Wind Power Time Series Research Articles

A novel hybrid BWO-BiLSTM-ATT framework for accurate offshore wind power prediction

Accurate wind power forecasting is pivotal in facilitating the efficient integration of renewable energy sources into existing power grids. This study proposes a novel hybrid framework, termed BWO-BiLSTM-ATT, which synergistically combines the strengths of a bidirectional long short-term memory (BiLSTM) network, a self-attention mechanism, and the Beluga Whale Optimization (BWO) algorithm. The BiLSTM architecture, with its unique ability to capture bidirectional temporal dependencies, effectively models the intricate dynamics present in wind power time series data. Integrating the self-attention mechanism further enhances the model's performance by discerning and emphasizing the most salient features within the input data. Furthermore, employing the BWO algorithm optimizes the model's hyperparameters, ensuring optimal configuration and enhancing its predictive accuracy and generalization capabilities. The proposed BWO-BiLSTM-ATT framework was rigorously evaluated using real-world data from an offshore wind farm in Yangjiang City, China. Comparative analyses were conducted against several baseline algorithms, including persistence, ARIMA, SVR, and vanilla LSTM models. The results demonstrate the superior performance of the BWO-BiLSTM-ATT framework, achieving higher R-squared and explained variance scores, coupled with lower error metrics such as MSE, RMSE, MedAE, and MAE. Statistical significance tests further corroborated the substantial performance improvements offered by the proposed framework. The combination of advanced deep learning techniques and metaheuristic optimization algorithms embedded in the BWO-BiLSTM-ATT framework provides a robust and accurate solution for wind power forecasting, enabling efficient management and seamless integration of renewable energy resources into existing power grids.

Interpretable wind power forecasting combining seasonal-trend representations learning with temporal fusion transformers architecture

Wind power forecasting plays a critical role in the efficient integration of wind energy into power systems. Accurate short-term WPF helps maintain grid stability, optimize power dispatch, and reduce operational costs. However, current deep learning-based WPF architectures rely on mapping spatial–temporal features into a single high-dimensional representation, compromising performance and interpretability. In this work, the interpretable seasonal-trend representation algorithm (ISTR) is proposed to learn a couple of disentangled latent representations that describe seasonal-trend of wind power time series. The ISTR is developed in a self-supervised manner with no need for prior knowledge, providing sufficient interpretable information while avoiding the feature entanglement problem. The ISTR is then integrated into the temporal fusion transformer (TFT) model through seasonal time-point embedding to perform accurate multi-horizon WPF with interpretable meanings. In the case study, ISTR-TFT outperforms the current state-of-the-art methods by achieving an 8.9% improvement in mean absolute percentage error (MAPE) compared to DeepAR and a 15.7% improvement in pinball score. The proposed ISTR-TFT also shows enhanced robustness to noise and missing data scenarios. The interpretability results generated by ISTR-TFT reveal the importance of various input features and the learned temporal patterns, offering valuable insights for decision-making in power system operations.

A bi-level mode decomposition framework for multi-step wind power forecasting using deep neural network

The proportion of wind energy in global energy structure is growing rapidly, promoting the development of wind power forecasting (WPF) technologies to solve the uncertainty and intermittence of wind power generation. However, the nonlinear and stochastic features of wind power time series restrain the accuracy of multi-step prediction performance. A multi-step WPF (MS-WPF) approach based on a time series bi-level empirical mode decomposition (BLEMD) method and BiLSTM neural network is proposed in this paper to improve the WPF accuracy of regional wind power generators. Since the uncertainty is always generated through coupled factors from both wind and weather-to-power conversion, the linearity feature is first introduced as an aspect apart from the frequency in the proposed approach to decompose the wind power time sequence data. The proposed BLEMD introduces Pearson product-moment correlation coefficient to evaluate the linearity of time series and a linearity-based decomposition algorithm is designed accordingly. To further enhance the precision and release computation burdens, a DL-based prediction strategy, including a BiLSTM network, a CNN-BiLSTM network, and a mean weight estimation method are implemented to predict the components separately. The proposed method only relies on local data, greatly reducing the data acquisition and computation cost. The precision of the proposed MS-WPF is verified by a 2.5 kW wind turbine with horizons from 5 s to 30 s, a 1.5 MW wind turbine with horizons from 10 min to 1 h, and a 51 MW wind farm with horizons from 1 h to 6 h. The comparative experimental results with other cutting-edge methods indicated that the proposed MS-WPF has superior prediction accuracy and stable performance for multi-step prediction.

A novel hybrid model based on multiple influencing factors and temporal convolutional network coupling ReOSELM for wind power prediction

Highly accurate wind power prediction is important for the operation and management of wind farms, energy utilization and power supply. In recent years, in order to further improve prediction accuracy, some researchers have considered utilizing factors such as wind speed and wind direction to establish prediction models. However, this often results in input redundancy due to a lack of technical means. In this paper, a multivariate prediction model for wind power based on variational mode decomposition (VMD), fuzzy entropy (FE), partial auto-correlation function and cross-correlation function (PACF&CCF, abbreviated as PC), improved artificial hummingbird algorithm (IAHA), temporal convolutional network (TCN), and regularized online sequential extreme learning machine (ReOSELM) is proposed. To address the nonlinearity and non-stationarity of the original wind power time series and decrease computational costs, the VMD combined with FE data preprocessing technique is employed. Considering the information redundancy in the multivariate data, the PC technique is utilized to conduct correlation analysis on the input data and select highly correlated features as inputs. When the input is multivariate and high-dimensional, the ReOSELM model struggles to achieve satisfactory prediction performance. Therefore, a new model based on TCN-ReOSELM is constructed. Additionally, to overcome the drawbacks of manual parameter adjustment, an IAHA algorithm is proposed, which integrates Latin hypercube sampling and chaotic local search strategies, to optimize the crucial parameters of TCN-ReOSELM. The experimental results demonstrate that the multivariate method proposed in this paper reduces the RMSE by 31% compared to the univariate method. Specifically, the proposed VMD-FE-PC-TCN-ReOSELM model reduces the RMSE, MAE, and SMAPE by 60–64%, 55–58%, and 10–36%, respectively, compared to the PC-TCN-ReOSELM model.

Ultra-Short-Term Offshore Wind Power Prediction Based on PCA-SSA-VMD and BiLSTM.

In order to realize the economic dispatch and safety stability of offshore wind farms, and to address the problems of strong randomness and strong time correlation in offshore wind power forecasting, this paper proposes a combined model of principal component analysis (PCA), sparrow algorithm (SSA), variational modal decomposition (VMD), and bidirectional long- and short-term memory neural network (BiLSTM). Firstly, the multivariate time series data were screened using the principal component analysis algorithm (PCA) to reduce the data dimensionality. Secondly, the variable modal decomposition (VMD) optimized by the SSA algorithm was applied to adaptively decompose the wind power time series data into a collection of different frequency components to eliminate the noise signals in the original data; on this basis, the hyperparameters of the BiLSTM model were optimized by integrating SSA algorithm, and the final power prediction value was obtained. Ultimately, the verification was conducted through simulation experiments; the results show that the model proposed in this paper effectively improves the prediction accuracy and verifies the effectiveness of the prediction model.

Short-term Wind Power Prediction based on CEEMDAN-TCN

In order to improve the accuracy of wind power prediction, a wind power prediction method based on time series decomposition and error correction is proposed in this paper. Firstly, the maximum information coefficient (MIC) method is used to select the features with strong correlation with wind power, so as to reduce the complexity of the original data; Then, according to the non-stationary characteristics of wind power, the wind power is decomposed into several stationary subsequences by using adaptive noise complete set empirical mode decomposition (CEEMDAN); Finally, the time convolution network (TCN) is used to dynamically model the multivariable time series of wind power; In order to further improve the prediction accuracy, a light quantization hoist (LightGBM) is introduced to correct the error of the prediction value. The simulation results show that the proposed method has higher short-term wind power prediction accuracy than other prediction models.

Chaotic time series wind power interval prediction based on quadratic decomposition and intelligent optimization algorithm

Wind power prediction plays a critical role in the economic deployment of power systems including wind farm. To improve the prediction accuracy of wind power, a chaotic wind power interval prediction method, which relies on quadratic decomposition and improved state transition algorithm, is proposed in this study. Firstly, the data are decomposed twice through the optimal variational mode decomposition, improved complete ensemble empirical mode decomposition and permutation entropy (OVMD-ICEEMDAN-PE) in combination, so as to deal with the volatility and complexity of wind power time series. Then, the decomposed data are used to build the prediction model for each component. In addition, the improved state transition algorithm is introduced to optimize the hyperparameters of the convolutional neural network and bidirectional long short-term memory (CNN-BiLSTM), time convolutional network (TCN) and gated recurrent unit (GRU) of the prediction models. With the optimal hyperparameters solved for each model, the linear weights are combined to achieve accurate prediction. Finally, through the combined prediction model, interval prediction is achieved. The results show that the proposed method outperforms other models in the accuracy of prediction.

Analysis of hybrid exploitation of wind and wave power in the Mediterranean and the Black Sea

The hybrid use of renewables enables resource complementation by other sources when one of them is inactive. This study focuses on wind and wave hybrids in the Mediterranean and the Black Sea with the aim of quantifying the potentials, directionality, correlation, and complementarity of the two sources. The investigation includes the spatial distributions and point-wise analysis capacity factors, the relationship between wave height and period, prevailing directions of wind and wave, bivariate probability analysis of wind and wave power, correlation of wind and wave power based on Pearson and Kendall’s Tau methods, and complementarity analysis by computing the active hours of one source when the other is inactive. The energy density, correlation, and complementarity analysis are realized considering the working conditions of a wind turbine with eight different wave energy converters by taking into account their power matrices. ERA5 dataset is used as a source for the wind and wave data in the basin with over six-decade long time series for each grid point. The results indicate that the wind turbine possesses higher capacity factors. For the wind turbine, the spatial maximum is 0.5, whereas for the wave energy converters, this value is 0.16 for the Pelamis device with the highest performance among the considered devices. The relationship between the wave climate parameters is investigated by fitting the wave height and period time series to an exponential function in all study area. Results are provided as spatial distributions of function coefficient which is hoped to be useful for the researchers when only one of the variables is known. Bivariate probabilities are computed for wind and wave power, and it is seen that for the selected analysis points, the wave power has a potential between 1 and 1.5 kW/m in the most probable axial locations. Directional analysis of wind and waves resulted in a map of the prevailing directions in the study area showing that the Northwest is the dominant direction. Correlation analysis between the wind and wave power time series of devices indicates that a positive correlation is evident in the study area up to + 0.69. The main result from the complementarity investigation accompanied by the correlation analysis showed that wind complementing wave condition is much more dominant in the region.

A new distributed decomposition–reconstruction–ensemble learning paradigm for short-term wind power prediction

The rapid and accurate forecast of wind power is crucial for power grid dispatching, because wind power time series are frequently nonlinear and noisy. Currently, typical decomposition and ensemble approaches can enhance prediction precision, but at the expense of computing efficiency. In addition, existing approaches rely primarily on a single model to forecast the deconstructed sequences, which may not correspond to the physical significance of each sequence. This research presents a wind power forecasting paradigm based on distributed decomposition-reconstruction-ensemble learning to address the aforementioned issues. Initially, the wind power data are decomposed using variational mode decomposition (VMD) to reduce data noise. Second, frequency, grey correlation degree, and Pearson coefficient are employed to cluster the decomposed data into high-frequency and low-frequency sequences in order to decrease the complexity of the model. Then, the reconstructed sequences are equally divided into 10 parts and distributed model training is performed using federated learning (FL), federated averaging (FedAvg) algorithm, and differential privacy Laplace (DPLA) method in a prediction pool composed of different machine learning models. For each sequence, an optimal machine learning model is selected and combined to form the optimal hybrid model. Finally, the final wind power forecast results are obtained by adding and summing the prediction results of each sequence. The experimental results show that the root mean square error (RMSE) of the proposed method is 10.230, mean absolute error (MAE) is 5.175, R-square (R2) is 0.990, and accuracy rate (ACC) is 0.644, which is better than all the 28 compared models. Moreover, the technique protects data privacy, solves the problem of data silos, provides a new concept for the conventional decomposition-ensemble prediction method, and is more in accordance with the physical relevance of wind power prediction.

A hybrid model based on LSTM neural networks with attention mechanism for short-term wind power forecasting

Wind power plants have gained prominence in recent decades owing to their positive environmental and economic impact. However, the unpredictability of wind resources poses significant challenges to the secure and stable operation of the power grid. To address this challenge, numerous computational and statistical methods have been proposed in the literature to forecast short-term wind power generation. However, the demand for more accurate and reliable methodologies to tackle this problem remains. In this context, this paper proposes a new hybrid framework that combines a statistical pre-processing stage with an attention-based deep learning approach to overcome the shortcomings of existing forecasting strategies in accurately predicting multi-seasonal wind power time series. The proposed ensemble model involves a data transformation stage that normalizes the data distribution, along with modeling and removing multiple seasonal patterns from the historical time-series. Considering these results, the proposed model further incorporates an LSTM Recurrent Neural Network (RNN) model with an attention mechanism, for each month of the year, to better capture the relevant temporal dependencies in the input residuals sequence. The model was trained and evaluated on hourly wind power data obtained from the Spanish electricity market, spanning the period from 2008 to 2019. Experimental results show that the proposed model outperforms well-established DL-based models, achieving lower error metrics. These findings have potential applications in energy trading, grid planning, and renewable energy management.

Ultra-short-term wind power prediction method based on FTI-VACA-XGB model

In order to predict wind power quickly and accurately and reduce the negative impact of wind power instability on the grid, this study proposes an ultra-short-term wind power prediction model based on financial technical indicators and parameter optimization algorithms. Firstly, historical wind power time series data calculates the financial technical indicators. Secondly, the Monte Carlo method and rank ant colony algorithm are used to optimize the parameters of financial technical indicators calculation. Finally, the future wind power is predicted based on the XGBoost algorithm combining financial technical indicators with historical power. The proposed model is validated on several wind power datasets from different countries and compared with various comparative models, leading to several important conclusions: (1) Fintech indicators can effectively indicate the intrinsic characteristics of wind power time-series data. (2) The variational ant colony algorithm can make the financial technical indicators better fit the wind power time-series data trends. (3) The proposed model has high prediction accuracy and speed and has similar prediction accuracy to the mainstream deep learning models. (4) The proposed model is not limited by meteorological, geographical, and seasonal factors and can make predictions by relying only on historical wind power data.

Wind power estimation on local scale—A case study of representativeness of reanalysis data and data-driven analysis

IntroductionWith hydropower being the dominant source of renewable energy in Austria and recent years being disproportionally dry, alternative renewable energy sources need to be tapped to compensate for the reduction of fossil fuels and account for dry conditions. This becomes even more important given the current geopolitical situation. Wind power plays an essential role in decarbonizing Austria's electricity system. For local assessments of historic, recent, and future wind conditions, adequate climate data are essential. Reanalysis data, often used for such assessments, have a coarse spatial resolution and could be unable to capture local wind features relevant for wind power modeling. Thus, raw reanalysis data need post-processing, and the results need to be interpreted with care. The purpose of this study is to assess the quality of three reanalysis data sets, such as MERRA-2, ERA5, and COSMO-REA6, for both surface level and hub height wind speed and wind power production at meteorological observation sites and wind farms in flat and mountainous terrain. Furthermore, the study aims at providing a first knowledge baseline toward generating a novel wind speed and wind power atlas at different hub heights for Austria with a spatial resolution of 1 × 1 km and for an experimental region with sub-km resolution. Thus, the study tries to answer (i) the questions if the reanalysis and analysis data can reproduce surface-level wind speed and (ii) if wind power calculations based on these data can be trusted, providing a knowledge base for future wind speed and wind power applications in complex terrain.MethodsFor that purpose, a generalized additive model (GAM) is applied to enable a data-driven gridded surface wind speed analysis as well as extrapolation to hub heights as a first step toward generating a novel wind speed atlas. In addition, to account for errors due to the coarse grid of the re-analysis, the New European Wind Atlas (NEWA) and the Global Wind Atlas (GWA) are used for correction using an hourly correction factor accounting for diurnal variations. For the analysis of wind power, an empirical turbine power curve approach was facilitated and applied to five different wind sites in Austria.Results and discussionThe results showed that for surface-level wind speed, the GAM outperforms the reanalysis data sets across all altitude levels with a mean average error (MAE) of 1.65 m/s for the meteorological sites. It even outperforms the NEWA wind atlas, which has an MAE of 3.78 m/s. For flat regions, the raw reanalysis matches the production data better than NEWA, also for hub height wind speeds, following wind power. For the mountainous areas, a correction of the reanalysis data based on the NEWA climatology, or even the NEWA climatology itself, significantly improved wind power evaluations. Comparisons between modeled wind power time series and real data show mean absolute errors of 8% of the nominal power in flat terrain and 14 or 17% in mountainous terrain.

Short-term wind power prediction based on modal reconstruction and CNN-BiLSTM

Accurate prediction of short-term wind power plays an important role in the safe operation and economic dispatch of the power grid. In response to the current single algorithm that cannot further improve the prediction accuracy, this study proposes a combined wind power prediction model based on data processing, signal decomposition, and deep learning. First, outliers in the original data can affect prediction accuracy. This study detects the outliers by the Z-score method and fills them with cubic spline interpolation to ensure the integrity of the data. Second, for the volatility of wind power, the wind power time series is decomposed using the complete ensemble empirical modal decomposition with adaptive noise (CEEMDAN). The component complexity of the decomposition is calculated using the sample entropy (SE), and the components are reconstructed according to the SE size to improve the prediction accuracy. Finally, the traditional convolutional neural network (CNN) structure is improved and the bi-directional long short-term memory (BiLSTM) network is used to further extract the feature links between wind power data and superimpose the prediction results of each reconstructed component to obtain the final wind power prediction value. The experimental results demonstrate that the hybrid model proposed in this study has better performance in terms of prediction performance.

Ultra-short-term wind power prediction method combining financial technology feature engineering and XGBoost algorithm

The input features of existing wind power time-series data prediction models are difficult to indicate the potential relationships between data, and the prediction methods are based on deep learning, which makes the convergence of the models slow and difficult to be applied to the actual production environment. To solve the above problems, an ultra-short-term wind power prediction model based on the XGBoost algorithm combined with financial technical index feature engineering and variational ant colony algorithm is proposed. The model innovatively applies financial technical indicators from financial time series data to wind power time series data and creates a class of model input features that can highly condense the potential relationships between time series data. A bionic algorithm is used to search for the best computational parameters for financial technical indicators to reduce the reliance on financial experts' experience. Taking the German power company Tennet wind power data set as an example, the prediction model proposed in this study has an mean absolute error of 0.859 and a root mean square error of 1.329, and it takes only 244 ms to complete the prediction. Thus, this study provides a new solution for ultra-short-term wind power prediction.

An intelligent hybrid method based on Monte Carlo simulation for short-term probabilistic wind power prediction

Precise estimation of wind power is exceedingly challenging due to the quick and random fluctuations in wind speed. Therefore, this article presents a probabilistic intelligent method for wind power prediction to minimize the risk caused by the uncertainty of the generated power. The proposed method is realized through four important steps: 1. Analyzing the wind power time series signals; 2. Creating training patterns; 3. Training patterns by applying machine learning; 4. Making a probabilistic prediction based on the Monte-Carlo simulation. To estimate the wind power more accurately, the time series of historical data on wind power are decomposed into different levels by appropriate signal analysis tools, i.e., the wavelet transform and variational mode decomposition. Then, the intelligent tools, i.e., extreme learning machine and support vector regression, are used as predictor cores due to their high speed and accuracy. The obtained results show that the presented method can accurately predict the generated wind power in 10-min intervals for the next hour. Moreover, the average value of the estimated data for 1000 repetition steps in probabilistic protection is very close to the real data. Thus, the proposed method can be effectively applied to both deterministic and probabilistic predictions.

Chaotic time series wind power prediction method based on OVMD-PE and improved multi-objective state transition algorithm

As the global wind power generation capacity is constantly increasing, the problems of safe operation and utilization after grid connection are becoming more and more prominent. Aiming at the problem of low accuracy and stability of wind power time series with chaotic characteristics, a time series prediction method combining chaotic characteristic processing and neural networks is proposed. First, optimal variational mode decomposition with permutation entropy (OVMD-PE) is used to decompose the original wind power time series and overcome the disadvantage of insufficient mode aliasing encountered by empirical and integrated empirical modes. Second, an improved multi-objective state transition algorithm is proposed to determine the weight coefficients among the neural networks and improve the accuracy of the reconstructed predictive neural networks. Finally, the combined prediction method is used to study and analyze the wind power data from a wind farm in Xinjiang, China, from the perspectives of multiple scenarios and multiple time scales. The experimental results show that OVMD-PE can successfully deal with chaotic characteristics and the improved algorithm has improved the prediction accuracy. Compared with other traditional prediction models, the combined prediction model has higher robustness and stability.

A Bayesian Optimization-Based LSTM Model for Wind Power Forecasting in the Adama District, Ethiopia

Renewable energies, such as solar and wind power, have become promising sources of energy to address the increase in greenhouse gases caused by the use of fossil fuels and to resolve the current energy crisis. Integrating wind energy into a large-scale electric grid presents a significant challenge due to the high intermittency and nonlinear behavior of wind power. Accurate wind power forecasting is essential for safe and efficient integration into the grid system. Many prediction models have been developed to predict the uncertain and nonlinear time series of wind power, but most neglect the use of Bayesian optimization to optimize the hyperparameters while training deep learning algorithms. The efficiency of grid search strategies decreases as the number of hyperparameters increases, and computation time complexity becomes an issue. This paper presents a robust and optimized long-short term memory network for forecasting wind power generation in the day ahead in the context of Ethiopia’s renewable energy sector. The proposal uses Bayesian optimization to find the best hyperparameter combination in a reasonable computation time. The results indicate that tuning hyperparameters using this metaheuristic prior to building deep learning models significantly improves the predictive performances of the models. The proposed models were evaluated using MAE, RMSE, and MAPE metrics, and outperformed both the baseline models and the optimized gated recurrent unit architecture.

The analysis of time-varying high-order moment of wind power time series

The efficient wind power forecasting is of great significance for the safety and stability of power grid. Analysis on the high order moment of wind power time series play an important role in improving the forecasting accuracy of wind power. With the rolling subsample generalized auto-regressive conditional heteroskedasticity with skewness and kurtosis (GARCHSK) model sequence, the analysis method for conditional skewness and conditional kurtosis of wind power time series is proposed. The dynamic curves of parameter are presented by estimating the characteristic parameters in the subsample models. Moreover, the time-varying structure of high order moment of the wind power time series is analyzed. A Generalized News Impact Curve (GNIC) is proposed, and the impact of news on the conditional skewness and conditional kurtosis are measured and evaluated by the third order GNIC and the fourth order GNIC. The result of case based on the wind power data of Jiangsu power grid demonstrates that the time-varying structure of high-order moments of wind power time series is steady relatively, and the time-varying characteristics of high-order moments are further verified by GNIC.

Short-term wind power prediction based on the combination of numerical weather forecast and time series

The accurate prediction of wind power has a huge impact on the grid connection and dispatching of the power system. In order to make the prediction accuracy of wind power higher, this paper proposes a combined forecasting model based on the combination of numerical weather prediction (NWP) and wind power time series, called gray wolf algorithm-wavelet neural network-variational mode decomposition-long short-term memory-Q-learning (GWO-WNN-VMD-LSTM-Q-learning). First, the wind power prediction (WPP) is implemented based on the NWP, and prediction result 1 is obtained. In this stage, the wavelet neural network (WNN), which is optimized by the gray wolf algorithm (GWO), is used for prediction. Then, the historical time series of wind power is subjected to variational mode decomposition (VMD), and the decomposed sub-sequences are predicted by long short-term memory (LSTM) networks, respectively, and the prediction results of each sub-sequence are summed to obtain the prediction result 2. Finally, the Q-learning algorithm is used to superimpose prediction result 1 and result 2 on the basis of optimal weight and get the final WPP results. The simulation results demonstrate that this model's prediction accuracy is high and that it has a substantially greater predictive impact than other traditional models that merely take time series or numerical weather forecasts into account.

Reduction method for multi-period time series scenarios of wind power

In the planning and operation of power systems containing wind power, it is of great significance to use a small number of representative wind power time series scenarios to accurately portray the stochastic characteristics of wind power. With the increase of the number of scenarios, how to form representative typical scenarios to balance computational efficiency and accuracy is an urgent problem to be solved. In this study, a bi-directional optimization method is proposed to generate daily wind power time series scenarios based on a single-period optimal scenario generation strategy and a multi-period scenario reduction strategy. First, the Wasserstein probability distance index is used to form a scenario model that is best approximated to the probability distribution of wind power, and the optimal scenarios of each period are generated. Secondly, a wind power scenario reduction strategy that integrates spatial distance and stochastic features is proposed. The improved taboo search method is used to selectively connect the representative scenarios of each period to form the representative daily wind power time series scenarios. Finally, the effectiveness and practicality of the proposed wind power time series scenario generation method are verified by simulations.