A hybrid deep learning architecture for wind power prediction based on bi-attention mechanism and crisscross optimization

Meng, Anbo; Chen, Shun; Ou, Zuhong; Ding, Weifeng; Zhou, Huaming; Fan, Jingmin; Yin, Hao

doi:10.1016/j.energy.2021.121795

Cited by 73 publications

(19 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, due to the intermittent and strong variability of wind power (Yin et al, 2021;Duan et al, 2022). Therefore, it is necessary to develop a method that can accurately forecast wind power, reduce the negative impact of wind power grid connection, ensure the safe and stable operation of the power system, and improve the utilization rate of wind power in the power system (Hu et al, 2021a;Lin and Zhang, 2021;Meng et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

A combined model based on secondary decomposition technique and grey wolf optimizer for short-term wind power forecasting

Zheng²,

Lu³

2023

Front. Energy Res.

View full text Add to dashboard Cite

Short-term wind power forecasting plays an important role in wind power generation systems. In order to improve the accuracy of wind power forecasting, many researchers have proposed a large number of wind power forecasting models. However, traditional forecasting models ignore data preprocessing and the limitations of a single forecasting model, resulting in low forecasting accuracy. Aiming at the shortcomings of the existing models, a combined forecasting model based on secondary decomposition technique and grey wolf optimizer (GWO) is proposed. In the process of forecasting, firstly, the complete ensemble empirical mode decomposition adaptive noise (CEEMDAN) and wavelet transform (WT) are used to preprocess the wind power data. Then, least squares support vector machine (LSSVM), extreme learning machine (ELM) and back propagation neural network (BPNN) are established to forecast the decomposed components respectively. In order to improve the forecasting performance, the parameters in LSSVM, ELM, and BPNN are tuned by GWO. Finally, the GWO is used to determine the weight coefficient of each single forecasting model, and the weighted combination is used to obtain the final forecasting result. The simulation results show that the forecasting model has better forecasting performance than other forecasting models.

show abstract

Section: Introductionmentioning

confidence: 99%

A combined model based on secondary decomposition technique and grey wolf optimizer for short-term wind power forecasting

Zheng²,

Lu³

2023

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…Vertical crossover is to exchange the dimensional information of a single search agent, which can facilitate escaping from the stagnancy in local optima without destroying other dimensions that may be the global optimum. Considering the advantages of CSO, some papers tried to apply this algorithm on handling complex problems [50][51][52][53][54][55][56][57]. In addition, the excellent search capability of the two crossover operators can exactly make up for the deficiencies of HHO mentioned above.…”

Section: Introductionmentioning

confidence: 99%

Crisscross Harris Hawks Optimizer for Global Tasks and Feature Selection

et al. 2022

View full text Add to dashboard Cite

Harris Hawks Optimizer (HHO) is a recent well-established optimizer based on the hunting characteristics of Harris hawks, which shows excellent efficiency in solving a variety of optimization issues. However, it undergoes weak global search capability because of the levy distribution in its optimization process. In this paper, a variant of HHO is proposed using Crisscross Optimization Algorithm (CSO) to compensate for the shortcomings of original HHO. The novel developed optimizer called Crisscross Harris Hawks Optimizer (CCHHO), which can effectively achieve high-quality solutions with accelerated convergence on a variety of optimization tasks. In the proposed algorithm, the vertical crossover strategy of CSO is used for adjusting the exploitative ability adaptively to alleviate the local optimum; the horizontal crossover strategy of CSO is considered as an operator for boosting explorative trend; and the competitive operator is adopted to accelerate the convergence rate. The effectiveness of the proposed optimizer is evaluated using 4 kinds of benchmark functions, 3 constrained engineering optimization issues and feature selection problems on 13 datasets from the UCI repository. Comparing with nine conventional intelligence algorithms and 9 state-of-the-art algorithms, the statistical results reveal that the proposed CCHHO is significantly more effective than HHO, CSO, CCNMHHO and other competitors, and its advantage is not influenced by the increase of problems’ dimensions. Additionally, experimental results also illustrate that the proposed CCHHO outperforms some existing optimizers in working out engineering design optimization; for feature selection problems, it is superior to other feature selection methods including CCNMHHO in terms of fitness, error rate and length of selected features. Supplementary Information The online version contains supplementary material available at 10.1007/s42235-022-00298-7.

show abstract

“…At present, signal-decomposition-based methods combined with NN have been widely used in the fields of predicting air quality, crude oil prices, wind power, and so on. For example, it was stated in reference [33][34][35][36][37][38][39][40] that it is hard to get a precise estimation for a single model because of the non-linearity and non-stationarity of the raw data. Aiming at this problem, Huang et al [33] proposed a fusion method of EMD-GRU for predicting PM2.5 concentration.…”

Section: Introductionmentioning

confidence: 99%

“…Lin et al [ 34 ] proposed a hybrid method that is a combination of the CEEMDAN and multi‐layer gated recurrent unit (ML‐GRU) NN for predicting crude oil prices. In reference, [ 35 ] the attention‐based deep residual GRU network is combined with ensemble empirical mode decomposition (EEMD) and cross optimization algorithm (CSO) to make multi‐step wind power prediction. Niu et al [ 36 ] proposed a fusion model of EEMD and RNN for landslide displacement prediction.…”

Section: Introductionmentioning

confidence: 99%

State‐of‐Health Prediction for Lithium‐Ion Batteries Based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise‐Gate Recurrent Unit Fusion Model

2022

View full text Add to dashboard Cite

State‐of‐health (SOH) estimation is one of the most critical battery management system (BMS) tasks. A challenge remains for the SOH prediction due to the complicated battery aging mechanism. The most common health indicator is the capacity of the lithium‐ion battery. The fluctuation of capacity caused by the capacity regeneration phenomenon can seriously affect the prediction performance. A new complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and gate recurrent unit (GRU) based fusion prediction model for SOH estimation is proposed to solve the problem effectively. First, the CEEMDAN algorithm decomposes the original SOH into local fluctuations and global degradation trends. Then, the GRU network and autoregressive integrated moving average model are used to predict the above trends, respectively. Next, a sliding window is designed to calculate an average value of the global degradation trend prediction residuals. Then, the second GRU algorithm can be used to correct prediction residuals. Finally, the prediction results of the aforementioned parts are combined to obtain the final SOH estimation. The proposed method is verified by experimental battery data from NASA and CALCE datasets. The results show that the fusion method has both higher estimation accuracy and stronger robustness than other methods.

show abstract

A hybrid deep learning architecture for wind power prediction based on bi-attention mechanism and crisscross optimization

Cited by 73 publications

References 46 publications

A combined model based on secondary decomposition technique and grey wolf optimizer for short-term wind power forecasting

A combined model based on secondary decomposition technique and grey wolf optimizer for short-term wind power forecasting

Crisscross Harris Hawks Optimizer for Global Tasks and Feature Selection

State‐of‐Health Prediction for Lithium‐Ion Batteries Based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise‐Gate Recurrent Unit Fusion Model

Contact Info

Product

Resources

About