Search

For the current development of green civil aviation, this study aims to optimize the green four-dimensional (4D) trajectory of commercial flight by taking into account conventional cost and environmental cost. Some fundamental models, efficient processing methodologies, and conventional objectives are proposed to construct the framework of trajectory optimization. Based on the environmental cost including greenhouse gas cost and harmful gas cost, green objective functions are presented. The $\text{A}^\ast $ algorithm and the trapezoidal collocation method are employed to optimize the lateral path and vertical profile for 4D optimization trajectory generation. A case study for the A320 from Barcelona Airport to Frankfurt Airport yields the results that the optimal costs can be obtained under different objectives and the total cost can be more optimized by adjusting the weights of environmental cost and conventional cost. The study builds an aided tool for 4D trajectory optimization and demonstrates that environmental factors and conventional factors should be taken into comprehensive consideration when constructing the flight trajectory in the future, as well as it can underpin the green and sustainable development of the air transport industry.

This paper is concerned with the control design for a class of stochastic nonlinear systems. Three uncertainties are considered; that is, nonlinear parameter uncertainty, matched uncertainty and stochastic disturbance. The nonlinear uncertainty contains some uncertain parameter and satisfies bound condition. Neither the exact value of the matched uncertainty nor its possible bound is known; its upper bound function satisfies certain concave condition. The stochastic disturbance is a standard Wiener process. Based on stochastic Lyapunov stability theory, the adaptive robust controller is designed, which renders the state variables of the closed-loop system bounded in probability, regardless of all uncertainties. The desired controller is constructed by the upper bound function and the saturation function, in which the upper bound function represents the magnitude of the control, while the saturation function indicates the control direction. The design of the adaptive robust controller is based on the minimum information of uncertainty, which is simple and can be easily realized in practical systems. Finally, a two-tank water level control example is used to demonstrate the effectiveness of our control design.

When the battery data is under-sampled, the state of charge (SOC), a significant parameter of a battery management system (BMS) in electric vehicles (EVs) is more difficult to be estimated with high accuracy. Long-short term memory (LSTM) network combined with empirical mode decomposition (EMD) and a compensation strategy is proposed to solve this issue. EMD is used to decompose the battery sequence data into low and high frequency sub-sequences first. Further, voltage residue, one of the low frequency subsequences extracted by EMD is compensated through average voltage on the internal resistance so as to improve the polymerization degree of different working conditions. Then, multiple sub-sequences of the battery sequence data are fed into a constructed LSTM for training and predicting. Finally, two driving cycles at −10°C, 0 °C, 10°C and 25 °C ambient temperatures are applied to verifying the proposed method. The maximums of root-mean square errors of SOC estimation at four ambient temperatures are 2.1%, 2.0%, 1.8% and 1.1%, respectively.

The flow corridor is a tube-shape class of airspace designed for the future air transportation system, which aims to reduce complexity, restructuring the airspace to provide more system capacity. In order to support operational procedures design towards increased operational efficiency in the flow corridor, an accurate assessment of alternative procedures is a pre-requisite. This paper proposes a dynamic stochastic simulation framework including various microscopic behaviors for the assessment of distributed self-separation procedures for the air traffic in flow corridors. We first specify three prominent self-separation modes which distinguish flow corridors from today’s airways system, and present detailed self-separation procedures and algorithms in a parallel-lane flow corridor incorporating self-separating, lane-passing and lane-switch behaviors based on the aircraft dynamic model and the proportional derivative control theory. Then, incorporating these self-separation algorithms, a dynamic stochastic simulation modeling framework is constructed to assess and compare the alternative distributed self-separation procedures. The framework is applied to a parallel-lane flow corridor deployed from Beijing nearby airports (ZBAA, ZBTJ and ZBNY) to Guangzhou nearby airports (ZGGG, ZGSZ and ZGSD) in China, and the self-separation procedures were thoroughly assessed with both realistic and simulated data for benefits assessment and sensitivity analysis. Results show that the speed-based operational procedure is more suitable for high-density operations while the other two procedures have more flexibility which can be used for air traffic flow contingency management and/or trajectory management.

Low-contrast defects recognition is a dramatically difficult issue in the field of image recognition. The traditional machine vision method is mainly suitable for defects with obvious feature differences. In recent years, machine learning techniques have been successfully applied to the image analyses, and the deep learning methods provide new solutions for challenging problems in many areas. In this paper, a deep learning network framework based on the low-order residual network is proposed to detect low-contrast defects. Especially, a low-order feature extraction module is designed in order to effectively extract target features with low contrast and small size. The low-contrast watermark defects on complementary metal-oxide-semiconductor transistor (CMOS) camera modules are collected as the test objects to validate the effectiveness of the proposed method. The gray differences between the watermark defects and their adjacent areas are generally several gray-levels. The experimental results show that compared with the existing advanced classification neural network algorithms, the proposed method can effectively identify the watermark defects with a recognition accuracy of over 89%.

With the popular application of direct part mark (DPM) technology, DPM code inspection has been a hot issue in the machine vision. It mainly consists of two steps, namely, localization and decoding. DPM code localization is a key and complex step in the DPM code inspection. However, the traditional localization methods suffer from complex imaging environment, involving various imaging background, illumination, imaging distance, and exposures. Furthermore, the target itself, i.e., the DPM code, could be severely polluted or worn. Aiming at improving the performance and robustness of DPM code localization, an efficient method with depthwise separable convolution is proposed in this paper. The optimized network model has the advantages of a few parameters, high computational efficiency, high precision localization, and good generalization ability. Meanwhile, the precision of the DPM code region is improved with the help of multi-scale prediction. The experiments on our DPM code localization database demonstrate the effectiveness and flexibility of the proposed method in comparison with the YOLOv3 network and the Tiny_YOLO network. Furthermore, the proposed method can estimate the exposure level of the DPM code region, which is benefiting to the DPM code recognition and enables the adaptive ability.

To reduce the negative impact on flight delay of weather, airport delay prediction based on the improved Weather Impacted Traffic Index (WITI) model is proposed in this paper. Gaussian Mixture Model and OTSU method are used to improve the weather influence coefficient in the WITI model firstly. Correlation analysis is used to compare the original and improved WITI model. The correlation between the improved WITI model and the airport departure delay flight and delay time is 58.68% and 26.72% higher than original model. Then airport delay prediction model based on improved WITI through the random forest prediction model is constructed. Data of Shanghai Pudong International Airport (ZSPD) in 2016 is used for experimental verification, and R-square score of airport delay prediction model based on improved WITI model is 91.67% higher than the model based on original WITI model and mean square error is reduced by 84.01%.

Highly efficient blue quantum-dot light-emitting diodes (QD-LEDs) have been fabricated by substituting Mg-doped ZnO (MgxZn1-xO) for ZnO as an electron transporting layer (ETL). The device performance can be enhanced by optimizing Mg-doped ratios for the device structure of ITO/PEDOT:PSS/PVK/QDs/MgxZn1-xO/Al. The maximum luminescence efficiency and maximum brightness of blue QD-LEDs increase from 1.1 cd/A and 1500 cd/m2 for ZnO to 2.3 cd/A and 2500 cd/m2 for 10% Mg-doped ZnO (Mg0.10Zn0.90O), respectively. The improved device performance should be attributed to balance electron and hole injection due to Mg-doped ZnO increasing the band gap and reduce exciton loss at the interface between the ETL and the emissive layer.

With the complexity of traffic avement and the increase of vehicles, the traffic accidents of fatigue driving still account for a large proportion of the total traffic accidents. Drivers will wear glasses that meet their needs in order to reduce the incidence of accidents, but the occlusion of glasses and the changes and jitter of light during driving can significantly affect the accuracy of facial information detection. Aiming at the above problems, in order to extract the facial features accurately, an adaptive compensation infrared acquisition system is used, and a fatigue state detection model of parallel convolution neural network is proposed. Based on the different detection characteristics of the same image, the convolution neural network is used to automatically complete the feature learning, so as to obtain a more comprehensive description of the fatigue driving characteristics. The support vector machine is used to train the characteristics and establish the classifier to judge whether the driver is tired or not. Compared with other existing algorithms, this method obtains higher accuracy, meets the requirements of real-time detection and has high robustness to complex driving environment.

The configuration of collaborative robots is analyzed and optimized based on the Kriging model. First, through analyzing configurations of the existing collaborative robots, it can be seen that the offset is an important factor affecting the configuration, which is divided into S type and Y type by the influence on the range of motion of the joint. Then, based on the Kriging model, the multi-objective optimization function is established, which takes the length and offset of each link as parameter variables and regards the global performance index, the workspace volume index and the workspace dexterity performance index as optimization objective. And the influence of the link and offset parameters of the robot on the robot performance index is analyzed. In the end, the pareto optimal solution set and the optimization configurations are obtained using multi-objective genetic algorithm, and the performance is improved, the solution provides theoretical basis for the configuration design of collaborative robots.

Automatic vessel enhancement is significant for automatic analysis of the X-ray angiographic image sequence. However, the image is often interrupted by motion artefacts and other anatomical structures. In this paper, we propose an intra-frame and inter-frame constraint principal component analysis model (IFC-PCA) for vessel enhancement in the X-ray angiographic image sequence. The proposed method not only considers temporal information of image sequence by iteratively updating variable value on the basis of previous image information, but also takes the spatial information of each image into consideration. Experiments are conducted on a series of clinical X-ray angiographic image sequences. It demonstrates that our method is effective and robust for vessel enhancement from image sequence.

China has experienced sustained and rapid economic growth since its reform in 1978. Researchers have cast lots of attentions to explain China’s economic growth and attributed it to the rapid growing manufacturing in the process of industrialization. Regional economic growth and industrialization process in China from 1978-2016 are shown in this paper. Using province panel data, this paper explores the relationship between total output growth and manufacturing growth based on Kaldor’s laws to answer whether manufacturing is the engine of economic growth. The regression results confirm all hypotheses in Kaldor’s growth laws. Positive relationship between manufacturing and GDP growth is confirmed. Increasing returns to scale is verified in manufacturing. Employment from non-manufacturing to manufacturing can improve the labor productivity and growth rate of total output.

In the era of "Big data" and "Web 2.0", the viral speed of information dissemination magnified the data problem of subjectivity and timeliness in student satisfaction (SS) study. In this study, we conducted an innovative SS study based on a new data collection instrument which makes the data objective, timely, and even massive. Combining traditional Student Satisfaction Inventory (SSI) with frontier Natural Language Processing technology, we conducted the SS study based on social media. Instead of asking students to participate in, we used data mining and sentiment analysis technology to mine the existed data. Further, drawing from SSI, we proposed an SS analysis model using these collected data. To verify the validity of this method, we gave an empirical study.

In recent years, unmanned aerial vehicle has shown significant increase as a typical representative of the “slow and small” aircraft, which is becoming a threat to sensitive areas' defense, public safety and privacy, etc. Hence, the management of UAVs has become an urgent issue. Except for suppression, there is no effective interference mode in our country at present. This paper attempts to study the remote control signal of UAV from signal collection and analysis. Firstly, remote control signals for the typical civilian small UAV are received and analyzed based on the GNU Radio and USRP X310. Afterwards, in order to get the related parameters such as bandwidth, period and frequency hopping sequence, the signal is analyzed in both time and frequency domain using MATLAB. Next, the modulation of the unknown remote control signal is identified by using wavelet transform algorithm. Finally, the command signal is parsed out. This paper provides both theoretical basis for the interference signal or the control signal and new ideas for UAV's smart jamming, which is of great practical significance.

To improve the positioning accuracy of collaborative robots, an elastic deformation modeling method of manipulator robots is proposed in this paper. This approach is able to consider flexibility of robot modules comprehensively, and can compensate the elastic deformation caused by external wrenches and the deadweight of the robot in addition to possessing advantages of low computational complexity, easy modification as well as high real-time. Firstly, stiffness matrices of each robot module with 36 degrees of freedom (DOFs) are obtained by using the finite element analysis. Then, the kinematic model of a robot is established which is adapted to stiffness modeling of robot modules, and the force condition can be achieved by the robot statics analysis with the deadweight in order to acquire the elastic deformation of module units, which will be used to compensate the positioning error resulted from manipulator compliance. Furthermore, the elastic deformation model is built by the differential transformation method, and can be used to compensate the positioning error. Lastly, the finite element simulation is carried out, and results of the simulation show the proposed modeling method is effective.

Continuous descent approach (CDA) is an aircraft operating technique enabling the execution of a flight profile optimized with low engine thrust and drag configuration, thereby reducing fuel burn and emissions during descent. However, due to lack of consideration of aircraft type, gross mass and meteorological parameters, the current generic design criteria does not allow the fuel saving potential of CDA to be efficiently utilized. In this paper, we propose an Aircraft-specific Trajectory Optimization Model to reduce CDA routes fuel consumption given different aircraft states and meteorological conditions. The model defines a transition airspace concept for generating alternative CDA routes, and incorporates a classic point mass flight dynamics model to determine the aircraft trajectory and fuel consumption. Subsequently, the simulated annealing algorithm with Monte Carlo simulation are used to evaluate the average fuel burn before and after optimization. To gain insight on fuel saving efficiency, a comprehensive transition area length and aircraft states sensitivity analysis is conducted. We found that the transition area plays an important role in CDA optimization, with the increasing of the transition area length from 5 nm to 20 nm, the average fuel savings percentages could be increased by 10.76% to 22.62%.

This paper presents a complete static stiffness model of the prismatic joint of a modular reconfigurable robot based on the flnite-element-and-analytical combined method. The stiffness of the mechanical drive system was analyzed, and its tension and compression stiffness, which can be expressed by a formula, was obtained. The stiffness of the mechanical structure was extracted by employing the finite element method. The additional stiffness due to translational motion of the joint was analyzed. Then the comprehensive stiffness of the joint was obtained by combing the above stiffness parts. A simulation, which consists of four working conditions, was conducted, and the results show that the proposed method is valid.

This paper present a prediction model for three different objection (the airflow rate, the carbonaceous biochemical oxygen demand (CBOD) of the effluent, and the total suspend solids (TSS) of the effluent. The model is built by the MLP neural network. The accurancy of the prediction result of MLP neural network is compared with the accurancy of the result of trational stational autoregressive model(AR). The conclution is that the percentage error prediction model of the MLP neural network (PE), the fractional deviation (FB), normalized mean square error (NMSE), the mean absolute error (MAE) and mean square error (MSE) prediction model of evaluation index is better than the AR model. In other words, the prediction model based on MLP neural network provides a reliable basis for reducing the energy consumption in the activated sludge process of industrial waste water treatment and further improving its effect on the treatment of industrial waste water.

As the soaring development of electric vehicles and distributed generation systems, lithium-ion battery has been commonly used for energy storage. Accurate estimation of state of charge (SOC) is crucial for charging or discharging the batteries safely and reliably. However, the SOC is immeasurable and nonlinearly varies with factors (e.g., current rate, battery degeneration, ambient temperature and measurement noise), a reliable and robust algorithm for SOC estimation is accordingly expected. In this paper, an optimal nonlinear observer (ONLO) for SOC estimation is proposed. The particle swarm optimization algorithm is employed to optimize parameters of the nonlinear observer. The proposed approach is verified by experiments performed on INR18650-25R lithium-ion batteries produced by SAMSUMG SDI. Experimental results indicate that the proposed ONLO can accurately estimate the battery SOC with a mean absolute error of 1.8% and a maximum error of less than 6.5%, which are both lower than that of the unscented Kalman filter (UKF). Furthermore, the computation cost of the ONLO is reduced to 30% compared with the UKF.

Accurate prediction of model parameters and state of charge (SOC) is crucial for lithium-ion batteries, especially for those used in wireless charging systems. For example, estimation of the battery model parameters is beneficial to adjust the system operation frequency, improving the system efficiency. In addition, SOC prediction can guarantee the battery safety by preventing the battery from over-charging, and can be used as an indicator for limiting the battery's energy in an optimal range, which is helpful to extend the battery lifespan. This paper presents a joint estimation method for model parameters and SOC of the lithium-ion battery. Battery model parameters are estimated online by recursive least square algorithm with forgetting factor. Based on the online battery model, the battery SOC is predicted using a nonlinear observer algorithm. Validity of the presented method is verified by experiments based on both new and aged lithium-ion battery. Experimental results indicate that the proposed approach can accurately predict the battery SOC with errors less than 0.5%.

The algorithm of Partial Transmit Sequence (PTS) technique has high computational complexity in Orthogonal Frequency Division Multiplexing (OFDM) system. In this paper, a new low-complexity combined subblock segmentation method for PTS OFDM is proposed. For analytic purposes, we derive computational complexity expressions for the proposed segmentation method and analyze the computational complexity of the proposed segmentation method compared with that of the random segmentation method which has the best Peak to Average Power Ratio (PAPR) reduction performance. The simulation results show that the PAPR reduction performance is degraded only slightly compared with random segmentation method.

By analyzing the configuration of a 6R manipulator based on global performance index, the problem is proposed in this paper. The performances of two configurations cannot be compared based on the global performance index, if the configurations have same index. To solve the problem, an improved performance index is proposed in this paper. Firstly, this paper points out the boundary distribution index (BDI) based on the distribution of the singular points in the workspace. Secondly, the effectiveness of BDI is verified by comparing with the existing performance index. In addition, an improved performance index is established with the combination of the existing performance index and BDI. Then, the effect of offset on the manipulator is analyzed based on the improved performance index. Finally, the optimized configuration is obtained by the improved performance index and contrasts with other configuration. Then, the effectiveness of the improved performance index is verified.

With the rapid development of electric vehicles, lithium-ion battery has been widely used in energy storage systems. Accurate estimation of state of charge (SOC) is crucial for charging or discharging the batteries safely and reliably, as well as for indicating the remaining driving range for drivers. In this paper, four popular model-based SOC estimation algorithms, namely extended Kalman filter, unscented Kalman filter (UKF), sliding mode observer and nonlinear observer (NLO) are compared in terms of prediction accuracy, tracking ability to initial SOC error, and computation complexity. Since most resonant topologies for wireless charging systems are featured as constant current, a charging cycle combined with a constant current stage and several pulsed current stages is developed to evaluate the performance of these algorithms. Experimental results indicate that the UKF method performs better than the other three methods in terms of prediction accuracy, while the NLO performs best in terms of tracking ability to initial SOC error and it has the similar accuracy with the UKF method.

One-class support vector machine (OCSVM) is one of the most well-known kernel methods, in which one attempts to use the hyperplane defined on feature space to separate the normal data from all other possible abnormal data. However, the standard OCSVM ignores the importance of the data distribution. With respect to the drawback of OCSVM, the information of data distribution is introduced to improve the robustness of algorithm and descript the normal data more reasonably. Meanwhile, combining with the solution of pre-image problem, we propose a novel method which takes the characteristic of data into full consideration for pattern denoising. In the paper, denoising procedures are completely presented and experimental results on a toy dataset and several real-word datasets explicitly indicate the applicability and effectiveness by comparing with some former pattern denoising methods.

This paper reports an autopilot design for a robotic unmanned surface vehicle in the control laboratory at DMU. A robust adaptive steering law is developed with the aid of a predictor, a tracking differentiator, neural networks, and a dynamic surface control technique. The developed controller is able to achieve satisfactory performance in the presence of model uncertainties, time-varying ocean disturbances, and measurement noises. Simulation results demonstrate the efficacy of the proposed method.

Battery is the major energy storage source of electric vehicles (EVs), which makes the accurate estimation of State-of-Charge (SOC) very crucial. Various SOC estimation methods have been proposed in recent years. Among of them, Ah method is widely used due to its easy implementation. This paper presents an energy method for SOC estimation of power lithium-ion battery, which is based on the law of energy conservation. The mathematical model is firstly built to describe its principle. A MATLAB-Simulink simulation model and a laboratory prototype are established to verify the performance of the energy method. Simulation and experimental results show that this SOC estimation method is more accurate than the Ah method, especially under high current and variable current conditions.

LiNi 1/3 Co 1/3 Mn 1/3 O 2 , well known as the ternary material, is currently considered as one of the promising cathode materials for lithium-ion battery (LIB). In application, accurate estimation for state of charge (SOC) is crucial for the performance and safety use of LIB. However, it is a hard work as the SOC cannot be directly measured and is affected by many factors, such as ambient temperature, current rate and cycle number. Moreover, the influence of electromagnetic interference on measurement results (current, voltage and temperature etc.) makes the situation more terrible. In this paper, an approach of SOC estimation for LiNi 1/3 Co 1/3 Mn 1/3 O 2 battery, which is based on the Unscented Kalman Filtering (UKF) is presented. An improved model for online SOC estimation under abominable operating conditions is presented, in which the battery rated capacity variation with ambient temperature and discharge current rate are taken into consideration. A modified two-order RC equivalent circuit model is employed to simulate the battery characteristics, based on which the state-space equation of battery is established. Experimental results show that the presented model can capture the battery characteristics very well and the UKF-based SOC estimator can ensure the estimation accuracy even if the measurement is incorrect.

Air traffic capacity is a prerequisite and foundation of air traffic flow management (TFM). The sector capacity is usually given a static value in traditional TFM. In fact, it is time varying because it is always affected by weather, airspace status, and many other factors. A more refined TFM requires dynamic sector capacity. How to accurately evaluate dynamic capacity is the difficulty and development trend of sector capacity evaluation. The paper firstly presents the definition and the influential factors of the sector dynamic capacity. Then the characteristic of dynamic sector capacity affected by the main dynamic factor of severe weather is analyzed. In considering sector structure, civil aviation regulation, controller workload and airborne weather radar, sector dynamic capacity estimation model is set up. At last a prototype example is given to evaluate the sector dynamic capacity. The results verify the model is valid and practical.

This paper introduces nine characteristics of service, which impact on customer satisfaction. Aiming at improving service quality, three service strategies for improving service quality are put forward. These are across-the-board service strategy, omnipotent service strategy and full service strategy. After analyzing the three ways including providing high quality products, offering perceived service in the pre-sales phase, improving the link of communication for enterprises to satisfy customer demands, the three strategies are subdivided into 9 detail strategies. Then the QFD is built on the basis of service characteristics as customer demands and service strategies as technical requirements. Finally, an evaluation method of service strategy is put forward based on the QFD. This model provides suggestion to enterprises on selecting service strategy to meet the different needs of customers.

A novel and efficient method to simulate the behavior of guidewires in the vascular system is proposed in this paper. The graph-theoretical method is based on the principle of minimal total potential energy. We formulate the total potential energy in the vascular interventional system as the summation of the elastic energy of the guidewire and the energy due to the deformation of the vessel wall. A graph is constructed with low complexity ensuring the efficiency of the single source shortest path. Compared to previous results, experiments in three phantoms have been conducted to evaluate the performance of the proposed method and the results demonstrate that our method can achieve 20% improvement with faster running time.

This paper mainly focuses on the design of a novel type of intelligent DDV actuator system for aeroplane. The function, structure principle and redundancy management of the intelligent DDV (Direct-Drive-Valve) servo actuator system for aeroplane is described in detail. The intelligent DDV servo actuator which is the important part of the system is also described in the paper, including the structure and performance characteristic, the key technique, and so on. An experiment is conducted to verify the dynamic performance and fault monitoring function of the new system. The hardware-in-loop experimental results prove that the design and application of the intelligent DDV servo actuator system on some aeroplane could solve the technical problems in real world. Also, the control precision, stability, and fault monitoring of the system can perfectly satisfy the aeroplane control system design requirements.