or producing units to provide frequency regulation of a power system. In particular, the invention relates to modern energy systems that are based on smart grids and on the control of loads and other power units located in the grid.

Background of the Invention A future energy system is likely to be based on C02-free electricity production and resource efficiency. There will be a gradual transition from traditional energy production based of finite energy sources and combustion fuels towards production forms based on infinite and zero- emissions solar and wind energy. The transition will take time in the capital-intensive energy industry and will advance in phases. Such new energy systems require smart grids and demand side management solutions. Smart grids may take into account much more energy production and consumption variations than until now and analyze the supply/demand scheme that adapts dynamic pricing, load aggregation and frequency control schemes to optimally meet the energy demand with a supply response.

For example, in the night time nuclear power and hydro (or wave) power runs, but load is low since people are sleeping. Then electric car batteries may then be charged at a lower price. In the morning when the wind power is increasing (sea breeze) the price may go further down, to later go up when the load is increasing. At noon the load is increasing further, whereby the price goes up and hydro power plants may be started. In the afternoon, less clouds will give more solar power and the energy price declines. In the night when the sun goes down and gives less power production, electric cars are started to being charged. The load increases and the price go up, until people go to sleep. In a scenario like this, the daily price variation is likely to increase.

Recent developments of data communication and computing has enabled smart grid solutions, such as aggregated load control. Controlling of loads according to the state of the power system may be referred to as Demand response (DR) or Demand side management (DSM). Demand response is a desired service for Transmission system operators (TSO) or Distribution system operator (DSO), which pay compensation to providers for frequency regulation. In a synchronized power system, the utility frequency is the same in every point. A single frequency value can therefore be used to control the virtual power plant operating on a synchronized power system. The significance of grid balancing on demand side has grown along the prevalence of renewable power sources, such as wind and solar power and a variety of load and storage types that can store energy. This kind of controllable asset is said to have flexibility, which can be used for example in frequency regulation.

Power unit types suitable for aggregation and aggregated control are for example electric heating, cooling and ventilation appliances and batteries, e.g. for electric vehicles. The presented inventive control logic is planned to operate with these types of units having flexible energy charge and discharge patterns. In known demand response solutions the response is performed according to the hourly prices of energy or on direct requests of the Transmission system operators (TSOs). Frequency regulation is mainly performed by electricity production, such as by ramping up and down hydro power and gas turbines. From the publication WO 2015/058279 is known a method for managing a group of energy consuming loads for frequency regulation purposes. The loads are switched on and off according to enablement decisions made in a sequence, in order to provide for the frequency regulation. The enablement to turn a load on is determined in a decentralized manner by a controller at each load. A communication protocol transfer the enablement requests to frequency regulating means which send out a frequency regulation signal, in response to which grouped load controllers try to optimize the group's ability to match the frequency regulation signal.

Summary of the Invention

In a centralized system as in the present invention, a fleet of power units are dynamically aggregated by a control system at the beginning of a time frame, based on a variety of estimated or measured parameters predicting the energy demand for each unit during the time frame. The logic divides the power units to up and down regulation reserves and commands them

accordingly. A target may be set for the logic to ensure that the size of the aggregated up and down regulation reserves are, for example, equal in size at a set utility frequency, e.g. 50 Hz. If the utility frequency increases, the size of the up regulation reserve is increased and down regulation is decreased, for example by changing the status of a load from being idle to consume electricity. When the utility frequency falls, the size of the down regulation reserve is increased and the up regulation reserve decreased. In this manner, it is possible to provide fast and gradual regulation both up and down for transmission system operators.

According to the invention, the power units are centrally monitored and controlled by the control system. A priority count for each power unit is established based on its present status, the desired utility frequency regulation reserves in both directions (up/down) and other possible factors. A revised aggregation of the power units is repeatedly calculated after certain time intervals. The power units are then re-aggregated in order to provide an optimized response to the demand of each power unit while maintaining the frequency regulation target.

Among the benefits of the invention, it brings more means for the transmission system operators (TSOs) to balance their power system. Also the aggregator and the owners of the power units gain benefit since they receive compensation out of frequency regulation. The invention also enables more solar and wind production since their unbalancing effect on the power system can be neutralized more easily.

Further advantages of the inventive method and system includes:

- frequency control can be done both up- and downwards;

- the control response is very fast;

- the regulation can be almost linear since individual power units are often small compared to the size of the power system;

- individual requirements that may have for their units may be taken into account in

determining the control priority;

- changes in circumstances (e.g.) temperature may be dynamically taken into account.

Definitions As used herein, the following definitions are provided for the terminology of this disclosure: Frequency regulation

To provide frequency regulation means in this context the tangible and technical aim of keeping or assisting in the keeping of the target utility frequency of a synchronous power system of e.g. at 50 Hz or 60 Hz. This service if often required by the Transmission system operators, which organizes the frequency regulation in order to maintain the balance of the production and consumption in the power system. If there is more production than consumption in the power system the utility frequency gets greater than the target utility frequency. When the consumption is greater than the production the utility frequency gets under the target utility frequency. The frequency value used in the regulation can be the actual real time frequency value or a frequency value that is for example an average frequency value of the last minute. Using this kind of moving average value prevents the brief changes in the frequency to affect the regulation.

Frequency regulation is a product, typically sold by an Energy utility company on a frequency regulation market to a Transmission system operator, and the regulation time frame is part of that product.

Smart Grid

Smart grid is a power system of the future, where supply and demand are balanced on different time frames, in addition to power plants, also by smart resources, that are connected to the grid. A fleet of power units, as described in this patent, is one form of smart resource. Development towards smart grids increases the number of participants in the system and requires sophisticated control methods. Communication in the smart grid between the aggregator and the power units can be done real-time in several fashions.

Power Unit

A power unit is a device that is able to consume electrical power from the power grid and/or produce electrical power to the grid. Such devices are for example electric heating boilers, AC- units, electric vehicles, diesel generator and batteries.

Power unit state

Power units can have up to three different states which are power consuming/charging state, idle state and power producing/discharging state. The inventive method commands power units to these different states according to the frequency value and local requirements that the power unit operation has. By default the power units consume and produce the full nominal power in the I charge and discharge state. In the idle state the consumed and produced electric power is zero.

Control System A central control computer with associated software, memory, databases and communications capabilities, which is configured to implement the inventive method and system, and therein at least to retrieve and receive information on-line and offline and to issue control commands over to power units connected to the communication network and capable of being remotely controlled.

Aggregation - re-aggregation

Individual power units are often extremely small compared to the size of the power system. With aggregation these units can be grouped so that the size of the aggregation fleet is the sum of the individual power units. Within the aggregation fleet, the units are controlled so that the desired frequency regulation effect is achieved by adjusting the size of the up and down regulation reserves. In the re-aggregation the required states of the power units are re-computed and updated.

Up regulation reserve URR & down regulation reserve DRR

Up regulation reserve consists of power units which can react to decrease of the utility frequency by changing their power unit state in such sense that the utility frequency rises. For example, aggregated group of electric boilers consuming electricity can operate as up regulation reserve.

Down regulation reserve again consists of power units which react to increase of the utility frequency by changing their power unit state so that the utility frequency decreases. For example, aggregated group of batteries which are discharging can operate as down regulation reserve. Aggregated group of batteries can be in both up and down regulation reserves if they have ability to both charge and discharge.

Energy demand

Energy demand means a predicted or otherwise calculated understanding of the forthcoming energy consumption or production of a power unit, or an aggregation of such units. For each unit, its energy demand is the basis for calculating the time it should be provided with energy supply or be discharged.

Given Time Frame The time period during which the power units subject to aggregation should be provided with energy supply so that they at the end of the time frame are adequately charged, heated or filled, the time frame for a water boiler being for example between 23.00 -07.00 hours.

Priority Count A value (e.g. an integer 1...N) dynamically assigned to each power unit during the given time frame that controls which power units may be provided with energy supply during the next re- aggregation cycle. The priority count may be mainly based on the time in the present status, and/or the amount of time the power unit is still in demand of charging or discharging power within a given timeframe. Under certain circumstances, frequency regulation issues, environmental changes and a device-specific priority classification may also affect the priority count. Such cases may include the distance to a unit(e.g. loss factors), loads dependent on airborne consumer power lines (e.g. before an imminent storm indicating possible power outages), and other weather-based priority overrides, e.g. that all car battery priorities are adjusted downwards in favor of house warming during a sudden cold. The specific embodiments of the invention defined in the appended claims. In the following, various exemplary embodiments of the invention are described while making reference to the attached drawings.

Brief description of the drawings Fig. 1 shows a schematic view of a fleet of aggregated power unit; Fig.2 shows an example of a frequency regulation control curve; Fig. 3 shows the principle of frequency regulation; Fig. 4 shows a workflow of an inventive system;

Fig. 5 shows a workflow of a control algorithm according to the present invention.

Detailed description of Embodiments In Fig. 1 is shown schematically a fleet 10 consisting of power units 1 ... N. In this case a subset of the power units (units 5 ... N) may be aggregated to a frequency regulation fleet 11. The assignment of power units to group 11 may be based solely on the suitability of the units to be commanded to perform the frequency regulation by an aggregator's control system 12 during the given time. The control system 12 is serving a frequency regulation market 13 where the aggregator is acting as a service provider, and provides frequency regulation by aggregating the power units 11 with the inventive method.

Fig.2 shows an example of a frequency regulation control curve, which defines the proportion of up and down regulation reserves that are required in function of frequency. In practice the frequency response is almost linear due to the large amount of controllable power units involved. A power supply network utility frequency is allowed to vary within narrow limits, the exemplary range being shown is between 49,9 - 50,1 Hz. The total regulating power Pw is the combined power production and power consumption of the regulating fleet. At any point, the regulation reserves DRR and URR are dependent on the corresponding frequency F. If the regulation reserves DRR and URR are expressed as percentages, DRR = 100% - URR. In terms of power ratings (kW), the Up regulation reserve = Pw*URR and the Down regulation reserve =

Pw*DRR.

In Fig. 3 is shown the principle of frequency regulation in a simple example. The total consumption in W of six aggregated electric heaters is in the example controlled over time in order to contribute to performing frequency regulation according to the frequency values 30. As can be seen from the energy consumption curve 31 , a half of the heaters are switched on at the 50 Hz level indicated by 32, having then an energy consumption of some 13 kW. When the frequency falls towards 49,9 Hz at 33, first one and then two heaters are switched off to a minimum power rating of 5 kW, in order to up-regulate the utility frequency 30.

It becomes clear from Fig. 3 that besides software and algorithms which run in the central control system, the implementation may also require a hardware which is located next to the power units, and which can at least perform the commands issued over a smart grid. The power output and input can be controlled for example with a relay switch or by communicating directly to the control mechanisms of the power unit, for example using Modbus communication protocol. This may require a local hardware device with a communication gateway. In addition, the unit-specific hardware may measure the actual power consumption of the load, the charge status or discharge power in case of a battery, the water temperature in case of a water boiler, etc. Such additional information alleviates the need for estimates and makes any calculations on power demand or priority count etc. for the unit more accurate, and the frequency regulation, which is a real time effort, less subject to errors and subsequent corrections measures. According to the invention, the power units need to be prioritized so that the units can be put and held in a state which they require. When the loads are in a certain state, their priority usually increases since they are more prepared to change to another state. This ensures that each unit will be charged or discharged at least enough taking into consideration the local need for the unit. A number of other parameters or circumstances may affect the prioritization and

aggregation of the units, some of which are explained in the examples below. However, as a general rule for the priority determination, charging or discharging of the aggregated units is distributed to make use of the whole time frame T (e.g. night time). Each unit type may have its own optimal charging or discharging window or time frame T, depending on the typical usage of the power unit. The charging time of each unit is read from a database or estimated, and all loads are then given an initial priority or charging order. The charging time can also be negative if the unit needs to discharge energy. Where the current charge status (e.g. water temperature, battery charge level) of the load may be read by the control system, that can be taken into account when a priority count for that load is determined, but in general the task of the initial priorities is just to cause a distributing effect, i.e. not to start charging all loads at the same time. Obviously, the overall task becomes more complex when different loads with different time frames or charge windows are involved, but for each load group with a common time frame T the procedure remains the same.

For example, turning again to Fig. 2, there is shown a point P, which marks the target utility frequency F of the system at 50,075 Hz. From the curve it can be seen that an up regulation reserve URR of 75% is needed, and accordingly a down regulation reserve DRR of 25 % of the total fleet. If the starting point situation of the down vs. up regulation fleets is 40/60%, the control system starts to re-aggregate the power units toward the desired 25/75 % reserve division.

Over time, when some loads are charged and other have been idle, the priority count starts to reflect the actual need for charging, in order for the load to become adequately charged at the end of its time frame T. The inventive system enables that frequency regulation can be done fast also with relay controlled consumption, storages and production in a near-to-linear fashion, and that the regulation can be provided both up and down.

In Fig. 4 is shown a workflow of the inventive system. At 40, the system is initiated for a new start of workflow, which can be started for example once a second. The current frequency regulation control curve 41 (see Fig. 2) and the present utility frequency value 42 are read into the central control unit at 43. The task of the control unit 43 is to determinate the share of up regulation reserve and down regulation reserve as described earlier. The unit 44 is to determine which power units (consisting of loads, storages and power generating units), considering their priority, are required to meet the desired up and down regulation reserves. Tasks 43 and 44 are computed in in a computer that runs the software and algorithms needed to retrieve information and to perform the actions for the blocks 40 - 47. Such a computer corresponds to a control system as explained in the definitions.

At 44, power units 47 are aggregated either to a up regulation reserve or to a down regulation reserve and thus controlled based on the priority of each unit and power impact on the regulation at 45, and on demand forecast and other parameters at 46. These will be explained in more detail later on. Unit aggregation and controlling is continued until the desired ration of up and down regulation is achieved (Fig. 2).

Up regulation reserve comprises loads that are switched on, batteries being charged or power production units set to idle. When the utility frequency needs to be brought upwards, a change in status of any of these will have the effect of increasing the utility frequency. Correspondingly, down regulation reserve comprises loads that are switched off, batteries being discharged, or power production units set to discharge energy into the grid. When the utility frequency needs to be brought downwards, a change in status of any of these will have the effect of decreasing the utility frequency.

Obviously, certain batteries being able to discharge energy to a grid may have a third, "idle" status as well, which can be changed in either direction with regard to frequency regulation.

Table 1 lists units that are in a down regulation reserve. A priority group column shows the priority of a unit. The higher the priority, the more the unit needs or "can afford" to be set to a different status in the next re-aggregation. The units may be re-aggregated once a second, for example. If a status change is not induced in the next re-aggregation period (e.g. a second), it can anyway happen in later re-aggregation periods in the order of the priority. The next column, the "time in current state" tells the time a unit has spent in its current state. This unit time in the current status is an important, but not necessarily the only factor when determining the priority of a unit. For example some power units can tolerate shorter times between the change of the state than others. The third column tells the current state of the unit. The fourth column,

"Action", displays what unit is in question and what action will be taken in order to put the unit in a new state, i.e. in an up regulation reserve. The fifth column tells the power effect of the action in kilowatts (kW). The action taken on a unit in Table 1 will put it in a state which will push the utility frequency downwards. At the same time, the unit leaves the down regulation reserve and enters the up regulation reserve, with which the utility frequency may be regulated upwards. For example, if Load 44 is set to charge, it will start consuming energy. This will lower the utility frequency. The Load44 will enter the up-regulation reserve, because when switched off again, it will shift the utility frequency upwards. The high priority (1) of Load44 means it has been deemed to have a need to be connected to power, i.e. "to charge" soon, in order to perform its duties e.g. as a water boiler in a household. A production plant "Production 3" switched from a discharge status, i.e. from feeding a grid, to an idle status will increase the burden on the remaining producers and thus shift the utility frequency downwards. Such a production unit then enters the up-regulation reserve, as the utility frequency will shift upwards when the status of the unit is reversed to discharge.

Priority Time in current Current

Action Rating

count state (s) state

(kW)

1 250 discharge Battery 333 to idle 7

1 125 idle Load 44 to charge 3

1 100 idle Load 109 to charge 5

2 380 discharge Production 3 to idle 6,5

2 60 idle Load 99 to charge 4

2 20 idle Battery 111 to charge 4

2 20 idle Load 55 to charge 5,5

3 280 discharge Production 2 to idle 7 3 150 idle Load 88 to charge 9

5 785 discharge Battery 777 to charge 11

5 582 idle Load 77 to charge 4

Table 1

Table 2 lists units that are in an up-regulation reserve. The action taken on a unit will put it in a state which will push the utility frequency upwards. At the same time, the unit leaves the up regulation reserve and enters the down regulation reserve, with which the utility frequency may be regulated downwards. For example, if Battery 222 is set to discharge, it will start producing energy to the grid. This will make the utility frequency to rise, and the Battery 222 will enter a down-regulation reserve, because when turned to idle or to charge status, it will shift the utility frequency downwards. A high priority of a battery in idle status means it is been deemed to be sufficiently charged, and may be set to discharge. A battery that is not connected to a grid as an energy producer, e.g. the battery of an electric car, can be set from the status charge to idle to produce an up-regulating effect. Correspondingly, a production plant "Production 5" switched from an idle status to a discharge status will increase the energy supply to the grid and thus shift the utility frequency upwards.

Priority Time in current Current

Action Rating

count state (s) state

(kW)

1 100 idle Battery 222 to discharge 5

1 50 charge Load 77 to idle 7,5

2 101 idle Production 5 to discharge 6

2 50 charge Load 22 to idle 3

2 35 idle Production 1 to discharge 4

4 350 charge Load 66 to idle 5

4 300 charge Battery 555 to idle 5

4 280 charge Load 33 to idle 11

5 250 charge Battery 888 to idle 2

5 120 charge Load66 to idle 9 Table 2

The function of the control algorithm running in the central control system is now described with reference to Fig. 5. In Fig.5, there are shown four stages of an inventive utility frequency regulation process, which is progressing in the direction of the horizontal arrows from left to right. The down-regulation reserve is marked DRR and the up-regulation reserve URR, respectively. 50a and 50b mark the start of a new cycle for the regulation control process, where 50a marks an initial size of the down regulation reserve DRR, corresponding to the size of the rectangle. Similarly 50b marks the initial size of the up regulation reserve.

At 5 la and 5 lb, the target sizes 51c and 5 Id illustrated by dashed lines are calculated for the down and up regulation reserves, respectively. The target sizes 51c and 5 Id of the regulation reserves are computed by the control algorithm out of a given target utility frequency and control curve at Figure 2, to create regulation reserves needed to achieve the target utility frequency F. In this example, the down regulation reserve 51a need to be increased 5 la -> 5 lc, and the up- regulation reserve 5 lb need to be reduced 5 lb -> 5 Id, respectively At 52a and 52b, the control algorithm transfers loads and other units between the up and down regulation reserves DRR and URR according to what has been described in connection with tables 1 and 2. Loads and units with the highest priority are first transferred, which means that some loads or units may also be transferred in the opposite direction from an optimization point of view. Let's assume that the target is to increase DRR with 20 kW and correspondingly to decrease the URR with 20 kW. First the priority 1 power units are transferred. According to table 1, transferring priority 1 loads and units means the up regulating reserve 52b is increased by 7 + 3 + 5 = 15 kW. According to table 2, the down regulation reserve 52a is increased by 5 + 7.5 = 12.5 kW. This is not enough, as the net effect is in fact an increase of the up regulation reserve with 15 - 12.5 = 2.5 kW. Thus 22.5 kW more DRR is now needed to meet the target. Looking in table 2 at power units having a priority 2, we will find that production 5, load 22 and production 1, all having a priority 2, represent together a DRR potential of 6+3+4 = 13 kW. The lacking power rating may here be found in priority 4 devices, e.g. load 66and battery 555 represent together a DRR potential of 5+5 = 10 kW. Switching the status of production 5, load 22, production 1, load 66 and battery 555 thus increases the DRR with 23 kW. This is 0,5 kW more than the target, but may be considered close enough for the present optimization cycle because not performing the final state change would have lead further from the target. As a result the target DRR and URR are achieved as in 53a and 53b.

It should be noted that the power units have also some local constraints in their operation. For example, a boiler may reach the maximum water temperature before its estimated time, and will not draw energy anymore as its own thermostat has switched it off. The same applies to a fully charged electric vehicle battery in which case the unit cannot operate as up or down regulation reserve. In these cases, if such device-specific status information is available to the grid, the control system 12 may substitute these units with new ones to join the frequency regulation fleet. It is also possible that in some locations all of the local units cannot be operating at the same time, due to bottlenecks in the electricity grid, and the control system has to take this into account.

In the following, the control and actions of various units are described by way of examples. Example 1 - Heating apparatus In this embodiment, the controlled load is an electric water heater (boiler, geyser). The heating is preferably performed during the night 23:00-7:00 when, in general, only little warm water is used and electricity is often cheap. In this case the time frame T given for heating the water to be at the target temperature in the morning is 8 hours. The control algorithm receives a forecast of the total time required to heat the water for this time period. The forecast can be calculated for example by dividing the total consumed energy by the water heater during the previous day with the nominal power rating of the load, e.g. 3 kW. This estimate can be adjusted for example with a weekday, a holiday or a temperature factor, if real-time information on the water temperature in the boiler can be transmitted to the control system.

After receiving the forecast, the control algorithm allocates the water heater to frequency regulation during wanted hours within the time frame T. The criteria for the schedule, in addition to the heating demand forecast, can be for example level of the compensation for the frequency regulation, the price of the electricity or any identified requirements of the power units.

This allocation is done in a similar manner for all of the controlled power units. Consequently, an aggregated fleet of devices performing frequency regulation (i.e. regulation fleet) is created for each time. This fleet, and the regulation capacity it provides, can be then sold to regulation markets.

At the beginning of each hour, the control algorithm may read which power units are part of the regulation fleet and starts to command these units according to the desired utility frequency and other priorities. The algorithm keeps track how much power the boiler still requires to satisfy its demand, and determines a priority count for them. It is also taken into account that the boiler does not change its state too often. If the boiler does not switch on despite the ON command, the algorithm may conclude that a thermostat of that boiler prevents additional heating. In this case, the control algorithm can replace the boiler with another to join the aggregation fleet and/or update the priority count of the load in such mean that it stays as long as possible in the idle state, for example to zero in the down regulation reserve.

Example 2 - Electric car batteries or accumulators.

In this embodiment, the controlled load is the battery pack of an electric car. Charging of the batteries may be controlled during weekdays from 19:00-7:00, when it is most likely the cars are parked at their charging post. Here the time frame T is 12 hours. During weekends, the time frame may be longer, and during vacation periods, when the car may be away long periods from its charging point, the load may not be taken into account in a frequency regulation scheme at all. Here it is assumed the charging post is a fixed installation e.g. at the home of the car owner.

The control algorithm may receive a measured and exact charge status of the batteries, and determine from that the total charge time needed to achieve a full charge at the end of the given time frame T. The charging time can be calculated from historical data, technical specifications (if available) concerning the particular battery pack, and also the outside temperature, which affects both the charge level attainable and the charging time. In some instances, the charging post may be provided with e.g. a selectable charging current, which then adds some flexibility and opportunities for further optimization of the charging event.

Having determined the required charging time, the control algorithm allocates the battery charger to frequency regulation and schedules the charging during the given time frame T. As in the previous example 1 , additional criteria for the schedule may exist, such as the price of the electricity (e.g. night vs. daytime pricing), change of the ambient temperature or the power rating caused by e.g. a cabin heater switched on in the car, which may change the requirements of the load.

At the beginning of the frequency regulation, the control algorithm reads which chargers are part of the regulation fleet and starts to command these loads to charge or to be idle according to the desired utility frequency and other priorities.

Example 3 - Mixed loads

A very likely scenario is that the controlled loads are of mixed constitution. In addition to the heaters and batteries already mentioned, suitable loads are at least those with at least a limited capacity of storing energy such as air conditioning units, ventilation units and heat pumps. Also smaller distributed power generating units, such as generators, wind and solar power plants can be controlled. They may not consume electric energy but their production may be controlled. Other distributed power unit can be batteries, these units can often be controlled in frequency regulation with three different status: charge, discharge and idle.

As different loads will have time frames T of different length and different charging

characteristics, the control algorithm need to read the information of all units and decide within the framework of its own frequency regulation time frame how to optimize the load control from a frequency regulation and unit requirement point of view. During mornings and evenings, when the consumers are using hot water and driving their cars, forming of frequency regulation fleets of such loads may not be possible. During daytime, some aggregation may again be possible, as electric cars are again parked at a secondary charging post at workplaces, and the boilers may need re-heating. Wind and especially solar energy are also available more at daytime, so when these power assets feed a significant amount of power into the grid, they can be aggregated and regulated as well.

It is to be understood that the embodiments of the invention disclosed are not limited to the particular structures, process steps, or materials disclosed herein, but are extended to equivalents thereof as would be recognized by those ordinarily skilled in the relevant arts. It should also be understood that terminology employed herein is used for the purpose of describing particular embodiments only and is not intended to be limiting.

Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment.

As used herein, a plurality of items, structural elements, compositional elements, and/or materials may be presented in a common list for convenience. However, these lists should be construed as though each member of the list is individually identified as a separate and unique member. Thus, no individual member of such list should be construed as a de facto equivalent of any other member of the same list solely based on their presentation in a common group without indications to the contrary. In addition, various embodiments and example of the present invention may be referred to herein along with alternatives for the various components thereof. It is understood that such embodiments, examples, and alternatives are not to be construed as de facto equivalents of one another, but are to be considered as separate and autonomous representations of the present invention.

Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided, such as examples of power ratings, priorities, etc., to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the invention.

While the forgoing examples are illustrative of the principles of the present invention in one or more particular applications, it will be apparent to those of ordinary skill in the art that numerous modifications in form, usage and details of implementation can be made without the exercise of inventive faculty, and without departing from the principles and concepts of the invention.

Accordingly, it is not intended that the invention be limited, except as by the claims set forth below.

Claims

1. A method for providing frequency regulation to a power system and for aggregating a fleet of power units by means of a central control unit, said units being connected to said power system and capable of consuming and/or producing electrical energy, the method comprising the steps of:

- in said central power system control unit at the beginning of a given frequency regulation period, reading:

- a utility frequency target value for said power system during said period, and

- from a data source functionally coupled to said control unit, at least one of the following: the current status of the power units, the power rating of each power unit, and/or an estimated or measured energy demand for each power unit;

- estimating the sizes of an up regulation reserve and a down regulation reserve required for the power system to reach said utility frequency target value;

- establishing a priority count for each power unit based at least on the amount of time a power unit has been in its present status, and/or its remaining power demand;

- monitoring said utility frequency of said power system during said frequency regulation period and in order to provide frequency regulation thereof transferring power units according to their priority from said down regulation reserve to said up regulation reserve if the utility frequency is rising, and power units from said up regulation reserve to said down regulation reserve, if the utility frequency is decreasing.

2. A method according to claim 1, wherein the determination of a priority count for a load-type power unit is based on parameters, including one or more of the following:

- the cumulative time the load has been connected to power from the beginning of said frequency regulation period; - the estimated or measured remaining energy demand for the power unit until the end of said frequency regulation period;

- power unit specific data and/or environmental factors.

3. A method according to claim 2, wherein the environmental factors are selected from one or more of the following: the outside temperature, calendar information, a weather forecast, the current price of electricity.

4. A method according to any of claims 1 - 3, wherein the priority count for a power unit not in response to a transfer command, is set to zero.

5. A method according to any of claims 1 - 4, wherein said frequency regulation has a time period such as 1 hour.

6. A method according to any of claims 1 - 5, wherein said transfer of power units from said down and up regulation reserves in response to a deviation of said utility frequency, is performed with predetermined time intervals, such as once a second.

7. A system for providing frequency regulation to a power system and for aggregation a fleet of power units, said units being connected to said power system and capable of consuming and/or producing electrical energy, the system comprising:

- a central control unit for said fleet of power units; - a data source functionally coupled to said control unit and adapted to collect and store information about the power units; said central control unit comprising a non-transitory computer readable medium having stored thereon a set of computer executable instructions for causing the data processing unit to carry out the steps of: - reading at the beginning of a given frequency regulation period a utility frequency target value for said power system during said period, and at least one of the following: the current status of the power units, the power rating of each power unit and/or an estimated or measured energy demand for each power unit;

- establishing a priority count for each power unit based at least on the amount of time a power unit has been in its present status, and/or its remaining power demand; and - monitoring said utility frequency of said power system during said frequency regulation period and in order to provide frequency regulation thereof, transferring according to their priority, power units from said down regulation reserve to said up regulation reserve if the utility frequency is rising, and power units from said up regulation reserve to said down regulation reserve, if the utility frequency is decreasing.

8. A system according to claim 7, wherein the control unit determines a priority count for a load- type power unit based on parameters, including one or more of the following:

- the cumulative time the load has been connected to power from the beginning of said frequency regulation period; - the time the power unit has been in its current state;

- the estimated or measured remaining energy demand for the power unit until the end of said frequency regulation period;

- load-specific data and/or environmental factors.

9. A system according to claims 8, wherein the control unit reads at least one environmental factor selected from one or more of the following: the outside temperature, calendar information, a weather forecast, the current price of electricity.

10. A system according to any of claims 7 - 9, wherein the priority count for a power unit not in response to a transfer command is set to zero.

11. A system according to any of claims 7 - 10, wherein said frequency regulation period is 1 hour.

12. A system according to any of claims 7 - 11, wherein said transfer of power units from said down and up regulation reserves in response to a deviation of said utility frequency, is performed with predetermined time intervals, such as once a second.