A hybrid model for forecasting wind speed and wind power generation

Chang, G.W.; Lu, H. J.; Hsu, Lewis; Chen, Y. Y.

doi:10.1109/pesgm.2016.7742039

Cited by 23 publications

(11 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Wind speed is highly variable in time and space, and that makes wind speed forecasting challenging for offshore applications such as O&M activity and wind farm installation. In the literature a variety of techniques to forecast short-term as well as long term wind speed, including physical models, have been proposed; for example [12] where numerical weather prediction (NWP) is mostly used; statistical methods [13] such as the ARIMA model; the intelligent models based on ANNs [14]; and the hybrid forecasting models [15], that include different types of approaches. Author of [16] carried out a performance comparison of ANN, ARIMA and hybrid models (the combination of ARIMA and ANN) for wind speed forecasting at different look-ahead times.…”

Section: Related Work Of Forecasting Using Data-driven Methodsmentioning

confidence: 99%

Data‐driven weather forecasting models performance comparison for improving offshore wind turbine availability and maintenance

Pandit

Kolios

Infield

2020

IET Renewable Power Generation

View full text Add to dashboard Cite

Wind energy is an attractive alternative to conventional sources of electricity generation due to its effectively zero carbon emissions. Wind power is highly dependent on wind speed and operations offshore are affected by wave height; these together called turbine weather datasets that are variable and intermittent over various timescales and signify offshore weather conditions. In contrast to onshore wind, offshore wind requires improved forecasting since unfavourable weather prevents repair and maintenance activities. Delayed repair results in increased downtime and reduced wind farm availability and energy yield. This paper proposes two data-driven models for long-term weather conditions forecasting to improve the wind farm availability and support operation and maintenance (O&M) decision-making process. These two data-driven approaches are Long Short-Term Memory Network, abbreviated as LSTM, and Markov chain. A LSTM is an artificial recurrent neural network (RNN), capable of learning long-term dependencies within a sequence of data and is typically used to avoid the long-term dependency problem. While, Markov is another data-driven stochastic model, which assumes that, the future states depend only on the current states, not on the events that occurred before. The readily available weather datasets are obtained from FINO3 database to train and validate the performance of these data-driven models. A performance comparison between these weather forecasted models would be carried out to determine which approach is most accurate and suitable for improving offshore wind turbine availability and support maintenance activities. The full paper outlines the weakness and strength associated with proposed models in relations to offshore wind farms operational activities.

show abstract

Section: Related Work Of Forecasting Using Data-driven Methodsmentioning

confidence: 99%

Data‐driven weather forecasting models performance comparison for improving offshore wind turbine availability and maintenance

Pandit

Kolios

Infield

2020

IET Renewable Power Generation

View full text Add to dashboard Cite

show abstract

“…In 2016, G.W. Chang et al used a hybrid ARIMA-NN model to predict wind power generation in Taiwan, pointing that this method was suitable for short-term prediction [40]. In 2017, some researchers combined FFNN [41], genetic algorithm (GA) [42] and adaptive neuro-fuzzy inference system (ANFIS) to predict photovoltaic power generation in southern Greece.…”

Section: Overview Of Forecasting Methodsmentioning

confidence: 99%

Forecasting India’s Electricity Demand Using a Range of Probabilistic Methods

Zhou

2019

Energies

View full text Add to dashboard Cite

With serious energy poverty, especially concerning power shortages, the economic development of India has been severely restricted. To some extent, power exploitation can effectively alleviate the shortage of energy in India. Thus, it is significant to balance the relationship between power supply and demand, and further stabilize the two in a reasonable scope. To achieve balance, a prediction of electricity generation in India is required. Thus, in this study, five methods, the metabolism grey model, autoregressive integrated moving average, metabolic grey model-auto regressive integrated moving average model, non-linear metabolic grey model and non-linear metabolic grey model-auto regressive integrated moving average model, are applied. We combine the characteristics of linear and nonlinear models, making a prediction and comparison of Indian power generation. In this way, we enrich methods for prediction research on electrical energy, which avoids large errors in trends of electricity generation due to those accidental factors when a single predictive model is used. In terms of prediction outcomes, the average relative errors from five models above are 1.67%, 1.62%, 0.84%, 1.84%, and 1.37%, respectively, which indicates high accuracy and reference value of these methods. In conclusion, India’s power generation will continue to grow with an average annual growth rate of 5.17% in the next five years (2018–2022).

show abstract

“…Then, Chang et al [35] proposed an ARIMA model to transform non-stationary wind speed and temperature series into stationary ones. The transformed series are fed to a radial basis neural network for wind speed forecasting.…”

Section: Machine Learning Methodsmentioning

confidence: 99%

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

A hybrid model for forecasting wind speed and wind power generation

Cited by 23 publications

References 13 publications

Data‐driven weather forecasting models performance comparison for improving offshore wind turbine availability and maintenance

Data‐driven weather forecasting models performance comparison for improving offshore wind turbine availability and maintenance

Forecasting India’s Electricity Demand Using a Range of Probabilistic Methods

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

Contact Info

Product

Resources

About