Abstract:Triangular rotor is a core component of small aviation Wankel rotary engine, subjected to complex load coupling effects such as temperature, inertial force and gas explosion pressure in the process of highspeed rotation, and more prone to failure and damage due to insufficient strength. In view of the stress concentration and strength evaluation of triangular rotor, a model of the engine's thermodynamic working process is established to obtain the change law of the cylinder temperature, cylinder pressure and heat transfer coefficient of the combustion chamber in a single cycle of the engine,and calculate the thermal boundary conditions everywhere in the rotor.Under conditions of mechanical stress, thermal stress and thermomechanical coupling, the finite element method is adopted respectively to simulate and analyze the temperature field, stress field and deformation of the triangular rotor. And optimization methods such as the processing of rounded corners at the edge of the round hole at the waist of the rotor and the arrangement of heat sinks at the cooling hole are proposed, and comparative analysis is carried out.The simulation results show that after the optimization, the stress at the edge of the circular hole at the waist of the rotor is reduced to 403.9 MPa from 687.0 MPa, which is about 58.7% of the original, and the stress value at the cooling hole is also reduced to 113.2 MPa from 202.6 MPa, which is 55.9% of the original. Cooling fins being arranged, a drop of rotor at temperature is more than 20 K on average, the temperature of the round hole at the waist of the rotor and the cooling hole dropped is about 40 K, and the deformation of the sealing groove tip is reduced to 0.15 mm, from 0.21 mm, i.e. a reduction of 27.7%.

Abstract:AbstractIn view of the consumption prediction of aviation guided ammunition, consumption point prediction and interval prediction models are established respectively.According to the characteristics of small sample, nonlinearity and strong randomness, the support vector regression model is adopted to predict the consumption data, and the optimal parameters are found by particle swarm optimization algorithm. Combining with the point prediction error data, the error uncertainty analysis is conducted by kernel density estimation to determine the error probability density curve, and determine the best confidence interval for a given confidence using highestdensity regions based on kernel density estimation.The results show that the proposed method can provide more accurate prediction results and uncertainty change interval for the use consumption data, and provide a reference for the subsequent use arrangement of aviation guided ammunition.

Abstract:This research is proceeding in all cases from the overall situation to continuously narrow the research scope and efficiently carry out the exploration of infrared anti jamming evaluation, a sequential uniform experimental design method is proposed to carry out the research. The infrared anti jamming evaluation test scheme is designed by adopting the method, and the test is carried out by the digital simulation system. In the process of experimental research, the detailed information of each stage can be mastered and adjusted appropriately according to the situation, until the final satisfactory test area and miss distance are obtained. The results show that the method is effective and feasible, and the research scope is gradually reduced. Compared with the other methods, this method is unique in advantages, fast on convergence, and good in effect, and provides a method for the study of infrared anti jamming evaluation.

Abstract:When aircraft with multi wheel and multi strut turns on the ground, both design methods are uncertain about loads determination and distribution. Based on elastic tire theory and the turning motion model of nose wheel landing gear, a ground turning motion model of multi wheel multi strut landing gear is established. The lateral loads and vertical loads distribution law of each single landing gear is analyzed. The harsh loading landing gears are picked out in combination with stable turning conditions. The analysis results are verified by the result obtained from flight test with the strain method. Lateral loads and vertical loads prediction models of the landing gear are established under two kinds of strategies, which are used to predict the limit turning loads, and finally the prediction result is verified by the measured result. The results showed that the loads direction is opposite between the front main landing gear and the latter landing gear. Compared with the middle landing gear, the magnitude of the main landing gear loads is much smaller than that of the other two. The vertical loads distribution is related to the filling of the landing gear stut, and the distribution ratio of lateral loads is different from the vertical loads. The established landing gear loads prediction models are accurate and can effectively reduce the risk of the aircraft ground limit turning test.

Abstract:In wide angle inverse synthetic aperture radar (ISAR) imaging, serious migration through range cells (MTRC) will lead to the defocus of ISAR image. A wide angle ISAR imaging method based on U-net convolutional neural network (U-net CNN) is proposed, Firstly, the echo data is preprocessed by fast Fourier transform to obtain a defocused ISAR complex value image as the training samples; Secondly, according to ISAR imaging characteristics, the u-net structure is improved, and an imaging network with good focusing ability is obtained after training. Simulation results show that compared with traditional wide angle ISAR imaging methods, the proposed method reduces the peak sidelobe ratio (PSLR) of ISAR image to less than 18 dB, has smaller image entropy and minimum mean square error (NMSE), and the imaging time is reduced to about 0.28 seconds. Under the condition of low signal to noise ratio (SNR), the proposed method can still achieve fast and accurate reconstruction of ISAR image.

Abstract:Aimed at the problems that the convolutional neural netwok (CNN) method in polarimetric SAR image classification is long in training time, and slow at convergence speed, and the original Softmax function cannot effectively deal with the intra class differences of polarimetric SAR images, a model based on finetuning and addinga polarimetric SAR image classification method is proposed by Additive Margin Softmax (AM-Softmax). This method improves the efficiency and classification accuracy of the CNN model through the overall fine tuning of the pre trained network, and then replaces the Softmax with AM-Softmax to solve the problem of large intra class variationin SAR images and further improve the classification accuracy. The experiments show that this method is fast on convergence and deal with the problems of large variation in polarization SAR images within a class, and the overall classification accuracy of the model reaches above 96%.

Abstract:he vortex electromagnetic wave being generable and the spinning target being imageable, a high resolution two dimensional imaging method based on the rotating antenna is proposed. A radar observation model based on the rotating antenna is constructed firstly, and then the target range information is obtained by using the range direction fast Fourier transform (FFT), and the sinusoidal expression of Doppler effect containing the target azimuth information is derived. Finally, the orthogonal matching pursuit (OMP) algorithm is applied to the sinusoidal signal to remake at an angle of azimuth to the target, effectively improving the azimuth angle resolution in the range of radar beam. The algorithm requires radar to be low in the performance, overcoming the dependence of image quality on pure multi mode vortex electromagnetic wave, and achieving high resolution two dimensional imaging of stationary targets. The simulation results show that the proposed algorithm is valid and robust.

Abstract:Iridium is widely used in low orbit satellite communication system currently, and its airborne transceivers have been using in a variety of civil aviation aircraft. Aimed at the problems that the Iridium satellite system being relatively earlier in design on no consideration of the interference of the newly developed satellite navigation system, and the airborne electronic environment being more and more complex, and the threat to Iridium user link signal being increased, more complete electromagnetic compatibility analysis is needed, a model of receiving Iridium downlink signal is established. The antenna isolation and bit error rate under different interferences are analyzed, and the simulation results are verified by experiments. The effects of wideband and narrowband interference with different frequencies and duty cycles on the quality of downlink signal received by the Iridium airborne equipment are tested, which is in good agreement with the simulation. The results show that under the same power level, the influence of wideband interference is greater than that of narrowband interference. The impact of different types of wideband interference is not much different. The impact of in band and out of band wideband interference is not much different, but the impact of narrowband interference decreases first and then increases. The impact of pulse interference is positively correlated with the duty cycle.

Abstract:Multi functional microwave devices play an important role in modern communication systems because they can achieve large capacity function integration in small sized devices. However, multifunctional devices often exhibits serious channel crosstalk due to the high integration, which significantly reduces the device efficiency. Here, a frequency multiplexing method with low channel crosstalk is proposed. A double Cshaped slot resonator and a double C shaped metal resonator are utilized in a complementary form on two dielectric boards for realizing a hybrid meta atom with double operation modes. The high Q value being achieved in hybrid resonator, the phase can be controlled independently at two operation bands with low frequency ratio. To verify the concept and explore possible applications, a bifunctional metasurface with two distinct functions are designed at f1=9.2 GHz and f2=11.2 GHz. The numerical results are in good agreement with the experimental ones, indicating a focused OAM beam with mode number l=3 at f1 and a zero order Bessel beam at f2. Compared with the previously reported wavelength multiplexing devices, the frequency ratio of dual operation modes here is only 1.2, and the efficiency is measured as 86.1% and 93% respectively. The wavelength multiplexing method with low frequency ratio and low channel crosstalk provides an effective avenue in high capacity integrated function application.

Abstract:Aimed at the problems that current research on the static deployment of software defined satellite network controllers is ignorence of the dynamic topology of satellite networks， and the user data traffic is unsteadly，a multi controller dynamic deployment method based on the modified whale optimization algorithm (MWOA) is proposed in comprehensive utilization of the dynamic characteristics of satellite networks by setting three threshold values and adopting switch migration in combination of the software defined satellite network architecture. The simulation results show that compared with other algorithms, MWOA has significant advantages of switching migration costs, controlling link delay, and load balancing. This method can further improve the processing performance of satellite networks and meet the needs of communication for users.

Abstract:With the rapid development of the information society, taking video sensors as the front end for acquiring information is of great significance in effectively finding specific objects through pedestrian re identification algorithms to protect people’s lives and property. This paper applies deep learning to the field of person re identification, and embeds the multi scale attention fusion module into the neural network for multi scale feature extraction and representation, effectively improving the recognition performance of the attention mechanism for deep learning networks. The paper proposes a multi scale channel attention fusion module based on SE block in combination with the ResNet50 convolutional neural network to extract features, further extract the feature sequence context information through the bidirectional LSTM network, and improve the model’s ability to extract important image features. At the same time, the attention to redundant features of images is reduced. Finally, the network model is jointly trained by the cascaded hard sampling triplet loss function and the cross entropy loss function, clustering the samples in the high dimensional feature space, and further improving the model recognition accuracy. Market1501 dataset and CUHK03 dataset are tested by the proposed algorithm respectively, and compared with other attention module algorithms under the same conditions. In order to further verify the function of each module, an ablation experiment is performed by the algorithm to verify the effectiveness of each module. The experimental results show that the proposed method can be effectively applied to person re identification

Abstract:Aimed at the problem that the demands of the modulation recognition algorithm based on deep learning for the labeled samples are too heavy, a multi task training strategy based on meta learning is adopted. This strategy obtains a ability of cross task signal recognition through a large number of different task training networks to make the network quickly adapt to new signal categories with a small number of samples. In order to extract the features from signal samples more comprehensively, a hybrid parallel feature extraction network is designed, completing the recognition task by measuring the distance between sample feature vectors. And a joint loss function is introduced to take both inter class and intra class distance into account, making the samples realize comparison more efficiently after feature extraction. The experimental results show that the algorithm can achieve a highest recognition accuracy of 88.43% when there are only five labeled signal samples.

Abstract:In view of the low prediction accuracy of ResNet, a method of DenseResNet (DRNet) for approximation of autonomous systems and series prediction is proposed based on the relationship between observed data and the system phase trajectory. Firstly, in order to strengthen the extraction and the circulation of ‘feature information’ contained in the data, all the outputs of previous layers in each hidden layer of feedforward neural network are concatenated as an input of this layer to form a dense block. Secondly, to avoid the ‘degradation’ phenomenon occurs when the depth of neural network increases, the residual mechanism is introduced to connect the input layer and output layer of the dense block to form the DRNet. Finally, DRNet is applied to the linear model, Damped single degree of freedom system and nonlinear models, SEIRS model and Logistic Volterra model. The results show that DRNet outperforms the ResNet, Back Propagation Neural Network (BPNN) and DenseNet in terms of model approximation and prediction accuracy on both datasets of 5 000 and 10 000. According to the four evaluation indexes on the nonlinear models, DRNet has high effectiveness on autonomous systems. The DRNet also shows good noise immunity for its better performance on the data with 5% noise.

Abstract:Aimed at the problems that the target image detection methods are low in accuracy and slow at detection speed of current UAV (Unmanned Aerial Vehicles, UAV), a target detection algorithm in combination with the lightweight network and the improved multi scale structure is proposed. Firstly, MobileNetV3 lightweight network is used to replace the backbone network of YOLOv4, reducing the model complexity and improving the detection speed. Secondly, the improved multi scale PANet network is introduced to enhance the flow superposition of high dimensional image features and low dimensional location features, and improve the classification and location accuracy of small targets. Finally, the K means method is introduced to optimize the parameters of the target anchor frame to improve the detection efficiency. Meanwhile, a new UAV target image dataset Drone dataset is constructed by combinating with the open dataset and the self shot images. The results show that the mAP and FPS of the proposed algorithm reach 91.58% and 55 f/s, and the parameter number of 44.39M is only 1/6 of the YOLOv4 algorithm and is superior to the mainstream SSD, the YOLO series and the Faster R CNN algorithms.

Abstract:A Lanchester model is proposed to describe the modern electronic fire attack. First of all, a model of incomplete information antagonism is proposed based on second linear law and the square law of Lanchester. Then, the model is extended to the cooperative confrontation model by bring in the factor of fire distribution. And then, based on the analysis of the effect of electronic attack on firepower, the damage probability and information factor are discussed, and the multivariate Lanchester model is given. Finally, the rationality of the proposed model is verified by simulation.

Abstract:To address the problem of low prediction accuracy of existing network security posture prediction models, a prediction method based on Stacking model fusion is proposed. In this method, the TCN network, WaveNet, GRU, and LSTM are integrated with the Stacking algorithm to explore the correlation among the situational data; after that, logistic regression is used to further predict the final situational values; the particle swarm optimization algorithm is used to optimize the parameters and improve the model performance. Based on two data sets for validation, the experiments show that the proposed prediction method has small mean square error and mean absolute error, fast convergence speed, and the fit degree can reach 0.999, which can well solve the problem of low prediction accuracy.

Abstract:A breakage phenomenon of a M6×30 35CrMnSiA fastening screw is found in the process of repair at a certain type of an aircraft. This case affecting the use safety of the aircraft, the chemical composition, macro and micro fracture, microstructure and mechanical properties of the broken fastening screw are verified and analyzed. The fracture form of the screw is confirmed as fatigue fracture, and there are no metallurgical defects, heat treatment defects and processing defects. The main reasons for the crack are preliminarily inferred by the machining mode of the screw. Subsequently, the fatigue performance of screws with the same material formed by screw rolling is verified. The same batch of materials as the broken fastening screws and the same processing technology except screw rolling are selected to conduct a comparative analysis of pull pull fatigue test for the reproduced fastening screws. Three parallel samples are selected for each processing method, and the test method is GJB 715.30A—2002 \[Fastener Test Method for Tensile Fatigue\]. The test results showed that The fatigue life of the rolling screw reaches 500,000 times without any damage, while the thread fracture occurs when the fatigue life of the machining thread is 50 200 times, and the fracture is similar to the fracture of the fastening screw in service. The fatigue performance of the screw formed by rolling wire is much better than that of the screw made by thread turning. In order to eliminate the follow up risks, the paper suggests to improve the screw forming process and stop using this batch of screws. The research results provide a basis for the optimization of screw and the improvement of technological method, and are of great significance for the improvement of aircraft safety.