Linkage of TIMES with Power Dispatch Models and Network Optimization
0 likes479 views
The document discusses linking the TIMES energy systems model with other models to provide more detailed spatial and temporal resolution and a holistic perspective of the energy system. Some challenges in linking models include different granularities, ensuring convergence, and managing computational time. The document presents examples of linking TIMES with power dispatch models and network optimization tools to better model high shares of variable renewable energy, district heating, and resource transportation costs. Developing a framework for properly linking models is important to avoid suboptimal solutions and make use of each model's strengths.
Linkage of TIMES with Power Dispatch Models and Network Optimization
- 1. Linkage of TIMES with power dispatch models and network optimization tools ETSAP Meeting, Zürich December 2017 PhD Student Amalia Pizarro Alonso aroal@dtu.dk
- 2. 2 Our motivation for linking TIMES with other models • Integration of high shares of Variable Renewable • Modelling of District Heating • Maintenance of Chronology • Modelling of Resources Transportation – i.e. endogenous fuel cost More detailed Spatial and Temporal Resolution, even for long-term energy planning
- 3. 3 Our Challenges when linking models The interaction is done by using the same common exogenous parameters and linking the common endogenous variables TIMES MODEL Input Parameters Input Parameters Common Input Parameters RESULTS RESULTS Convergence System Boundaries Common Endogenous Variables
- 4. 4 Our Challenges when linking models The interaction is done by using the same common exogenous parameters and linking the common endogenous variables TIMES MODEL Input Parameters Input Parameters Common Input Parameters RESULTS RESULTS Convergence System Boundaries Common Endogenous Variables What happens when the spatial and temporal granularity are different? How do we ensure we have achieved convergence? Keep computational time low!!
- 5. Linkage of TIMES with the power dispatch model Balmorel Analysis of the implications of long-term decarbonization goals in the Mexican power sector In collaboration with, Baltazar Solano Rodriguez (UCL)
- 6. 6 What is Balmorel? A brief introduction Simultaneous optimization of investments and power dispatch, including power transmission: Flexible spatial and temporal resolution, including the possibility to model chronology. Flexible degree of foresight within a year and between years.
- 7. 7 Other links of Times & Balmorel Nordic Energy Technology Perspectives 2016 – IEA: Chapter 3. Flexibility in the system and VRE integration In the Mexican Study, there is a different integration approach
- 8. 8 Analysing low decarbonization scenarios in Mexico TIMES Optimization of the full integrated energy system, to achieve desired GHG targets. Which level of electrification is required? How many GHG emissions are allocated to power generation? (or fossil fuel consumption) 16 time slices and 5 regions 5 sectors: agriculture, industries, services, residential and transport Full foresight until 2050
- 9. 9 Analysing low decarbonization scenarios in Mexico TIMES Optimization of the full integrated energy system, to achieve desired GHG targets. Which level of electrification is required? How many GHG emissions (fossil fuel) are allocated to power generation? Balmorel What is the actual potential for integration of high shares of VRE? What are the flexibility requirements? Optimal location of plants and investments in transmission lines. 16 time slices and 5 regions 5 sectors: agriculture, industries, services, residential and transport Full foresight until 2050 288 time slices and 53 regions Only modelling of electricity demands: classic and transport Full foresight until 2050
- 10. 10 Natural Gas Allocated for Power Generation - TIMES 0 200 400 600 800 1000 1200 1400 2025 2030 2035 2040 2045 2050 NaturalGasConsumption allocatedtoElectricity (PJ/year) Low Ambition - TIMES Medium Ambition - TIMES High Ambition - TIMES
- 11. 11 Natural Gas Allocated for Power Generation - TIMES 0 200 400 600 800 1000 1200 1400 2025 2030 2035 2040 2045 2050 NaturalGasConsumption allocatedtoElectricity (PJ/year) Low Ambition Low Ambition - TIMES Medium Ambition Medium Ambition - TIMES High Ambition High Ambition - TIMES
- 12. 12 Natural Gas Allocated for Power Generation - TIMES 0 200 400 600 800 1000 1200 1400 2025 2030 2035 2040 2045 2050 NaturalGasConsumption allocatedtoElectricity (PJ/year) Low Ambition Low Ambition - TIMES Medium Ambition Medium Ambition - TIMES High Ambition High Ambition - TIMES With 16 time slices, 5 regions and the use of the peak equation in TIMES, the cost- efficient potential for RE integration was underestimated. Re-adjustment of the time slices and/or peak equation and RE potentials needed
- 13. 13 Accumulated Investments in VRE Energy by 2050 RE potentials in TIMES should be re-adjusted, considering where it is more likely plants will be located: High RE potential vs. electricity demand vs. transmission capacity Solar PV Wind
- 14. 14 Flexibility Requirements by 2050 in the Mexican power system How to avoid the use of CCS technologies to achieve the toughest decarbonization target by 2050?
- 15. 15 Further Work 16 time slices and 5 regions 5 sectors: agriculture, industries, services, residential and transport Full foresight until 2050 288 time slices and 53 regions Only modelling of electricity demands: classic and transport Full foresight until 2050 Is the convergence of models really needed? The Nordic ETP uses a myopic approach, the use of rolling horizons might be a preferred option, to include certain degree of uncertainty towards the future and to minimize considerably the computational time
- 16. Linkage of TIMES with the Network Optimization model OptiFlow and the power model Balmorel Maximizing the socioeconomic value of straw in the Danish Energy System
- 17. 17 Scattered availability of biomass resources The role of Biomass in the future energy system Biomass is a valuable resource in the future energy system, with competing uses: • Back-up power generation • Heat • Transport Their gate cost, will depend on the distance transported: Higher Transport vs. Economy of Scale Optimal location and size of biomass plants?
- 18. 18 Integration of biorefineries in the energy system Biorefineries might provide significant amount of excess heat; however they benefit from economy of scale and they could provide it in a District Heating grid which cannot absorb it. Competition of base-load district heating: • Waste incineration plants • Industrial Excess Heat • Excess Heat from Biorefineries Inter-seasonal heat storages (e.g. water pits) might allow to maximize the value from the excess heat, if there is land available to construct it.
- 19. 19 What is OptiFlow? Process 1 Process 2 Process 3 Flow 1 Flow 2 … … Data-driven network model that consists in Processess (nodes) and Flows (arcs). … … Process 1 Process 2 Process 3 Flow 1 Flow 23 Process 4 Flow 24
- 20. 20 What is OptiFlow? Data-driven network model that consists in Processess (nodes) and Flows (arcs). Model fully hard-linked with Balmorel
- 21. 21 Linkage of TIMES with OptiFlow & Balmorel TIMES-DK Sectors: supply, power & heat, residential, industry, transport Time: 32 time slices per year Geography: 2 regions, 3 DH areas OptiFlow & Balmorel Sectors: Supply, Power & District Heating and Hydrogen Time: hourly time slices per representative week of each season Geography: 10x10km2 areas for resource transport and 34 DH Areas INPUT DATA End-use demands: electricity, heat & transport Techno-economic data for plants Emission coefficients Capacity of existing plants Biomass availability Fuel prices MODEL OUTPUT • Optimal Use of Biomass • Optimal location and sizing of Biorefineries • Etc. Allocation of Biomass use o Transport cost of biomass for each type of plant o Amount of excess heat for DH that can be potentially recovered o Electricity Trade INPUT DATA Electricity, District Heating and Hydrogen Demand Techno-economic data for plants Emission coefficients Capacity of existing plants Biomass availability Fuel prices
- 22. 22 Conclusions • A more detailed spatial and temporal representation might be desired to represent challenges in representation of high shares of VRE and integrating energy systems. • Linkage of TIMES with different modeling tools allows to provide a holistic perspective of the energy system, but keeping the computational time low. In addition, TIMES dataset and characterization might be improved. • A proper framework for linking TIMES with other tools is needed, to ensure there are not sub-optimal scenarios, being aware of strengths and limitations of each approach.
- 23. Thank you for your attention Amalia Pizarro Alonso aroal@dtu.dk
- 24. A modelling framework to link a TIMES model with detailed upstream, power and macroeconomic models A case study for Mexico Baltazar Solano Rodriguez (UCL) Amalia Pizarro Alonso (DTU) Cecilia Martín del Campo (UNAM)
- 25. 25 Designing a coherent framework for integrating energy models Identify and describe the role that the TIMES model should play in strategic long-term energy planning within a set of modelling tools. Develop a framework with best- practices for proper linkage of models. Show to the Mexican Ministry of Energy (and other governments in LAC) the institutional collaborations facilitated by ETSAP.