Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization

Ning Zhou; Bowen Shang; Mingming Xu; Lei Peng; Guang Feng

doi:10.1016/j.gloei.2024.10.005

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization

Clean Energy Development | 更新时间：2026-10-05

- Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization
- Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization
- 全球能源互联网(英文) 2024年7卷第5期页码：667-681
- 作者机构：
  
  1. State Grid Henan Electric Power Research Institute,Zhengzhou,PR China,450052
  2. State Grid Henan Electric Power Corporation,Zhengzhou,PR China,450052
- 作者简介：
- 基金信息：
- DOI：10.1016/j.gloei.2024.10.005
  中图分类号：
- 网络首发：2024-11-07，
  
  纸质出版：2024-11-07
- 稿件说明：
移动端阅览
Ning Zhou, Bowen Shang, Mingming Xu, 等. Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization[J]. 全球能源互联网(英文), 2024,7(5):667-681.

Ning Zhou, Bowen Shang, Mingming Xu, et al. Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization[J]. Global Energy Interconnection, 2024, 7(5): 667-681.
Ning Zhou, Bowen Shang, Mingming Xu, 等. Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization[J]. 全球能源互联网(英文), 2024,7(5):667-681. DOI： 10.1016/j.gloei.2024.10.005.

Ning Zhou, Bowen Shang, Mingming Xu, et al. Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization[J]. Global Energy Interconnection, 2024, 7(5): 667-681. DOI： 10.1016/j.gloei.2024.10.005.

摘要

Abstract

Improving the accuracy of solar power forecasting is crucial to ensure grid stability

optimize solar power plant operations

and enhance grid dispatch efficiency. Although hybrid neural network models can effectively address the complexities of environmental data and power prediction uncertainties

challenges such as labor-intensive parameter adjustments and complex optimization processes persist. Thus

this study proposed a novel approach for solar power prediction using a hybrid model (CNN-LSTM-attention) that combines a convolutional neural network (CNN)

long short- term memory (LSTM)

and attention mechanisms. The model incorporates Bayesian optimization to refine the parameters and enhance the prediction accuracy. To prepare high-quality training data

the solar power data were first preprocessed

including feature selection

data cleaning

imputation

and smoothing. The processed data were then used to train a hybrid model based on the CNN-LSTM-attention architecture

followed by hyperparameter optimization employing Bayesian methods. The experimental results indicated that within acceptable model training times

the CNN-LSTM-attention model outperformed the LSTM

GRU

CNN-LSTM

CNN-LSTM with autoencoders

and parallel CNN-LSTM attention models. Furthermore

following Bayesian optimization

the optimized model demonstrated significantly reduced prediction errors during periods of data volatility compared to the original model

as evidenced by MRE evaluations. This highlights the clear advantage of the optimized model in forecasting fluctuating data.

关键词

Keywords

references

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

Two-stage photovoltaic power forecasting method with an optimized transformer

Bayesian optimized support vector regression with a Gaussian kernel for accurate prediction of the state of health of lithium-ion batteries used for electric vehicle applications

Negative electricity prices in electricity markets: causes, impacts,and response strategies

EV＄{7*7}AL

Distribution network data asset protection considering multiple topology and multi-dimensional knowledge graph