Abstract:Wind power ramp event is a small probability and large hazard event, for effectively avoiding its impact on the grid, an offshore wind power ramp prediction method based on optimal combination model is proposed. First, on the basis of the output power data of existing offshore wind turbines, the data preprocessing of the wind power series is carried out by using the Variational Modal Decomposition method optimized by the novel Seagull Optimization Algorithm, which is decomposed into several Intrinsic Mode Function components and one residual component to extract the salient features of the power signal and reduce the data fluctuation. Second, the prediction models of Extreme Learning Machine and Bayesian optimized Long Short Term Memory network are built, respectively, while the optimal combined prediction method is adopted to combine the two single models for obtaining the optimal combined model. Individual components are treated as the inputs of the model, then the final offshore wind power is obtained by superimposing and integrating the prediction results of each component. Eventually, adopting the definition method to predict and characterize the offshore wind power ramp. The research shows that, compared with BP, RNN, LSTM and single model, the wind power prediction error of the combined model proposed in this paper is smaller, with RMSE of 71.10 kW, MAE of 50.26 kW and SMAPE of 0.01%, and it can completely predict 7 ramp events with a higher accuracy. Meanwhile, the applicability of the proposed method is verified by varying the time resolution. The final results show that the proposed method in this paper has a high accuracy in offshore wind power ramp prediction.

A direct prediction method for wind power ramp events considering the class imbalanced problem

Wind Power Ramp Events Prediction Considering Wind Propagation

Two-stage Prediction Framework for Wind Power Ramps Considering Probability Distribution Distance Measurement

An Algorithm for Forecasting Day-Ahead Wind Power Via Novel Long Short-Term Memory and Wind Power Ramp Events

Short-term Wind Power Ramp Forecasting with Empirical Mode Decomposition Based Ensemble Learning Techniques

Offshore wind power ramp prediction based on optimal combination model

Ramp events forecasting based on long‐term wind power prediction and correction

Wind Power Ramp Events Prediction Considering Time-frequency Characteristics

Short-term prediction of wind power and its ramp events based on semi-supervised generative adversarial network

Wind power ramp prediction algorithm based on wavelet deep belief network

Short-term wind speed prediction of wind farms based on improved particle swarm optimization algorithm and neural network

Selection of Time Window to Predict Wind Power Ramp Events

An Integrated Evaluation Method of the Wind Power Ramp Event Based on Generalized Information of the Source, Grid, and Load

A wind power ramp event detection method based on eigenvalue correction and trend integration

Ramp Event Forecast Based Wind Power Ramp Control with Energy Storage System

An Adaptive Weight Ensemble Learning Method for Probabilistic Wind Power Prediction

Wind power prediction model based on deep learning neural network

A novel network training approach for solving sample imbalance problem in wind power prediction

Research on Wind Power Ramp Events Prediction Based on Strongly Convective Weather Classification

Integrating wind variability to modelling wind-ramp events using a non-binary ramp function and deep learning models

Wind ramp event prediction with parallelized Gradient Boosted Regression Trees