With the continuous advancement of China’s“peak carbon dioxide emissions and Carbon Neutrality”process,the proportion of wind power is increasing.In the current research,aiming at the problem that the forecasting m...With the continuous advancement of China’s“peak carbon dioxide emissions and Carbon Neutrality”process,the proportion of wind power is increasing.In the current research,aiming at the problem that the forecasting model is outdated due to the continuous updating of wind power data,a short-term wind power forecasting algorithm based on Incremental Learning-Bagging Deep Hybrid Kernel Extreme Learning Machine(IL-Bagging-DHKELM)error affinity propagation cluster analysis is proposed.The algorithm effectively combines deep hybrid kernel extreme learning machine(DHKELM)with incremental learning(IL).Firstly,an initial wind power prediction model is trained using the Bagging-DHKELM model.Secondly,Euclidean morphological distance affinity propagation AP clustering algorithm is used to cluster and analyze the prediction error of wind power obtained from the initial training model.Finally,the correlation between wind power prediction errors and Numerical Weather Prediction(NWP)data is introduced as incremental updates to the initial wind power prediction model.During the incremental learning process,multiple error performance indicators are used to measure the overall model performance,thereby enabling incremental updates of wind power models.Practical examples show the method proposed in this article reduces the root mean square error of the initial model by 1.9 percentage points,indicating that this method can be better adapted to the current scenario of the continuous increase in wind power penetration rate.The accuracy and precision of wind power generation prediction are effectively improved through the method.展开更多
Reliable and accurate ultra-short-term prediction of wind power is vital for the operation and optimization of power systems.However,the volatility and intermittence of wind power pose uncertainties to traditional poi...Reliable and accurate ultra-short-term prediction of wind power is vital for the operation and optimization of power systems.However,the volatility and intermittence of wind power pose uncertainties to traditional point prediction,resulting in an increased risk of power system operation.To represent the uncertainty of wind power,this paper proposes a new method for ultra-short-term interval prediction of wind power based on a graph neural network(GNN)and an improved Bootstrap technique.Specifically,adjacent wind farms and local meteorological factors are modeled as the new form of a graph from the graph-theoretic perspective.Then,the graph convolutional network(GCN)and bi-directional long short-term memory(Bi-LSTM)are proposed to capture spatiotemporal features between nodes in the graph.To obtain highquality prediction intervals(PIs),an improved Bootstrap technique is designed to increase coverage percentage and narrow PIs effectively.Numerical simulations demonstrate that the proposed method can capture the spatiotemporal correlations from the graph,and the prediction results outperform popular baselines on two real-world datasets,which implies a high potential for practical applications in power systems.展开更多
A physical approach of the wind power prediction based on the CFD pre-calculated flow fields is proposed in this paper. The flow fields are obtained based on a steady CFD model with the discrete inflow wind conditions...A physical approach of the wind power prediction based on the CFD pre-calculated flow fields is proposed in this paper. The flow fields are obtained based on a steady CFD model with the discrete inflow wind conditions as the boundary conditions, and a database is established containing the important parameters including the inflow wind conditions, the flow fields and the corresponding wind power for each wind turbine. The power is predicted via the database by taking the Numerical Weather Prediction （NWP） wind as the input data. In order to evaluate the approach, the short-term wind power prediction for an actual wind farm is conducted as an example during the period of the year 2010. Compared with the measured power, the predicted results enjoy a high accuracy with the annual Root Mean Square Error （RMSE） of 15.2% and the annual MAE of 10.80%. A good performance is shown in predicting the wind power＇s changing trend. This approach is independent of the historical data and can be widely used for all kinds of wind farms including the newly-built wind farms. At the same time, it does not take much computation time while it captures the local air flows more precisely by the CFD model. So it is especially practical for engineering projects.展开更多
针对传统风电功率预测仅考虑气象因素,且无法计及风电机组真实出力状态导致预测精度较差问题,本文提出一种计及风机状态的超短期风电功率动态预测方法。首先,为能够精确评估风机状态,将BP(error back propagation, BP)算法引入层次分析...针对传统风电功率预测仅考虑气象因素,且无法计及风电机组真实出力状态导致预测精度较差问题,本文提出一种计及风机状态的超短期风电功率动态预测方法。首先,为能够精确评估风机状态,将BP(error back propagation, BP)算法引入层次分析法(analytic hierarchy process, AHP)的评估结构中,构建BP-AHP风机状态评估模型,实现单台风机状态评估;然后,综合考虑地形及机组排布等因素,将风电场所有风机的状态取均值作为风电场状态,利用皮尔逊相关系数衡量所评估状态与功率之间的相关性以验证评估模型合理性,并采用XGBoost构建计及风机状态的动态预测模型;最后,以陕西地区某风电场实测数据进行算例分析,验证了所提方法的可行性及有效性。展开更多
针对最小二乘支持向量回归(Least Square Support Vector Regression,LSSVR)建模风电功率时变特性的局限性,提出了一种基于动态集成LSSVR的超短期风电功率预测模型.首先利用风电场监测控制与数据采集(Supervisory Control And Data Acqu...针对最小二乘支持向量回归(Least Square Support Vector Regression,LSSVR)建模风电功率时变特性的局限性,提出了一种基于动态集成LSSVR的超短期风电功率预测模型.首先利用风电场监测控制与数据采集(Supervisory Control And Data Acquisition,SCADA)与数值天气预报(Numerical Weather Prediction,NWP)系统的历史数据建立离线单体LSSVR模型库,然后根据预测时段与训练时段NWP序列的相似度从单体LSSVR模型库中动态选择候选集成成员,再后综合考虑正确性与多样性确定集成成员.最后由预测时段与训练时段NWP序列间的相似度分配集成LSSVR成员的权重.通过对湖南省某风电场输出功率进行预测,验证了动态集成LSSVR预测模型的有效性,与持续法、自回归求和移动平均法、单体LSSVR模型、常权重LSSVR组合模型及BPNN动态集成模型相比,动态集成LSSVR模型具有更高的精度,在天气非平稳变化阶段更加明显.展开更多
基金funded by Liaoning Provincial Department of Science and Technology(2023JH2/101600058)。
文摘With the continuous advancement of China’s“peak carbon dioxide emissions and Carbon Neutrality”process,the proportion of wind power is increasing.In the current research,aiming at the problem that the forecasting model is outdated due to the continuous updating of wind power data,a short-term wind power forecasting algorithm based on Incremental Learning-Bagging Deep Hybrid Kernel Extreme Learning Machine(IL-Bagging-DHKELM)error affinity propagation cluster analysis is proposed.The algorithm effectively combines deep hybrid kernel extreme learning machine(DHKELM)with incremental learning(IL).Firstly,an initial wind power prediction model is trained using the Bagging-DHKELM model.Secondly,Euclidean morphological distance affinity propagation AP clustering algorithm is used to cluster and analyze the prediction error of wind power obtained from the initial training model.Finally,the correlation between wind power prediction errors and Numerical Weather Prediction(NWP)data is introduced as incremental updates to the initial wind power prediction model.During the incremental learning process,multiple error performance indicators are used to measure the overall model performance,thereby enabling incremental updates of wind power models.Practical examples show the method proposed in this article reduces the root mean square error of the initial model by 1.9 percentage points,indicating that this method can be better adapted to the current scenario of the continuous increase in wind power penetration rate.The accuracy and precision of wind power generation prediction are effectively improved through the method.
文摘Reliable and accurate ultra-short-term prediction of wind power is vital for the operation and optimization of power systems.However,the volatility and intermittence of wind power pose uncertainties to traditional point prediction,resulting in an increased risk of power system operation.To represent the uncertainty of wind power,this paper proposes a new method for ultra-short-term interval prediction of wind power based on a graph neural network(GNN)and an improved Bootstrap technique.Specifically,adjacent wind farms and local meteorological factors are modeled as the new form of a graph from the graph-theoretic perspective.Then,the graph convolutional network(GCN)and bi-directional long short-term memory(Bi-LSTM)are proposed to capture spatiotemporal features between nodes in the graph.To obtain highquality prediction intervals(PIs),an improved Bootstrap technique is designed to increase coverage percentage and narrow PIs effectively.Numerical simulations demonstrate that the proposed method can capture the spatiotemporal correlations from the graph,and the prediction results outperform popular baselines on two real-world datasets,which implies a high potential for practical applications in power systems.
基金Project supported by the National Natural Science Foundation of China(Grant No. 51206051)
文摘A physical approach of the wind power prediction based on the CFD pre-calculated flow fields is proposed in this paper. The flow fields are obtained based on a steady CFD model with the discrete inflow wind conditions as the boundary conditions, and a database is established containing the important parameters including the inflow wind conditions, the flow fields and the corresponding wind power for each wind turbine. The power is predicted via the database by taking the Numerical Weather Prediction （NWP） wind as the input data. In order to evaluate the approach, the short-term wind power prediction for an actual wind farm is conducted as an example during the period of the year 2010. Compared with the measured power, the predicted results enjoy a high accuracy with the annual Root Mean Square Error （RMSE） of 15.2% and the annual MAE of 10.80%. A good performance is shown in predicting the wind power＇s changing trend. This approach is independent of the historical data and can be widely used for all kinds of wind farms including the newly-built wind farms. At the same time, it does not take much computation time while it captures the local air flows more precisely by the CFD model. So it is especially practical for engineering projects.
基金国家自然科学基金资助项目(6147212861173108)+1 种基金National Natural Science Foundation of China(6147212861173108)
文摘针对最小二乘支持向量回归(Least Square Support Vector Regression,LSSVR)建模风电功率时变特性的局限性,提出了一种基于动态集成LSSVR的超短期风电功率预测模型.首先利用风电场监测控制与数据采集(Supervisory Control And Data Acquisition,SCADA)与数值天气预报(Numerical Weather Prediction,NWP)系统的历史数据建立离线单体LSSVR模型库,然后根据预测时段与训练时段NWP序列的相似度从单体LSSVR模型库中动态选择候选集成成员,再后综合考虑正确性与多样性确定集成成员.最后由预测时段与训练时段NWP序列间的相似度分配集成LSSVR成员的权重.通过对湖南省某风电场输出功率进行预测,验证了动态集成LSSVR预测模型的有效性,与持续法、自回归求和移动平均法、单体LSSVR模型、常权重LSSVR组合模型及BPNN动态集成模型相比,动态集成LSSVR模型具有更高的精度,在天气非平稳变化阶段更加明显.
