Construction of artificial intelligence load forecasting model for new energy grid connection and regulation strategy for stable operation of power grid
Xiaoping Chen
Communication University of China, Nanjing
Mingyu Wei
Communication University of China, Nanjing
Shiqian Luo
Communication University of China, Nanjing
DOI: https://doi.org/10.59429/esta.v13i2.14558
Keywords: new energy grid-connection; artificial intelligence load forecasting; power grid stability regulation
Abstract
Aiming at the prominent load fluctuation, insufficient accuracy of traditional forecasting methods and great difficulties in stable power grid regulation under new energy grid-connection, this paper constructs an artificial intelligence load forecasting model and proposes supporting regulation strategies. The Long Short-Term Memory (LSTM) network is adopted to adapt to nonlinear time-series features, the 3σ criterion is applied for data preprocessing, and the Particle Swarm Optimization (PSO) algorithm is introduced to optimize the key parameters of the model so as to improve forecasting accuracy. The comparison results show that the PSO-LSTM model has lower errors and better fitting performance than the single LSTM model. Based on high-precision forecasting results, a coordinated regulation system of source-grid-load-storage as well as a voltage and frequency stability regulation system is established, which can effectively suppress new energy fluctuations and reduce operation risks. It provides technical support for accurate forecasting and safe and stable operation of distribution power grids with high penetration of new energy.
References
[1] Dong Q, Huang R, Cui C, et al. Short-Term Electricity-Load Forecasting by deep learning: A comprehensive survey[J]. Engineering Applications of Artificial Intelligence, 2025, 154110980-110980.
[2] Li L, Ma W, Liu C. Optimization of power load forecasting based on big data and artificial intelligence: Enhancing power system stability and operational efficiency through a systematic study[J]. Advances in Resources Research, 2025, 5(2): 645-665.
[3] Li X, Zhao H, Feng Y, et al. Research on key technologies of high energy efficiency and low power consumption of new data acquisition equipment of power Internet of Things based on artificial intelligence[J]. International Journal of Thermofluids, 2024, 21100575-.
[4] Dankir S, Puig V, Lasri R, et al. Empowering Microgrid Energy Management with Artificial Intelligence and Model Predictive Control[J]. IFAC PapersOnLine, 2024, 58(13): 436-441.