Efficient Method for Photovoltaic Power Generation Forecasting Based on State Space Modeling and BiTCN DOI Creative Commons
Guowei Dai,

Shuai Luo,

Long‐Qing Chen

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(20), P. 6590 - 6590

Published: Oct. 13, 2024

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 (

Language: Английский

Using Crested Porcupine Optimizer Algorithm and CNN-LSTM-Attention Model Combined with Deep Learning Methods to Enhance Short-Term Power Forecasting in PV Generation DOI Creative Commons

Yiling Fan,

Zhuang Ma, Wanwei Tang

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(14), P. 3435 - 3435

Published: July 12, 2024

Due to the inherent intermittency, variability, and randomness, photovoltaic (PV) power generation faces significant challenges in energy grid integration. To address these challenges, current research mainly focuses on developing more efficient management systems prediction technologies. Through optimizing scheduling integration PV generation, stability reliability of can be further improved. In this study, a new model is introduced that combines strengths convolutional neural networks (CNNs), long short-term memory (LSTM) networks, attention mechanisms, so we call algorithm CNN-LSTM-Attention (CLA). addition, Crested Porcupine Optimizer (CPO) utilized solve problem generation. This abbreviated as CPO-CLA. first time CPO has been into LSTM for parameter optimization. effectively capture univariate multivariate series patterns, multiple relevant target variables patterns (MRTPPs) are employed CPO-CLA model. The results show superior traditional methods recent popular models terms accuracy stability, especially 13 h timestep. mechanisms enables adaptively focus most historical data future prediction. optimizes network parameters, which ensures robust generalization ability great significance establishing trust market. Ultimately, it will help integrate renewable reliably efficiently.

Language: Английский

Citations

8
Photovoltaic power prediction based on multi-scale photovoltaic power fluctuation characteristics and multi-channel LSTM prediction models DOI
Fengpeng Sun,

Longhao Li,

Dun-xin Bian

et al.

Renewable Energy, Journal Year: 2025, Volume and Issue: unknown, P. 122866 - 122866

Published: March 1, 2025

Language: Английский

Citations

0
Enhancing Latent Defect Detection in Built‐In Spindle Assembly Lines Through Vibration Data Analysis DOI Creative Commons
Kuohao Li,

Chao‐Nan Wang,

Yaochi Tang

et al.

Shock and Vibration, Journal Year: 2025, Volume and Issue: 2025(1)

Published: Jan. 1, 2025

This study proposed a novel machine learning–driven methodology for detecting potential defects in computer numerical control (CNC) spindle manufacturing. The methodology, which analyzes 13 real‐world built‐in spindles, employs t ‐distributed stochastic neighbor embedding ( ‐SNE) data visualization and enhances k ‐means++ clustering with the Davies–Bouldin Index (DBI) automatic selection of optimal number clusters, significantly surpassing traditional inspection methods identifying subtle yet critical defects. utilized fast Fourier transform (FFT) precise feature extraction. integration these advanced algorithms accurately identified categorized them, thus optimizing manufacturing processes. inclusion DBI algorithm facilitated an objective evaluation cluster quality, ensuring that selected clusters represents underlying patterns. automated value enhanced stability reliability defect detection process. substantially reduced yield defective spindles by addressing before installation CNC machines. proactive intervention system rectified failures at early stage improved overall quality approach operational efficiency reliability, rework warranty claims costs, aligned industrial needs while gap academic research. contributes to manufacturing, high‐quality production outcomes bridging important gaps both application

Language: Английский

Citations

0
Advancements and Challenges in Photovoltaic Power Forecasting: A Comprehensive Review DOI Creative Commons
Paolo Di Leo, Alessandro Ciocia, Gabriele Malgaroli

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(8), P. 2108 - 2108

Published: April 19, 2025

The fast growth of photovoltaic (PV) power generation requires dependable forecasting methods to support efficient integration solar energy into systems. This study conducts an up-to-date, systematized analysis different models and used for prediction. It begins with a new taxonomy, classifying PV according the time horizon, architecture, selection criteria matched certain application areas. An overview most popular heterogeneous techniques, including physical models, statistical methodologies, machine learning algorithms, hybrid approaches, is provided; their respective advantages disadvantages are put perspective based on tasks. paper also explores advanced model optimization methodologies; achieving hyperparameter tuning; feature selection, use evolutionary swarm intelligence which have shown promise in enhancing accuracy efficiency models. review includes detailed examination performance metrics frameworks, as well consequences weather conditions affecting renewable operational economic implications performance. highlights recent advancements field, deep architectures, incorporation diverse data sources, development real-time on-demand solutions. Finally, this identifies key challenges future research directions, emphasizing need improved adaptability, quality, computational large-scale By providing holistic critical assessment landscape, aims serve valuable resource researchers, practitioners, decision makers working towards sustainable reliable deployment worldwide.

Language: Английский

Citations

0
Distributed photovoltaic ultra-short-term power prediction using whole-sky images and multi-source data DOI

Qinlong Zhang,

Beiping Hou,

Wen Zhu

et al.

Electrical Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: April 16, 2025

Language: Английский

Citations

0
Efficient Method for Photovoltaic Power Generation Forecasting Based on State Space Modeling and BiTCN DOI Creative Commons
Guowei Dai,

Shuai Luo,

Long‐Qing Chen

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(20), P. 6590 - 6590

Published: Oct. 13, 2024

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 (

Language: Английский

Citations

2