The ultra-short term power prediction of wind farm considering operational condition of wind turbines

Fang, Ruiming; Wang, Yandong; Shang, Rong; Yin, Liang; Wang, Li; Peng, Changqing

doi:10.1016/j.ijhydene.2016.03.173

Cited by 27 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prisco and Duitama [95] also proposed OCSVM for detecting intrusions on SCADA network. In a related work, Fang et al [96] modelled a support vector regression for predicting a SCADA monitoring data. Terai et al [97] and Waghmare et al [73] developed SVM models for detecting intrusions and achieved remarkable results.…”

Section: Support Vector Machine (Svm)mentioning

confidence: 99%

A Review of Research Works on Supervised Learning Algorithms for SCADA Intrusion Detection and Classification

et al. 2021

View full text Add to dashboard Cite

Supervisory Control and Data Acquisition (SCADA) systems play a significant role in providing remote access, monitoring and control of critical infrastructures (CIs) which includes electrical power systems, water distribution systems, nuclear power plants, etc. The growing interconnectivity, standardization of communication protocols and remote accessibility of modern SCADA systems have contributed massively to the exposure of SCADA systems and CIs to various forms of security challenges. Any form of intrusive action on the SCADA modules and communication networks can create devastating consequences on nations due to their strategic importance to CIs’ operations. Therefore, the prompt and efficient detection and classification of SCADA systems intrusions hold great importance for national CIs operational stability. Due to their well-recognized and documented efficiencies, several literature works have proposed numerous supervised learning techniques for SCADA intrusion detection and classification (IDC). This paper presents a critical review of recent studies whereby supervised learning techniques were modelled for SCADA intrusion solutions. The paper aims to contribute to the state-of-the-art, recognize critical open issues and offer ideas for future studies. The intention is to provide a research-based resource for researchers working on industrial control systems security. The analysis and comparison of different supervised learning techniques for SCADA IDC systems were critically reviewed, in terms of the methodologies, datasets and testbeds used, feature engineering and optimization mechanisms and classification procedures. Finally, we briefly summarized some suggestions and recommendations for future research works.

show abstract

Section: Support Vector Machine (Svm)mentioning

confidence: 99%

A Review of Research Works on Supervised Learning Algorithms for SCADA Intrusion Detection and Classification

et al. 2021

View full text Add to dashboard Cite

show abstract

“…A large number of researchers have conducted research on ultra short term wind power prediction. There are three main methods for predicting ultra short term wind power: statistical methods, physical methods, and artificial intelligence methods [13][14][15]. The ultra short term wind power prediction based on statistical methods establishes statistical models based on historical wind speed and power data, such as time series analysis, regression analysis, etc.…”

Section: Introductionmentioning

confidence: 99%

An Ultra-Short-Term Wind Power Forecasting Model Based on EMD-EncoderForest-TCN

Sun,

Yang,

Zhang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Accurate wind power prediction helps to stabilize the operation of the power system, improve the utilization rate of renewable energy, reduce dependence on traditional energy, and achieve sustainable energy development. An ultra short-term wind power prediction method integrating EMD-EncoderForest-TCN is proposed to address the difficulty of predicting wind power due to frequent changes in wind speed. Firstly, the time-series input data of the model is decomposed into high-frequency and low-frequency components using Empirical Mode Decomposition. Then, based on the EncoderForest model and TCN model, differential information extraction is performed on the low-frequency and high-frequency components. The EncoderForest model regularizes low-frequency information and captures trend patterns in the data. The TCN model models the high-frequency components of time series to capture complex patterns and structures in wind power. Finally, based on convolutional neural networks, the output results of each part are calculated to achieve accurate prediction of wind power. Based on the operational data of an actual wind farm, conduct a case study analysis. The results show that the proposed model can achieve accurate prediction of short-term wind power, with a prediction accuracy improvement of 2.57%.

show abstract

“…Recently, there has also been a surge in the historical data available for modern wind farms, making machine learning (ML) methods gain traction, as these can significantly reduce runtimes after training, while providing accurate results. Various ML methods have been studied, such as support vector regressors [6], fuzzy logic [7] and k-nearest neighbour algorithms [8], with a plethora of studies focusing on deep learning (DL) [9]. Yan et al [10] used a multilayer perceptron (MLP) model to predict power at different wind speeds and directions for a real wind farm.…”

Section: Introductionmentioning

confidence: 99%

Probabilistic Wind Park Power Prediction using Bayesian Deep Learning and Generative Adversarial Networks

Bentsen

Warakagoda²,

Stenbro³

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

The rapid depletion of fossil-based energy supplies, along with the growing reliance on renewable resources, has placed supreme importance on the predictability of renewables. Research focusing on wind park power modelling has mainly been concerned with point estimators, while most probabilistic studies have been reserved for forecasting. In this paper, a few different approaches to estimate probability distributions for individual turbine powers in a real off-shore wind farm were studied. Two variational Bayesian inference models were used, one employing a multilayered perceptron and another a graph neural network (GNN) architecture. Furthermore, generative adversarial networks (GAN) have recently been proposed as Bayesian models and was here investigated as a novel area of research. The results showed that the two Bayesian models outperformed the GAN model with regards to mean absolute errors (MAE), with the GNN architecture yielding the best results. The GAN on the other hand, seemed potentially better at generating diverse distributions. Standard deviations of the predicted distributions were found to have a positive correlation with MAEs, indicating that the models could correctly provide estimates on the confidence associated with particular predictions.

show abstract

The ultra-short term power prediction of wind farm considering operational condition of wind turbines

Cited by 27 publications

References 10 publications

A Review of Research Works on Supervised Learning Algorithms for SCADA Intrusion Detection and Classification

A Review of Research Works on Supervised Learning Algorithms for SCADA Intrusion Detection and Classification

An Ultra-Short-Term Wind Power Forecasting Model Based on EMD-EncoderForest-TCN

Probabilistic Wind Park Power Prediction using Bayesian Deep Learning and Generative Adversarial Networks

Contact Info

Product

Resources

About