Against the backdrop of high-penetration renewable energy integration and increasingly intelligent power communication infrastructures, electricity spot market prices exhibit complex nonlinear fluctuations that directly affect the operational costs and scheduling efficiency of energy-intensive industrial systems. Accurate price forecasting is therefore essential for coordinated energy management and reliable information transmission in modern smart grids. This paper proposes a time-series prediction algorithm integrating fractal analysis and deep learning to capture both long-term memory characteristics and local abrupt variations in electricity price sequences. The method first quantifies long-range correlations using the Hurst exponent derived from rescaled range analysis, then extracts multi-scale fluctuation features through multifractal detrended fluctuation analysis, and finally embeds fractal characteristics into the loss function of a Long Short-Term Memory (LSTM) network to construct a fractal-aware prediction model. A dualwindow strategy is adopted to reconcile the statistical requirements of fractal estimation with temporal responsiveness. Experimental results based on actual clearing data from five domestic spot market pilot regions demonstrate that the proposed model consistently outperforms conventional ARIMA, standard LSTM, VMD-LSTM, CEEMDAN-BERT-LSTM, and Transformer-based methods in both MAPE and RMSE. The proposed framework provides an effective decisionsupport tool for intelligent energy management and offers potential reference value for signal-aware forecasting and information processing in advanced electromagnetic and communication-enabled power systems.
M. Song, C. Su, G. Gao et al.· Advanced Electromagnetics· 0 citations
The intermittent and volatile characteristics of new energy generation, together with the increasing demand for stable power supply in intelligent industrial systems, make accurate forecasting a critical issue for grid dispatch and electromagnetic energy management. This study systematically reviews the technological evolution of time series analysis methods for wind and photovoltaic (PV) power forecasting and establishes a comparative framework covering classical statistical models, intelligent learning algorithms, and hybrid modeling strategies. Based on two years of operational data collected from an actual wind farm and PV station in East China, the forecasting performance of ARIMA, exponential smoothing, Support Vector Regression (SVR), Long Short-Term Memory (LSTM) networks, and Transformer architectures is comprehensively evaluated, while hybrid approaches based on Empirical Mode Decomposition (EMD) are further investigated. The results demonstrate that model selection should jointly consider forecasting horizon, data characteristics, and computational constraints. Classical statistical methods remain robust under stable operating conditions but are less effective in capturing extreme fluctuations, whereas deep learning approaches exhibit superior capability in modeling long-range temporal dependencies despite reduced interpretability. Decomposition-based hybrid strategies achieve a more balanced performance across diverse scenarios and show enhanced robustness under extreme weather conditions. The study further proposes a structured model selection guideline by matching data characteristics with operational requirements, providing theoretical support for forecasting system design in renewable-energy-driven power networks and offering useful references for electromagnetic energy utilization and intelligent industrial applications.
M. Song, C. Yang, Z. Heng et al.· Advanced Electromagnetics· 0 citations
