Multi-Sensor Data Fusion for Wind Farms Based on KF-SBOA-BiLSTM

To effectively improve the accuracy of multi-sensor data fusion in wind farms, a data fusion method based on KF-SBBOA-BILSTM is proposed, which gives the fusion data high precision. First, for the problem of more noise and outliers in the sensor data, the Kalman filtering algorithm is used to perform filtering operations on the original sensing data to improve the data quality. Secondly, the Secretary Bird Optimization Algorithm (SBOA) combines the bidirectional long and short-term memory network (BiLSTM). Parameters such as the number of hidden layers and the individual weights between hidden layers of BiLSTM are optimized by SBOA to further improve the fusion accuracy of BiLSTM. The experimental results show that the fusion data with the proposed method have higher accuracy, and the proposed method has higher reliability and Ruban properties and can predict the sensing data of wind farms more accurately.