With the rapid development of renewable energy, the share of wind and solar energy in the energy sector is gradually increasing. However, the volatility of their generating power poses challenges to the stable operation of power grids. Wind and solar farms pose new challenges to the stable operation of the grid due to their volatility. The aim of this study is to predict the volatility of wind (wind) power generation using the Arima model so as to enhance the grid's ability to accommodate renewable resources. In this study, historical power generation data from wind farms and solar power plants in a certain region are used, and the triple standard deviation method is applied to deal with outliers, combined with the Lagrangian method to fill in the missing values, and the Arima model is built to make predictions. The results show that the model can better capture the fluctuation trend of power generation. Predictions are made 300 seconds before a significant increase or 120 seconds before a significant decrease in the PV power generation rate value, and the prediction accuracy is obtained by comparing the predicted value with the true value. The predictions show that the model has a high prediction accuracy. The study shows that the Arima model is suitable for short-term prediction of wind power generation, which can provide effective data support for grid scheduling and help to improve the stability and reliability of new energy power generation systems.

Jipeng Han, Siyuan Hao, Shijie Chen. Prediction Study of Wind Power Generation Power Based on Arima Model. International Journal of New Developments in Engineering and Society (2025), Vol.9, Issue 2: 51-56. https://doi.org/10.25236/IJNDES.2025.090209.

[1] S. Rong, L. Qiang, L. Jiankun, et al. Short-Term Wind Power Forecasting Based on Multi-Scale Fusing Deep Learning[J]. 2021 IEEE Sustainable Power and Energy Conference (iSPEC), 2021: 596-601.

[2] M. Lydia, S. Suresh Kumar, A. Immanuel Selvakumar, et al. A comprehensive review on wind turbine power curve modeling techniques[J]. Renewable and Sustainable Energy Reviews, 2014, 30: 452-460.

[3] Mantas Marčiukaitis, Inga Žutautaitė, Linas Martišauskas, et al. Non-linear regression model for wind turbine power curve[J]. Renewable Energy, 2017, 113: 732-741.

[4] Borowski, Sylwester & Migawa, Klaudiusz & Sołtysiak, et al. Regression model in the operation of wind turbines[J]. MATEC Web of Conferences, 2021, 351(01021). 

[5] P. Sivaprakash, M. Sankar, R. Chithambaramani, et al. Regression Method-Based Prediction of Wind Power Generation[J]. 2023 International Conference on System, Computation, Automation and Networking (ICSCAN), 2023: 1-6.

[6] Zahraa Tarek, Mahmoud Y. Shams, Ahmed M. Elshewey, et al. El-dosuky,Wind Power Prediction Based on Machine Learning and Deep Learning Models[J]. Computers, Materials and Continua, 2022, 74(1): 715-732.

[7] Wu, Linhan and Yu, Jizhou and Dai, Yuxin, et al. Photovoltaic Power Generation Forecasting Based on TCN-Transformer Model[J]. 2024 5th International Conference on Artificial Intelligence and Electromechanical Automation (AIEA), 2024: 620-626.

[8] X. Gong, F. Zhang, T. Lu, W. You. Comparative Analysis of three outlier detection methods in univariate data sets[J]. 2022 3rd International Conference on Electronic Communication and Artificial Intelligence (IWECAI), 2022: 209-213.

[9] P. Manembu, A. Kewo, B. Welang. Missing data solution of electricity consumption based on Lagrange Interpolation case study: IntelligEnSia data monitoring[J]. 2015 International Conference on Electrical Engineering and Informatics (ICEEI), 2015: 511-516.

[10] Tianhao Liu, Kunye Lv, Fengjie Chen, et al. Wind power curve model combining smoothed spline with first-order moments and density-adjusted wind speed strategy[J]. Energy, 2024(133628).

[11] F. Mahia, A. R. Dey, M. A. Masud, et al. Forecasting Electricity Consumption using ARIMA Model[J]. 2019 International Conference on Sustainable Technologies for Industry 4.0 (STI), 2019: 1-6.