3. Working with Technologies

Important

The interface must be linked to the model before executing any of the steps in this section. See Linking the interface to your model file.

A technology in the model is a power-producing unit or a combination of such units or a transmission line with specific parameters such as, maximum capacity, capacity factor, CAPEX, FOM, fuel cost etc.

3.1. Technology Types in SPLAT

A technology can be site specific (a specific plant or transmission line with known parameters) or generic (a technology with generalised parameters). In high level (Level 0), all the technologies in SPLAT are classified into four main types:

1. Existing technologies: The technologies which are already in place as of the reference year.

2. Committed technologies: The technologies which are currently under construction, or whose implementation has been approved. The deployment of these technologies may be considered guaranteed in the model results.

3. Candidate technologies: The technologies which may or may not be commissioned in the long term and thus are part of the optimisation process. The characteristics of these projects (size, location, technology, etc.) must be consistent with existing infrastructure, price trends and energy resource potential. The model will select them if they show a net benefit to the modelled power system, which implies that building, operating and maintaining them should minimise the total cost of power system expansion.

The SPLAT Africa model used in the Continental Power System Master Plan (CMP) project makes used of energy model “supply regions” (MSR). MSR comprises of pre-loaded georeferenced solar PV, solar CSP, wind onshore and offshore supply options designed by IRENA through a systematic geospatial analysis explained in this paper. These variable renewable energy technologies are also considered as candidate technologies.

4. Generic technologies: These technologies are not site specific. They have generalised parameters and hence, they cannot be assigned specific name/description. Generic technologies appear in the model results when the candidate and other technologies are not sufficient to close the gap between supply and demand.

The generic technologies are normally set up in a way that they don’t come online in reference scenaerios. One of the main motivations to have generic technologies in the model is to make the model tech-agnostic, which in turn increases acceptance among stakeholders.

The technologies in SPLAT are further sub-classified into different levels as given in Technology naming in the SPLAT model. The sections below describe the ways to view, add and change technologies using the SPLAT Excel Interface.

3.2. Viewing technology inputs

1. In any of the technologysheets (except the tabs containing information of specific zones), enter the scenario to be queried in the cell Choose Scenario

2. Click on Refresh Sheet. The technologies and their parameters in the scenario will be shown in the same sheet. Only data for the country(s) loaded (in the Main sheet) will be displayed.

3. Refer to Defining some key technologies for steps to retrieve the information of renewable zones (Solar PV, Solar CSP, Onshore Wind, Offshore Wind).

3.3. Adding a technology

1. Refresh the sheet SpecificTech for the scenario selected.

2. Add new specific tech name and parameters in the table. Ensure the that technology code is unique and not repeated.

3. Click on Add New Techs. A technology is added with parameters from the underlying generic tech.

4. Refresh sheet to see the new technology added. Refresh other sheets before modifying relevant parameters.

Note

Battery and pump storage technologies need to be defined separately in Battery&PumpStorage sheet in SPLAT interface.

3.4. Renaming a technology

1. Enter the old and new technology names in RenameTechFacility and click on Rename Techs in List.

2. To confirm that the technology has been renamed, refresh the relevant tabs (GenericTech or SpecificTech) to see the updated names. Multiple technologies can be renamed.

3.5. Deleting a technology

If a technology is not relevant in the current model run but may be relevant in the future, one trick to deactivate the technology in the current model run is to specify the year after the end of model time horizon as the First Year. e.g., if the model time horizon ends at 2050 and we don’t want a technology to appear in the results, it’s First Year can be specified as 2051. This can be changed in the future if we want the technology to be the part of the model simulation.

In case we want to delete a technology, the procedure is as follows:

1. Enter the technology names in DeleteTechFacility and click on Delete Techs in List.

2. To confirm that the technology has been deleted, refresh the relevant tabs (GenericTech or SpecificTech) to see the update. Multiple technologies can be deleted.

3.6. Changing a technology

1. In any of the technologysheets (except the tabs containing information of specific zones), click on Refresh Sheet to get the data saved in the model for the scenario chosen.

2. Make changes to the technologies in the sheet.

3. Click on Update Model Data to update the model with the new data.

3.7. Defining some key technologies

3.7.1. Solar PV, CSP, onshore and offshore Wind

As part of the features of SPLAT starter modelling kits for users, the SPLAT models have been pre-loaded with special supply options (called ‘zones’) for four generation technologies: solar PV (??SOPCZ), concentrated solar power CSP, wind onshore (??WDOCZ), wind offshore (??WDOCZ). For CSP, given the fact that there can be many plant design possibilities with varying thermal storage and solar field size; only two specific plant-design based zones are available in SPLAT which are namely, the solar multiple 2 plant with 6 hour storage (??SOTNZ) and solar multiple 4 plant with 12 hour storage (??SOTSZ). Each zone name carries a suffix representing its id (e.g. 001, 002 …).

SPLAT zones have site specific cost and performance assumptions, derived from an exogenous GIS assisted analysis that identified thousands of georeferenced Model Supply Regions (MSRs) across Africa and clustered them into SPLAT zones (see MSR methodology paper for more details).

Technology wise, the SPLAT zones vary in count. For CSP and wind offshore, given the relatively less impact of site weather on the production profiles, only two zones are included per country. For solar PV and wind onshore, given their high future prospects in almost all African countries and high impact of site weather on the production profiles, higher number of zones are included in SPLAT models which vary across three country groups, described below.

Unlike other generation technologies, the per kW overnight capital costs assumed for zones, include an additional offset which represents following:

All MSR technologies except Wind offshore: the cost of typical grid tie infrastructure and the connecting road, required to connect each kW of capacity, with the existing transmission grid and existing road respectively

Wind offshore: the cost of typical grid tie infrastructure, involving offshore and onshore parts, required to connect each kW of capacity with the existing transmission grid

Additionally in wind onshore zones, the overnight cost assumption is set by subjecting this assumption adopted for generic wind onshore case, to a multiplication depending on the appropriate wind class per zone, estimated in the MSR analysis. The multiplication factors are set as 1x, 1.16x and 1.36x; for class-1, class-2 and class-3 wind respectively (classes as per NREL wind toolkit-2014).

The MSR analysis has also derived the following parameters required by SPLAT as input per zone:

Sheets: PVZones, WindZones, OffshoreWindZones, CSP6hrZones, CSP12hrZones

• Capacity potential MW (upper bound on total installed capacity i.e. bdi up c)

• Location info (Longitude & latitude)

• Overnight cost offsets

Sheet: TimeSlices

• Zone specific representative hourly profiles (the hourly values available in .tit file in each subregion model directory can be reviewed and re-aggregated to model time slices (load regions) if needed)

The representation of the above stated georeferenced zones, allow capturing of the following two important aspects in SPLAT models:

• Significant influence of the location and site specific weather on the cost and performance of these supply options respectively

• To optimize the selection of each zone capacity based on a more elaborate accounting of the complementarities of production patterns of these supply options with demand, dispatchable hydropower dam technologies, run of river generators and cross-border energy system resources

In rare cases, when refinements are necessary, the SPLAT user has the ability to modify the above stated zone parameters (see PVZones, WindZones, OffshoreWindZones, CSP6hrZones and CSP12hrZones & Defining Time). However, in normal use case, the user is required to just review/revise the ‘first year’ for the zones only (see GenericTech and SpecificTech).

VRE technologies can be defined in two ways - either as generic technologies or site-specific technologies. Below is an example for adding offshore wind, first as a generic technology, then as zones.

1. In the GenericTech tab, add technology “XXWDLCO00” (XX being country ID, for e.g. DZ) with tech description “Offshore generic tech”. Use add new tech button. The macro will update the underlying files and reload at the end.

2. Go to RenameTechFacility sheet. Change the newly added offshore techs to appropriate generic tech name i.e. XXWDOC00. The macro will update the underlying files and reload at the end.

3. Go to OffshoreWindZones sheet. Add new techs in each country. Click on Add New Techs. The macro will update the underlying files and reload at the end.

4. Locate the .tit file of the model and open as excel, it will ask you about delimit parameter. Select comma. The generic wind offshore and newly added offshore zones will have same profiles. Now, got to OffshoreWindZones sheet. Give address to the file that contains the profiles, in the section MSR data file. This will update the zone profiles in .tit file. Currently, the wind offshore generic tech has same profile as wind generic. But remember, wind onshore generic tech has been ousted from model by setting first year=2050

5. The updated profiles in the .tit file needs to be inserted in model files. Go to TimeSlices sheet, press Update Files. The profiles refer to the capacity factor in the case of solar and wind technologies. In case of hydro technologies, the profiles refer to normalized peak monthly river flow rates.

3.7.2. Hydro Dam

SPLAT CMP model characterizes the dam-based hydropower plants by accounting river and dam specific resource conditions. Their dispatch is optimized while synergizing with other renewable supply options (i.e. the solar photovoltaic, wind, concentrated solar power and run of river based hydropower) that are given fixed and exogenously determined generation profiles.

The dam hydropower plants are represented as a combination of three elements: river, dam and a generator. The river and the generator are inserted as ‘technology’ while the dam is inserted as a ‘storage’ (for details, see MESSAGE manual). As explained ahead, the SPLAT naming convention requires these three elements to carry a common name but different prefixes.

The river technology is characterized with the exogenously determined maximum monthly inflow assumption in MW units and a normalized monthly flow profile. Similarly, the dam storage is characterized with a maximum volume in MWyr units. These assumptions are mainly derived from IRENA’s AfREP Hydropower database. In rare cases, when refinements are necessary, the SPLAT user has the ability to modify these characteristics. However, in normal use case, the user is required to just review/revise the generator side characteristics only (see GenericTech and SpecificTech).

The SpecificTechHydroDams sheet manipulates the hydro dams in the model.

1. Click on Refresh Sheet button to extract the technologies that belong to the TechSetL2: Large Hydro Dams.

2. Create River Tech+Storage Constraint button adds a technology and a storage constraint for each dam.

A new dummy technology for each hydro station with Dam is added to model the river inflows to the dam. The naming convention of the dummy technology is XXRIDM_rivername, for example CMRIDM_LAGDO (using LAGDO as an example). The output is set to the existing dummy elc energy form.

A new storage constraint is added, example D_LAGDO with short name DXXX. The storage constraint is linked to CMRIDM_LAGDO with +1 coefficient, so each MWyr flow from CMRIDM_LAGDO increases the storage content by 1 MWyr.

The storage constraint is linked to CMHYDM_LAGDO with -1 coefficient (meaning that each MWyr flow from CMHYDM_LAGDO decreases the storage content by 1 MWyr). It would be possible in theory to do cascade modelling by linking the output of upstream plants to storage constraints downstream (rather than a river technology). The coefficients would have to be scaled by the relative “Energy per unit volume (MJ/m3)” of the upstream and downstream plants. This functionality will need a revisit as a new development task if there is a pressing need for it.

The user has to specify 2 parameters, whose values can be calculated in the right-most table and copy pasted.

3. Once this is done the user can click on Update Model Data:

The capacity is set to max flow (in MW, m3/s max flow scaled by design flow). The capacity is specified as a capacity limit on the River Technology (bdi) .

The storage constraint max volume is set to Max volume in MWyr as per table.

The user then has to add a time series in the csv file under the tech CMRIDM_LAGDO and Update Timeslices in the Timeslice sheet. The values in the csv file must be monthly average flow divided by “max flow” that was used to set the “River Capacity”, using the same max flow value regardless of the scenario. If the user wants to simulate different rainfall scenarios without a full time series, they could use plant factor to scale up or down the profile in the SpecificTech sheet. It is currently not possible to specify a different seasonal profile by scenario, but this feature is on the todo list for the near future.

3.7.3. Batteries and Pump Storage

SPLAT interface allows the user to characterize one battery technology per country. This technology represents a 4 hour grid connected storage resource, whose capacity is optimized. In the modelled energy system, the batteries would charge and discharge when it makes least cost sense. Their contribution to ReserveMarginConstraint is also allowed.

The inherent modelling of ‘storages’ in MESSAGE can appropriately represent the characteristics of hydro dams, which can store water resources for long durations up to seasonal scale. In contrast, the batteries can store only a few hours of charge which, in practice, can be retained up to few days at most. As a result, the inclusion of battery storage model in MESSAGE is not straight forward and required insertion of several elements and constraints. The user doesn’t have to deal with these elements and constraints in the normal use cases. These are briefly described and illustrated below just for context:

1. SPLAT model entails a main ‘technology’ (??ELST04) that represents battery and a ‘storage’ (SS_??ELST04) that represents the reservoir of charge connected with the main technology.

2. SPLAT model entails a proxy ‘technology’ (??ELPT04) that is constrained – via a constraint called PC_??ELST04 - to have the same installed MW as the main technology (??ELST04) and is linked with storage (SS_??ELST04) – via a constraint called PS_??ELST04 - to enforce a constant relationship between installed MW and the charge reservoir size (MWh). In simple words, this relationship can be described as ‘every MW battery installed would expand the charge reservoir size by 4 MWh’. This relationship is enforced by activating an exogenously determined capacity factor (CF) profile on the proxy technology (??ELPT04) using a formula given in the diagram. Keeping in view the shorter storage duration limits of grid batteries (vs hydropower dam), the CF value in the last time slice of every season is set to 0. This means that whatever charge that is left in the storage (SS_??ELST04) at the end of the season is discarded (because of PS_??ELST04 constaint), or in other words, the batteries cannot retain charge for long periods of seasonal scale.

3. Dummy technologies are inserted to complete the battery model. Dummy technology ensures that the main battery technology accounts the charge left in the reservoir in the end time slice (end of the day), by shifting it into the beginning time slice (beginning of the day). Separate dummy technology is required for each season. SPLAT naming convention sets the dummy technology name as ‘??ELDT04_??’, where the suffix preceded by underscore represents the season number. This means, that the count of dummy technologies will be equal to the count of seasons selected for the model run.

In SPLAT models, the pumped hydropower plant is represented using the same modelling concept as the battery technology. However, the user can insert multiple pumped hydropower plants and control their type (i.e. committed or candidate). Since, each of such technology requires insertion of several extra technologies as described above, usually, the user cannot insert more than 6 or 7 pumped hydropower technologies in any single country due to inherent MESSAGE software limitations. The way around for this is therefore to aggregate multiple pumped hydropower plants in one technology.

Batteries and pump storage technologies can be added and modified in the standard way through the SPLAT excel interface:

1. In Battery&PumpStorage sheet: create the technology with techname convention: xxELST?? for a battery (the suffix ?? should be set as storage size in hours e.g. 04) or xxELSTPS[site/group name] for pump storage (e.g. ZAELSTPSDrakensberg); where xx is the country code.

2. Reload Global.

3. In the same Battery&PumpStorage sheet click Refresh and then specify hours of storage and storage cycle efficiency.

4. In the SpecificTech sheets specify the other usual parameters first year, total capacity upper limit, lifetime, etc.

3.8. Technology naming in the SPLAT model

The naming convention of various technologies including technology set levels 1 and 2, and generic technologies are given in the table below. The “??” in the technology code in the front refers to the two-letter country code (alpha-2). And the * provides further information about the technology.

The naming of technology set level 1 follow the following conventions in the SPLAT model:

Level 1 Sets	Code
Large Hydro PP	??HY*,!??HYMI*
VRE	??WD*,??SOP*
Other RE	??HYMI*,??SOT*,??BMST*,??GO*
Fossil&Nuclear PP	??CO*,??DS*,??HF*,??NG*,??NUPW*,??ELBS00*
Electricity Net Imports	??EL00i00*
Hydrogen Plant	??HGNEL*
Electricity for Hydrogen	??ELHGT*
Non Hydro Storage	????ST*,!*B?ST*
Fuel Imports/Extraction	??????i00,??????x00,??????I00,??????X00,!??EL00i00
Dom.T&D	??EL00T00*,??EL00TD*
Demand	??EL00TU*
TxBorder	??ELEX*,??ELNC*,??ELNP*,??ELNG*
Electricity Trade Outside Region	??EL00i00*,!??EL00i00

The naming of technology set level 2 follow the following conventions in the SPLAT model:

Level 2 Sets	Code
Coal PP	??COSC*
Coal w CCS PP	??COCS*
Diesel PP	??DSRC*,??DSSC*,??DSCC*
Diesel gen.small PP	??DSRI*,??DSRB*,??DSRU*,??DSRR*
HFO PP	??HF*
Natural Gas PP	??NG*
Nuclear PP	??NUPW*
Pump Storage PP	??EPPT*
Pumping	??EPPP*
Large Hydro Dam PP	??HYDM*
Large Hydro ROR PP	??HYRO*
Small Hydro PP	??HYMI*
Solar Thermal PP	??SOT*
Solar PV - Utility PP	??SOPC*
Solar PV - Rooftop PP	??SOPB*,??SOPR*,??SOPU*,??SOPS*
Geothermal PP	??GO*
Wind PP	??WD*
Biomass PP	??BMST*,??BWST*
Transmission	??EL00T00
Distribution	??EL00TD*
Demand	??EL00TU*
Coal Fuel	??CO00*
Diesel Fuel	??DS00*
HFO Fuel	??HF00*
Natural Gas Fuel	??NG00*
Nuclear Fuel	??NU00*
Biomass Fuel	??BM00*
Electricity for Hydrogen	??ELHGT*
Hydrogen Production	??HGNEL*
Electricity Net Imports	??EL00i00
Electricity Trade Outside Region	??EL00i00*,!??EL00i00
Hydrogen Imports/Exports	??HG00i00*
River Techs	??RIDM*
BatteryStorage	??ELST*
PumpStorage	??ELSTPS*
ProxyTech for Battery Storage	??ELPT*
DummyTech for Battery Storage	??ELDT*
Backstop	??ELBS00*
RE Curtailment	??RECURTAIL
RE Supply	??WD00*,??HY00*,??SO00*
TxBorderNonGeneric	??ELEX*,??ELNC*,??ELNP*
TxBorderGeneric	??ELNG*

The naming convention of generic technologies is given in the table below:

Generic Technology Codes	Code
Biomass Bagasse CogenBagasseTransmission	??BMST00*
Biomass Bagasse CogenBagasseTransmission	??BMST00*
Biomass Wood CogenWood residueTransmission	??BWST00*
GeothermalNATransmission	??GOCV00*
Hydro DamNATransmission	??HYDM00*
Hydro SmallNADistribution	??HYMI00*
Hydro Run of RiverNATransmission	??HYRO00*
Solar PVNATransmission	??SOPC00*
Solar CSP no StorageNATransmission	??SOTN00*
Solar CSP with StorageNATransmission	??SOTS00*
WindNATransmission	??WDLC00*
CoalCoalTransmission	??COSC00*
Coal w CCSCoalTransmission	??COCS00*
Diesel 100 kW system (industry)DieselOff-grid	??DSRI00*
Diesel EngineDieselTransmission	??DSRC00*
Diesel Open CycleDieselTransmission	??DSSC00*
Diesel/Gasoline 1kW system (commercial buildings)DieselOff-grid	??DSRB00*
Diesel/Gasoline 1kW system (urban)DieselOff-grid	??DSRU00*
Diesel/Gasoline 1kW system (rural)DieselOff-grid	??DSRR00*
Gas Combined CycleNatural GasTransmission	??NGCC00*
Gas Open CycleNatural GasTransmission	??NGSC00*
Gas EngineNatural GasTransmission	??NGRC00*
HFO EngineHFOTransmission	??HFRC00*
BagasseDomestic	??BM00X00*
Wood residueDomestic	??BW00X00*
CoalImport	??CO00I00*
DieselImport	??DS00I00*
HFOImport	??HF00I00*
Natural Gas (Domestic)Domestic	??NG00X00*

3.9. Country Code in the SPLAT model

The two-letter or three-letter country codes used in the SPLAT model are based on ISO 3166-1 standard.