Distributed Multi-wind Farm Short-term Power Prediction for Data Privacy-preserving

Abstract

Purposes Deep learning based on centralized data can effectively improve the prediction accuracy of wind power, but the serious consequences of data leakage make wind farms constantly pay attention to the confidentiality of their own data, which hinders data-driven wind power prediction methods. Methods To solve the above problems, a short-term power prediction method was proposed for distributed multi-wind farms oriented to data privacy protection, which uses the Horizontal Federated Learning framework to complete the power prediction task of wind farms. First, wind farms adopt the distributed training method, and use the Temporal Pattern Attention (TPA) mechanism and Long Short-Term Memory (LSTM) network to form a TPA-LSTM local model to complete local data training. Then, the parameters of these local models are aggregated, and the weight values of the models are introduced to improve the contribution rate of the local models with good fitting effect. Finally, the update of global model parameters is implemented. Findings The results of experiments show that the global model obtained by this method has good prediction performance and generalization ability in multiple scenarios under the premise of ensuring the privacy of wind farm data.

Keywords

• data privacy
• horizontal federated learning
• long short-term memory
• temporal pattern attention
• wind power forecasting