Chenguang Zhang

In recent years, preventive maintenance has emerged as a focal point of research in the aerospace field. The concept of equipment maintenance, exemplified by prognosis and health management (PHM), has permeated every aspect of development and design. Extracting degradation features presents a fundamental and challenging task for health assessment and remaining useful life prediction. To facilitate the efficient operation of the incipient fault diagnosis model, this paper proposes a data-driven feature extraction process for converters, which consists of two main stages. First, feature extraction and comparison are conducted in the time domain, frequency domain, and time–frequency domain. By employing wavelet decomposition and the Hilbert transform method, a highly correlated time–frequency domain feature is obtained. Second, an improved feature selection approach that combines the ReliefF algorithm with the correlation coefficient is proposed to effectively minimize redundancy within the feature subset. Furthermore, an incipient fault diagnosis model is established using neural networks, which verifies the effectiveness of the data-driven feature extraction process presented herein. Experimental results indicate that this method not only maintains fault diagnosis accuracy but also significantly reduces training time.

Owing to peak power demands of pulsed power load like radar and beam weapon is much larger than capability of generator, researches about energy storage equipment sizing optimization have been extensively carried out. However, these researches are mainly considered from static perspective without taking dynamic characteristic into consideration. To cover this gap and meet strict weight requirements of airborne platform, this paper proposed a hybrid energy storage system sizing process to satisfy peak power demand while dynamic standard set up by MIL-STD-704F is also studied by means of spectrum and impedance analysis. The overall filter capacitor on DC BUS is considered as a part of energy storage system for improving system dynamic characteristics. The research content is divided into four parts. First, the power waveform is discussed through spectrum analysis to give full play to hybrid energy storage system. Second, the influence of energy storage equipment on system dynamic characteristics is analyzed and the results are taken as constraints for optimization. Then, combined with dynamic and static constraints, a hybrid energy storage system sizing process depend on Non-dominated Sorting Genetic Algorithm II and decision-making method are proposed to obtain the most suitable result. Finally, simulation model is established in MATLAB/Simulink environment for verification.

The distributed arrangement of airborne equipment is an important feature of more electric aircraft. The health status assessment system and method need to be designed for the distributed feature. This paper proposes an intelligent distributed PHM health state assessment method and system for more electric aircraft. The system adopts distributed architecture and optical fiber-wireless hybrid communication to improve the reliability and flexibility of the assessment system. Furthermore, an expert weight determination method based on grey synthetic incidence analysis is proposed to calculate the corresponding weights for different distributed assessment results, which can effectively improve the assessment accuracy.