A Novel Framework for Co-Expansion Planning of Transmission Lines and Energy Storage Devices Considering Unit Commitment

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. Once an abbreviation is given, repeating its original version again and again is not needed.
2. In the introduction, while the previous studies have been discussed, only the coverage of the study is mentioned. Limitations about these studies are not discussed. This might help readers to better understand the contribution of authors’ work.
3. “Contribution” section should be rewritten? What is the main contribution(s)? Using GA cannot be a target. Using GA should be proposed to make some improvement. What are the new methods or techniques applied and what are the benefits obtained?
4. Are “Thermal generation units” important units to focus on while Transmission Network Expansion Planning? What is their ratio within the power system?
5. What are the variables used in Eqs. 6-35?
6. Is evaluating the operational cost of the system over a one-year horizon sufficient? This investment will be active for years.
7. Please provide some tangible results in Conclusion.

Author Response

1. Once an abbreviation is given, repeating its original version again and again is not needed.
Response: Thank you for the observation. We have revised the manuscript accordingly and removed redundant repetitions of previously defined abbreviations throughout the text.

2. In the introduction, while the previous studies have been discussed, only the coverage of the study is mentioned. Limitations about these studies are not discussed. This might help readers to better understand the contribution of authors’ work.
Response: Thank you for the comment. We would like to clarify that Table 1, already included in the manuscript, was specifically designed to help readers identify not only the scope but also the key limitations of the reviewed studies. Among the aspects we highlight as limitations are the absence of uncertainty modeling and the lack of unit commitment (UC) considerations.
For instance, only [5] explicitly incorporates a unit commitment model for thermal units. However, its formulation is overly simplified, relying on a single binary variable to represent generator status while neglecting crucial operational constraints such as minimum up/down times and startup/shutdown ramps. Our work addresses these gaps by including a more realistic UC model and modeling uncertainties through probabilistic representative days. [5]. Qiu, T.; Xu, B.; Wang, Y.; Dvorkin, Y.; Kirschen, D.S. Stochastic multistage coplanning of transmission expansion and energy storage. IEEE Transactions on Power Systems 2016, 32, 643–651. 482

3. “Contribution” section should be rewritten? What is the main contribution(s)? Using GA cannot be a target. Using GA should be proposed to make some improvement. What are the new methods or techniques applied and what are the benefits obtained?
Response: Thank you for this question. We have revised the “Contribution” section to make it clearer and more objective. The main contribution of our work lies in the formulation of a comprehensive co-planning model for transmission expansion and energy storage systems, considering realistic technical constraints and operational characteristics.

4. Are “Thermal generation units” important units to focus on while Transmission Network Expansion Planning? What is their ratio within the power system?
Response: Thank you for this question. Thermal generation units are crucial in Transmission Network Expansion Planning and must be considered to ensure realistic and reliable planning outcomes. These units provide essential operational flexibility, allowing the system to adjust generation in real time to compensate for fluctuations in load and intermittent renewable generation.
In practical systems, thermal generation often constitutes a substantial share of firm capacity. For example, in the Brazilian power system, thermal units represent about 25–30% of the firm energy supply, playing a vital role during peak demand and hydrological scarcity.

5. What are the variables used in Eqs. 6–35?
Response: Thank you for this question. To clarify this point, we have explicitly stated in the manuscript that all variables, indices, sets, and parameters used in the mathematical model are defined in the Nomenclature section included in the paper.
6. Is evaluating the operational cost of the system over a one-year horizon sufficient? This investment will be active for years.
Response: Thank you for this question. Evaluating the system's operational cost over a one-year horizon is a standard practice in the literature, as it allows planners to capture seasonal variability in generation and demand while keeping the problem tractable. Furthermore, investment costs for new assets (e.g., storage
units and transmission lines) are annualized using a discount rate to account for their multi-year lifespan. This annualization approach aligns with methods used in several studies.

7. Please provide some tangible results in Conclusion.
Response: Thank you for the suggestion. We have revised the Conclusion section to include tangible results from our simulations. Specifically, we now highlight: the percentage reduction in operational costs; the number and location of selected expansion assets; and the improvement in system reliability indicators.

Reviewer 2 Report

Comments and Suggestions for Authors

Overall the paper has good analysis. Here are few of the comments:

1. The abstract lacks specific details on results/methodology. A line can be added, regarding the simulation tools used for this proposed model.
2. Are there any challenges while using unit commitment constraints and uncertainties in load demand and wind generation?
3. What are the limitations for using MINLP and decomposition-based approach?
4. Figures 3 and 4 are not clear
5.  The future work is not discussed.

Author Response

Overall the paper has good analysis. Here are few of the comments:
1. The abstract lacks specific details on results/methodology. A line can be added, regarding the simulation tools used for this proposed model.
Response: Thank you for the suggestion. The abstract has been revised to include a concise statement about the methodology and simulation environment.

2. Are there any challenges while using unit commitment constraints and uncertainties in load demand and wind generation?
Response: Thank you for your question. Indeed, the inclusion of unit commitment (UC) constraints in models that account for uncertainties in load demand and wind generation introduces several significant challenges, both from operational and computational perspectives.
When uncertainties in load and wind generation are introduced, especially due to the intermittent nature of renewable sources, the problem becomes more dynamic and volatile. For instance:
● Sudden drops in wind generation may require rapid ramp-up from thermal generators, which may violate ramp limits or fail to meet minimum up-time constraints.
● Conversely, unexpected increases in renewable generation may cause overgeneration, forcing thermal units into inefficient or infeasible operating conditions due to their minimum generation levels or slow ramp-down rates.

3. What are the limitations for using MINLP and decomposition-based approach?
Response: Thank you for your question. The Transmission Network Expansion Planning problem, even in its simplest form, is a non-convex, combinatorial, and multimodal optimization problem. When formulated as a Mixed-Integer Nonlinear Programming (MINLP) model including nonlinear power flow equations, binary investment decisions, and operational constraints, the problem becomes extremely hard to solve directly due to its high computational complexity.

4. Figures 3 and 4 are not clear
Response: Thank you for pointing this out. To improve clarity, we have revised the figure captions and added explanatory text in the results section, providing a more detailed interpretation of the figures. Additionally, we are prepared to adjust the visual design (e.g., axis labels, legends, or color scheme) to enhance readability and ensure the conveyed information is accessible to the reader.

5. The future work is not discussed.
Response:
Thank you for the suggestion. A dedicated paragraph on future work has now been added to the Conclusion section. Potential directions for future research include:
● Integration of hydroelectric power plants, which introduce long-term storage and seasonal planning challenges.
● Extension of the current framework to consider AC power flow constraints, improving the electrical accuracy of investment decisions.
● Extend the methodology to larger systems

Reviewer 3 Report

Comments and Suggestions for Authors

Integration of energy storage systems (ESSs) and transmission co-expansion under uncertainty with a decomposition-based solution approach combining MILP is innovative and relevant to current challenges in power systems planning. The work has great importance but lacks in many parts, as mentioned in the following comments;

1. The authors did not properly discuss the methodology, how they model and simulations work on the proposed model, and how the steps are being followed. There must be a flowchart showing the significance of the work.
2. The authors considered the IEEE-24 bus system, but there are no details of that system, like their power values, their branches with the resistance or inductance values. Also, the details of buses (load, power, and other). Also, how it supports transmission network as per different IEEE bus systems.
3. Results are not properly discussed, like Table#3-6 and Figures 4- 5. What do the tables and figures show? It is not clear that co-planning reduces wind curtailment, fuel costs, and total investment costs compared to transmission-only expansion. 
4. There are many acronyms used, but a table is not added for easy understanding.
5. References are also incorrect and not in sequence. Like in the sentence "The k-means clustering technique is applied to group the data and assign weighted probabilities to each scenario. The complete system data are available in [? ]". What does this question mark show?
6. The abstract and conclusion are also not properly written. Both sections lack the proper discussion and the research work carried out as per the standards. 

Author Response

Integration of energy storage systems and transmission co-expansion under uncertainty with a decomposition-based solution approach combining MILP is innovative and relevant to current challenges in power systems planning. The work has great importance but lacks in many parts, as mentioned in the following comments;

1. The authors did not properly discuss the methodology, how they model and simulations work on the proposed model, and how the steps are being followed. There must be a flowchart showing the significance of the work.
Response: Thank you for the suggestion. We have addressed this point by adding a detailed description of the proposed methodology, outlining the modeling structure, key components, and simulation workflow. A flowchart has also been included to clearly illustrate the sequence of steps followed in the co-optimization process of transmission expansion and energy storage systems under uncertainty.

2. The authors considered the IEEE-24 bus system, but there are no details of that system, like their power values, their branches with the resistance or inductance values. Also, the details of buses (load, power, and other). Also, how it supports transmission network as per different IEEE bus systems.
Response: Thank you for the comment. All system data used in this study, including network topology, generation units, load profiles, transmission line parameters, and scenario inputs, have been made publicly available in an online repository referenced as [25].
[25] Nepomuce, L.; de Oliveira, E.; de Oliveira, L.; de Paula, A. Data for the paper "A Novel Framework for Co-Expansion Planning of Transmission Lines and Energy Storage Devices Considering Unit Commitment", 2025.

3. Results are not properly discussed, like Table#3-6 and Figures 4- 5. What do the tables and figures show? It is not clear that co-planning reduces wind curtailment, fuel costs, and total investment costs compared to transmission-only expansion.
Response: Thank you for the question. We have revised the manuscript to improve the clarity and depth of the results’ discussion.
● Table 4 presents the optimal investment decisions for each planning case (A: transmission-only, B: co-planning). It shows the total investment cost in transmission lines and storage, as well as the annual operational cost under uncertainty. As shown, Case B (co-planning) leads to lower total system costs by balancing investments between transmission and storage assets.
● Table 5 presents the expected values of thermal generation, energy charged and discharged by the storage systems, and wind curtailment for the best solutions obtained in each case. These values are computed as the weighted sum across all scenarios, considering each scenario's probability of occurrence. The results indicate that Case B (co-planning) leads to a notable reduction in both thermal generation and wind curtailment compared to Case A. This improvement is attributed to the additional operational flexibility provided by the energy storage systems, which enhances the integration of renewable sources and contributes to a significant reduction in system operational costs.
● Tables 6 and 7 provide the optimal hourly dispatch of thermal generation units for the first scenario, corresponding to the best solutions obtained for Case A and Case B, respectively. These tables are important to illustrate how thermal units operate under the expansion plans and how their dispatch decisions are influenced by unit commitment constraints. They highlight the impact of investments on the operational behavior of thermal generators throughout the day
● Figures 4 and 5 illustrate the hourly operation of energy storage devices, particularly the charging and discharging profiles for a representative scenario in Case B. The figures highlight that:
○ Charging occurs during low-demand or high-renewable periods (typically late night or early morning),
○ Discharging occurs during peak demand hours to offset thermal generation.

4. There are many acronyms used, but a table is not added for easy understanding.
Response: Thank you for the suggestion. A table of acronyms has been added in the Nomenclature section to facilitate understanding and improve the clarity of the manuscript.

5. References are also incorrect and not in sequence. Like in the sentence "The k-means clustering technique is applied to group the data and assign weighted probabilities to each scenario. The complete system data are available in [? ]". What does this question mark show?
Response: Thank you for pointing this out. It has been corrected, and all references have been revised to ensure correct sequencing and proper citation throughout the manuscript.

6. The abstract and conclusion are also not properly written. Both sections lack the proper discussion and the research work carried out as per the standards.
Response: Thank you for the comment. Both the Abstract and Conclusion sections have been thoroughly revised to reflect the objectives clearly, methodology, key results, and contributions of the study.

Reviewer 4 Report

Comments and Suggestions for Authors

This study proposes a methodology for planning a simultaneous expansion of transmission lines and energy storage devices, considering generator start-up and shutdown constraints and uncertainties in load demand and wind power generation. The study results are interesting because they show that planning transmission line expansion and ESS installation simultaneously reduces wind power generation output constraints, fuel costs, and total investment costs compared to expanding transmission lines alone. Developing an integrated plan that combines ESS investment with transmission line expansion is a realistic approach needed to expand renewable energy.  However, I suggest considering the following points:

(1) Unfortunately, only the results for the IEEE 24-bus system are presented. Analysis of computation time and convergence at the actual system scale (e.g., 100 buses or more) is required.

(2) The impact of neglected factors (e.g. voltage stability, losses) on the results compared to the AC model should be qualitatively discussed.

(3) There is insufficient quantitative explanation of the causal relationship between the charge/discharge pattern in Figure 3-4 and the reduction in fuel costs. (e.g., “Save $X when discharging 1MWh during peak demand hours”)

(4) Increased maintenance costs or life cycle impacts due to ESS investment were not considered, which should be reflected in long-term net benefit calculations.

Author Response

This study proposes a methodology for planning a simultaneous expansion of transmission lines and energy storage devices, considering generator start-up and shutdown constraints and uncertainties in load demand and wind power generation. The study results are interesting because they show that planning transmission line expansion and ESS installation simultaneously reduces wind power generation output constraints, fuel costs, and total investment costs compared to expanding transmission lines alone. Developing an integrated plan that combines ESS investment with transmission line expansion is a realistic approach needed to expand renewable energy. However, I suggest considering the following points:

1. Unfortunately, only the results for the IEEE 24-bus system are presented. Analysis of computation time and convergence at the actual system scale (e.g., 100 buses or more) is required.
Response: Thank you for the comment. We fully agree that analyzing the computational performance and convergence behavior of the proposed model on larger-scale systems (e.g., 100 buses or more) would enrich the study and enhance its practical relevance.
Our choice of the IEEE 24-bus system was motivated by several factors. Firstly, it is a widely used benchmark in the literature on Transmission Network Expansion Planning, especially for studies proposing methodological extensions, such as the co-optimization of transmission and energy storage or the incorporation of uncertainty. This is supported by the review summarized in Table 1, which compiles recent publications in the field using similar system sizes when new modeling features are introduced.
Secondly, the proposed formulation incorporates detailed operational constraints, including unit commitment of thermal generators, multi-scenario stochastic modeling for both load and wind power uncertainty, and binary investment decisions for both lines and storage systems. These elements significantly increase the dimensionality of the model and data requirements.
Nonetheless, we acknowledge the importance of testing scalability. As a part of our future work, we plan to extend the methodology to larger systems by either adapting existing datasets or collaborating with utility partners to construct realistic case studies.
We believe that the IEEE 24-bus test case still provides meaningful insights into the qualitative behavior of the proposed model and is consistent with the modeling scope adopted in similar pioneering works in the area.

2. The impact of neglected factors (e.g. voltage stability, losses) on the results compared to the AC model should be qualitatively discussed.
Response: Thank you for this observation. In this work, we adopt a DC optimal power flow (DC-OPF) formulation with approximated losses, which is a widely accepted modeling approach in Transmission Network Expansion Planning studies, especially when the focus is on investment decisions under uncertainty and long-term strategic planning. The DC-OPF approximation allows for improved tractability and scalability, which is critical when modeling complex operational features such as unit commitment, energy storage dynamics, and multi-scenario uncertainty.
A detailed quantitative comparison between DC and AC-based TEP formulations is beyond the scope of the present work but is an important topic for future research. Such an analysis could explore the trade-offs between solution accuracy and computational efficiency, as well as the operational feasibility of DC-based plans under full AC power flow validation.

3. There is insufficient quantitative explanation of the causal relationship between the charge/discharge pattern in Figure 3-4 and the reduction in fuel costs. (e.g., “Save $X when discharging 1MWh during peak demand hours”)
Response: Thank you for the comment. The text and figures were fixed.
In this context, energy storage plays a critical role in shaving peaks and filling valleys of the net load (defined as total demand minus renewable generation). In the early morning and late evening hours, when demand is low and renewable generation may exceed load, storage systems are charged with the surplus energy. These charging patterns are clearly illustrated in Figure 3.
During peak demand periods, typically occurring in the afternoon and early evening, storage systems discharge the previously stored energy. This displaces the need to commit additional thermal generation units or reduces the dispatch of high-cost peaking plants. As a result, the system avoids fuel consumption from thermal units that would otherwise operate inefficiently or be cycled on and off. These discharge patterns can be observed in Figure 4.
In the revised manuscript, we have expanded the discussion to clarify this causal relationship and referenced Figures 3 and 4 directly to support the explanation.

4. Increased maintenance costs or life cycle impacts due to ESS investment were not considered, which should be reflected in long-term net benefit calculations.
Response: Thank you for this comment. We acknowledge that maintenance costs and life cycle degradation of energy storage systems (ESS), such as battery replacement, efficiency losses over time, and operational and environmental costs, are critical components that affect the long-term economic viability of storage investments.
In this study, our primary objective was to assess the strategic value of co-planning ESS and transmission investments under operational constraints and uncertainty. To maintain tractability in the optimization model, we adopted a simplified cost structure, where ESS investment costs were annualized based on typical capital expenditures and a fixed lifetime, following common practices in preliminary expansion studies.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Authors should highlight why thermal generation units are crucial in Transmission Network Expansion Planning.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have incorporated all the comments. It should be considered for final approval.