Machine Learning Techniques for Supporting Renewable Energy Generation and Integration: A Survey

Perera, Kasun S.; Aung, Zeyar; Woon, Wei Lee

doi:10.1007/978-3-319-13290-7_7

Cited by 73 publications

(38 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These techniques attempt to discover underlying relationships, without an a priori structural hypothesis that relates wind power with several historical meteorological variables [2,25]. In this category, we can mention the work of Olaofe and Folly [26], who present a recurrent neural network to estimate the wind power of a turbine from 1-288 h ahead.…”

Section: Machine Learning Methodsmentioning

confidence: 99%

“…end if 18: end while Output: Net's Weights Later, in order to update the output layer weights W [o] , the weights that solve the multi-linear regression y = x [o] • W [o] are chosen in Step 4 of Algorithm 1, called TrainNet_by_ESN(). This can be done, for example, using ridge regression as in Equation (2) or by means of quantile regression [54]. The latter approach gets a more robust estimation of y.…”

Section: Algorithm 2 Trainnet_by_lstm()mentioning

confidence: 99%

“…In many countries around the world, renewable energies are gradually replacing traditional energy sources, mainly because of the unprecedented and irreversible environmental damage caused by fossil and nuclear fuels and the continuing reduction of renewable technology costs [1,2]. Among the main sources of renewable energies, wind energy has presented an important development, being an attractive alternative in terms of its costs and low environmental impact [3].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Wind Power Forecasting Based on Echo State Networks and Long Short-Term Memory

et al. 2018

View full text Add to dashboard Cite

Wind power generation has presented an important development around the world. However, its integration into electrical systems presents numerous challenges due to the variable nature of the wind. Therefore, to maintain an economical and reliable electricity supply, it is necessary to accurately predict wind generation. The Wind Power Prediction Tool (WPPT) has been proposed to solve this task using the power curve associated with a wind farm. Recurrent Neural Networks (RNNs) model complex non-linear relationships without requiring explicit mathematical expressions that relate the variables involved. In particular, two types of RNN, Long Short-Term Memory (LSTM) and Echo State Network (ESN), have shown good results in time series forecasting. In this work, we present an LSTM+ESN architecture that combines the characteristics of both networks. An architecture similar to an ESN is proposed, but using LSTM blocks as units in the hidden layer. The training process of this network has two key stages: (i) the hidden layer is trained with a descending gradient method online using one epoch; (ii) the output layer is adjusted with a regularized regression. In particular, the case is proposed where Step (i) is used as a target for the input signal, in order to extract characteristics automatically as the autoencoder approach; and in the second stage (ii), a quantile regression is used in order to obtain a robust estimate of the expected target. The experimental results show that LSTM+ESN using the autoencoder and quantile regression outperforms the WPPT model in all global metrics used.

show abstract

Section: Machine Learning Methodsmentioning

confidence: 99%

Section: Algorithm 2 Trainnet_by_lstm()mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Wind Power Forecasting Based on Echo State Networks and Long Short-Term Memory

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Ciaramella et al [28] proposed a Bayesian-Based Neural Network model for solar photovoltaic power forecasting. Perera et al [29] surveyed machine learning techniques for the supporting of renewable energy generation and integration. The studies reviewed in this paper analyzes the different machine learning techniques used for supporting the generation of renewable energy and more importantly their integration into the power grid.…”

Section: Literature Reviewmentioning

confidence: 99%

Fuzzy Multi-Criteria Decision-Making Model for Supplier Evaluation and Selection in a Wind Power Plant Project

Wang

Yang²,

Cheng

2019

Mathematics

View full text Add to dashboard Cite

In order to meet ambitious growth targets in the medium term, Vietnam must continue exploiting traditional energy sources. In the longer term, Vietnam has to develop a strategy and roadmap for the development of new energy sources. In these new energy sources, wind energy has emerged as a viable option. Given the geographic conditions of a locality with a long coastline and high winds that are fairly distributed all year, many wind-power plants are being built in Vietnam. One of the most important pieces of equipment in a wind-power plant is the wind turbine. The wind turbine suppliers’ selection is a complex and multicriteria decision-making (MCDM) process that can reduce the costs of procuring equipment and aid in receiving products on time. Many studies have applied the MCDM model to various fields of science and engineering. One of the fields that the MCDM approaches have been applied to is the supplier selection problem. Supplier selection is an important issue of the MCDM model. Especially in a renewable energy project, decision-makers have to evaluate both natural and society factors. Although some researchers have reviewed the applications of the MCDM model in wind turbine supplier selection, limited work has focused on this problem in a fuzzy environment. Therefore, in this work, the authors propose a fuzzy MCDM model for the wind turbine supplier selection process under fuzzy environment conditions. In the first step, all factors for wind turbine supplier selection are identified by supply chain operations reference (SCOR) metrics and the results from a review of the literature. A fuzzy analytic network process (FANP) model is applied for determining the weight of all the criteria in the second stage, and the technique for order preference by similarity to an ideal solution (TOPSIS) model is used to rank all the potential suppliers in the final stage. As a result, Decision-Making Unit 010 (DMU010) becomes an optimal option for the wind turbine supplier selection processes. The contribution of this research is to develop new hybrid fuzzy MCDM approaches for wind turbine supplier selections. Furthermore, this work presents useful guidelines for wind turbines as well as provides a guideline for supplier selection in other industries.

show abstract

“…Treiber, Heinermann, and Kramer (2016) proposed a model using a multitude of machine learning algorithms for short-term wind power prediction (Treiber, Heinermann, and Kramer, 2016). Perera, Aung, and Woon (2014) provide a survey on different machine learning techniques to predict the amount of power generated in the future (Perera, Aung, and Woon, 2014). Consequently, smart energy systems bring to the fore typical characteristics of big data scenarios.…”

mentioning

confidence: 99%

Big Data in Smart Energy Systems: A Critical Review

Codal

Ari

Ilter

2020

AJIT-e

View full text Add to dashboard Cite

Climate change is an undeniable fact. Considering that two-thirds of greenhouse gas emissions originate from the energy sector, it is expected that the world's energy system will be transformed with renewable energy sources. Energy efficiency will be continuously increased. Reducing energy-related carbon dioxide emissions is the heart of the energy transition. Big data in energy systems play a crucial role in evaluating the adaptive capacity and investing more smartly to manage energy demand and supply. Indeed, the impact of the smart energy grid and meters on smart energy systems provide and assist decision-makers in transforming energy production, consumption, and communities. This study reviews the literature for aligning big data and smart energy systems and criticized according to regional perspective, period, disciplines, big data characteristics, and used data analytics. The critical review has been categorized into present themes. The results address issues, including scientific studies using data analysis techniques that take into account the characteristics of big data in the smart energy literature and the future of smart energy approaches. The manuscripts on big data in smart energy systems are a promising issue, albeit it is essential to expand subjects through comprehensive interdisciplinary studies

show abstract

Machine Learning Techniques for Supporting Renewable Energy Generation and Integration: A Survey

Cited by 73 publications

References 34 publications

Wind Power Forecasting Based on Echo State Networks and Long Short-Term Memory

Wind Power Forecasting Based on Echo State Networks and Long Short-Term Memory

Fuzzy Multi-Criteria Decision-Making Model for Supplier Evaluation and Selection in a Wind Power Plant Project

Big Data in Smart Energy Systems: A Critical Review

Contact Info

Product

Resources

About