Abstract:Based on the differential manifold theory, a design method of robust dynamic boundary protection system for ice formation on aircraft is proposed. First, the dynamic boundary of preice formation on aircraft is calculated by the differential manifold theory, and its accuracy and efficiency are verified by Monte Carlo method. Secondly, taking the influence model of icing factor as foundation, the variation rule of aerodynamics boundary under different icing degree is studied and analyzed. Then a robust dynamics boundary of aircraft is proposed, and a robust dynamic boundary protection system for aircraft under ice condition is designed. Compared to the traditional dynamic boundary, it is not necessary to obtain accurate ice type information and ice severity when the aircraft encounters icing adverse flight environment. In consideration of the current ice detection technology, the method proposed in this paper has more extensive engineering application significance. Before the aircraft goes beyond the dynamic boundary, the boundary protection system can keep the flight state within the safety threshold, ensuring the safety and controllability of the aircraft.

Abstract:The influence of icing on the aerodynamics of aircraft is a focal point of aircraft icing research contents. An aircraft model on the background of a typical geometry of large passenger aircraft is constructed. By the RANS method, the complex flow field structure and aerodynamic characteristics of the aircraft under condition of wing icing are studied. The research results show that the wing icing mainly affects the aerodynamic characteristics near the stall point of the background aircraft, and the departure of the airfoil flow separation is greatly in advance and the separation ladder disappears completely, which is the direct reason for the deterioration of the stall characteristics and the reduction of the aerodynamic boundary. The study can provide full support to deeply understand the aerodynamic flow mechanism of the effect of aircraft icing on aerodynamics, and provide a basis to analyze aerodynamic characteristics and flight dynamics of large aircraft after icing.

Abstract:The quantitative assessment of flight risk under condition of ice formation on wings can provide reference indicators for aircraft icing airworthiness verification. A quantitative risk assessment method of the extreme value theory combined with the Copular model for wings under condition of ice formation is proposed. Monte Carlo simulation experiments being made based on the complex dynamics model of pilotaircraftenvironment, the extreme values of key flight parameters (airspeed, roll angle and AOA) are extracted. The goodness of fit test is made by using the identified parameters. Joe Copula model can more accurately describe the joint distribution of threedimensional extreme value parameters in verification. The flight risk probability is calculated through the established risk probability criterion. The analysis shows that icing risk events are often to the accompaniment of varying degrees of overrun and coupled effects of key flight safety parameters. The proposed method can provide a theoretical reference for the quantitative assessment and predictive control of flight risk under icing conditions.

Abstract:Taking asymmetric ice formation on wings caused to happen by troubles in single wing in deicing system as background, the stability in the aircraft's lateraldirectional system is studied. Firstly, a complex lateraldirectional model composed of dynamics model, actuator model, asymmetrically icing model, and the control law of stability augmentation is constructed. Secondly, by using the Monte Carlo algorithm, the calculation nodes are divided by sideslip angle β, roll velocityp, yaw velocityr. The stability points of each calculation node are obtained by dynamic simulation, and the stability regions near the stability points are obtained by the continuousadvancing algorithm proposed in this paper. and, the multiparameter coupled stability domain of aircraft's lateraldirectional system is constructed. Finally, the differences between stability regions in the case of noice and asymmetrical severe ice are analyzed. The results show that the asymmetric ice on wing may cause the deviation and contraction of the stability region in aircraft's lateraldirectional system, and the stability region contracts by under condition of asymmetric severe ice formation. Therefore, the aircraft deviates from predetermined trajectory under condition of asymmetric ice formation on wing, pilot must avoid rough operation surface, or aircraft will easily cross the stability region to induce flight accidents.

Abstract:Aimed at the problems that the traditional power converter fault diagnosis method is poor in versatility, the additional hardware is needed, and the opencircuit devices fail in location, taking an asymmetric halfbridge power converter as the research object, the installation position of the traditional current sensor is improved, through analyzing the characteristics of the current slope under different faults, a diagnostic scheme based on the current slope is proposed. A mathematical model of fault variables is built up based on the input of logic signals, detection current and its slope taken as input by the scheme. Through the output value of the model, the open and shortcircuit faults of the power tube can be quickly located, and the diagnosis of each phase is independent of the motor phase. The number and control method are limited, and the versatility is strong; no additional hardware is required, and the diagnosis function is rich and the controller has a small burden. The scheme is valid.

Abstract:Aimed at the problems that the traditional fault detection and diagnosis methods based on fixedthreshold have a high false alarm rate and fail to realize effective fault detection and isolation, an adaptive fault detection and isolation method for hydraulic servo system based on multistage observer is presented. The firststage RBF neural network is adopted as a fault observer of the hydraulic servo system, and the residual error signal is generated by comparing the estimated observer output with the actual measurements. The secondstage RBF neural network is employed as an adaptive threshold producer, realizing the adaptive fault detection. Features of the residual error signal are extracted by using wavelet packet analysis, and the system fault isolation is made by using the thirdstage RBF neural network. The experimental results show that the multistage observer is effective in detecting and isolating the failure in the hydraulic servo system.

Abstract:In order to investigate the influence of flat tail deflection Angle on the plane's landing and taxiing performance, through aerodynamic data by using the method of computational fluid dynamics reserves and by using the multibody dynamics modeling method, this paper builds up a model able to reflect the aircraft aerodynamic characteristics changes over tail peace at an Angle of attack Angle of aircraft landing run dynamics under condition of different plane landed flat tail deflection for taxiing simulation calculation. The results show that the dynamic model can well reflect the dynamic characteristics of aircraft at different horizontal tail deflection angles during landing and taxiing, and the horizontal tail leading edge downward deflection (pull rod) can effectively shorten the landing and taxiing distance.

Abstract:Mixed Reality (MR) maintenance is characterized by the unique advantages of low cost, low risk and high efficiency, and is a research hotspot in the field of military equipment support. The concept of MR is introduced, and the composition and the key technology of MR maintenance system are described in this paper. The applications of MR in military equipment teaching and training, battlefield emergency repair and intellectualized maintenance are introduced. The existing problems are analyzed, and the corresponding solutions are given out. Finally, the development trend of intellectualization, standardization and modularization in the systems is pointed out, and the suggestions for its application in PLA are put forward.

Abstract:Aimed at the problems that the prediction of aero-engine exhaust gas temperature is low in precision, and the mining of the time dimension information of flight data by traditional RNN neural networks is inadequate, an exhaust gas temperature prediction model is proposed in combination with Autoencoder neural network and temporal convolutional network (TCN). First, the features related to the exhaust gas temperature (EGT) from the flight data are extracted by the Autoencoder method, and then the reduceddimensional features are taken as the input to build up a TCN network deep learning model. Secondly, the exhaust gas temperature of the aeroengine is used as the output to fully mine the flight data dimensional information, improving model accuracy. Finally, the actual flight data is selected to be an experiment. The results show that compared BP to LSTM, the average absolute percentage error is reduced from 13.035%, 9.593% to 3.369% respectively, effectively improving the prediction accuracy of the model.

Abstract:Aimed at the problems that the cognitive limitation and decision preference of decision makers leave out of consideration, and the individual regret and group value are under no consideration synthetically in the current air combat threat assessment, a threat assessment model is constructed based on the three reference points prospect theory and the visekriterijumska optimizacija I kompromisno resenje (VIKOR). First, according to the independence between the schemes to be evaluated, a planning model of maximizing the foreground value of the evaluation scheme set is established to calculate the index weight, avoiding the uncertainty brought about by the subjective weighting method. In order to overcome the problem of information loss and the decrease of the resolution of decisionmaking results caused by the simple reference point of the index average value, the reference approach technique for order performance by similarity to ideal solution (TOPSIS) is introduced, and the positive and negative ideal points are used as the parameters, the fuzzy analysis method is used to determine the preference weight of decision makers for difference reference points. Finally, in order to balance the group value and the individual regret, the subjective preference of the decisionmaker is on reflection, and the prospect value is combined by using the VIKOR method to rank the threat of the target. The method is feasible and valid.

Abstract:In order to solve the problem of the different distribution between the testing samples and training samples, this paper proposes a modulation recognition method based on the domain adaptive neural network. Firstly, VGG 16 deep convoluted neural network is utilized for extracting the features of wavelet transform images.Then the high dimensional features are reduced by using the autoencoder, and the CORAL losses between training samples and testing samples are calculated.Finally, the optimal classification loss and the CORAL loss are combined to optimize the model.The simulation results show that under condition of different signal categories or different channel environments, the recognition accuracy of signals tested can be improved more than 5% by introducing domain adaptation technology.

Abstract:Aimed at the problem that the maximum exposure time of the server in the dynamic defense platform is hard to determine, this paper utilizes the fuzzy controlrelated theories and technologies for designing a fuzzy controller, determining the maximum exposure time of the virtual server and minimizing defense costs due to change. Thepaper assumes that the defender knows the attack surface and the attacker cannot complete the attack within a certain period of time. Firstly, Starting from the platformlevel moving target defense principle, the mathematical expectation of the system attack surface is measured. Secondly, the system vulnerability of each virtual server is quantitatively analyzed, and these two indicators are used as inputs to the fuzzy controller to ensure the security of the system. Finally, the maximum exposure time of the virtual server is calculated, providing an important reference for determining the time point during the virtual server switching process.

Abstract:Aimed at the problems that the inter symbol interference is severe and the bit error rate is high at the receiving end of slow decaying variable air multipath channels,on the basis of SC-FDE system, the sparsity of airborne multipath channels is analyzed, the airborne channel estimation model is taken as arecovery problem for sparse signals, and a PN-SAMP channel estimation algorithm is proposed based on a compressionaware framework by using PN sequences to construct a deterministic measurement matrix with SAMP as the recovery algorithm. The estimated mean square errors of compressionaware algorithms are compared to that of the traditional PN and LS algorithms, and the error performance of several estimation methods applied in SC-FDE systems is compared in conjunction with MMSE equalization. The simulation results show that the estimation errors and the bit error rates of compressionaware class algorithms are smaller and lower than that of the traditional PN and LS algorithms respectively, and the bit error rate of the compressionaware class algorithms is less than at a signaltonoise ratio of 20dB. The robust of the PN-SAMP algorithm is better than that of the OMP algorithm in the case of unknown sparsity, further meeting the estimation needs of sparse channels in the timevarying slowdecay aviation.

Abstract:In order to suppress the artifacts in the reconstructed target image obtained by the conventional back projection methods, a 2D scaled Wiener postfilter based method is proposed in this paper. Firstly, for a steppedfrequency continuouswave (SFCW) radar system, the frequency domain BP (FDBP) imaging method is formulated as a 2D distortionless beamformer. Then, the 2D Wiener postfilter with a scaling factor is adopted to minimize the mean square error of the beamforming result. At last, the influence of the scaling factor on the imaging performance for both single target and multiple targets is analyzed. Coherence factor (CF) based methods, which have been used for artifact suppression, can be interpreted as the special cases of the proposed method with different scaling factors. By selecting a proper scaling factor, the limitations of CF based methods can be overcome. Simulation results demonstrate that, both in range direction and azimuth direction, the artifacts in the reconstructed target image can be effectively suppressed by the proposed method.

Abstract:A magnetoelectric (ME) dipole antenna with broadband characteristics simple in structure is designed, fabricated, and measured in this paper. This antenna being different from the common dipole antenna, two vertical aluminum plates are added in the traditional Г shaped feeding structure to improve impedance matching and impedance bandwidth. These plates form a set of capacitive loading to compensate the inductance effect of the original feeding structure at high frequencies, effectively broadening the impedance bandwidth of the antenna. The experimental results show that an impedance bandwidth of 92.5% for |S11|<-10 dB, ranging from 1.36 GHz to 3.7 GHz, is achieved. Besides, symmetrical radiation patterns with low cross polarization levels, high fronttoback ratios (FBRs) and gains with maximum value of 11dBi are realized within a range of the entire working band.

Abstract:Aimed at the problems that the current centralized network update method remains to be long in update time and excessively dependent on the controller, and in order to save the network update time and slow down the controller load, a distributed update method geared to the airborne network of aviation swarm is proposed. A modified update message is designed. The controller carries out the construction of update message by generating the update dependency graph, encodes the update dependency relationship into the update message, and sends it to the data plane. And then the switch distributed update operation is made by using the query and notification message to update the order coordination, achieving the distributed network update in the data plane. The simulation results show that compared with the existing centralized update methods, the proposed distributed update method can reduce the update coordination time by reducing the number of node interactions during the update period, saving much update time of the business flow and the entire network.

Abstract:Damage Assessment and Rush Repair (BDAR) in battlefield is the most important means by improving aircraft survivability and aviation unit's combat capability. In view of the damage modes of military aircraft Electrical Wiring Interconnection System (EWIS), this paper analyzes the corresponding effects caused to happen. While introducing the experience and practices of foreign military and civil aviation, the paper makes a suggestion on the damage detection and the emergency repair of EWIS for military aircraft.