Educational theory claims that integrating learning style into learning-related activities can improve academic performance. Traditional methods to recognize learning styles are mostly based on questionnaires and online behavior analyses. These methods are highly subjective and inaccurate in terms of recognition. Electroencephalography (EEG) signals have significant potential for use in the measurement of learning style. This study uses EEG signals to design a deep-learning-based model of recognition to recognize people’s learning styles with EEG features by using a non-overlapping sliding window, one-dimensional spatio-temporal convolutions, multi-scale feature extraction, global average pooling, and the group voting mechanism; this model is named the TSMG model (Temporal-Spatial-Multiscale-Global model). It solves the problem of processing EEG data of variable length, and improves the accuracy of recognition of the learning style by nearly 5% compared with prevalent methods, while reducing the cost of calculation by 41.93%. The proposed TSMG model can also recognize variable-length data in other fields. The authors also formulated a dataset of EEG signals (called the LSEEG dataset) containing features of the learning style processing dimension that can be used to test and compare models of recognition. This dataset is also conducive to the application and further development of EEG technology to recognize people’s learning styles.

The wind energy exploitation technique has been developed very quickly in recent years. The vertical axis wind turbine is a hot research domain due to several advantages: low noise, flexible for installation, ease of maintenance, great safety and credibility, etc. The aerodynamic performances of different forms of airfoils including an active deformation airfoil and a fluid-solid coupling passive airfoil with two-dimensional (2D) and three-dimensional (3D) cases have been investigated numerically in this paper. Firstly, the aerodynamic performances of the airfoils with the maximum deformation amplitudes of their cambers which are 3%, 5% and 7% of the chord length have been discussed, respectively, with the angles of attack in the range of 0° and 20°. Secondly, for the angle of attack set at 18°, the two-way fluid-solid coupling simulations with the Young’s Modulus of 1 Mpa and 2 Mpa have also been investigated. Results show that: (1) for the pseudo 3D and real 3D single active deformation airfoil cases, the lift coefficients increase as the maximum deformation amplitudes augment from 3% to 7% of the chord length at the same angle of attack. With the same maximum deformation amplitude, when the angles of attack increase from 0° to 20°, the lift coefficients which increase firstly and then decrease are bigger than that of the original NACA0012 airfoil. When the maximum deformation amplitude of the pseudo 3D airfoil reaches 5% of the chord length, a relatively good aerodynamic performance with better inhibition effect of vortex generation can be obtained. The 3D vortex distribution demonstrates that the deformable airfoil has a better vortex generation controlling effect at the middle cross-section along the spanwise direction than the non-deformable airfoil. (2) From the aspect of fluid-solid coupling, the lift increases and the drag decreases so that the lift to drag ratio has a big improvement when the Young’s Modulus is equal to 1 Mpa and 2 Mpa. The deformable airfoil can inhibit the generation and the shedding of the surface vortex when the fluid-solid coupling effect is considered.

The adaptive learning environment provides learning support that suits individual characteristics of students, and the student model of the adaptive learning environment is the key element to promote individualized learning. This paper provides a systematic overview of the existing student models, consequently showing that the Elo rating system has greater potential as compared to the other models regarding application in the online learning environment. Based on the Elo model, this study proposes the EELO, an enhanced Elo rating system, in consideration of the application scenarios of polychotomously scored items and multi-dimensional granularity evaluations not covered by the basic Elo rating system. The EELO model estimating students’ cognitive abilities and predicting their future performances on unknown questions is evaluated based on one public set (Assigment2) and one proprietary dataset (HSK), and achieved an AUC of 0.92 for Assigment2 and 0.84 for HSK, which shows that the EELO model has the best performance regarding the above-mentioned objectives as compared with the latest extensions of the IRT and BKT models. Subsequently, the EELO model was tested and applied successfully in a real large-scale online learning environment to demonstrate the potential of the EELO model in adaptive learning applications.

Water jet propulsion is widely used in various military and civilian fields due to its advantages of simple structure and high propulsion efficiency. The process of mooring involves utilizing specially designed equipment to secure a ship at a designated berth. During the process of water jet propulsion, the single propeller operates within a complex and turbulent three-dimensional flow. Hence, studying the coupling between the water jet propeller and the hull is critical to comprehending the characteristics of the device and the distribution of the flow field in detail. Firstly, we conducted computational fluid dynamics (CFD)-based self-propulsion calculations to evaluate the interaction between the hull and the propeller. We subsequently analyzed the propeller's performance and the forces acting on the hull to understand how the presence or absence of the hull influenced the water jet propeller. Finally, we performed calculations and analysis of the cavitation characteristics of the coupling between the hull and the water jet propeller, considering different rotational speeds and water depths at the bottom of the pool. The study demonstrated that the presence of the hull boundary layer under the hull-propeller coupling condition led to reduced uniformity of propeller inlet flow and lower efficiency of the propulsion pump. However, it also increased the bias toward low-flow conditions. Additionally, increasing the impeller speed led to a gradual increase in the cavitation volume within the water jet propeller, resulting in a gradual decrease in the propeller's performance. This research provides the technical support required for effective design and operation of water jet propulsion systems. This paper involves studying and analyzing the performance and flow field of the coupling between the hull and propeller under mooring conditions with a specified hull model.

Thermal radiation is dominant in oxy-fuel combustion technology. When considering the difference between conventional air combustion and burgeoning oxy-fuel combustion caused by diverse components of fuel gas, it is critical to improve the accuracy of numerical simulation for radiation by CFD. The line-by-line model is considered as the most accurate approach to simulate thermal radiation while it will cost vast calculation resources on the other hand. This approach is applied as a benchmark for the validation of other approximate models in most cases. The corrected-K distribution method (CK) has attracted considerable attentions in recent years. the NBCk model always has an excellent performance when the temperature gradients are small. However, when the temperature gradient gets higher, the correlation assumption breaks down and leads to non-negligible errors. To improve the accuracy of the CK model, we introduced a new method called the multi spectral corrected-K distribution model based on the functional data analysis (FDA) in this work. The objective of this MSCK method is to group together wavenumbers according to the spectral scaling functions defined as the ratio between spectra in different thermophysical states. Over these intervals, the correlation assumption can be considered as nearly exact. Results show that the MSCK model always achieves a better performance on calculation accuracy while sharing the same computational cost as the traditional CK model.