Hybrid deep learning models for time series forecasting of solar power (2024-02-22)

TL;DR

This paper introduces and investigates novel hybrid deep learning models for solar power forecasting using time series data and shows that the CNN–LSTM–TF hybrid model outperforms the other models, with a mean absolute error (MAE) of 0.551%.

Abstract

Forecasting solar power production accurately is critical for effectively planning and managing renewable energy systems. This paper introduces and investigates novel hybrid deep learning models for solar power forecasting using time series data. The research analyzes the efficacy of various models for capturing the complex patterns present in solar power data. In this study, all of the possible combinations of convolutional neural network (CNN), long short-term memory (LSTM), and transformer (TF) models are experimented. These hybrid models also compared with the single CNN, LSTM and TF models with respect to different kinds of optimizers. Three different evaluation metrics are also employed for performance analysis. Results show that the CNN–LSTM–TF hybrid model outperforms the other models, with a mean absolute error (MAE) of 0.551% when using the Nadam optimizer. However, the TF–LSTM model has relatively low performance, with an MAE of 16.17%, highlighting the difficulties in making reliable predictions of solar power. This result provides valuable insights for optimizing and planning renewable energy systems, highlighting the significance of selecting appropriate models and optimizers for accurate solar power forecasting. This is the first time such a comprehensive work presented that also involves transformer networks in hybrid models for solar power forecasting.

Authors

Diaa Salman

1 Paper

Cem Direkoglu

1 Paper

M. Kuşaf

1 Paper

TL;DR

Abstract

Authors

References36 items

Attention is All you Need

A novel long term solar photovoltaic power forecasting approach using LSTM with Nadam optimizer: A case study of India

Forecasting Solar PV Output Using Convolutional Neural Networks with a Sliding Window Algorithm

Forecasting Renewable Energy Generation with Machine Learning and Deep Learning: Current Advances and Future Prospects

Designing Solar Power Generation Output Forecasting Methods Using Time Series Algorithms: Global Warming Affected Weather Conditions

Solar Energy Production Forecasting Based on a Hybrid CNN-LSTM-Transformer Model

Hybrid deep learning models for multivariate forecasting of global horizontal irradiation

Short-Term Load Forecasting Based on the Transformer Model

Photovoltaic power prediction of LSTM model based on Pearson feature selection

Transformer based prediction method for solar power generation data

Deep learning models for solar irradiance forecasting: A comprehensive review

A Battle of Network Structures: An Empirical Study of CNN, Transformer, and MLP

A Review on Deep Learning Models for Forecasting Time Series Data of Solar Irradiance and Photovoltaic Power

Time series forecasting of solar power generation for large-scale photovoltaic plants

Time-series well performance prediction based on Long Short-Term Memory (LSTM) neural network model

Deep Transformer Models for Time Series Forecasting: The Influenza Prevalence Case

Long-term Solar Radiation Forecasting using a Deep Learning Approach-GRUs

Understanding LSTM - a tutorial into Long Short-Term Memory Recurrent Neural Networks

Multiple convolutional neural networks for multivariate time series prediction

Sunshine Duration Prediction Using 1D Convolutional Neural Networks

Day-Ahead Solar Irradiance Forecasting for Microgrids Using a Long Short-Term Memory Recurrent Neural Network: A Deep Learning Approach

1D Convolutional Neural Networks and Applications: A Survey

A Blind Spectrum Sensing Method Based on Deep Learning

Statistical and Machine Learning forecasting methods: Concerns and ways forward

Photovoltaic and solar power forecasting for smart grid energy management

Time series prediction for output of multi-region solar power plants

Photovoltaic Power Forecasting With a Hybrid Deep Learning Approach

Evaluation Metrics and Evaluation

Modelling atmospheric ozone concentration using machine learning algorithms

Historical data in simulation (Data in France)

2023) Solar power forecasting using hybrid model

LSTM models are primarily developed to capture long-term dependencies in historical data. They have cells with memories that can store and retrieve information over long periods

The TF model has a self-attention feature that lets it pick up on global dependencies and connections in the sequence of inputs

2022 State-of-the-art solar energy forecasting approaches: critical potentials and challenges

When dealing with large-scale time series data or deep architectures,

for them to model temporal dependencies explicitly

Field of Study

Journal Information

Name

Volume