A combined model for short-term wind power forecasting based on the analysis of numerical weather prediction data

He, Boyu; Pei, Ming; Lu, Peng; Dai, Binhua; Li, Zhuo; Wang, Kaifeng

doi:10.1016/j.egyr.2021.10.102

Cited by 44 publications

(16 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The direct influence of wind speed on the energy power curve of wind turbines and the stochastic and intermittent nature of wind have made its prediction a recurring research topic [14]. According to the type of techniques used, metocean (meteorological and oceanographic) variables forecasting models can be classified into (1) naive, (2) physical, (3) statistical and (4) intelligent models [15,16].…”

Section: Related Workmentioning

confidence: 99%

Use of State-of-Art Machine Learning Technologies for Forecasting Offshore Wind Speed, Wave and Misalignment to Improve Wind Turbine Performance

Sacie

Santos

López

et al. 2022

JMSE

View full text Add to dashboard Cite

One of the most promising solutions that stands out to mitigate climate change is floating offshore wind turbines (FOWTs). Although they are very efficient in producing clean energy, the harsh environmental conditions they are subjected to, mainly strong winds and waves, produce structural fatigue and may cause them to lose efficiency. Thus, it is imperative to develop models to facilitate their deployment while maximizing energy production and ensuring the structure’s safety. This work applies machine learning (ML) techniques to obtain predictive models of the most relevant metocean variables involved. Specifically, wind speed, significant wave height, and the misalignment between wind and waves have been analyzed, pre-processed and modeled based on actual data. Linear regression (LR), support vector machines regression (SVR), Gaussian process regression (GPR) and neural network (NN)-based solutions have been applied and compared. The results show that Nonlinear autoregressive with an exogenous input neural network (NARX) is the best algorithm for both wind speed and misalignment forecasting in the time domain (72% accuracy) and GPR for wave height (90.85% accuracy). In conclusion, these models are vital to deploying and installing FOWTs and making them profitable.

show abstract

Section: Related Workmentioning

confidence: 99%

Use of State-of-Art Machine Learning Technologies for Forecasting Offshore Wind Speed, Wave and Misalignment to Improve Wind Turbine Performance

Sacie

Santos

López

et al. 2022

JMSE

View full text Add to dashboard Cite

show abstract

“…The datasets messiness evokes a separate research subfield, data pre-processing, and various pre-processing techniques are used, for example, in order to eliminate outliers 159 or classify different weather types. 160 None of the datasets that are presented in Table 2 is described as being pre-processed. We, therefore, assume that the datasets contain non-processed data.…”

Section: Data Qualitymentioning

confidence: 99%

“…However, incomplete and noisy data usually represent the data measured in the real world better than consistent and complete datasets, although we usually attribute a higher quality to the latter. The datasets messiness evokes a separate research subfield, data pre‐processing, and various pre‐processing techniques are used, for example, in order to eliminate outliers 159 or classify different weather types 160 . None of the datasets that are presented in Table 2 is described as being pre‐processed.…”

Section: Data Qualitymentioning

confidence: 99%

A collection and categorization of open‐source wind and wind power datasets

Effenberger

Ludwig

2022

Wind Energy

View full text Add to dashboard Cite

Wind power and other forms of renewable energy sources play an ever more important role in the energy supply of today's power grids. Forecasting renewable energy sources has therefore become essential in balancing the power grid. While a lot of focus is placed on new forecasting methods, little attention is given on how to compare, reproduce and transfer the methods to other use cases and data. One reason for this lack of attention is the limited availability of open‐source datasets, as many currently used datasets are non‐disclosed and make reproducibility of research impossible. This unavailability of open‐source datasets is especially prevalent in commercially interesting fields such as wind power forecasting. However, with this paper, we want to enable researchers to compare their methods on publicly available datasets by providing the, to our knowledge, largest up‐to‐date overview of existing open‐source wind power datasets, and a categorization into different groups of datasets that can be used for wind power forecasting. We show that there are publicly available datasets sufficient for wind power forecasting tasks and discuss the different data groups properties to enable researchers to choose appropriate open‐source datasets and compare their methods on them.

show abstract

“…Vladislavleva et al (2013), (Wu et al, 2021), and (Taylor et al, 2009) used output predictions based on weather data to analyze relevant parameters and their correlation with energy output. He et al (2022) proposed a short-term WPF model based on NWP analysis (He et al, 2022). Several factors were selected from the NWP multivariate data by using the criteria of the minimum redundancy maximum correlation (MRMR) algorithm, and the weather patterns were divided into different types according to these characteristics.…”

Section: Introductionmentioning

confidence: 99%

A development potential assessment method for clean energy stations

Chao¹,

Wu²,

Li³

et al. 2022

Front. Energy Res.

View full text Add to dashboard Cite

Clean energy is expected to enter a new stage of large-scale development along with the growing demand for building regional clean energy stations. However, as many regional clean energy stations comprise multiple stations with different output characteristics and complementary coupling, the development potential of these stations cannot be simply based on the superposition of outputs, as this method lacks reasonable assessment results. This study proposes a method of combining Grey relational analysis (GRA), artificial neural network (ANN), and XGBoost algorithm for the potential assessment of clean energy stations. First, GRA and ANN are used for the relational analysis between the output of clean energy stations and meteorological factors. Second, the meteorological factors with high correlation and the existing historical data are used to predict the future outputs of new clean energy stations via XGBoost. Finally, according to the predicted output, an assessment method that includes available capacity coefficient (AOC) and other evaluation indicators is proposed. The case studies in this research prove the effectiveness and applicability of the proposed method.

show abstract

A combined model for short-term wind power forecasting based on the analysis of numerical weather prediction data

Cited by 44 publications

References 27 publications

Use of State-of-Art Machine Learning Technologies for Forecasting Offshore Wind Speed, Wave and Misalignment to Improve Wind Turbine Performance

Use of State-of-Art Machine Learning Technologies for Forecasting Offshore Wind Speed, Wave and Misalignment to Improve Wind Turbine Performance

A collection and categorization of open‐source wind and wind power datasets

A development potential assessment method for clean energy stations

Contact Info

Product

Resources

About