6 June 2018 Bogota, Colombia

1. Long-term planning and modelling with high shares of variable renewable power Focus on Latin America 6 June 2018 Bogota, Colombia

2. About IRENA International organization supported by governments of member countries • HQ: Abu Dhabi, UAE • Members: 180 countries (157 ratified) • Mandate: Sustainable deployment of six RE resources (Biomass, Geothermal, Hydro, Ocean, Solar, Wind) Programme of Work: • 1. Centre of Excellence for Energy Transformation • Transformation pathways Including planning for VRE • 2. Global Voice of Renewables • 3. Network Hub • 4. Source of Advice and Support

3. Agenda for Planning theGlobal Energy Transformation Agenda 1: Tap into the strong synergies between energy efficiency and renewables Agenda 2:Plan a power system with high shares of renewable energy Agenda 3: Increase the use of electricity in transport, buildings and industry Agenda 4: Scale up sustainable supply of bioenergy Agenda 5: Integrate SDG’s into energy planning

4. Agenda • Long-term planning and modelling with high shares of VRE • Focus on LATAM: IRENA 2017 regional planning workshop in Buenos Aires

5. Why focus on long-term planning with variable renewable energy (VRE)? • IRENA was frequently approached by member state representatives with questions on this topic, e.g. • How can we be sure our long-term modelling of solar and wind is accurate? • What models and methods are available to support this? • “Deploying variable renewables (VRE) is beneficial.” • “Our country should adopt ambitious long-term VRE targets.” • “VRE’s short-term variability endangers power system reliability” • “There is an upper limit of X% VRE” Government System operators Energy planning officials

6. IRENA’s “Addressing VRE in long-term planning (AVRIL)” project was started Key questions have been answered: • How does long-term generation expansion planning need to change when aiming for a high share of VRE? • Identify the planning impacts of VRE’s distinct features • What needs to change? • Institutional aspects (the planning process) • Techno-economic assessment methodologies (modelling) Based on expert inputs • IEW 2014, 2015 • AVRIL expert meeting • Expert consultation and interviews Part 1: Planning the transition to variable renewables Part 2: Long-term energy models for transition planning

7. What do we mean by long-term planning? Capacity expansion planning • Ministry of Energy • Planning agency • Utility Geo-spatial planning • Government planning office • Planning agency • Utility • TSO Dispatch simulation • Utility • Regulators • TSO Technical network studies • TSO • Regulator • Project developer

8. What types of models are commonly used? MESSAGE Quantum GIS MARKAL/TIMES ArcGIS NEPLAN IRENA FlexTool PLEXOS-LT PLEXOS-ST BALMOREL Power Factory Grid-View PSSE SDPP OPT-GEN WASP WASP GT-Max GT-MAX Cap. expansion Geo-spatial Dispatch Static grid Dynamic grid

9. What is the typical scope of long-term planning? • Planning time horizon • Long-term planning (planning ahead of 20-40 years) • Short-term planning (planning ahead of one year) • Real-time planning • Planning time frame • Years • Weeks • Snap shot • Planning time resolutions • Seasonal • Sub-daily • Hourly • Sub-hourly • Sub-seconds Lower detailWider scope Higher detailNarrower scope

10. What is the purpose of long-term planning? Examples from LATAM Colombia: Basis for policy making, establishing signals for investmentand capacity expansion needs Brazil: To be used as a basis for formulating public policies Uruguay: To design policies to support technologies to promote and investment needs Argentina: To establish a framework of discussion for the design of new policies and for the discussion with actors of the sector.

11. What needs to change? Unique issues for solar and wind representation • Rapid cost reductions • Firm capacity / capacity credit • System flexibility • Transmission investment needs • System stability impacts Good representation of VRE in long-term planning and modelling should reflect the consideration of: Typically not well- covered in “traditional” generation expansion planning models and methodologies

12. Some examples of modelling solutions • A cross-cutting solution • Increasing temporal and spatial resolution • Representing firm capacity • Better calibration of time slice using VRE data • Adding capacity credit constraints • Representing transmission capacity • Linking grid investment needs with VRE expansion • Site specific representation of generation and transmission needs • Representing flexibility • Incorporating constraints on flexibility provision (e.g. ramping, start up times, minimum load levels, partial load efficiency) • Validating flexibility balance in a system with external software • Coupling with production cost models • Representing stability constraints

13. The planning/modelling process also needs to become more dynamic Feedback from all the levels Long-term generation expansion models (time resolution: hours – seasons) High detail/ Narrow scope Firm capacity Geo-spatial planning models (time resolution: hours – seasons) Generation and network capacities New transmission Production cost models (time resolution: minutes – hours) Network topology Flexibility Static grid models (time resolution: single point) Highly-resolved dispatch and operational details Transmission enhancement Dynamic grid models Steady state grid currents and voltages Stability Low detail/ Wide scope Source: IRENA (2017), Planning for the Renewable Future: Long-term modelling and tools to expand variable renewable power in emerging economies

14. How can modeling incorporate this kind of feedback? Some examples • Creation of a “super-model” • Model coupling • Representation of the key elements in a simplified manner • Input data preparation (better temporal and spatial resolution RE generation data) • Constraints

15. Two methodological guides from IRENA contain more detail Long-term energy planning models (time resolution: hours – seasons) Defining long-term capacity mix and transmission infrastructure Geo-spatial planning models (time resolution: hours – seasons) Production cost models (time resolution: minutes – hours) Static grid models (time resolution: single point) Grid integration studies for a given capacity and transmission infrastructure Dynamic grid models (time resolution: milliseconds – minutes)

16. Agenda • Long-term planning and modelling with high shares of VRE • Focus on LATAM:IRENA 2017 regional planning workshop in Buenos Aires

17. 2017 LATAM Regional workshop on long-term planning with VRE • Co-organised by IRENA and Argentina’s Ministry of Energy and Mining in Buenos Aires • International and regional representatives from NREL, OLADE, and the World Bank • National planning representatives from ten Latin American countries – Argentina, Brazil, Bolivia, Chile, Colombia, Ecuador, Mexico, Paraguay, Peru, and Uruguay • Extensive national planning surveys completed beforehand Workshop objectives: • Exchange experience in best practice for long-term power system planning and modelling with high shares of VRE • Identify opportunities for improvement Workshop link: http://www.irena.org/events/2017/Aug/Incorporating-variable-renewables-into-long-term-energy-sector-planning-in-South-America

18. 2017 LATAM Regional workshop on long-term planning with VRE Opportunity for a follow-up technical workshop: • Participants confirmed there is a need for such dialogue focused on planning & modelling for VRE • Workshop summary report contains surveys andextensive list of working-level topics that countries in the region are eager and able to follow up on - the opportunity for collaboration is significant • The most cited areas for a potential follow-up workshop are: • System Flexibility • Capacity credit of VRE • RE data improvement: e.g. costs and resources • Topics beyond the modelling: e.g. grid codes and the remuneration of energy system services (e.g. flexibility and transmission investment) Workshop link: http://www.irena.org/events/2017/Aug/Incorporating-variable-renewables-into-long-term-energy-sector-planning-in-South-America

19. Some high-level conclusions • The paradigm in the region is changing – in 30 years the region’s power system/s are expected to be very different • Although all country contexts are different, concerns are similar "we all worry about the same things...it's good to know that planners in the region are not alone" • Modelling renewable costs:translating international cost projections into local contexts is a challenge • Modelling generation adequacy (i.e. capacity credit): there’s difficulty with hourly historical VRE generation and load data, uncertainty around future load profiles, and realistic depiction of VRE-hydro dynamics • Modelling flexibility:representing newer flexibility options such as storage and demand-side response is challenging • Modelling transmission:VRE projects can be spread to smooth variable production (e.g. wind generation in Argentina), but co-optimisation of generation and transmission planning remains a challenge • There is extensive global experience with power systems and VRE, and good practices from outside the region can be shared and adapted • International institutions such as IRENA should serve as a platform to identify specific needs, and countries that have already resolved them to facilitate exchange

20. Detailed insight: Areas identified for future improvement

21. Detailed insight: Areas identified for future improvement

22. AppendixAdditional results of 2017 technical workshop

23. Detailed insight: Tool coverage

24. The region is preparing for major changes in the power sector Some examples: • Argentina has a target to meet 20% of demand with renewable sources by 2025, up from 8% in 2017-18 • Brazil explores scenarios with increases of 120% in the installed capacity of non-conventional renewable sources like biomass, solar and wind in 10 years (from 29 to 63 GW) • Chile considers scenarios with shares of up to 90% renewable energy, even some with 75% VRE, in 30 years • In Colombia there are scenarios that include more than 2 GW of wind power in the next 15 years • Bolivia expects a transition from thermal generation to a system based on renewable energies • In Uruguay, the share of wind energy in total generation jumped from 1% to 27% between 2012 and 2016 (1.4 GW installed). In 2017 the share of renewable energy was 98%.

25. High-level conclusions: RE and VRE characteristics in the region • RE shares are currently high across the region, but that is mainly driven by hydropower, and there remain several countries with significant thermal capacity (e.g. Argentina, Bolivia, Chile, and Mexico) • Hydro is set to grow in some countries (e.g. Bolivia with large export plans, Ecuador, possibly Peru), but remain stable or decline elsewhere (e.g. Brazil due to environmental concerns) • VRE shares are currently low, with the exception of Uruguay, which has significant shares of wind generation, and certain regions within Brazil and Chile • Nevertheless, in many cases national plans to expand VRE capacity represent rapid increases relative to current shares, over a short period of time • The benefits of new VRE capacity heavily depend on complementarity between VRE generation, existing resources, and demand – e.g. in Brazil, wind generation complements hydro seasonality, and solar can complement biomass availability and an increase in summer air conditioning demand • Uncertainty around future resource/demand complementarity is a key challenge, e.g. in the same Brazilian context, GWs of diesel capacity – rather than solar – are being used to meet demand growth amidst declining hydro production, since peak demand shift from evening to summer daytime is not certain

26. Detailed insight: Regional representation of VRE investment costs

27. Detailed insight: Regional representation of generation adequacy

28. Detailed insight: Regional representation of system flexibility

29. Detailed insight: Regional representation of VRE siting and zoning

30. Detailed insight: Regional representation of stability-related operational constraint

31. AppendixFurther discussion: CEM campaign

32. A new CEM campaign • Long-term energy scenarios for the clean energy transition (LTES) • Launch: May 2018 at the 9th CEM meeting, Copenhagen • Duration: one year (possible extension to multiple years) • Lead countries: Denmark, Germany • Operating agent: IRENA • Goal: promote the wider adoption and improved use of long-term energy scenarios for clean energy transition Potentially joining campaign:

33. Context • The complex, interconnected, and long-term nature of the energy transition is posing new challenges for decision makers • A long-term vision is required to avoid risks of making poor, short-sighted decisions, e.g.: • Underestimating rapid renewable uptake, leading to stranded assets • Failing to utilize disruptive innovations • Long-term visions need to represent transformational changes, e.g.: • Very high shares of variable renewable energy in the power sector • Disruptive innovations in end-use sectors • Digitalization and its impacts

34. Three focus areas Use of scenarios for policy making Development of scenarios for clean energy transition Capacity building and enhancement • Share experience in the use of energy scenarios for national and regional policy planning • Identify ways to improve scenarios to make them relevant to private sector investment decisions • Share experience in building capacity within your country • Identify channels to build and enhance capacity in countries with limited experience • Showcase new tools & methods to address new, disruptive elements of the transition • Improve modelling of end-use sectors, sector coupling, and variable renewable energy integration • Identify research gaps

35. Campaign details • Proposed activities • Dedicated CEM campaign events, for best practice exchange among partners and the development of proposals to enhanced modelling capability (first major event tentatively scheduled March 2019) • Participation by IRENA and partners in other relevant events to both inform the campaign and to promote the best practice use of scenarios • Production of reports and recommendations, informed by events and by additional analytical work by IRENA and partners

36. Campaign details cont’d • Partner engagement State actors – officially designated by the national CEM PoC • Policy makers who are or wish to use scenarios in their strategic planning • Expert teams who support policy makers through modeling and scenario building. Non-state actors – technical institutions, IGO, and private companies • Technical institutions that are specialized in this field • Companies utilising scenarios in their business planning

37. Campaign details cont’d • Partner contributions Expected partner contributions • To attend one or more workshops • To provide best practice experience in the indicated areas of interests (use of scenarios, development of scenarios, and capacity building) • To contribute to discussions and debate to inform recommendations. Potential additional partner contributions • To host/offer support to organize a workshop • To represent the Campaign in other events • To support analytical work done jointly with IRENA

38. Campaign details cont’d • Benefits of engagement • Gain insights into how other countries and private-sector bodies are using scenarios for decision making and how to improve your own • Compare and contrast your own approaches with others and discuss strengths and weaknesses with fellow experts • Share details of how your country or organisation has built planning capacity and explore ways to share that expertise with those with less experience • Discuss your needs for improved modelling tools and develop shared recommendations that can influence further work and collaborations.