Uncontrolled appliances are appliances with no energy flexibility. The appliance must execute its functionality directly after receiving a trigger to start. Typical examples are brown goods and renewable energy generation units.
Examples Appliances: PV, TV, Computer, Wind Turbine
This section contains a description for each data element that parameterises the uncontrolled EFI. The first three subsections contain a UML representation and a parameter description of the three messages: registration, update and allocation message.
Finally two examples on how to use the uncontrolled EFI are provided.
The following figure represents the data elements of an uncontrolled control sp ace registration message. Table 17 reveals the description of the used parameters.
The UncontrolledRegistration object registers the appliance driver to the energy app, the message is describes the commodities that are consumed or produced by the uncontrolled appliance with the Commodity attribute. Furthermore if the modelled appliance has features that allow curtailing the consumption or production of the device, the curtail options can be expressed in a ConstraintList for every commodity.
| Attribute | Description |
|---|---|
supportedCommodityCurtailments |
A map of every applicable Commodity for the appliance as key and a ConstriantList representing the list of possible curtail steps as an value. The value of the map is optional and will only be provided if the appliance supports curtailing, otherwise it must be null. |
Table 14: Attribute Descriptions of the UncontrolledRegistration class
This is a utility class that efficiently stores values for each type of Commodity. This map has less overhead than the HashMap or TreeMap implementations, because it just stores one reference for each commodity type.
It is parameterized so that it can contain any class as a value.
| Attribute | Description |
|---|---|
gasValue |
This attribute allows direct access to the value that corresponds with the gas commodity key. |
electricityValue |
This attribute allows direct access to the value that corresponds with the electricity commodity key. |
heatValue |
This attribute allows direct access to the value that corresponds with the heat commodity key. |
entrySet |
This attributes holds the complete set of Entry objects. An Entry has a commodity as a key and can hold any value type. |
Table 15: Attribute Descriptions of the CommodityMap class
This class represents a single entry in the CommodityMap<V>. It has a commodity as a key and can hold any value type.
| Attribute | Description |
|---|---|
commodity |
This attribute holds the commodity that this Entry holds a value for. |
value |
The value can be of any type. |
Table 16: Attribute Descriptions of the Entry class
This class is derived from CommodityMap<V>. In the case of a ConstraintListMap the CommodityMap is parameterized with a ConstrainList<Q>. This is not directly visible in the figure, but is ensured via builder classes.
In the UncontrolledRegistration object the ConstraintList describes the curtail levels supported by the modelled appliance. For example, if the power consumption of an air-conditioning unit can be curtailed to 1500 Watt, 750 Watt and 0 Watt. The ConstaintList contains those three values.
A Constraint describes a value or a range of values from a certain Quantity (i.e. Watt). If it represents a single value then LowerBound and UpperBound will be equal.
| Attribute | Description |
|---|---|
LowerBound |
The lower bound of the range or the value. |
UpperBound |
The upper bound of the range or the value. |
Table 17: Attribute Descriptions of the Constraint class**
The uncontrolled control space measurement update message is used for updating the consumption/production information of an appliance form the appliance driver to the energy app. Since the energy behaviour of an uncontrolled appliance can be very unpredictable (for instance the power output of a PV inverter), this measurement update message only contains a measured value and does not provide any information on how long this value is valid. Typically the appliance driver must send a new measurement update message to the energy app whenever the value of the measurement is updated.
| Attribute | Description |
|---|---|
measurable |
This attribute is of the type CommodityMeasurables. It contains the actual measurements for an uncontrolled appliance. |
Table 18: Attribute descriptions of UncontrolledMeasurement class.
The following figure represents the data elements of an uncontrolled control space update forecast message. This message is optional and can only be used if there is a mechanism in the appliance driver that can calculate a consumption/production forecast.
Marker class for updates for uncontrolled appliances.
The UncontrolledForecast class contains a consumption or production forecast. This information is represented by an instance of CommodityForecast.
The CommodityForecast class does not specify a startime for the forecast profile. For this purpose the ControlSpaceUpdate.validFrom is used, it marks the start time of the CommodityForecast profile.
| Attribute | Description |
|---|---|
| forecast | This attribute is of the type CommodityForecast and contains the actual forecast profile. |
Table 19: Attribute descriptions of UncontrolledUpdate class.
The following figure represents the data elements of an uncontrolled allocation message. An allocation will only be send to appliances which support curtailing production/consumption.
After receiving an UncontrolledAllocation from an energy app the appliance driver should follow one or more curtailmentProfiles (each commodity may have its own curtailment profile). This message contains these profiles and their startTime.
| Attribute | Description |
|---|---|
startTime |
The time at which the curtailmentProfiles should start. |
curtailmentProfiles |
This attribute is of the ConstraintProfileMap type. For every applicable commodity it contains a ConstraintProfile. |
Table 20: Attribute descriptions of UncontrolledAllocation class.
As the name of this class implies it is a map of ConstraintProfile<Q> objects which uses Commodity<BQ,FQ> as a key.
| Method | Description |
|---|---|
get(Commodity<BQ,FQ>): ConstraintProfile<Q> |
This method returns a ConstraintProfile<Q> for a particular commodity. The Commodity<BQ,FQ> of interest is given as an argument. |
Table 21: Method descriptions of ConstraintProfile class.
The ConstraintProfile<Q> is extended from the Profile class. It is parameterized with a quantity Q, which is the flow quantity (FQ) from the commodity that needs to be curtailed.
The ConstraintProfile<Q>.value is of the type Constraint<Q>. This is the same flow quantity as described above. Although this is not visible in the UML diagram, it is enforced in the builder classes that are used to construct a ConstraintProfile<Q>.
The lower - and upper bound attributes of Constraint<Q> represent the curtailment levels. If for instance a PV panel needs to be curtailed at 2000 Watt, the lower bound would be -2000 and the upper bound 0 (production is interpreted as negative consumption).
ConstraintProfile<Q> has no additional attributes.
This section provides two examples on how to use the uncontrolled EFI. In the first example the use of an uncontrolled EFI is explained when representing a desktop computer. In the second example the use of the uncontrolled EFI is explained when representing a PV inverter with grid emergency shutoff features.
The standard way in which a desktop computer operates offers no energy flexibility. The appliance must be supplied with power whenever the end-user wishes to use the unit. It has no method of storing energy once the electricity supply has been switched off. As such it can be modelled as an uncontrolled appliance.
The control space registration message which commodities are consumed by the appliance and if it is able to curtail it power consumption. For a desktop computer the Commodity will typically be Electricity and it won’t support features to curtail its power consumption, therefore the ConstraintList will be empty.
The control space update consists of a CommodityForecast which for a desktop computer describes the amount of electricity consumption and could, for example, take a constant value over time. This will have a unique resourceMessageId associated with it.
Typically a desktop computer runs uncontrolled, which means it is not controlled by a smart grid service and the user is able to switch it on and off at any time. Therefore the desktop computer does not receive allocations.
Photovoltaics (PV) are used to generate electricity directly from solar radiation. The energy generated by these devices is uncontrolled as the level of generation is variable. It is however possible to estimate the level of generation in the future to produce a forecast (see below).
The control space registration message which commodities are consumed by the appliance and if it is able to curtail it power consumption. For a PV Inverter the Commodity will typically be Electricity. In this example it supports features to curtail its power consumption, therefore the ConstraintList will contain the values 1500, 1000, 500 and 0 Watt.
The control space consists of a UncontrolledMeasurement which for a PV inverter describes the amount of electricity produced at the time of creation of the control space. This will have a unique resourceMessageId associated with it.
Typically a PV inverter will run uncontrolled and produce as much energy as possible, however if the inverter supports remote dimming this feature can be used by sending an allocation to the device.
In a grid emergency situation, it is possible that the energy service desires the PV inverter to reduce power output. An allocation would be sent, which would request the appliance to change its energy production. The allocation consists of a resourceMessageId which matches the control space id (generated in 1) and curtailmentProfiles that should be followed by the PV inverter.
The application driver of the PV inverter has the ability to generate a production forecast based on weather forecast. The forecast is generated once every day once at 6 AM. In order to send the forecast to an energy app the Uncontrolled Forecast Update is used (Section 3.1.3). The forecast itself can be defined in the UncontrolledForecast class (Section 3.1.3.2). The forecasted profile created at 6 AM is depicted in Figure 16. It contains a start time (validFrom in ControlSpaceUpdate class) and a number of bins CommodityForecast.Element with a duration (duration in CommodityForecast.Element) and parameterized with the CommodityUncertainMeasurables class which is able to express uncertainty.
