Pith. sign in

REVIEW 2 minor 38 references

Reference-Free Heterogeneous Multi-Agent Reinforcement Learning for Grid-Friendly Tie-Line Power Shaping in Industrial Microgrids

T0 review · 0 major / 2 minor · reviewed 2026-06-25 · grok-4.3

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A sequential heterogeneous-agent coordination framework allows industrial microgrids to shape tie-line power without predefined references.

desk verdict SHAC gives a practical multi-agent RL design for reference-free tie-line shaping in steel microgrids, with large reported gains on real data but thin method details. read the letter →

arxiv 2606.25599 v1 pith:MO3EFC2O submitted 2026-06-24 eess.SY cs.SY

Daniyaer Paizulamua , Lin Cheng , Fashun Shi , Haoyu Zheng , Pengfei He , Haiwang Zhong This is my paper
classification eess.SYcs.SY
keywords multi-agentreinforcementlearningindustrialmicrogridstie-linepowershapingheterogeneousagentsgrid-friendlyoperationreference-freecontrolsteelindustrymicrogrid
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper develops a sequential heterogeneous-agent coordination framework to manage process loads, hydrogen storage, and battery storage in a steel industrial microgrid for grid-friendly tie-line power shaping. These resources are treated as functionally heterogeneous agents that share cross-role observations but operate on asynchronous intervals with role-specific rewards and critics. This setup is intended to handle differing temporal effects on the tie-line trajectory and reduce credit-assignment problems. If the approach works, it removes the need for preset reference trajectories and supports adaptive one-minute decisions that respect strict process safety rules while lowering grid purchase costs and ramp violations. Numerical tests with real renewable and market data report large reductions in cost and exceedance metrics alongside zero production failures and sub-millisecond computation.

What carries the argument

Sequential heterogeneous-agent coordination (SHAC) framework using role-specific rewards, cross-role observations, asynchronous decision intervals, action masking, and role-aware multi-timescale actor-critic training.

What would settle it

Deploy the trained SHAC policies on the physical steel microgrid using live renewable generation and electricity market data for an extended period and check whether production failures stay at zero while measured cost, exceedance time, and ramp excess reductions approach the reported 91 percent, 98 percent, and 96 percent levels.

Watch

Extended reading notes

Core claim

The sequential heterogeneous-agent coordination (SHAC) framework models process loads, hydrogen storage, and battery storage as functionally heterogeneous agents with cross-role observations, asynchronous decision intervals, role-specific rewards and critics. This captures heterogeneous temporal effects on the tie-line power trajectory and alleviates ambiguous credit assignment and weak inter-agent coordination. Process-knowledge-based action masking and feasibility projection are embedded in policy execution, while a role-aware multi-timescale actor-critic training scheme handles agents with different action structures. The resulting policies enable reference-free adaptive one-minute online

Load-bearing premise

Modeling process loads, hydrogen storage, and battery storage as functionally heterogeneous agents with cross-role observations, asynchronous decision intervals, role-specific rewards and critics accurately captures the real system dynamics, temporal effects, and interdependencies in the steel industrial microgrid.

Editorial extensions

If this is right

  • Eliminates dependence on predefined reference trajectories for tie-line power.
  • Enables adaptive one-minute online decision-making with zero production failures.
  • Reduces total grid purchase cost by 91.27 percent relative to original operation.
  • Reduces contract-demand exceedance time by 98.64 percent and cumulative ramp excess by 96.91 percent.
  • Maintains average computational time of 0.4 milliseconds per decision step.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same heterogeneous-agent structure could be tested in other energy-intensive sectors such as chemical plants or data centers that must coordinate multi-timescale loads and storage.
  • Removing reference dependence may improve robustness when renewable output or market prices deviate sharply from historical patterns used to build the training data.
  • A field trial could measure whether communication latency or sensor noise between agents degrades the reported performance gains.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

0 major / 2 minor

Summary. The manuscript proposes a sequential heterogeneous-agent coordination (SHAC) framework for reference-free tie-line power shaping in a steel industrial microgrid. Process loads, hydrogen storage, and battery storage are modeled as functionally heterogeneous agents with cross-role observations, asynchronous decision intervals, role-specific rewards/critics, action masking, and feasibility projection. A role-aware multi-timescale actor-critic training scheme supports real-time execution. Numerical studies on real renewable and market data report zero production failures, 0.4 ms average computation per step, and reductions versus original operation of 91.27% in grid purchase cost, 98.64% in contract-demand exceedance time, and 96.91% in cumulative ramp excess.

Significance. If the reported numerical improvements and real-time feasibility hold under the heterogeneous-agent modeling, the work would demonstrate a practical advance in applying multi-agent RL to industrial microgrids with strict process constraints and multi-timescale resources. The elimination of predefined reference trajectories and the achieved grid-friendliness metrics could inform control designs for renewable-integrated industrial systems.

minor comments (2)
  1. The abstract states large percentage reductions but does not specify the exact baseline operation details or the time horizon over which the metrics are aggregated; this should be clarified in the results section for reproducibility.
  2. Notation for agent roles, observation spaces, and reward structures is introduced in the abstract without explicit definitions; a dedicated notation table or early section would improve readability.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for reviewing our manuscript on the SHAC framework. The report provides a concise summary and notes potential significance if the numerical results hold, with an 'uncertain' recommendation. However, no specific major comments are listed under the MAJOR COMMENTS section. We therefore provide no point-by-point responses below and remain available to address any additional questions or clarifications the referee may have.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity detected

full rationale

The paper proposes a SHAC multi-agent RL framework for microgrid tie-line power shaping and validates it through numerical studies on external real-world renewable and market data. Performance metrics (cost reductions, zero failures, 0.4 ms computation) are reported as empirical outcomes of the simulation, not as quantities derived by fitting parameters to the target metrics themselves or by self-referential definitions. No equations or claims reduce the reported results to the modeling assumptions by construction, and no self-citation chains are invoked as load-bearing uniqueness theorems. The derivation chain is therefore self-contained against external benchmarks.

Assumptions & free parameters 0 free parameters · 0 assumptions · 0 invented entities

Review based solely on abstract; no specific free parameters, axioms, or invented entities can be extracted. The framework relies on standard multi-agent RL assumptions and power-system modeling choices that are not detailed here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reference-Free Heterogeneous Multi-Agent Reinforcement Learning for Grid-Friendly Tie-Line Power Shaping in Industrial Microgrids." pith.science (2026). https://pith.science/paper/MO3EFC2O

@misc{pith2026260625599,
  author       = {Pith},
  title        = {Pith review of: Reference-Free Heterogeneous Multi-Agent Reinforcement Learning for Grid-Friendly Tie-Line Power Shaping in Industrial Microgrids},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MO3EFC2O}},
  note         = {Machine review of arXiv:2606.25599}
}
read the original abstract

Tie-line power (TLP) shaping is a key requirement for the grid-friendly operation of industrial microgrids (IMGs). This paper studies the coordination of multi-timescale heterogeneous adjustable resources in a steel IMG to shape a grid-friendly TLP trajectory considering multiple objectives. A sequential heterogeneous-agent coordination (SHAC) framework is proposed, where process loads, hydrogen storage, and battery storage are modeled as functionally heterogeneous agents with cross-role observations, asynchronous decision intervals, role-specific rewards and critics. This design captures the heterogeneous temporal effects of different resources on the TLP trajectory and alleviates ambiguous credit assignment and weak inter-agent coordination. To ensure feasible real-time execution, process-knowledge-based action masking and feasibility projection are embedded into policy execution, and a role-aware multi-timescale actor--critic training scheme is developed for agents with different action structures and decision intervals. Numerical studies using real renewable generation and electricity market data show that SHAC effectively eliminates the dependence on predefined reference trajectories and enables adaptive 1-min online decision-making, achieving zero production failures with an average computational time of only 0.4 ms per step. Compared with the original operation, SHAC reduces the total grid purchase cost, contract-demand exceedance time, and cumulative ramp excess by 91.27\%, 98.64\%, and 96.91\%, respectively. These results demonstrate that the proposed framework improves the economic efficiency and grid friendliness of industrial microgrid operation while satisfying strict process-safety constraints and real-time computational requirements.

Figures

Figures reproduced from arXiv: 2606.25599 by the authors.

Figure 1
Figure 1. Architecture of the studied steel industrial microgrid with process loads and hybrid energy storage. units. (2) The load side is dominated by high-power industrial equipment and automated production lines, exhibiting nonlinear, impulsive, and strongly process-coupled characteristics. (3) The operating scale typically ranges from tens to hundreds of MW, with possible interconnection at higher voltage levels and cover… view at source ↗
Figure 2
Figure 2. Illustrative comparison of grid-unfriendly and grid-friendly IMG tie-line power trajectories under renewable generation and electricity price signals. 𝜂 ex= ∑ 𝑡∈ [−𝑃 TL 𝑡 ] +Δ𝑡 ∑ 𝑡∈ 𝑃 RES 𝑡 Δ𝑡 ≤𝜂 ex , (2b) where 𝑃 RES 𝑡 denotes the available renewable power at time 𝑡, 𝑃 RES 𝑡 denotes the locally accommodated renewable power, and [−𝑃 TL 𝑡 ] + denotes the exported power. 𝜂 RES imposes the lower limit on the local re… view at source ↗
Figure 3
Figure 3. Overall workflow of the proposed SHAC framework. 𝜅𝑚(𝑡)=𝑚 ⌊ 𝑡 𝑚 ⌋ , (9) then the PL scheduling action is retained over [𝜅5 (𝑡),𝜅5 (𝑡)+4], the HS action is retained over [𝜅15(𝑡),𝜅15(𝑡)+14], and the BS action is updated at each minute-level period. To preserve the Markov property under asynchronous action retention, the global state 𝑥𝑡∈ includes operating states, exogenous forecasts, retained actions, and temporal pha… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: (b) presents the duration curves of renewable generation and real-time price, showing the uneven distribution of high-/low-renewable periods and price spikes [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]
Figure 5
Figure 5. Figure 5: Representative daily profiles of renewable generation and real-time electricity price under four RES–price scenarios. 5.1.2. System Configuration and Parameters The test system represents a short-process steel industrial microgrid with three parallel production lines, …
Figure 6
Figure 6. Figure 6: (c) shows the training reward of the HS agent and its regulation effect on the 15-min TLP residual. The mean HS contribution gradually increases and stabilizes at approximately 35–45 MW, while the 15-min TLP residual after HS regulation decreases from more than 140 MW …
Figure 7
Figure 7. Figure 7: Representative day operational performance of the SHAC framework: (a) RES power and electricity price, (b) PL-agent power and rule-based PL power, (c) HS and BS power responses with SOC trajectories, (d) TLP under SHAC and rule-based PL operation. 5.4. Ablation Studies…
Figure 8
Figure 8. Figure 8 [PITH_FULL_IMAGE:figures/full_fig_p019_8.png]
Figure 9
Figure 9. Figure 9: Normalized diagnostic validation reward trends of SHAC, SC-SHAC, and P-SHAC during training [PITH_FULL_IMAGE:figures/full_fig_p019_9.png]
Figure 10
Figure 10. Figure 10: Mean completed production heats during validation for SHAC and SC-SHAC. N o rm a l i z e d B S Va l i d a t i o n R e w a r d N o rm a l i z e d B S Va l i d a t i o n R e w a r d Episode ( 104 ) 0 0.5 1 1.5 2 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Episode ( 104 ) 0 …
Figure 11
Figure 11. Figure 11: HS- and BS-agent validation reward comparison between SHAC and P-SHAC. 5.5. Comparative performance evaluation To further evaluate the advantages of SHAC in grid-friendly TLP shaping, it is compared with three representative coordination methods: rule-based control (R…
Figure 12
Figure 12. Figure 12: presents the TLP trajectories of different methods on a representative testing day. Due to its fixed production rhythm, RB-Control leads to pronounced peak-valley variations and rapid ramping in TLP. HRTC improves the TLP profile in some periods through reference trac…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

38 extracted references · 1 canonical work pages

  1. [1]

    Toward sustainable and low-carbon industrial transformation: Insights from industrial green microgrids

    Yufei Xi, Xinyi Wang, Lin Cheng, Meng Chen, Ioannis Lestas, Fashun Shi, and Kaidi Huang. Toward sustainable and low-carbon industrial transformation: Insights from industrial green microgrids. Applied Energy, 417:128055, 2026

    2026

  2. [2]

    A review of industrial load flexibility enhancement for demand-response interaction

    Jiubo Zhang, Bowen Zhou, Zhile Yang, Yuanjun Guo, Chen Lv, Xiaofeng Xu, and Jichun Liu. A review of industrial load flexibility enhancement for demand-response interaction. Sustainability, 17(11):4938, 2025

    2025

  3. [3]

    Multi-objective scheduling of a steelmaking plant integrated with renewable energy sources and energy storage systems: Balancing costs, emissions and make-span

    Pengfei Su, Yue Zhou, and Jianzhong Wu. Multi-objective scheduling of a steelmaking plant integrated with renewable energy sources and energy storage systems: Balancing costs, emissions and make-span. Journal of Cleaner Production, 428:139350, 2023

    2023

  4. [4]

    Hybrid lithium-ion battery and hydrogen energy storage systems for a wind-supplied microgrid

    Michael Anthony Giovanniello and Xiao-Yu Wu. Hybrid lithium-ion battery and hydrogen energy storage systems for a wind-supplied microgrid. Applied Energy, 345:121311, 2023

    2023

  5. [5]

    A demand response energy management scheme for industrial facilities in smart grid

    Yue Min Ding, Seung Ho Hong, and Xiao Hui Li. A demand response energy management scheme for industrial facilities in smart grid. IEEE transactions on industrial informatics, 10(4):2257–2269, 2014

    2014

  6. [6]

    Optimal industrial load control in smart grid

    Armen Gholian, Hamed Mohsenian-Rad, and Yingbo Hua. Optimal industrial load control in smart grid. IEEE Transactions on Smart Grid, 7(5):2305–2316, 2015

    2015

  7. [7]

    A real-time decision model for industrial load management in a smart grid

    Mengmeng Yu, Renzhi Lu, and Seung Ho Hong. A real-time decision model for industrial load management in a smart grid. Applied energy, 183:1488–1497, 2016

    2016

  8. [8]

    Demand response of ancillary service from industrial loads coordinated with energy storage

    Xiao Zhang, Gabriela Hug, J Zico Kolter, and Iiro Harjunkoski. Demand response of ancillary service from industrial loads coordinated with energy storage. IEEE Transactions on Power Systems, 33(1):951–961, 2017

    2017

Show all 38 references
  1. [9]

    Demand-side management via optimal production scheduling in power-intensive industries: The case of metal casting process

    Danial Ramin, Stefano Spinelli, and Alessandro Brusaferri. Demand-side management via optimal production scheduling in power-intensive industries: The case of metal casting process. Applied Energy, 225:622–636, 2018. Page 22 of 23

    2018

  2. [10]

    Robust self-scheduling of operational processes for industrial demand response aggregators

    Hessam Golmohamadi, Reza Keypour, Birgitte Bak-Jensen, Jayakrishnan R Pillai, and Mohammad Hassan Khooban. Robust self-scheduling of operational processes for industrial demand response aggregators. IEEE Transactions on Industrial Electronics, 67(2):1387–1395, 2019

    2019

  3. [11]

    Cost-effective scheduling of steel plants with flexible eafs

    Xiao Zhang, Gabriela Hug, and Iiro Harjunkoski. Cost-effective scheduling of steel plants with flexible eafs. IEEE Transactions on Smart Grid, 8(1):239–249, 2016

    2016

  4. [12]

    Quantifying flexibility provisions of the ladle furnace refining process as cuttable loads in the iron and steel industry

    Jiayang Wang, Qiang Wang, and Wenqiang Sun. Quantifying flexibility provisions of the ladle furnace refining process as cuttable loads in the iron and steel industry. Applied Energy, 342:121178, 2023

    2023

  5. [13]

    Cost-effective scheduling of a hydrogen-based iron and steel plant powered by a grid-assisted renewable energy system

    Pengfei Su, Yue Zhou, Hongyi Li, Hector D Perez, and Jianzhong Wu. Cost-effective scheduling of a hydrogen-based iron and steel plant powered by a grid-assisted renewable energy system. Applied Energy, 384:125412, 2025

    2025

  6. [14]

    Distributionally robust chance-constrained energy management of steel industrial microgrid with energy storage in distribution market

    Linbo Fu, Houhe Chen, Rufeng Zhang, Tao Jiang, Guoqing Li, and Rui Qu. Distributionally robust chance-constrained energy management of steel industrial microgrid with energy storage in distribution market. Applied Energy, 400:126574, 2025

    2025

  7. [15]

    Efficient scheduling of discrete industrial processes through continuous modeling

    Ruike Lyu, Xiangbo Su, Ershun Du, Hongye Guo, Qixin Chen, and Chongqing Kang. Efficient scheduling of discrete industrial processes through continuous modeling. IEEE Transactions on Smart Grid, 2025

    2025

  8. [16]

    Smoothing tie-line power fluctuations for industrial microgrids by demand side control: An output regulation approach

    Chen Zhang, Wei Lin, Deping Ke, and Yuanzhang Sun. Smoothing tie-line power fluctuations for industrial microgrids by demand side control: An output regulation approach. IEEE Transactions on Power Systems, 34(5):3716–3728, 2019

    2019

  9. [17]

    Suppressing active power fluctuations at pcc in grid-connection microgrids via multiple besss: A collaborative multi-agent reinforcement learning approach

    Wangli He, Chengyuan Li, Chenhao Cai, Xiangyun Qing, and Wenli Du. Suppressing active power fluctuations at pcc in grid-connection microgrids via multiple besss: A collaborative multi-agent reinforcement learning approach. Applied Energy, 373:123858, 2024

    2024

  10. [18]

    Multi-time-scale energy management of renewable microgrids considering grid-friendly interaction

    Shenglin Li, Jizhong Zhu, Hanjiang Dong, Haohao Zhu, Fengji Luo, and Alberto Borghetti. Multi-time-scale energy management of renewable microgrids considering grid-friendly interaction. Applied Energy, 367:123428, 2024

    2024

  11. [19]

    Flat tie-line power scheduling control of grid-connected hybrid microgrids

    Zhao-xia Xiao, Josep M Guerrero, Jia Shuang, Dezso Sera, Erik Schaltz, and Juan C Vásquez. Flat tie-line power scheduling control of grid-connected hybrid microgrids. Applied Energy, 210:786–799, 2018

    2018

  12. [20]

    A demand response and battery storage coordination algorithm for providing microgrid tie-line smoothing services

    Dan Wang, Shaoyun Ge, Hongjie Jia, Chengshan Wang, Yue Zhou, Ning Lu, and Xiangyu Kong. A demand response and battery storage coordination algorithm for providing microgrid tie-line smoothing services. IEEE Transactions on Sustainable Energy, 5(2):476–486, 2014

    2014

  13. [21]

    Data center holistic demand response algorithm to smooth microgrid tie-line power fluctuation

    Ting Yang, Yingjie Zhao, Haibo Pen, and Zhaoxia Wang. Data center holistic demand response algorithm to smooth microgrid tie-line power fluctuation. Applied Energy, 231:277–287, 2018

    2018

  14. [22]

    Multi-data center tie-line power smoothing method based on demand response

    Ting Yang, Yuxing Hou, Shaotang Cai, Jie Yu, and Haibo Pen. Multi-data center tie-line power smoothing method based on demand response. IEEE Transactions on Cloud Computing, 12(4):983–995, 2024

    2024

  15. [23]

    A novel rolling optimization strategy considering grid-connected power fluctuations smoothing for renewable energy microgrids

    Shenglin Li, Jizhong Zhu, Hanjiang Dong, Haohao Zhu, and Junwei Fan. A novel rolling optimization strategy considering grid-connected power fluctuations smoothing for renewable energy microgrids. Applied Energy, 309:118441, 2022

    2022

  16. [24]

    A coordinated multitimescale model predictive control for output power smoothing in hybrid microgrid incorporating hydrogen energy storage

    Muhammad Bakr Abdelghany, Ahmed Al-Durra, Hatem H Zeineldin, and Fei Gao. A coordinated multitimescale model predictive control for output power smoothing in hybrid microgrid incorporating hydrogen energy storage. IEEE Transactions on Industrial Informatics , 20(9):10987–11001, 2024

    2024

  17. [25]

    Multi-agent deep reinforcement learning based demand response for discrete manufacturing systems energy management

    Renzhi Lu, Yi-Chang Li, Yuting Li, Junhui Jiang, and Yuemin Ding. Multi-agent deep reinforcement learning based demand response for discrete manufacturing systems energy management. Applied Energy, 276:115473, 2020

    2020

  18. [26]

    Multi-agent deep reinforcement learning based demand response and energy management for heavy industries with discrete manufacturing systems

    Atit Bashyal, Tina Boroukhian, Pakin Veerachanchai, Myanganbayar Naransukh, and Hendro Wicaksono. Multi-agent deep reinforcement learning based demand response and energy management for heavy industries with discrete manufacturing systems. Applied Energy , 392:125990, 2025

    2025

  19. [27]

    Energy management based on multi-agent deep reinforcement learning for a multi-energy industrial park

    Dafeng Zhu, Bo Yang, Yuxiang Liu, Zhaojian Wang, Kai Ma, and Xinping Guan. Energy management based on multi-agent deep reinforcement learning for a multi-energy industrial park. Applied Energy, 311:118636, 2022

    2022

  20. [28]

    Coordination for multienergy microgrids using multiagent reinforcement learning

    Dawei Qiu, Tianyi Chen, Goran Strbac, and Shengrong Bu. Coordination for multienergy microgrids using multiagent reinforcement learning. IEEE Transactions on Industrial Informatics, 19(4):5689–5700, 2022

    2022

  21. [29]

    Physics-model-free heat-electricity energy management of multiple microgrids based on surrogate model-enabled multi-agent deep reinforcement learning

    Sichen Li, Weihao Hu, Di Cao, Zhe Chen, Qi Huang, Frede Blaabjerg, and Kaiji Liao. Physics-model-free heat-electricity energy management of multiple microgrids based on surrogate model-enabled multi-agent deep reinforcement learning. Applied Energy, 346:121359, 2023

    2023

  22. [30]

    Renewable energy integration and microgrid energy trading using multi-agent deep reinforcement learning

    Daniel JB Harrold, Jun Cao, and Zhong Fan. Renewable energy integration and microgrid energy trading using multi-agent deep reinforcement learning. Applied Energy, 318:119151, 2022

    2022

  23. [31]

    Multi-agent reinforcement learning for energy management in microgrids with shared hydrogen storage

    David Toquica, Kodjo Agbossou, and Nilson Henao. Multi-agent reinforcement learning for energy management in microgrids with shared hydrogen storage. International Journal of Hydrogen Energy, 144:1019–1027, 2025

    2025

  24. [32]

    Multi-agent hierarchical reinforcement learning for energy management

    Imen Jendoubi and François Bouffard. Multi-agent hierarchical reinforcement learning for energy management. Applied Energy, 332:120500, 2023

    2023

  25. [33]

    Collaborative optimization of multi-energy multi-microgrid system: A hierarchical trust-region multi-agent reinforcement learning approach

    Xuesong Xu, Kai Xu, Ziyang Zeng, Jiale Tang, Yuanxing He, Guangze Shi, and Tao Zhang. Collaborative optimization of multi-energy multi-microgrid system: A hierarchical trust-region multi-agent reinforcement learning approach. Applied Energy, 375:123923, 2024

    2024

  26. [34]

    Trust region policy optimisation in multi-agent reinforcement learning

    Jakub Grudzien Kuba, Ruiqing Chen, Muning Wen, Ying Wen, Fanglei Sun, Jun Wang, and Yaodong Yang. Trust region policy optimisation in multi-agent reinforcement learning. arXiv preprint arXiv:2109.11251, 2021

    2021

  27. [35]

    Heterogeneous-agent reinforcement learning

    Yifan Zhong, Jakub Grudzien Kuba, Xidong Feng, Siyi Hu, Jiaming Ji, and Yaodong Yang. Heterogeneous-agent reinforcement learning. Journal of Machine Learning Research, 25(32):1–67, 2024

    2024

  28. [36]

    Guidelines for the Construction and Application of Industrial Green Microgrids (2026–2030)

    Ministry of Industry and Information Technology of the People’s Republic of China et al. Guidelines for the Construction and Application of Industrial Green Microgrids (2026–2030). https://www.miit.gov.cn/zwgk/zcwj/wjfb/tz/art/2026/art_ 44c5364d80b748e4b797bc115388c6aa.html, 2026

    2026

  29. [37]

    Long-term energy management for microgrid with hybrid hydrogen-battery energy storage: A prediction-free coordinated optimization framework

    Ning Qi, Kaidi Huang, Zhiyuan Fan, and Bolun Xu. Long-term energy management for microgrid with hybrid hydrogen-battery energy storage: A prediction-free coordinated optimization framework. Applied Energy, 377:124485, 2025

    2025

  30. [38]

    Data Miner 2: Real-Time Five Minute LMPs

    PJM Interconnection, L.L.C. Data Miner 2: Real-Time Five Minute LMPs. https://dataminer2.pjm.com/feed/rt_fivemin_hrl_ lmps/definition, 2026. Page 23 of 23

    2026

Pith tools

Reviewed June 25, 2026 · model on record in the stance chip above.