Scrutinizing the Intermittency of Renewable Energy in a Long-Term Planning Model via Combining Direct Integration and Soft-Linking Methods for Colombia’s Power System
Round 1
Reviewer 1 Report
This paper addresses a very interesting and relevant topic and is well written. The topic of the paper is very important for researchers dealing with power sector development all over the world, given the obvious trend towards an increase of the share of renewable energy sources in the future. It’s obvious that conventional long-term energy system optimization models need to be upgraded to adequately model main feature of variable renewable sources – their intermittency. This could be done either by improvement of temporal and spatial resolution of the long-term model itself or by coupling with stand-alone power dispatch model. In this paper authors suggest combining both approaches which looks promising and is in line with current research trends.
In my opinion, overall the paper is good, but there are some comments/suggestion from my side.
Please consider adding a short explanation why simulation power system model (PowerPlan) was used. In my opinion it could be beneficial to use simplified optimization dispatch model like IRENA’s Flextool to take into account load following constraints and cycling costs.
It’s not clear for me was the capacity margin constraint used in TIMES model and what was a capacity credit (value) for VRES technologies. In my experience introducing such a constraint could notably increase adequacy of chosen power mix taking into account the need for proper reserving intermitting VRES.
Please provide an explanation for this statement at page 18, lines 492-495: “The projected supply of solar and wind power by TIMES–CO–BBE is overestimated by 41% and 10%, respectively, relative to PowerPlan Colombia because of the shortcoming of aggregated time slices compared to the hourly simulation of VRES supply and electricity demand”. Does it mean that so much of the annual VRES generation should be curtailed according to PowerPlan simulations?
Please consider adding a figure with a simulated load schedule for a day of a maximum VRES curtailment. In my opinion, this will significantly improve the analysis as well as reader’s understanding of why so much VRES generation is curtailed.
Please explain why you didn’t include storage technologies (batteries or pumped hydro) as an alternative for fossil-based generation in PowerPlan simulations. Most probably it would allow notably reducing VRES curtailment and limiting usage of fossil fuels and CO2 emissions.
Minor comments:
1) Pages 4 and 15. Text formatting is broken.
2) Figures 1 and 5. Font size is too small and hard to read in some parts of these figures.
3) Figure 1. I can’t see capacities per technology type and electricity demand as main inputs for PowerPlan model. I believe they should be there.
4) Page 8, lines 200-202. The statement “However, its technical detail is higher…” sounds like in TIMES technical details are set at the power plant level. Please rephrase.
5) Page 10, lines 287-290. Seems these lines don’t belong here or some lines are missing.
Author Response
Dear reviewer, thank you very much for your feedback on our manuscript and your comments. We addressed them to the best of our ability and revised our manuscript as follows. We hope to have responded to all your comments sufficiently. Please note that all page and line numbers mentioned below are based on viewing the document in Track Changes All Markup mode.
Response 1: An explanation has been provided in page 17 line 256-269: “Compared to operational optimization dispatch power system models … “.
Response 2: Thank you for the insights. We introduced the following peak constraints: 100%, 40% and 10% for thermal flexible technologies, wind power and solar PV, respectively. The constraints were not a limitation to the optimal solution, where the contribution of VRES to the peak has not exceeded 18%. We amended the contribution of VRES technologies to the peak in the scenarios with highest VRES penetration in Appendix D.1 (extended results). We referred to these additional results in the main manuscript in page 33 lines 566-567: “More details on the capacity margin for VRES is reported in Appendix D.1.”.
Response 3: Thank you for raising this point. In our analysis the role of curtailment was rather limited and did not exceed 250 GWh per year in any of the scenarios. The above-mentioned statement can be mainly explained by factors considered in PowerPlan model to enhance the realism of the system, namely the degeneration of the efficiency of solar panels and to some extent unplanned outage. To support this explanation, we simulated two additional scenarios (the BioLo dry scenarios with conventional D12 and alternative V36 time slicing techniques) while overriding the realism assumptions. We found that ignoring these factors can reduce the gap in solar power generation between TIMES and PowerPlan model to 10%. Under such conditions, the dispatch of fossil power is projected to be reduced, from 8-10% to 6% of generated electricity, and the need for additional flexible generation capacity is marginally reduced from 16 GW to 14-14.25 GW. While ignoring these realism assumptions can slightly reduce the total cost and emissions of a reliable system, we conclude that the main results of the analysis and conclusions over the approach remain robust.
Please note that we addressed this point as follows: we moved the position of the statement “The projected supply of solar ….” to page 38 line 655-665 (after introducing to the reader the outcome of the mismatch analysis). Then we provided the explanation mentioned above. We also referred the reader to Appendix D.3 where we added a visual comparison between the results with and without using the realism assumptions. We hope to have responded to this comment sufficiently.
Response 4: Thanks for your suggestion. As mentioned in the response to point 3, the level of curtailment in the simulations is relatively low (in the order of 250 GWh per year), which can be explained by the flexibility of projected technologies in the mix (hydropower dams and BIGCC-CCS technologies). On the other hand, large shortage is observed during peak load, in the absence of additional flexible generation capacity. We added a new figure (Figure 9) to show this dynamic. We referred to this figure in page 38 line 647-654: “Figure 9 shows the hourly simulation of the power mix from a sample week …”.
Response 5: Please note that the level of curtailment in PowerPlan simulations was too low to consider storage technologies as a viable solution. The low level of curtailment can be explained by the large expansion of low carbon flexible power generation capacity, namely hydropower reservoirs and BIGCC-CCS technologies. Such capacity expansion is an output of the TIMES-CO-BBE model. The projected capacity expansion in the scenarios corresponded to reserve factors between 2.2 and 2.7 (after correction for mismatch in system reliability). We highlighted this explanation in the discussion section in page 49 line 832-840: “Note that the power mixes projected by TIMES-CO-BBE .. ”.
Response 1 (minor): Regarding page 4 (page 10 in the revised manuscript), a similar comment was also raised by reviewer 2 (point 3). To address this issue, bullet points were removed to align the list of sections with the text. Considering page 15, we noticed a paragraph in blue ink in the version formatted by the journal. However, this issue is not the case in the version used for responding to the comments. We kindly ask the production team to help us address this issue.
Response 2 (minor): Figures 1 and 5 have been updated with enlarged font.
Response 3 (minor): Figure 1 has been adjusted to clarify the abovementioned information.
Response 4 (minor): Within the context of responding to major point 1, this statement has been rephrased.
Response 5 (minor): We rephrased the sentence and added more information to elaborate the meaning of this part. (Page 22 line 375-379: “While reference [15] considered two Representative Concentration Pathway (RCP) scenarios (4.5 and 8.5), here we only focused on the case where efforts are exerted to stabilize radiative forcing at 4.5 W/m2.”.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear authors, interesting article, but there are parts to be corrected or added:
1) The title of the paper is too long. I recommend that you reduce it
2) Standardize the font and size in the paper. Furthermore, in the text, avoid highlighting some words or abbreviations in bold type.
3) Paragraph 2: Align the list of sections with the text
4) Figure 1: Describe in the paper how figure 1 should be read. There are several inputs and it is not clear how the "modeling framework" works. Summarize, citing other references if necessary.
5) A description is missing which goes from figure 1 to paragraph 2.1. To describe well, it is not very clear
6) Figure 2: a description of the abbreviations in Figure 2 is missing from the text of the paper. To be added
Best Regards
Author Response
Dear reviewer, thank you very much for your comments and suggestions. We addressed them in the manuscript as follows. Please note that all page and line numbers mentioned below are based on viewing the document in Track Changes All Markup mode.
Response 1: A reduced title is suggested accordingly.
Response 2: Font and size were standardized throughout the manuscript. Bold formatting of words and abbreviations was removed.
Response 3: A similar comment was also raised by reviewer 1 (Minor comment 1). To address both comments, bullet points were removed to align the list of sections with the text.
Response 4: Figure 1 has been adjusted and divided into three parts, where each part is linked to the corresponding paragraph in the Methodology section.
Response 5: Paragraph 2.1 has been adjusted to better describe the different parts of Figure 1 and how they link to the corresponding paragraphs in the Methodology section. We hope to have clarified this sufficiently.
Response 6: A description of the abbreviations was added to the figure caption.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Dear authors, thank you very much for responding to all my comments.
Good luck in your further research
