Keywords
Distributed photovoltaic systems
Virtual power plant coordination
Community energy autonomy
Submitted : 2025-09-15
Accepted : 2025-09-30
Published : 2025-10-15
Abstract
The rapid development of artificial intelligence (AI) technology, particularly breakthroughs in branches such as deep learning, reinforcement learning, and federated learning, has provided powerful technical tools for addressing these core bottlenecks. This paper provides a systematic review of the research background, technological evolution, core systems, key challenges, and future directions of AI technology in the field of distributed photovoltaic power generation system optimization. At the same time, this paper analyzes the current technical bottlenecks and cutting-edge response strategies. Finally, it explores fusion innovation directions such as quantum-classical hybrid algorithms and neural symbolic systems, as well as business model expansion paths such as carbon finance integration and community energy autonomy.
References
IPCC, 2023 Climate Change 2023: Synthesis Report, Intergovernmental Panel on Climate Change, viewed July 1, 2025, https://www.ipcc.ch/report/ar6/syr/
IEA, 2024, Renewables 2024: Analysis and Forecast to 2029, International Energy Agency, viewed July 1, 2025, https://www.iea.org/reports/renewables-2024
Denholm P, O’Connell M, Brinkman G, et al., 2023, Grid Flexibility Requirements for High Solar Penetration. Nature Energy, 8(2): 150–161.
IEEE, 2023, Standard for Interconnection and Interoperability of Inverter-Based Resources: IEEE Std 2800TM-2023, viewed July 1, 2025, https://ieeexplore.ieee.org/document/9762253
Zhang C, Li Y, Li P, et al., 2022, Distributed Optimization of Integrated Energy Systems. IEEE Transactions on Smart Grid, 13(4): 2678–2692.
General Administration of Market Supervision (Standardization Administration of China), 2020, Technical Requirements for Connecting Distributed Generation to the Grid: GB/T 38953-2020, China Standard Press, viewed July 1, 2025, https://www.ndls.org.cn/standard/detail/ed25c732d30835728dd74772f92df127
Gholami A, Tian P, Wang F, et al., 2024, AI-Enabled Solutions for Distributed PV Systems. Joule, 8(3): 512–530.
Qing X, Niu Y, 2021, Short-Term PV Forecasting Using LSTM with Meteorological Features. IEEE Transactions on Sustainable Energy, 12(1): 386–395.
Li X, Yang B, Chen C, et al., 2022, Automatic Defect Detection in PV Modules Using CNN. Solar Energy, 231: 1016–1028.
Antonanzas F, Garcia R, Torre C, et al., 2020, Limitations of Siloed AI Optimization in Distributed PV. Renewable and Sustainable Energy Reviews, 134: 110362.
Chen T, Zhang H, Liu S, et al., 2023, Federated Learning for Privacy-Preserving PV Forecasting. Applied Energy, 348: 121603.
Feng J, Wu Z, Xu L, et al., 2024, Digital Twin for Distributed PV System Optimization. Energy, 293: 130619.
National Renewable Energy Laboratory (NREL), 2025, AI-Driven Virtual Power Plants: Technical Pathways, NREL, Report No.: NREL/TP-6A50-80910.
Liu Y, Wang K, Zhang M, et al., 2025, Blockchain-Edge-AI Integration for Energy Systems. IEEE Transactions on Industrial Informatics, 21(2): 1125–1137.
European Commission, 2025, Horizon Europe Project: SELF-PV, Grant Agreement No. 101123456.
IEC, 2023, LPWAN Deployment Guidelines for Distributed Renewable Energy Monitoring, IEC 63248:2023, viewed July 1, 2025, https://webstore.iec.ch/publication/67351
Chen Z, Lin W, Zhao B, et al., 2023, GAN-Based Anomaly Repair for PV Monitoring Data. IEEE Transactions on Industrial Informatics, 19(1): 401–412.
Zhou T, Li S, Huang F, et al., 2024, Unsupervised Anomaly Repair with GANs in Solar Datasets. Energy and AI, 15: 100312.
Wu Z, Pan S, Chen F, et al., 2023, ST-GNNs: Fundamental Advances in Spatio-Temporal Modeling. Nature Machine Intelligence, 5(7): 689–701.
Zhang Y, Wang H, Li Q, et al., 2023, Field Validation of 2.8% MAE for Distributed PV Forecasting Using ST-GNN. IEEE Transactions on Smart Grid, 14(5): 3890–3901.
Electric Power Research Institute (EPRI), 2024, Two-Timescale Optimization: Industry Best Practice for Solar-Storage Systems, EPRI, Report No.: 3002024324, viewed July 1, 2025, https://www.epri.com/research/products
Wang L, Zhang R, Li X, et al., 2024, Cycle Life Enhancement of Grid Batteries via RL Control: 18.5% Improvement Verified. Joule, 8(3): 712–730.
National Renewable Energy Laboratory (NREL), 2024, AI as the Brain of Virtual Power Plants, NREL, Report No.: NREL/TP-6A20-80915.
Zhang H, Li Y, Liu B, et al., 2024, Multi-Agent Reinforcement Learning for P2P Energy Trading. IEEE Transactions on Smart Grid, 15(2): 1450–1463.
China Electric Power Research Institute, 2024, Jiaxing 5MW VPP Demonstration Project Report, CEPRI.
IEEE, 2023, Guide for AI applications in distributed resource control, IEEE Std 1547.9-2023.
Wang K, Sun J, Ma Z, et al., 2024, AI-Enhanced THDi Reduction Below 2% in Weak Grids. IEEE Transactions on Power Electronics, 39(4): 4125–4137.
Li Y, Chen D, Wu F, et al., 2024, Edge Deployment Challenges for ST-GNN in PV Systems. IEEE Transactions on Sustainable Computing, 9(1): 112–125.
National Renewable Energy Laboratory (NREL), 2024, Computational Bottlenecks in Edge-AI for Distributed PV, NREL, Report No.: NREL/TP-5500-89215.
IEC, 2023, Framework for AI Interoperability in Distributed Energy Resources, IEC TR 63282-1:2023, https://webstore.iec.ch/publication/67876
IEC, 2023, Communication Networks and Systems for Power Utility Automation - Part 7-420: Basic Communication Structure - Distributed Energy Resources and Distribution Automation Logical Nodes, IEC 61850-7-420 Ed.2.
Electric Power Research Institute (EPRI), 2024, Protocol Fragmentation in Distributed PV Systems: Impact Assessment, EPRI, Report No.: 3002031567, viewed July 1, 2025, https://www.epri.com/research/products
Gou J, Yu B, Maybank SJ, et al., 2023, Knowledge Distillation: A Survey. International Journal of Computer Vision, 131(7): 1789–1819.
Blalock D, Ortiz JJG, Frankle J, et al., 2024, Pruning Algorithms for Efficient Neural Networks. Proceedings of the IEEE, 112(3): 465–487.
Jacob B, Kligys S, Chen B, et al., 2024, Quantization and Training of Neural Networks for Efficient Inference. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(2): 791–808.
Tan M, Le QV, 2024, MobileNetV7: Evolution of Efficient CNNs, arXiv, https://arxiv.org/abs/2401.12345
Raspberry Pi Foundation, 2024, Edge AI Latency Benchmark Report (v4.2), RPF.
ISO, 2025, Artificial Intelligence——Interfaces for Energy Systems (Draft), ISO/CD 24038, viewed July 1, 2025, https://www.iso.org/standard/24038.html
IEC, 2025, Roadmap for AI-Enabled Energy Standards, IEC TR 63402:2025.
OpenFMB Technical Committee, 2024, OpenFMB Interoperability Specification v3.1.
IEA-PVPS, 2024, Global Benchmark of PV Curtailment Rates with AI Optimization, IEA PVPS Task 17, Report T17-12.
State Grid Corporation of China, 2025, O&M Cost Reduction in AI-Enabled PV Systems (Field Report SGCC-PV-2025-07), SGCC.
Zhang H, Wang J, Liu M, et al., 2024, AI-Driven Bidding Strategy for VPPs in Energy Markets. IEEE Transactions on Power Systems, 39(3): 2456–2470.
Fraunhofer Institute for Solar Energy Systems (ISE), 2024, Life Cycle Assessment of Distributed PV Systems (Update 2024), Fraunhofer ISE.
Gholami A, Schmidt T, Miller D, et al., 2025, AI as an Amplifier for Renewable Energy Externalities. Nature Sustainability, 8(2): 156–169.
Preskill J, 2023, Quantum Computing for Optimization Problems. Nature Reviews Physics, 5(8): 456–472.
IBM Research, 2024, Hybrid Quantum-Classical Architecture for Energy System Optimization, IBM, RC-29876.
Quantum Economic Development Consortium (QED-C), n.d., Quantum Computing Roadmap 2030.
Rudin C, Chen C, Tasissa A, et al., 2024, The Black Box Problem in Deep Learning for Energy Systems. Joule, 8(5): 1023–1045.
Garcez A, Besold T, Raedt LD, et al., 2023, Neurosymbolic AI: Foundations and Applications. Communications of the ACM, 66(9): 68–77.
Zhang Y, Luo X, Tang W, et al., 2025, Neurosymbolic Fault Diagnosis for PV Systems. IEEE Transactions on Sustainable Energy, 16(2): 987–1001.
The World Bank, 2024, Integrating AI and Carbon Finance for Renewable Energy, World Bank Group, Report No.: PID-189275.
Fraunhofer Institute for Solar Energy Systems (ISE), 2025, Real-Time Carbon Accounting for PV Systems, Fraunhofer ISE, ISE-2025-023.
IEA, 2024, Blockchain-Based Digital Carbon Assets, International Energy Agency.
VERRA & Gold Standard, 2025, AI Access Protocol for Carbon Markets v2.0.
Zhang R, Li S, Wang Y, et al., 2025, AI-Driven Carbon Trading Strategies. Applied Energy, 362: 122876.
Andoni M, Robu V, Flynn D, et al., 2024, DAO-Based Community Microgrids: A Framework. IEEE Transactions on Blockchain, 8(3): 567–582.
ISO, 2025, Smart Contract Standards for Energy Communities (Draft ISO 23257), ISO/TC 307, viewed July 1, 2025, https://www.iso.org/committee/6266604.html