The electro-mechanical actuators (EMAs) play an important role in the new-generation aircraft, which makes the fault diagnosis of EMA become a hot topic in the industry. However, the EMA signals usually have nonlinear characteristics and seasonal tendency, which bring great challenge to the fault diagnosis. Furthermore, detecting faults in the early stage helps reduce the risk of serious damage to EMA, but most studies are focusing on the situation that the EMA faults are well-developed. To tackle the challenge, we present an innovative algorithm which combines a hybrid-spatial and temporal attention-based gated recurrent unit (HSTA-GRU) with Seasonal-Trend decomposition procedures based on Loess (STL) to predict multiple time-series data for more failure information. The STL extracts the seasonal factor for mitigating the influence of seasonal fluctuation, and the HSTA-GRU captures the spatio-temporal relationships among multivariate EMA sensors for a long-term prediction of multiple time-series data. Then, for the predicted time series, a similarity measure (SM) function based on dynamic time warping (DTW) is used to classify the fault types without training, so as to reduce the accumulated error and enhance the efficiency of classification. Ultimately, the analysis result on an experimental EMA fault dataset demonstrates that the proposed arithmetic can provide a superior performance not only in the time series prediction, but also for EMA fault diagnosis.

Due to the dynamic changes of network data over time, static intrusion detection systems cannot adapt well to the behavioral characteristics of the input network data, resulting in reduced detection accuracy. In addition, continuous input data streams will bring huge challenges to resource storage and computing costs. Therefore, we propose an intrusion detection method of dynamic ensemble incremental learning (DEIL-RVM), and realize a dynamically adjusted ensemble intrusion detection model. In which a new overall misclassification probability weight value (OMPW) based on incremental set or data chunk is designed as the basis for updating the ensemble model, and it can be used to prune and replace the poor base component in the ensemble model. We presented a probabilistic decision function taking into account the posterior probability of each base RVM model dividing the sample into each category. The RVM with high sparsity is used as the base component to obtain the good balance between the accuracy, robustness and resource consumption, which can sacrifice less time and storage cost in ensemble incremental learning while achieving higher detection accuracy and stability in network data streams.

For commercial aircraft, real-time fault detection is essential for condition monitoring of rotating engine components, which can improve aviation safety and reduce maintenance cost for airline companies. In this article, based on the adaptive kernel principal component analysis method, a real-time fault detection algorithm is proposed for turbine engine disk condition monitoring. A sample reduction strategy based on the k-nearest neighbors method is presented to speed up the kernel principal component analysis approach while still guaranteeing correct results. To efficiently detect fault, the fault detection model is updated timely to suit the working process of turbine engine disk. Sample clusters are obtained through the k-mean method, and the parameter of the kernel function is adaptively adjusted by minimizing the within-cluster distance and maximizing the between-cluster distance in the feature space. Experiments have demonstrated the superiority of the proposed approach in fault detection for turbine engine disk.

Measurement of the combustion temperature field is an extremely important issue in industrial production. Temperature is one of the key parameters in combustion studies. With the temperature field distribution of the combustion field obtained, heat transfer, heat convection, and heat radiation can be calculated directly and efficiently. Traditional background oriented Schlieren (BOS) is an effective method for non-axisymmetric temperature field measurements, but it requires simultaneous Schlieren imaging at multiple angles for tomographic reconstruction, which will greatly limit its application. In this paper, the compressive sensing algorithm is introduced into the temperature field reconstruction, which establishes the system of equations between the deflection angle and the refractive index gradient. Then, the reconstruction of the non-axisymmetric temperature field is realized by solving the underdetermined system of equations by the method of solving the sparse solution through the compressive sensing. First, light offsets across the non-axisymmetric temperature field are measured by the under-angled BOS system and image processing method. Second, the spatial refractive index field is reconstructed by the compressive sensing BOS method proposed in this paper. Finally, the spatial temperature field is obtained. The experimental results show that by comparing the iRadon reconstruction algorithm and the compressive sensing reconstruction algorithm, the temperature field reconstructed by the compressive sensing under the condition of the under-angled sampling of projection data had a higher accuracy than that reconstructed by the tomographic reconstruction algorithm under the same condition. The average error of the temperature field was reduced from 34.6 to 29.7 K under the same measurement conditions; however, the accuracy is better maintained by using the compressive sensing algorithm under the condition of undersampling projection.

To track the problem of aircraft fuselage defect detection in complex environments and reduce aviation safety hazards such as careless observation and delayed reporting due to objective factors, a semi-supervised aircraft fuselage defect detection network was proposed. Firstly, we constructed a new baseline model that extends one-stage detector with dynamic head and partial convolution named as dynamic decoupled detector, which enhances the representation capability of the model and improves the detection accuracy of small defects. Secondly, to address the issue of inconsistent pseudo-label distribution in semi-supervised learning, we propose a class-aware adaptive pseudo-label assignment strategy that adaptively obtains the pseudo-label filtering threshold during the training iteration to further optimize the pseudo-label assignment process. Finally, to validate the effectiveness of the proposed model, we construct a data set for aircraft fuselage defect detection for semi-supervised training. Experimental results show that the proposed semi-supervised aircraft fuselage defect detection network outperforms the current state-of-the-art semi-supervised object detection framework on the aircraft fuselage defect dataset. At the same time, the proposed model has better generalization performance and provides more reliable support for real-time visualization of aircraft fuselage defects.