Addressing multi-factor coupling, progressive degradation, and time-varying operating conditions in substation equipment under high-penetration renewable energy integration, this paper proposes a Time-varying Hidden Markov Model (T-HMM) for transformer fault diagnosis using multi-source heterogeneous data. Unlike conventional HMM with fixed transition matrices and initial parameter sensitivity, the proposed framework introduces a forgetting-factor-driven online transition matrix updating mechanism, enabling adaptive state tracking under varying operating conditions. A multi-dimensional feature system is constructed incorporating fundamental state, coupling correlation, and temporal evolution characteristics, with state-dependent sliding window standardization and adaptive wavelet denoising to enhance early-stage fault discriminability. Parameter optimization employs pre-clustering and multi-start strategies to circumvent local optima in the Baum–Welch algorithm, while an improved decoding strategy achieves real-time health state identification and multi-step probability prediction. Experimental results demonstrate that the proposed method accurately tracks state evolution trends and effectively identifies fault categories, achieving significantly superior state recognition accuracy and multi-step prediction hit rates compared to conventional HMM, with substantially reduced mean absolute error. The method effectively captures subtle precursors during normal-to-fault transitions, providing reliable theoretical foundations for early fault warning and condition-based maintenance.
Wei Li, Shanyun Gu, Lei Shen et al.· Energies· 0 citations
: The accelerating penetration of variable renewable generation into electrical grids necessitates advanced forecasting paradigms capable of addressing stochastic intermittency challenges. This paper engineers a novel hybrid intelligent framework—integrating Improved Lotus Effect Algorithm (ILEA), Variational Mode Decomposition (VMD), and ensemble deep learning—specifically designed for ultra-short-term wind power prediction in energy dispatch applications. Unlike conventional approaches relying on static parameterization, the proposed methodology employs elite chaotic opposition-based learning to autonomously optimize VMD decomposition levels, thereby decoupling non-stationary wind power sequences into stationary sub-components without empirical intervention. Local temporal feature extraction is subsequently performed via one-dimensional Convolutional Neural Networks (1D-CNN), wherein sliding convolutional kernels operate along the temporal axis to capture localized patterns—such as ramp events, gradient transitions, and short-term fluctuations—embedded within the univariate wind power sequence. Temporal dependencies are subsequently modeled through Bidirectional Long Short-Term Memory (BiLSTM) networks enhanced with attention mechanisms. An Adaptive Boosting (AdaBoost) ensemble strategy further aggregates multiple weak regressors to fortify prediction robustness against ramp events and turbulent meteorological conditions. Comprehensive validation utilizing Belgian grid operational data demonstrates that the proposed architecture achieves substantial error reductions, attaining Root Mean Square Error (RMSE) of 2.2496 MW, Mean Absolute Error (MAE) of 1.9897 MW, and Symmetric Mean Absolute Percentage Error (SMAPE) of 5.26%, with a Coefficient of Determination (R) 2 coefficient approaching unity (0.9962). These results underscore the framework’s practical efficacy for grid stabilization, load frequency control, and energy management decision support systems operating under high renewable penetration scenarios.
Lei Shen, Qifeng Xiang, Q. Gao et al.· Energy Engineering· 0 citations