N effective strategy to reduce carbon emissions is to replace traditional fossil fuels by developing clean renewable Traditional federated learning (FL) offers an effective solution energy sources. Among renewable energy sources, wind to data privacy disclosure issue in centralized data-driven energy stands out as one of the most significant, alongside methods. Under the FL framework, each turbine contributes hydropower [1]. Therefore, the efficient operation of wind its own data to jointly train a global model without direct turbines is crucial to maximize energy output. To optimize data exchange [10]. This collaborative learning method avoids the harnessing of wind energy, wind farms are commonly centralized data storage and protects the privacy and security established on ridges, mountaintops, or other elevated areas. of data. FL has already been first applied to detect blade icing The low-temperature climate in these areas can lead to blade in wind turbines using a heterogeneous framework [11].

The popularity and diffusion of wearable devices provides new opportunities for sensor-based human activity recognition that leverages deep learning-based algorithms. Although impressive advances have been made, two major challenges remain. First, sensor data is often incomplete or noisy due to sensor placement and other issues as well as data transmission failure, calling for imputation of missing values, which also introduces noise. Second, human activity has multi-scale characteristics. Thus, different groups of people and even the same person may behave differently under different circumstances. To address these challenges, we propose a multilevel heterogeneous neural network, called MHNN, for sensor data analysis. We utilize multilevel discrete wavelet decomposition to extract multi-resolution features from sensor data. This enables distinguishing signals with different frequencies, thereby suppressing noise. As the components resulting from the decomposition are heterogeneous, we equip the proposed model with heterogeneous feature extractors that enable the learning of multi-scale features. Due to the complementarity of these features, we also include a cross aggregation module for enhancing their interactions. An experimental study using seven publicly available datasets offers evidence that MHNN can outperform other cutting-edge models and offers evidence of robustness to missing values and noise. An ablation study confirms the importance of each module.