A solar forecasting framework based on federated learning and distributed computing

Wen, Haoran; Du, Yang; Lim, Eng Gee; Wen, Huiqing; Yan, Ke; Li, Xingshuo; Jiang, Lin

doi:10.1016/j.buildenv.2022.109556

Cited by 20 publications

(8 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The existing federal learning framework is based on the unconditional participation of all mobile devices in federal learning, but in fact, each mobile device will generate corresponding training costs in the training process [33] , [34] . To tackle the existing problems of FL, several extensions of the FL framework were proposed in recent years [35] , [36] , [37] . The current work shows another elegant extension of the existing FL framework with a de-centralized edge network and edge devices for sentiment analysis combating COVID-19.…”

Section: Literature Reviewmentioning

confidence: 99%

Federal learning edge network based sentiment analysis combating global COVID-19

Liang

Chen

Huang

et al. 2023

Computer Communications

Self Cite

View full text Add to dashboard Cite

Section: Literature Reviewmentioning

confidence: 99%

Federal learning edge network based sentiment analysis combating global COVID-19

Liang

Chen

Huang

et al. 2023

Computer Communications

Self Cite

View full text Add to dashboard Cite

“…To address the problem of the large datasets required for PV prediction, Haoran Wen used federal learning to aggregate historical PV data from various locations using four training strategies with much higher prediction performance [26]. Huaizhi Wang adapted a neural network to provide a clear interpretation of the relationship between prediction model inputs and outputs for PV prediction [27].…”

Section: Related Workmentioning

confidence: 99%

“…Alternatively, a combination of different models can be used. The state of the weather has an extremely important impact on the efficiency of solar power production, mainly solar irradiance and temperature [18], and as such can be divided into two main categories based on weather conditions: direct prediction methods for PV power generation [19][20][21][22][23][24][25][26][27][28], and indirect prediction methods that predict the solar irradiance in order to derive PV power generation. Datasets for solar energy forecasting mainly consist of time series, as weather conditions are strongly correlated with time [19].…”

Section: Introductionmentioning

confidence: 99%

An Improved Method for Photovoltaic Forecasting Model Training Based on Similarity

et al. 2023

View full text Add to dashboard Cite

Photovoltaic (PV) power generation is the most widely adopted renewable energy source. However, its inherent unpredictability poses considerable challenges to the management of power grids. To address the arduous and time-consuming training process of PV prediction models, which has been a major focus of prior research, an improved approach for PV prediction based on neighboring days is proposed in this study. This approach is specifically designed to handle the preprocessing of training datasets by leveraging the results of a similarity analysis of PV power generation. Experimental results demonstrate that this method can significantly reduce the training time of models without sacrificing prediction accuracy, and can be effectively applied in both ensemble and deep learning approaches.

show abstract

“…This approach also allows centers to control the information shared and increase the privacy and security of sensitive data. The broad applications of federated learning have recently been extended to the weather forecast [24] and air quality control using historical data and edge devices [25]. In this paper, We model a neural network on tabular weather data collected by the Bureau of Meteorology in Australia to predict rain in a centralized and federated learning framework.…”

Section: Introductionmentioning

confidence: 99%

Federated Learning with GANs-based Synthetic Minority Over-sampling Technique for Improving Weather Prediction from Imbalanced Data

Jafarigol

Trafali̇s

2023

Preprint

View full text Add to dashboard Cite

Detecting rare events is a challenging task among machine learning practitioners, motivating them to navigate and further improve data processing and algorithmic approaches to find accurate and computationally efficient methods for imbalanced learning. Imbalanced data is common in weather prediction, where the massive size of data poses storage and computational challenges. To learn from imbalanced data, the algorithms must strive to learn each class precisely to be able to classify both the minority and the majority class. While creating complex models is somewhat necessary for such problems, the machine learning algorithm must have adequate generalization capabilities. In recent years, federated learning has been used as an environmentally-friendly approach, which can produce accurate results in a distributed setting. Although privacy is not a concern of this study, inherent characteristics of federated learning are beneficial in weather prediction. Deep learning combined with data augmentation is explored in the framework of federated learning. We compare multiple data augmentation methods in a centralized and federated learning framework. Federated learning has great potential for weather prediction tasks. By leveraging data from multiple sources, we were able to improve the accuracy and generalization of the classifier. Addressing the issue of imbalanced data is an essential step. Incorporating the two remarkable approaches of Generative Adversarial Networks and Synthetic Minority Over-sampling Technique is a suitable solution for tabular weather data, and our experimental results confirm the effectiveness of this approach.

show abstract

A solar forecasting framework based on federated learning and distributed computing

Cited by 20 publications

References 35 publications

Federal learning edge network based sentiment analysis combating global COVID-19

Federal learning edge network based sentiment analysis combating global COVID-19

An Improved Method for Photovoltaic Forecasting Model Training Based on Similarity

Federated Learning with GANs-based Synthetic Minority Over-sampling Technique for Improving Weather Prediction from Imbalanced Data

Contact Info

Product

Resources

About