Search

The increasing range of ancillary products offered by airlines is making existing, static frameworks obsolete. The changing expectations of customers have created a need for more dynamic and reactive offers. In order to tailor an offer to an individual journey, it is possible to leverage similar journeys and the observed outcomes in a semi-supervised approach. In this paper, a multi-stage deep learning framework, namely Deep Ancillaries Prediction (DAP), is developed to understand personalised demand for airline ancillaries and improve pricing strategies. DAP aims to solve the overlapping distribution problem and class imbalances observed in real-world airline datasets. The framework incorporates a contrastive learning module to learn richer feature embeddings and an autoencoder for semi-supervised learning into one framework, and outperforms current ancillary prediction systems. The modules can be trained separately and hence, are suitable for an online learning setting. This framework is designed to be transferable to different prediction tasks in the airline industry. Significant performance enhancements are attained compared to the current state-of-the-art algorithms.

Rapid and accurate fault diagnosis under limited data has become one of the most important abilities of photovoltaic (PV) power generation systems. This study proposes an improved K-Nearest-Neighbor(KNN) method, which is based on the current at the maximum power point, voltage at the maximum power point and weather data. By using this method, short circuit, open circuit and shading of a PV string can be diagnosed quickly based on data obtained by inverters and weather monitor. Finally, a large number of data was obtained through a credible model whose data is proved to be consistent with the measured data greatly. Diagnosis result of the method was evaluated through these data.

In this article, a compact dual-band bandpass filter based on quad-mode substrate integrated waveguide (SIW) resonators with controllable frequency is proposed. The bandwidth of the filter can be adjusted by changing the position and angle of the metallized vias in the SIW resonant cavity. The X-shaped slot on the resonant cavity extends the transmission path of the surface current, which can realize the controllability of the filter frequency.

Plasma enhanced chemical vapor deposition (PECVD) is commonly known to be used in the field of silicon thin-film solar systems for the application of nanocrystalline silicon (nc-Si:H) film. The chemical deposition is a rather lengthy process, and it is difficult to determine the crystallization and crystalline phase of the thin film prior to X-ray diffraction (XRD) measurements. In this study, we are trying to analyze the spectral data collected by optical emission spectroscopy (OES) to find out there is any correlation between OES data and crystalline status. We used machine learning onto an in-situ detection tool to forecast this correlation. The collected large-scale OES spectral data obtained via principal component analysis (PCA) was used for the prediction of the crystalline phase in films without necessary experiments performed afterwards. Therefore, this method can be applicable to the field of thin film deposition for the detection of properties on thin films.

For a class of time-delay discrete systems with constraints, actuator faults and polytopic-type uncertainties, the problem of robust fault-tolerant control is researched in this paper. Firstly, a new state space model with output error is introduced for reference trajectory tracking control, which provides more scope for controller designs and improves control effect. Then based on the predictive control theorem, a min-max robust model predictive control (RMPC) algorithm which considers time delay and the fault model is proposed. Using linear matrix inequality (LMI) technology and Lyapunov theory, the min-max optimization problem is transformed into a minimization problem and the asymptotic stability of closed-loop system is proved. Finally, the algorithm comparison experimental results on the quadrotor semi-physical simulation platform validate the feasibility and effectiveness of the proposed control scheme.

In this study, pulsed dc reactive sputtering of aluminum nitride (AlN) thin films was investigated. The aluminum nitride thin films were deposited on Si (100) using a reactive direct current (DC) unbalanced magnetron sputtering system. The DC reactive sputtering was used in sputtering the aluminum targets in a mixture of argon (Ar) and nitrogen (N 2 ) plasma. Processes of aluminum target sputtering were carried out in an atmosphere of a mixture of Ar and N 2 . However, pulsed DC reactive sputtering of aluminum targets was carried out at total pressures with N 2 :Ar ratios from 7:30 to 45:15. In-situ optical emission spectrometry (OES) was applied to obtain the optimal deposition rate and the highest sputtering yield from the effects of flow nitrogen/argon (N2:Ar) ratio and pulse frequency on OES intensity. Thus, we have compared Fourier-transform infrared spectroscopy (FTIR) spectra and X-ray diffraction (XRD) patterns of AlN films deposited on Si (100) by DC reactive sputtering with an Al target in the mixture of Ar and N 2 . FTIR and XRD investigated the quality of the films and the preferred orientation.

In this paper, active fault tolerant control (AFTC) of unmanned aerial vehicles (UAVs) with bounded disturbances and an actuator fault is studied. Firstly, the system state variable and an unknown actuator fault are obtained through adaptive fault estimation (FE), then, it is proved that the estimation error of the dynamic system is asymptotically stable. Secondly, an active fault-tolerant control (AFTC) scheme combining integral sliding mode (ISM) and fault estimation is proposed. Under this scheme, the influence of disturbance and unknown actuator fault can be eliminated simultaneously in a finite time. Finally, we can illustrate the effectiveness of AFTC scheme by an example.

The paper is concerned with UAVs formation system in case of single UAV actuator fault. For distributed UAVs formation, a new fault diagnosis method based on the distributed unknown input observer is proposed. According to the hierarchical concepts of distributed control system, the controllers for single UAV and multi-UAV formations are designed, respectively. When the single UAV actuator fault occurs, an effective fault detection method for distributed systems with unknown input observers is presented. Then, augmented system is proposed to estimate fault. In Simulation, using FDC toolbox to build the model of UAV. According to known model parameters, numerical simulation can be built to diagnose the actuator fault. Simulation results show the efficiency of the proposed method.

This paper presents an interval observer-based fault detection (IOFD) scheme in UAVs formation system. Firstly, for the formation system in healthy case, the interval observer is constructed based on the known boundary information of the disturbance term and relative output estimation error. Then, the residual errors that can be used for detecting the actuator faults are developed by the output estimation error. Different from the traditional FD schemes, it is not necessary to design the threshold generators in the IOFD scheme. At last, MATLAB proves the correctness and feasibility of the IOFD scheme.

Fault detection and isolation (FDI) plays an important role in the safe operation of flight control system. For a complex flight control system, it is difficult to build an accurate physical model, so data-driven method has emerged. In this paper, an extreme learning machine (ELM) method optimized by genetic algorithm (GA) is proposed. Compared with traditional feed-forward neural network, ELM has great generalization performance at a high learning speed. Furthermore, GA is used in this method to adjust hidden nodes parameters to find the most suitable parameters that minimizes the ELM model error. What is more, different fault models are built respectively, so as to achieve the purpose of fault isolation. All in all, the proposed method can isolate fault while optimizing the extreme learning machine.

For a class of time-delay discrete systems with constraints, actuator faults and polytopic-type uncertainties, the problem of robust fault-tolerant control is studied in this paper. Firstly, a new state space model with output error is introduced for reference trajectory tracking control, which provides more scope for controller designs and improves control effect. Then based on the predictive control theorem, a min-max robust model predictive control (RMPC) algorithm is proposed, which considers time delay and the fault model. Using linear matrix inequality (LMI) technology, the min-max optimization problem is transformed into a minimization problem. Finally, the asymptotic stability of close-looped system is proved and the algorithm comparison experimental results on the quadrotor semi-physical simulation platform validate the feasibility and effectiveness of the proposed control scheme.

This paper investigates the problem of fault-tolerant consensus control of leader-following non-linear multi-agent systems (MAS) subject to actuator faults. Based on the relative states of neighbors, the consensus error dynamic is given. Then, a sliding mode surface based on consensus errors is proposed and the stability of the states on the sliding mode surface is proved by using Lyapunov theory. Furthermore, a sliding mode fault-tolerant consensus protocol is proposed to compensate the actuator faults. According to the sliding mode control theory the new proposed sliding mode fault-tolerant controller ensures that the consensus of MAS can be reached in finite time. Finally, an example of leader-following multi-robot systems is presented to demonstrate the effectiveness of the proposed controller.

This paper proposes a new interpolation design method to generate smooth feedrate profile for tool center point (TCP) trajectory. The developed method plans the TCP and B, $\pmb{C}$ axes velocities according to several criteria such that the maximum acceleration and jerk within the blocks are all constrained. Simulation results are conducted to validate the proposed method.

Adaptive-based sliding mode control scheme is developed to solve the problem of mismatched uncertainty that exists in communication channels and external disturbance for a class of distributed large-scale systems. A linear sliding manifold is adopted to guarantee asymptotic stability of each subsystems and achieve the global robustness. Meanwhile, by adjusting the parameters estimation of adjustable gains via proposed adaptive scheme, the upper bound of the unknown uncertainty can be obtained. Then an distributed adaptive sliding mode controller is designed combining with the relative information of neighboring subsystems. It is shown that the presented controller has the capability to achieve the asymptotic stability of each closed-loop subsystems via Lyapunov theory. Finally, a simulation example applied to a network of two quad-rotor dynamical systems is given to exhibit the effectiveness of the proposed control scheme.

In this paper, a sliding mode observer based fault diagnosis method was proposed for a class of linear systems with the disturbance and model uncertainty to detect the actuator faults. The difficulty on the sliding mode observer method can be directly solved by using linear matrix inequalities (LMIS). The stability of the designed sliding mode observer can be guaranteed by Lyapunov stability theory. The proposed actuator fault diagnosis method is tested in a 3-DOF aircraft simulation platform to illustrate the performance of the proposed observer and the effectiveness of the fault diagnosis algorithm.

Haptics is a new and important interaction method for virtual reality. As an important haptic interaction device, force feedback device can improve the reality of interactive operation. In this paper, the force feedback interface (OpenHaptics) is fully analyzed. Physical models are constructed with physical attributes such as friction, stiffness, inertia and gravity added. The principle of simulated physical properties of force feedback device is discussed in detail. The principle of simulated physical properties by force feedback device is described. Taking reducer as an example and using Visual Studio 2008 as a development platform, combined with OpenHaptics library, complete the integration of physical attributes with geometric models and interactive operation simulation of force feedback device.

We reported the perovskite based field effect transistors (CH3NH3PbI3-FETs) using HfO2 as the bottom gating dielectrics. A typical hysteresis behavior has been observed in the transfer characteristics of CH3NH3PbI3-FETs with a large hysteresis window. The modulation of gate voltage on the hysteresis behaviors has been demonstrated by applying different gate voltage sweeping ranges. The shift of hysteresis window in the transfer characteristics of CH3NH3PbI3 based transistor has been also investigated under the illumination, which may contribute to a fast photoresponse.

Polarity modulation of single-layer WSe2 field effect transistor is investigated by using hydrazine as a solution-processable and effective n-type dopant for WSe2. Compared to the intrinsic hole-dominant ambipolar behaviors, highly effective n-type doping characteristics are achieved after hydrazine treatment. It is found that the on-current improves obviously by one order of magnitude and the metal-WSe2 contact resistance decreases at the same time. After hydrazine being removed, the electron doping effect disappears which indicates hydrazine treatment is a reversible doping process. This work provides a possibility for transition metal dichalcogenides-based logic device applications in the future.

A flexible rGO-perovskite hybrid photodetector has been fabricated in a solution-based method. X-ray diffraction and UV-absorption characterization demonstrate the high quality of the CH3NH3PbI3 thin films. Through the electrical measurements, it is found that low dark current and representative time-dependent photoresponse to different wavelength is obtained at the same power density. Fast raise/decay time, high on-off ratio and good flexibility make it possible to be applied in the future potential optoelectronic devices.

For a class of time-varying state delay systems with uncertainties and additive actuator faults, a fuzzy global sliding mode controller (FGSMC) is proposed, which eliminates the reaching mode; A sufficient condition is given with Finsler's lemma and linear matrix inequality technique which guarantees the asymptotic stability of the sliding mode; Fuzzy logic is used to estimate the upper bound of uncertainties and faults. The simulation results show that the proposed FGSMC can alleviate chattering and overcome the influence of uncertainties and additive actuator faults.

In this paper, a control method using sliding mode variable structure control combined with adaptive control is proposed. Based on Lyapunov stability theory, a sliding mode variable structure controller is designed to ensure that the closed-loop system is asymptotically stable in system faulty cases. Then, a modified control law is designed to cope with the situation that without knowing the upper bound of the perturbation. At last, The fault tolerant control is carried out on the fault of the system failure of the flight control system. Finally, simulation results demonstrate the feasibility and effectiveness of the proposed theory by using MATLAB software.

The paper is concerned with development of sliding mode for uncertain linear time delay systems. A robust adaptive fault-tolerant compensation control system design method with actuator faults and exogenous disturbances is proposed. A delay-ahead predictor is used to handle system's time delay. Matrix full-rank factorization technique with sliding surface design is used to solve the actuators faults. An adaptive sliding mode controller replaces a fault detection and isolation mechanism to update automatically for compensating the effects of actuator faults. The asymptotic stability of closed-loop systems is guaranteed by mean of adaptive H 2 performance. Numerical examples are given to illustrate the validity of the results.

In this paper, a new feed drive model which considers the longitudinal and base vibration models is developed to estimate the elastic deformation. The effect of vibration force feedback to the torque is included. The dynamic model is integrated with the servo drive system. To obtain the dynamic parameter, the peak-picking method is adopted. It is shown from the experimental results that the developed model can predict the elastic deformation quite accurately. Furthermore, by properly designing the acceleration time, one can reduce the base vibration for the feed drive system significantly.

According to the different types of faults of the complicated system of civil aircraft, this paper designs a kind of fault diagnosis expert system based on the fault tree analysis. The designing method of each module of knowledge base is given in detail, and the process of calculation is simplified through direct binary decision diagrams (DBDD) method which analyzes component probability importance. The problem of confliction of rules is also solved effectively by introducing a priority coefficient combined with component probability importance and fault diagnosis time. Landing gear retraction system is an example in order to analyze and explain the feasibility of the design of fault diagnosis expert system for civil aircrafts.

This paper study the adaptive robust sliding mode controller design of a class of uncertain systems with time-varying delay. The uncertainty is norm-bounded, but the upper bound is unknown. Firstly, an integral sliding surface is designed so that the reaching phase has been eliminated. A sufficient condition such that the sliding mode is asymptotically stable is given in term of LMIs(Linear Matrix Inequality). The condition involves a few matrix variables but could reduce the conservatism. A new adaptive law is presented to estimate the unknown upper-bound of the uncertainty. Then a sliding mode based controller is proposed so that the system state stay on surface. Finally, an example is given to show the effectiveness of this method.

As the key equipment in photovoltaic system, the operating reliability of PV array influences on the security and stability of photovoltaic system deeply. To diagnose the fault of PV array effectively, the paper proposes an improved wavelet neural network algorithm for fault diagnosis of PV array, in which uses Gaussian function as activation function and adds additional momentum method and adaptive learning rate method to the training algorithm. The proposed algorithm is applied to fault diagnosis of the actual PV array, the experimental results demonstrate the effectiveness of the algorithm.

When doing post-competition analysis in team games, it can be hard to figure out if a team members' character attribute development has been successful directly from game logs. Additionally, it can also be hard to figure out how the performance of one team member affects the performance of another. In this paper, we present a data-driven method for automatically discovering patterns in successful team members' character attribute development in team games. We first represent team members' character attribute development using time series of informative attributes. We then find the thresholds to separate fast and slow attribute growth rates using clustering and linear regression. We create a set of categorical attribute growth rates by comparing against the thresholds. If the growth rate is greater than the threshold it is categorized as fast growth rate; if the growth rate is less than the threshold it is categorized as slow growth rate. After obtaining the set of categorical attribute growth rates, we build a decision tree on the set. Finally, we characterize the patterns of team success in terms of rules which describe team members' character attribute growth rates. We present an evaluation of our methodology on three real games: DotA,1 Warcraft III,2 and Starcraft II.3 A standard machine-learning-style evaluation of the experimental results shows the discovered patterns are highly related to successful team strategies and achieve an average 86% prediction accuracy when testing on new game logs.

Training and handling working dogs is a costly process and requires specialized skills and techniques. Less subjective and lower-cost training techniques would not only improve our partnership with these dogs but also enable us to benefit from their skills more efficiently. To facilitate this, we are developing a canine body-area-network (cBAN) to combine sensing technologies and computational modeling to provide handlers with a more accurate interpretation for dog training. As the first step of this, we used inertial measurement units (IMU) to remotely detect the behavioral activity of canines. Decision tree classifiers and Hidden Markov Models were used to detect static postures (sitting, standing, lying down, standing on two legs and eating off the ground) and dynamic activities (walking, climbing stairs and walking down a ramp) based on the heuristic features of the accelerometer and gyroscope data provided by the wireless sensing system deployed on a canine vest. Data was collected from 6 Labrador Retrievers and a Kai Ken. The analysis of IMU location and orientation helped to achieve high classification accuracies for static and dynamic activity recognition.

von Willebrand Factor (vWF) is secreted as the ultra-large (UL) multimeric glycoprotein by endothelial cells and megakaryocytes. It can be cleaved into small multimers by metalloprotease ADAMTS-13 following its secretion. As a member of a disintegrin and metalloproteinase with thrombospondin family, ADAMTS-13 plays a crucial role in thrombosis regulation. Many studies have revealed that unsuccessful cleavage of UL-vWF by ADAMTS-13 induced thrombosis. Here, we have purified A domain, which is the most important activity unit of vWF, and the full-length ADAMTS-13. Results of SDS-PAGE and Western blot confirmed two purified proteins have correct molecular weights. Flow cytometry analysis showed that the microspheres were functionalized via coating purified ADAMTS-13 on their surfaces. Using binding assay, microspheres coated with ADAMTS-13 can adhere on petri dishes coated with immobilized vWF A1A2A3 under static conditions, which were resulted from the binding between ADAMTS-13 and vWF A1A2A3. These results demonstrated purified ADAMTS-13 and vWF A1A2A3 have functional activities and can be used as an important tool for exploring the mechanism of interaction between vWF A1A2A3 and ADAMTS-13 in further studies for thrombosis regulations.