Impact of Forecast Time-Step on PV Production Accuracy Using Machine Learning for Micro-Grid Efficiency

P.R. Sudha Rani; S. Ajit; M. Shunmugasundaram; S. Gangadharan; Swetha Mareddy; R. Senthamil Selvan; M. Prabha

doi:10.63278/10.63278/mme.v31.1

Authors

P.R. Sudha Rani Professor, Department of CSE, Shri Vishnu Engineering College for Women, India
S. Ajit Professor, Department of MBA, St. Joseph's College of Engineering, India
M. Shunmugasundaram Assistant Professor, Department of MBA, St Joseph's College of Engineering, India
S. Gangadharan Associate Professor, Department of Master of Business Administration, St. Joseph's College of Engineering OMR, India
Swetha Mareddy Department of Electrical and Electronics Engineering, Assistant professor, Annamacharya University, India
R. Senthamil Selvan Associate Professor, Department of ECE, Annamacharya Institute of Technology and Sciences, India
M. Prabha Assistant Professor, Department of ECE, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, India

DOI:

https://doi.org/10.63278/10.63278/mme.v31.1

Keywords:

Micro-Grid, PV Generation, Accuracy Forecasting, Machine Learning, Energy Control System.

Abstract

Efficient energy management solutions are becoming more important with emergence of micro-grids that include photovoltaic generation & storage. Their prediction of energy output over the near to long term is an important part of their work. An essential parameter influencing the forecast's accuracy, optimum control time discretisation, efficiency, and computing load is the forecast time-step. This trade-off is measured by putting four machine learning (ML) forecast methods through their paces on two different sites, with time-steps ranging from 2 to 60 minutes as well as horizons from 10 minutes to 6 hours. The methods are evaluated on both horizontal and tilted global irradiance charts, depending on the availability of data. All of the methods show comparable findings, which show that for predictions less than an hour and between one and six hours, the error measure may be decreased by up to 1.9% every minute on the time-step, and by up to 2.8% every ten minutes. Additionally, it is demonstrated that for short-term horizons, it could be beneficial towards make high-resolution forecasts & then average results at time-step required by energy control scheme.

References

Patty, Sukriti, and Tanmoy Malakar. "Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty." Results in Control and Optimization 15 (2024): 100407.

Wazirali, Raniyah, et al. "State-of-the-art review on energy and load forecasting in microgrids using artificial neural networks, machine learning, and deep learning techniques." Electric power systems research 225 (2023): 109792.

Buonanno, Amedeo, et al. "Machine learning and weather model combination for pv production forecasting." Energies 17.9 (2024): 2203.

Ajaegbu, David Chukwudi. Forecasting PV-Diesel Microgrid Campus Load Using Machine Learning: The University of the Free State QwaQwa Campus Microgrid. MS thesis. Norwegian University of Life Sciences, 2024.

Mbuya, Benson. Short-term forecast techniques for energy management systems in microgrid applications. Diss. NM-AIST, 2023.

Bouquet, Pierre, et al. "AI-based forecasting for optimised solar energy management and smart grid efficiency." International Journal of Production Research 62.13 (2024): 4623-4644.

Kumar, T. Rajasanthosh, G. Laxmaiah, and S. Solomon Raj. "A Framework of Intelligent Manufacturing Process by Integrating Various Function." AI-Driven IoT Systems for Industry 4.0. CRC Press 241-254.

Notton, Gilles, et al. "Profitability and performance improvement of smart photovoltaic/energy storage microgrid by integration of solar production forecasting tool." Intelligent Learning Approaches for Renewable and Sustainable Energy. Elsevier, 2024. 73-102.

Mewada, Shivlal, et al. "Smart diagnostic expert system for defect in forging process by using machine learning process." Journal of Nanomaterials 2022.1 (2022): 2567194.

Mohamad Radzi, Putri nor Liyana, et al. "Review on the application of photovoltaic forecasting using machine learning for very short-to long-term forecasting." Sustainability 15.4 (2023): 2942.

Panda, Subrat Prasad, et al. "Methods for mitigating uncertainty in real-time operations of a connected microgrid." Sustainable Energy, Grids and Networks 38 (2024): 101334.

Dong, Weichao, et al. "Forecast-driven stochastic optimization scheduling of an energy management system for an isolated hydrogen microgrid." Energy Conversion and Management 277 (2023): 116640.

Mansourlakouraj, Mohammad, et al. "multi-stage volt/var support in distribution grids: Risk-aware scheduling with real-time reinforcement learning control." IEEE Access 11 (2023): 54822-54838.

Aziz, Mujtaba. Time Series Forecasting of Reactive Power Support from Smart Converters in SDN Using Machine Learning. MS thesis. UiT Norges arktiske universitet, 2023.

Elazab, Rasha, et al. "Reviewing the frontier: modeling and energy management strategies for sustainable 100% renewable microgrids." Discover Applied Sciences 6.4 (2024): 168.

Zaben, Muiz M., et al. "Machine learning methods for fault diagnosis in ac microgrids: A systematic review." IEEE access 12 (2024): 20260-20298.

Das, Avijit, Zhen Ni, and Xiangnan Zhong. "Microgrid energy scheduling under uncertain extreme weather: Adaptation from parallelized reinforcement learning agents." International Journal of Electrical Power & Energy Systems 152 (2023): 109210.

Gokulraj, K., and C. B. Venkatramanan. "Advanced Machine Learning-Driven Security and Anomaly Identification in Inverter-Based Cyber-Physical Microgrids." Electric Power Components and Systems (2024): 1-18.

Yaghoubi, Elnaz, et al. "A Real-Time and Online Dynamic Reconfiguration against Cyber-Attacks to Enhance Security and Cost-Efficiency in Smart Power Microgrids Using Deep Learning." Technologies 12.10 (2024): 197.

Jannah, Nurul, et al. "Recent Advances and Future Challenges of Solar Power Generation Forecasting." IEEE Access (2024).