The cross-domain diagnosis method based on transfer learning (TL) provides a solution for fault diagnosis of the aircraft fuel pump. However, due to the fuel pump working in harsh conditions, unknown faults or similar faults that are difficult to distinguish are likely to occur. The diagnosis results determine whether effective maintenance measures can be taken. To overcome these weaknesses, in this article, a class adaptive transfer learning network (CATLN) is proposed for detecting similar and unknown faults of aircraft fuel pumps. Firstly, according to the characteristics of the collected vibration signals, a multi-branch feature extractor is constructed to extract multi-scale features in time domain and frequency domain. Secondly, a class adaptive method is designed to reduce the inter-class distance of the same class by dividing the core area of the sample, and increase the inter-class distance of different classes to distinguish similar faults and unknown faults. The fault data collected in the ground experiment of the fuel pump are used for transfer tasks. The results demonstrate the effectiveness of the CATLN in the cross-domain fault diagnosis task containing similar or unknown fault classes.
The study of ethical leadership has emerged as an important topic for understanding the effects of leadership in the public sector. Existing literature finds that ethical leadership can significantly affect the desirable outcomes of subordinates, but most scholars only research from a single theoretical perspective of social learning or social exchange. In addition, the boundary conditions of the downward effects of ethical leadership have not been fully studied, and there is still a lack of clear theoretical explanations for why ethical leadership cannot effectively affect employee behaviors. This research examines the mechanisms and boundary conditions of ethical leadership influencing civil servants’ discretionary work behaviors. A total of 256 civil servants have participated in this study. Results revealed that ethical leadership can indirectly affect civil servants’ helping behavior and extra effort behavior through moral efficacy, and can also indirectly affect civil servants’ extra effort behavior through felt obligation. Social distance weakened the relationship between ethical leadership and moral efficacy, while power distance strengthened the relationship between ethical leadership and felt obligation. These findings broaden ethical leadership research in the public sector by differentiating the processes through which it affects civil servants’ discretionary work behaviors.
Research on workplace helping behavior highlights the need for a more balanced perspective that acknowledges both the positive and negative consequences of receiving help. The purpose of this study is to investigate how the mechanisms through which we receive autonomous and dependent help differentially impact recipient task performance, as well as the boundary condition for such effects. Drawing on social information theory, we examined the mediating role of task- and self-focused processes, and the moderating role of perceived prosocial motivation. Through a two-wave and two-source field survey, we collected matched data from 350 employees and their direct supervisors. We examined our hypothesized model with path analysis using Mplus 7.4. Results indicated that receiving autonomous help improved task performance by leading recipients into task-focused processes, and perceived prosocial motivation further strengthened this positive indirect relationship. In contrast, receiving dependent help reduced task performance by eliciting recipients to engage in self-focused processes, and perceived prosocial motivation further augmented this negative indirect relationship. Overall, we spotlight the differential consequences of receiving autonomous and dependent help on recipients and encourage further inquiry about the role of social information processing in the helping literature.