A deep reinforcement learning-based adaptive multivariate state estimation fault diagnosis method for a photovoltaic storage grid-connected system under a high proportion of new energy integration
DOI:
https://doi.org/10.4108/ew.12145Keywords:
deep agent reinforcement learning, high proportion of new energy integration, photovoltaic storage grid connected system, fault diagnosis, adaptive multivariate state estimationAbstract
With the increasing penetration of photovoltaic and other renewable energy sources in power systems, the operational characteristics of photovoltaic storage grid-connected systems have become increasingly complex. The stochastic nature of renewable energy output and the alternating charging/discharging states of energy storage systems due to diurnal cycles lead to significant time-varying operational conditions, under which the performance of conventional multivariate state estimation methods for fault diagnosis markedly deteriorates. To address this, targeting high-penetration renewable energy integration scenarios, this paper proposes an adaptive state estimation-based fault diagnosis method for photovoltaic-storage grid-connected systems using deep reinforcement learning. Within the deep reinforcement learning framework, an agent is deployed to adaptively match and update modeling samples according to the real-time system state, thereby enhancing the adaptability of state estimation and the fault diagnosis accuracy. Test results under multivariate typical operational scenarios demonstrate that, compared to existing advanced fault diagnosis methods, the proposed approach achieves an average improvement of 10% in fault detection rate and reduces the missed detection rate by 3%, validating its superior adaptability and diagnostic performance in systems with complex time-varying characteristics. This research provides a new technical pathway for ensuring the safe and stable operation of photovoltaic-storage grid-connected systems under high-penetration renewable energy integration, contributing to enhanced intelligent operation and maintenance capabilities and improved fault response.
Downloads
References
[1] Hsu HW, Fan ZW. Multi-faceted procurement with mixed integer linear programming for corporate 100 % renewable energy goal. Energy, 2025, 320.
[2] Efthymiou C, Khan A, Assimakopoulos MN, Santamouris M. Urban pavement-mounted photovoltaics as renewable energy systems for energy generation and microclimate control. Solar energy, 2025, 299.
[3] Sanda MG, Emam M, Ookawara S, Hassan H. Techno-enviro-economic evaluation of on-grid and off-grid hybrid photovoltaics and vertical axis wind turbines system with battery storage for street lighting application. Journal of cleaner production, 2025, 491.
[4] Wesonga R, Tutesigensi A, Moodley K. Advancing net zero carbon construction: A techno-economic and environmental analysis of onsite microgrids and prosumer energy adoption. Applied energy, 2025, 398.
[5] Bai WW, Fang YY, Li XY, Li WW. Additional self-disturbance rejection control strategy for battery storage in VSG-based wind power grid-connected system.New energy power control technology, 2024, 46(6).
[6] Jiang SQ, Zhang HF, Fu G, Xin YC, Wang LX. Coordinated Control Strategies for Enhancing Frequency Stability of Photovoltaic and Storage Networking Systems. Electric Power Construction, 2025, 46(8).
[7] Zhang ZW, Jiao ZH, Li YJ, Shao MY, Dai XJ. Intelligent fault diagnosis of bearings driven by double-level data fusion based on multichannel sample fusion and feature fusion under time-varying speed conditions. Reliab Eng Syst Saf 2024, 251.
[8] Xue JY, Zhang TS, Ye H. KPI-oriented process monitoring based on causal-weighted partial least squares. Inform Sciences 2025, 689.
[9] Obanya PO, Coetzer RLJ, Olivier CP, Verster T. Variable contribution analysis in multivariate process monitoring using permutation entropy. Comput Ind Eng 2024, 190.
[10] Chen C, Wang T, Lu KJ, Liu Y, Cheng LL. Compact convolutional transformers- generative adversarial network for compound fault diagnosis of industrial robot. Eng Appl Artif Intell 2024, 138.
[11] Yang C, Li Y, Chen QJ. A novel two-Stage fault-detection method based on constrained RVM and integrating LDA with minimax probability machine. IEEE Trans Ind Informat 2023, 19(3).
[12] Jin CC, Chen X. An end-to-end framework combining time-frequency expert knowledge and modified transformer networks for vibration signal classification. Expert Syst Appl 2021, 171.
[13] Dong YN, Qin SJ. A novel dynamic PCA algorithm for dynamic data modeling and process monitoring. J Process Control 2018, 67.
[14] Harrou F, Kini KR, Madakyaru M, Sun Y. Sensor fault detection and diagnosis in photovoltaic systems using Hellinger Distance and Individual Conditional Expectation analysis. Solar energy 2025, 298.
[15] Xue JY, Zhang TS, Ye H. KPI-oriented process monitoring based on causal-weighted partial least squares. Inform Sciences 2025, 689: 121470.
[16] Liu HB, Yang J, Zhang YC, Yang C. Monitoring of wastewater treatment processes using dynamic concurrent kernel partial least squares. Process Safety and Environmental Protection 2021, 147: 274-282.
[17] Su SY, Sun YC, Li LB, Peng C, Zhang H, Zhang TT. Risk warning for aircraft bleed air system with multivariate state estimation technique. J Aerosp Inform Syst 2022.
[18] Zhang S, Wang F, Tan S, Wang S, Chang Y. Novel Monitoring Strategy Combining the Advantages of the Multiple Modeling Strategy and Gaussian Mixture Model for Multimode Processes. Ind Eng Chem Res 2015, 54(47).
[19] Gao GL, Zhong YM, Gao SS, Gao BB. Double-Channel Sequential Probability Ratio Test for Failure Detection in Multisensor Integrated Systems. IEEE Trans Instrum Meas 2021, 70.
[20] Lv Y, Fang F, Yang TT, Romero CE. An early fault detection method for induced draft fans based on MSET with informative memory matrix selection. ISA Trans 2020, 102.
[21] Gao XY, Zhang Y, Zhou JF. Improved dynamic kernel PCA based on local preserving projections and its application for electric submersible pump fault diagnosis. Canadian Journal of Chemica Engineering. 2023, 101: 8.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Yiguo Guo, Zhiwei Wang, Fumu Lu, Yimu Fu, Zhiqing Wang, Jinghui Meng
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open-access article distributed under the terms of the Creative Commons Attribution CC BY 4.0 license, which permits unlimited use, distribution, and reproduction in any medium so long as the original work is properly cited.
