Efficient Method for Photovoltaic Power Generation Forecasting Based on State Space Modeling and BiTCN

• DOI: 10.3390/s24206590
• Publication Date: 2024-10-16T11:58:32Z
• Authors: Guowei Dai; Shuai Luo; Hu Chen; Yulong Ji
• Keywords: time series prediction; intelligent fusion; state space model; Chemical technology; photovoltaic power forecasting; deep learning; TP1-1185; Article

As global carbon reduction initiatives progress and the new energy sector rapidly develops, photovoltaic (PV) power generation is playing an increasingly significant role in renewable energy. Accurate PV output forecasting, influenced by meteorological factors, essential for efficient management. This paper presents optimal hybrid forecasting strategy, integrating bidirectional temporal convolutional networks (BiTCN), dynamic convolution (DC), long short-term memory (BiLSTM), a novel mixed-state space model (Mixed-SSM). The mixed-SSM combines state (SSM), multilayer perceptron (MLP), multi-head self-attention mechanism (MHSA) to capture complementary temporal, nonlinear, long-term features. Pearson Spearman correlation analyses are used select features strongly correlated with output, improving prediction coefficient (