Abstract:In order to master the current development status and development trends of memristor based neuromorphic chips, the existing memristor based neuromorphic chips and architectures were investigated. The memristor array structure and integration process, anterior and posterior neuron circuits, multi-array interconnection topology and data transmission strategy used in the chip, as well as the system simulation and evaluation methods used in the chip design process were compared and analyzed. It is concluded that the current circuit design of memristor based neuromorphic chips still need to solve the problems of limited resistance states, large device parameter fluctuation, complex array peripheral circuits, small integration scale, etc. It is pointed out that the actual application of this type of chip still faces challenges such as the improvement of memristor production process, improvement of development tool support, special instruction set development, and determination of typical traction applications.

Abstract:To solve shortcomings of traditional hierarchy directed graph method when applied to satellite ground station fault diagnosis, the EHDG (enhanced hierarchy directed graph) fault diagnosis method was proposed. In process of modeling, considering large number of fault symptoms and complex modeling, nodes of the same kind were consolidated according to the fault propagation paths of devices working status to reduce the scale of the model. Furthermore, the node effectiveness enabling function was introduced into the model to overcome the problem of remodeling in the traditional hierarchy directed graph model when the system structure was changed caused by the active/standby switching of device. In process of fault diagnosis reasoning, the search space of fault source was reduced by the combination of backtracking and forward reasoning method. The fault probability was calculated with the number of search hits of the nodes in order to enhance the efficiency. Fault diagnosis method in single fault and multiple faults scenarios of BeiDou radio determination satellite service ground station system was verified, and the results demonstrate that the EHDG method improves the accuracy and comprehensiveness in fault diagnosis.

Abstract:Because of the high correlation between the forward jamming and the radar transmitting waveform, it is easy to cause serious jamming to the radar after being received by the radar main lobe. In order to solve the main lobe interference problem, the forward interference signal model of OFDM-MIMO (orthogonal frequency division multiplexing-multiple input multiple output) radar main lobe was established, and the interference mechanism of OFDM-MIMO radar was analyzed. At the same time, based on the adaptive method processing theory, the analytical formula of OFDM-MIMO radar adaptive processing weight vector was derived, and an OFDM-MIMO radar forward main lobe interference adaptive suppression method based on range-dependent beam was proposed. Theoretical research and simulation results show that the proposed method can improve the output signal to interference plus noise ratio and effectively suppress the forward main lobe interference.

Abstract:Aiming at the multi-channel power allocation problem of the communication system with a fix transmitting rate, which is undergoing intelligent jamming attack, an asymmetric Colonel Blotto game model was proposed. Under the condition of complete information, the equivalent optimal power allocation strategies at a single channel were deduced for the communication system and the jammer. Moreover, the only mixed Nash equilibrium strategy of them was proved, and the Nash equilibrium benefit was acquired. Based on the equivalent optimal power allocation strategies at a single channel, a direct column element exchange algorithm by multiple scanning was proposed to construct the multi-channel mixed power distribution matrix, which could adapt to more channels and wider power distribution range than the linear programming method. Through numerical simulation, the effectiveness of the multi-channel mixed power allocation matrix construction algorithm and the optimality of the proposed power allocation strategies were verified.

Abstract:According to the time-frequency distribution difference between SSC(spectrum spread and compression) shift-frequency jamming and echo, a jamming suppression method based on generalized S transform and Tsallis cross entropy threshold segmentation was proposed. The jamming principle of SSC shift-frequency jamming was analyzed and the jamming signal model after dechirp processing was constructed . And the generalized S transform, which is of good time-frequency concentration performance, was utilized to acquire the time-frequency image of received signals after dechirp processing. Based on the related gray-scale image of time-frequency image, the minimum Tsallis cross entropy was utilized to obtain the optimal segmentation threshold and the time-frequency filter was constructed according to the segmentation threshold to suppress jamming. The simulation results show that the proposed method is of good jamming suppression performance to the false targets produced by SSC shift-frequency jamming and the jamming suppression ratio can reach more than 30 dB.

Abstract:Matrix transpose is one of the common matrix operations, which is widely employed in various fields such as signal processing, scientific computing, and deep learning. With the popularization of Phytium heterogeneous multi-core DSPs(digital signal processors) developed by National University of Defense Technology, there is a strong demand for high-performance matrix transpose implementations for Phytium multi-core DSPs. Based on the architecture of multi-core DSPs and the characteristic of matrix transpose operations, a parallel matrix transpose algorithm (called ftmMT) for matrices with different element bit widths (8 B, 4 B, and 2 B) was proposed. In ftmMT, the main optimizations include vectorization based on vector Load/Store functions, core-level parallelization based on matrix blocking, and overlapping between vectorization and memory access through implicit ping-pong methods. The experimental results show that ftmMT can significantly improve the performance of matrix transpose operations, and achieve a speedup of up to 8.99 times in comparison with the open-source transpose library HPTT running on CPU.

Abstract:In order to achieve the miniaturization and low power consumption of the low-frequency electromagnetic transmitter, a system design and performance verification method based on multi-physics coupling simulation was proposed for the theoretical research and technical practice of RMBMA(rotating-magnet based mechanical antennas). A RMBMA system scheme including a permanent magnet source, a high-efficiency drive motor and its controller was designed. By developing an experimental prototype, the system performance and near magnetic field characteristics were tested experimentally, and the feasibility and effectiveness of the proposed design method and system scheme were verified. Results show that the main indexes of the prototype, such as magnetic field intensity and size, meet the requirements of the penetrating radio frequency technical area in the AMEBA(a mechanically based antenna) project of the United States, which provides a feasible idea for the scheme research and prototype design of RMBMA.

Abstract:Aiming at the high difficulty and high false alarm rate of infrared dim small moving target detection in low altitude and complex scenes, a precise detection method based on full convolution network and a low-latency parallel processing method based on FPGA (field programmable gate array) were proposed to meet the real-time processing application requirements of high frame rate images in the detection system. A lightweight full convolutional network was used to detect dim small targets in spatial domain, and time domain trajectory correlation on suspected targets in adjacent frames was performed to further reduce false alarms. The experimental results show that the detection rate and false alarm rate are significantly improved compared with the five traditional methods. And the real-time processing of 100 Hz image is completed on a single FPGA with the processing delay less than 1.8 ms. This method realizes high-precision, high-robust and fast real-time detection of dim small targets in complex low altitude scene.

Abstract:The matrix multiplication-based convolutional algorithm, which can efficiently implement convolutions with different parameters, is the first choice of convolution performance optimization for a given chip. Based on the architecture of Phytium heterogeneous multi-core DSPs(digital signal processors) developed by National University of Defense Technology and the characteristic of the matrix multiplication-based convolutional algorithm, a parallel implementation of the matrix multiplication-based convolutional algorithm (called ftmEConv) for different convolutions on multi-core DSPs was proposed. The ftmEConv consists of four parallelized parts(input feature maps transformation, filter transformation, matrix multiplication, and output feature maps transformation), all of which were optimized for multi-core DSPs, and the performance of each part was improved by effectively exploiting the potential of all functional units in DSP cores. The experimental results demonstrate that ftmEConv achieves computational efficiency of up to 42.90%. Compared with other implementations of the matrix multiplication-based convolutional algorithm on heterogeneous chips, ftmEConv gets a speedup of up to 7.79 times.

Abstract:In order to reveal the aging mechanism of solid propellant grain during the long-term storage, a HTPB (hydroxyl-terminated polybutadiene) propellant rocket motor which was moulded by casting and stored naturally for 19 years was dissected. Samples were taken along the radial position of the propellant grains, and were tested and analyzed by means of SEM(scanning electron microscope), EDX(energy dispersive X-ray spectrometer), infrared absorption spectroscopy and cross-linking density. Results show that the aging degree of propellant is not the same at different positions of grain. The closer the position is to the non-metallic case, the more serious the aging degree of propellant is. Further analysis shows that, in addition to the effects of CC double bond oxidation chain breaking and condensation of —OH on propellant aging, Al³⁺ ions produced by the oxidation of Al powder catalyzes the addition reaction of HTPB′s CC double bond with Cl₂ or water to form CCl double bond and secondary alcohol, and the uneven migration and distribution of aluminum ions will cause the uneven aging of grain. The research results are of great significance to the improvement of propellant aging performance and the research of anti-aging measures.

Abstract:To improve the survival ability of cold launch missile in silo, the research on blast resistant and shock absorption design had been performed. Finite element models were established for missile-canister system in different shock absorption ways. The dynamic responses under blast-induced ground shock were calculated. The vibration reducing performances of several shock absorption systems were compared, including vertical suspension mode, lower supporting mode and slant suspension mode. Meanwhile, the influences of stiffness and damping on the shock absorbing performance were investigated. The results show that the vibration reducing performance of vertical suspension mode is slightly better than that of lower supporting mode, and that of slant suspension mode is worst. The vibration reducing performance of vertical suspension shock absorption system can be significantly improved by properly decreasing its stiffness and damping.

Abstract:The thermodynamic and calorific of the gas in wind tunnel will be imperfect when the wind tunnel operates under cryogenic temperature. In this case, the real gas effects should be considered for calculating flow parameters. The change laws of the compressibility factor and specific heart ratio for nitrogen gas under the ranges of pressures(100～450 kPa)and temperatures(100～323 K) were presented. By introducing the isentropic expansion coefficient into one-dimensional perfect gas flow equations, the computational analysis model for cryogenic wind tunnel flow parameters was developed, and the operating envelope of transonic wind tunnel considering the gas liquefaction temperature and pressure under high speed was obtained. The analysis results show that for the ranges of pressures (115～450 kPa) and temperatures (110～323 K) that cover the range of operation of transonic cryogenic wind tunnel, the maximum deviation of the various isentropic parameters from the ideal values is about 1% or less, and this deviation would be insignificant for the aerodynamic designing of the cryogenic transonic wind tunnel.

Abstract:Turbine disk is one of the main components of aero-engine. Once a critical failure occurs, it will lead to serious consequences. On the basis of fully considering the uncertain factors affecting high-low cycle complex fatigue life of turbine disks, co-simulation platform of reliability optimization with high-low cycle compound fatigue life for turbine disk was designed by MATLAB. The uncertain factors affecting high-low cycle complex fatigue life of turbine disks are full considered in the platform. Based on the common requirements of lifetime function and lifetime reliability analysis limit state function, a strategy of sharing training sample points in adaptive construction of lifetime function Kriging model and lifetime reliability limit state surface Kriging model in the process of optimization iteration was proposed. Meanwhile, a learning function for constructing Kriging model of lifetime function was proposed. The high-low cycle complex fatigue life reliability optimization of turbine disk center and mortise were completed using the co-simulation platform. The results show that the local maximum stress of the optimal result is significantly reduced, the average life-cycle is increased, and the reliability constraints are satisfied.

Abstract:In view of the propagation characteristics of Q-band communication links that were severely affected by conditions such as rainfall, fog and atmospheric absorption, based on the analysis of the propagation loss of the Q-band geostationary satellite orbit downlink, the influence of rainfall attenuation on the I/N distribution under the interference scenario of the earth station co-location was studied. A method to calculate the propagation loss of Q-band downlink combined with rainfall distribution was proposed, and a functional model for I/N combined p% that the percentage of time when rainfall attenuation exceeds a threshold value was established. The influence of p% on the distribution of I/N was analyzed, and the average annual interference distribution was forecast. This method was used to analyze the influence of the CHINASAT-G-115.5E satellite network which was registered by the ITU, and the results show that the lower the p%, the more obvious the influence of rain attenuation on the communication link. When p＜0.5, the attenuation threshold changes faster. When the location of earth station and the orbit longitude of the interfering satellite are fixed, the smaller the p%, the smaller the minimum value of I/N in the annual average 1-p% time. This method can be applied to study the interference distribution of other orbital satellites, and provide a reference for interference avoidance.

Abstract:The electric field distribution of submarine is affected by the degradation of coating properties during service. With material polarization as boundary condition, three electrostatic field models were established by using boundary element method,including submarine coating performance deterioration, coating penetration corrosion and coating local damage polarization. The propeller disturbance electric field model was analyzed from time domain and frequency domain characteristics. The results show that electrochemical polarization state has a significant effect on the potential and harmonic of propeller disturbance electric field. Under the three boundary conditions of constant potential, linear polarization and nonlinear polarization, when the propeller rotates at the same frequency to disturb the electrostatic field, the proportion of frequency components of each model is significantly different. Under the nonlinear boundary condition, the difference between the disturbance frequency and the ratio of twice disturbance frequency is the largest. Under the linear boundary condition, the difference between the disturbance frequency and the ratio of twice disturbance frequency is the smallest. Under three kinds of boundary conditions, the harmonic range of the electric field model above three times of the disturbance frequency will change. The harmonic range of the electric field model is the smallest under the constant potential boundary condition, and the harmonic range is the largest under the nonlinear boundary condition.

Abstract:For Cerenkov-type devices, the thermal desorption caused by the temperature rise of the high-frequency structure can cause radio frequency breakdown, which in turn leads to a decrease in output power and a shortened pulse. Based on this, the heating temperature rise effect of electromagnetic pulses on the high-frequency structure was studied, the temperature rise formula of the high-frequency structure was deduced, the scope of application of the theoretical formula was pointed out, and a numerical method to solve the temperature rise of the specimen was given. Taking the two-period 1 MV·cm-1 high-frequency structure as an example, the numerical calculation and simulation study of temperature rise were carried out. Results show that the temperature rise effect of a single pulse with a pulse width of 100 ns on the stainless steel specimen is significantly higher than that of other materials. When the system is working at a high repetition frequency, the temperature rise of electromagnetic pulses may cause the metal material to reach the threshold of gas desorption and cause the gas to desorb to form a local high pressure. The research in the thesis can provide reference for the research of gas desorption and radio frequency breakdown in high power microwave sources.

Abstract:The sabot discard process of shell piercing projectile has significance on the flight stability and strike efficiency of it. Aiming at the six degrees of freedom movement process of shell piercing projectile after leaving the chamber, a measurement method of sabot discard angle based on the principle of binocular vision was proposed. By setting some markers on the surface of projectile and sabot, the recognition and tracking of the markers were realized with the image processing and tracking algorithms. The spatial locations of these marker were calculated by combining with the binocular calibrated space parameters, and the discard angle of sabot relative to the projectile was obtained. Meanwhile, the accuracy of the method above was validated by experiment which was proved to be 2%. Taking the shell piercing projectile launching at initial velocities of 1 550 m/s and 1 750 m/s conducted in the laboratory as an example, the discard trajectories under different muzzle velocity were measured and analyzed. The results show that the faster the discard initial velocity is, the closer the discard trajectory is to the projectile.

Abstract:Based on the concept of the intelligent combat of UAV (unmanned aerial vehicle) swarms, the UAV swarms intelligent combat simulation environment was established. Aiming at the problem that it is difficult to accurately control the speed and attack angle of UAVs in the confrontation process through reward signals in traditional reinforcement learning algorithms, the RIC-MADDPG (rule and intelligence coupling constrained multi-agent deep deterministic policy gradient) algorithm was proposed. The algorithm uses rules to constrain the actions of UAVs in reinforcement learning. The simulation results show that the wining-rate of red UAV swarm, trained by the method based on the RIC-MADDPG, can be improved from 53% to 79%. This proves that the strategy of "agent training—problem finding—rule making—agent training again—problem finding again—rule making again" is effective for the optimization of agent combat strategy. The research results can be a reference for establishing the training system of the intelligent combat strategy of UAV swarms and conducting the research of swarm tactics coupling rule and intelligence.

Abstract:In order to improve the machining accuracy of the metal shaft parts, the vibrating abrasive belt polishing method was innovatively applied to the deterministic figuring of high-precision shaft parts based on the principle of optical deterministic figuring. In this method, the abrasive belt covered the outer surface of the contact wheel, and the elastic contact wheel contacted the shaft workpiece under a certain pressure to form a rectangular grinding area. The removal of materials could be achieved by the axial vibration of the contact wheel. The contour of the shaft′s surface can be measured by cylindricity meter, and the error′s distribution could be obtained. The dwell time of the contact wheel at different positions on cylindrical surface was calculated by pulse iteration method. The material removal quantity at different positions of workpiece was different by servo control of the machine tool′s spindle, thus the cylindricity error could be corrected deterministically. After the simulated machining, the deterministic figuring experiment was carried out on the cylindrical surface of a 45# steel shaft. Experimental results show that the average roundness error of the workpiece converges from 0.42 μm to 0.11 μm, and the cylindricity error of the workpiece improves from 0.76 μm to 0.35 μm. The shape accuracy after machining is better than that of the ultra-precision cylindrical grinder, which verifies the feasibility of high-precision deterministic figuring on the cylindrical surface of shaft parts.

Abstract:Aiming at the problem that the wavelet packet adaptive control subband contains a large number of convolution and correlation operations, which leads to the slow convergence speed of the algorithm, the WPHB-FxLMS (wavelet packet based on Hartley block filtered-x least mean square ) algorithm was proposed. Fast Hartley transform was introduced into block algorithm to realize fast convolution and correlation operation in frequency domain. Hartley block algorithm was used to generate control signals in the subband, and WPHB-FxLMS algorithm was proposed through the overlapping reservation method. The vibration isolation performance and control effect under fixed frequency and fluctuant frequency working conditions were studied by simulation and experiment. The results show that the WPHB-FxLMS algorithm can not only greatly reduce the convergence time, but also significantly improve the control precision, which has good robustness and stability, and can be applied in engineering practice well.

Abstract:In order to further study the electromagnetic force and energy consumption characteristics of the solenoid valve of the electronic-controlled booster pump, a 3D static magnetic field simulation model of the high-speed solenoid valve of the electronic-controlled booster pump was established by using the finite element method, and the accuracy of the model was verified by the experimental data. Through numerical simulation analysis, the characteristics of the electromagnetic force and power loss of the solenoid valve of the electric-controlled booster pump were studied by driving current and structural parameters (the number of coil turns, magnetic pole′s radius, working air gap), and the weight influence of each parameter on the electromagnetic force and power loss of the solenoid valve was obtained and the quantitative analysis was made. The results show that the influence of the radius of major and vice magnetic poles on electromagnetic force accounts for 38.15%, followed by the driving current which accounts for 31.08%, the number of coil turns which accounts for 17.06%, and the working air gap which accounts for 13.71%. From the perspective of power loss, the ratio of driving current, coil turns, radius of major and vice poles and working air gap of electric-controlled booster pump solenoid valve were analyzed. The ratio of coil turns(54.85%) is the highest, followed by the driving current(44.99%), and the minimum ratio of the radius of major and vice magnetic poles and working air gap is only 0.16%.

Abstract:To predict the unsteady thrust of the interaction between pump jet rotor and turbulence, on the basis of the Sears function, and considering the influence of the turbulence, the prediction formula of propeller broadband unsteady excitation thrust was derived, which was verified by comparison with experimental values. Furthermore, considering the situation that the rotor of the pump-jet propeller works in the wake field of the leading vane, the turbulence wavenumber spectrum of the blade wake was established through the Gauss wake model, and the parameters of the blade wake field were calculated by empirical formulas. The prediction formula of the unsteady broadband force of the pump jet propeller rotor was derived. The broadband unsteady excitation thrust calculated by the formula was similar to the numerical calculation result. The influence of changes in flow field parameters on the prediction results of the propeller rotor broadband radiation noise was analyzed. Research shows that the turbulence intensity only affects the amplitude of the broadband unsteady excitation thrust prediction results, and the turbulence integral scale has a greater impact on the amplitude and spectrum shape.

Abstract:As a common energy storage component in fuse, spring is an important part which affects the reliability of fuse. The failure of fuse will cause serious hidden danger. In order to accurately predict the change of reliability of the spring for fuse during storage, a multivariate normal reliability evaluation method based on modified Arrhenius model was proposed. Taking the stress loss rate and permanent deformation rate as performance parameters, the constant stress accelerated test was designed to obtain the degradation data of spring at 120 ℃, 130 ℃, 150 ℃ and 160 ℃. Anderson darling statistics were used to verify that the optimal distribution of degradation was normal distribution. According to the modified Arrhenius model, the degradation trajectory curve of spring at storage temperature was obtained. Based on multivariate normal distribution, the storage reliability model of spring was established, and the change characteristics of spring reliability during storage were analyzed. Compared with other reliability models, the prediction results of this model are more accurate, which provides a theoretical basis for the condition-based maintenance of fuse.

Abstract:The amphibious bionic robot is an unmanned system which can work both underwater and on land, which has been widely used in the fields of disaster rescue, environmental detection and resource exploration. An amphibious robot compounded with wheel and fin with the ability of self-adaptive climbing was proposed in this paper. The kinematic and dynamic mechanics of the self-adaptive climbing process was analyzed. The torque required of the critical obstacle crossing point was set as the objective function, and the optimized design structural and operational parameters were obtained by applying the genetic algorithm. Meanwhile, the climbing ability of the amphibious robot in this work was compared with others. The results illustrate that the required torque of the amphibious robot was reduced by 718.4 N·mm. The robot compounded with wheel and fin can climb the vertical obstacle of a larger height. The self-adaptive climbing process of the optimized robot was simulated. The simulation results illustrate that the variation of the propulsive velocity, the displacement and the torque in the processes of moving forward and climbing the obstacles. The experiments of the obstacle climbing was investigated for verifying the structural and operational parameters design.