For cargo airlines, in the face of a large number of complicated air cargo demands, formulating reasonable route planning and distribution schemes can reduce transportation costs and improve transportation effciency. We established a mixed integer programming model that integrates transport capacity allocation and cargo flow distribution and took Air China Cargo as an example to solve the model, in which airlines can use their self-owned cargo aircraft or outsourcing service for transportation. We proposed two data processing ways: pre-allocating cargo flow according to different transportation distances and pre-specifying transit airports according to OD pairs, which not only reduces the search space, but also saves 85.12% of the calculation time compared to the situation without data processing. The total length of freight transportation path and the turnover of transit freight flow are also reduced by 17.19% and 28.99% compared with non-preprocessing, which reduces a large number of detours. Compared with the model without outsourcing, the integrated model saves 4.87% of the cost. Then the outsourcing price coeffcient was introduced for sensitivity analysis. As the outsourcing price coeffcient decreases from 1.2 to 0.8, the outsourcing cargo volume increases. At the same time, the flight segments operated by small aircraft decrease, and the flight segments allocated to large aircraft do not change. Finally, we analyzed the curves of unit transportation cost varying with transportation volume on the same flight segment for different modes of transportation, which provides a reference for airlines to respond to the temporary segment demand, and provides a better solution according to curves and residual transport capacity without changing the overall plan.

In order to effectively solve the problem of low SNR adaptability in the blind identification process of STBC-OFDM signals, an identification algorithm for constructing feature sequences in time domain is proposed under the premise of known OFDM block size. This algorithm deduces the time-domain features and the fourth-order eigenvectors of the STBC received signals and constructs the feature sequences. By detecting the feature sequences, the four STBC-OFDM signals are identified. The derivation and simulation results show that this algorithm does not need priori information such as channel, noise, modulation mode and starting position of OFDM block, and has good recognition performance under low SNR conditions. It has good robustness to frequency offset, Doppler frequency shift and time delay, requires low calculation amount, and thus has high practical value.

A multi-target automatic initiation and tracking algorithm suitable for avian radar data is proposed to realize the statistical analysis of the number of bird targets in the hot spots of bird activities around the airport. The algorithm estimates the probability of association between the measurement and all possible events by data association, including the birth, continuation and extinction of the target, as well as the elimination of clutter. The smooth trajectory of each target is estimated by the Kalman filter and smoothing method, and the whole life cycle management of the target is realized. The simulation results show that the algorithm can realize multi-target automatic initiation and tracking in clutter environment, estimate the start and end time of each target correctly, and count the change of the number of targets, and it is obviously better than the traditional logic method in the timeliness of target initiation. Finally, the algorithm is applied to the airport avian radar data to estimate the number of birds around the airport, leading the airport to carry out targeted bird strike prevention measures.

As an important fundamental application of intelligent traffic system, the safe and stable operation of Vehicular Ad Hoc Network (VANET) is indispensable for the transportation system and even the sustainable development of social economy. The largest connected degree, average size of connected components and global network efficiency were used as the vulnerability evaluation metrics for VANET. Based on complex network theory, the VANET simulation model was established through vehicle simulation software (VanetMobiSim). Then, the relationship between the quantitative index of vulnerability and the node removal ratio under random attacks and intentional attacks was analyzed in detail. Furthermore, the influence of the density of nodes, signal radiation radius and patterns of attack on VANET vulnerability was analyzed by simulation experiment. The experimental results show that the VANET has a high vulnerability under intentional attacks; the intentional attacks based on node betweenness destroyed the networks most strongly. The smaller node density and signal radiation radius are, the worse VANET connected degree is and the more vulnerable the network is. The research methodology and results provide the theoretical basis for VANET topology control optimization and network management decision-making.

To analyze the influence of earth-atmosphere radiation on imaging characteristics of space object, first, the visible light imager mounted on Geosynchronous Orbit (GEO) satellite was treated as an observation platform, and the motion scenes of Highly Elliptical Orbit (HEO) and Low Earth Orbit (LEO) objects were designed by the Satellite Tool Kit (STK). Then, the equivalent magnitude model of space object and earth-atmosphere background, and the calculation formulation for the Signal-to-Noise Ratio (SNR) of space object were derived by adopting infinitesimal method, according to the relative relationship among the space object, the sun, the earth and the observation platform. The effects of the distance and angle variables on the equivalent magnitude of different orbital objects and earth-atmosphere, as well as the SNR were analyzed. The simulation results indicate that the equivalent magnitude of earth-atmosphere background is lower than space object while the object is far away from the observation platform, which means that the earth-atmosphere radiation is stronger than the target signal. The observation window between the time when the earth-atmosphere radiation enters and leaves the target detecting field of view is the best time for target detection since the SNR is the largest. Moreover, the value of SNR obtained by simulation provides a theoretical calculation basis for the detection and recognition of space object.

In soil moisture observation using Global Navigation Satellite System Reflectometry (GNSS-R) technique, actual antennas' directionalities will bring about the biases in the correlation power measurements of the direct and reflected GNSS signals. For eliminating the cosine-like oscillatory bias of the correlation power caused by antenna directionality in ground-based scenario, this paper proposes a correlation power correction method based on the polynomial fitting. In order to verify the validity of this method, the ground-based GNSS-R soil moisture observation experiment is carried out. The experimental results show that the correlation power correction using polynomial fitting can remove the cosine-like oscillation of the correlation power waveform, and then improve the efficiency of measurement data and the accuracy of retrieval results.

In order to provide input parameters for the design of new types of oxygen-consuming inerting system components, based on the proposed low-temperature controllable oxygen-consuming catalytic inerting system flow, the system mathematical model is established based on the mass conservation and energy conservation equations with the suction flow rate at the outlet of fuel tank as the benchmark. Taking the central fuel tank as the object, the important performance changes of the inerting system under the full flight envelope and the influence of key parameters on it are simulated. The results show that: inerting system can effectively reduce the oxygen volume fraction. For example, under the condition of initial full load, 0.5 catalytic efficiency and 60 L/min suction flow, the oxygen volume fraction will reach below 12% after 24 minutes. During the flight, the volume fraction of gas phase oxygen in the fuel tank rises during the declining and approaching phases, while it is decreasing in other phases. The higher the catalytic efficiency is and the larger the fan flow is, the shorter the inerting time is required. When the catalytic efficiency is fixed, the same inerting time is achieved, and the maximum fan suction flow is required when there is no fuel load. Therefore, the oxygen-consuming catalytic inert system should be designed according to the most unfavorable no fuel load working conditions.

Due to the pathological property and ill-posedness of sensitivity matrix in Electromagnetic Tomography (EMT), the quality of the reconstructed image is relatively low. To improve the imaging quality and imaging speed, this paper proposes an optimized Landweber iterative fast iteration image reconstruction algorithm. Firstly, dimension reduction algorithm is used to decrease the sensitivity matrix dimension to eliminate the redundant information of sensitivity matrix and reduce the calculation load of each iteration. Secondly, Seeker Optimization Algorithm (SOA) is used to optimize the dimension-reduced sensitivity matrix. This optimization operation can reduce the condition number and improve the morbidity degree of sensitivity matrix. Finally, Landweber iteration algorithm and preprocessed sensitivity matrix are used to reconstruct image. Simulation experimental results show that, under the same experimental conditions, compared with Landweber iteration algorithm, the proposed algorithm increase the quality of reconstructed image and decrease the calculation load of image reconstruction.

Most of the existing bearing failure researches simplified the failure to regular shapes such as rectangular grooves or circular pits, which were quite different from the actual failure morphology, taking the aero-engine rotor system as the research object, starting from the actual fault morphology of the rolling bearing and the objective reality of the main shaft bearing being prone to failure in the complex rotor system, a method for characterizing the irregular bearing fault was proposed and introduced into the single rotor-bearing system dynamic model, and the irregular failure model of bearing inner and outer rings was established. Using the method of numerical calculation, the vibration response of the rotor system with faults was analyzed, and the influence of the circumferential width and depth of the faults on the system vibration was studied when the system bearings contained rectangular faults and irregular faults in the inner and outer rings. Finally, for the fault damage existing in the inner and outer rings of the rolling bearing, fault bearings with different positions and sizes were made and introduced into the rotor system to conduct experimental research, and the system vibration data at different rotation frequencies and fault sizes were collected. The comparison with the numerical simulation results fully verified the correctness of the irregular bearing failure dynamic model.

During high speed flight, the temperatures of the vehicle can reach extremely high values in the critical parts. To solve the problem of thermal protection for the critical parts, series of transpiration cooling tests using different materials as porous plate and water as coolant were carried out. The experimental platform which was used for the transient measurement in transpiration cooling process was developed. The cooling effects of different material porous plates under different heat flux were evaluated by measuring the inter and outer wall temperature.The results of the experiment indicate that transpiration cooling greatly reduces the temperature of the inner and outer walls of the porous plate, which plays an effective role in active thermal protection. For nickel and copper metal porous plates, the coolant flow rate is kept at about 3.5 g/s and the temperature of inner and outer wall is stable at about 30℃-50℃ when the heat flux is less than 120 kW/m². And for ceramic porous plates, the coolant water flow rate is kept at about 0.32 g/s, and the temperature of inner and outer wall is basically stable at about 30℃-40℃ when the heat flux is less than 220 kW/m². Moreover, for nickel, copper and ceramic porous plates, the temperature of the inner wall changes little under the condition of high heat flux of 315 kW/m² during transpiring cooling, and the outer wall temperature stabilizes at about 260℃, 110℃ and 130℃, respectively. The coolant on the outer wall surface is in a completely vaporized state, and the vaporized phase transition position of the coolant is inside the porous plate. In addition, the temperature of the inner and outer walls of the porous plate rises rapidly when there is no transpiration cooling, and its equilibrium temperature is greatly increased compared with the transpiration cooling situation, which further shows the enormous application potential of transpiration cooling.

Focusing on the problem that the performance of traditional adaptive beamformer declines sharply when the covariance matrix contains the target signal component and the target steering vector mismatch, a robust beamforming algorithm based on double-layer reconstruction of interference-plus-noise covariance matrix is proposed in this paper. Firstly, sparse reconstruction method is used to estimate interference-plus-noise covariance matrix. The interference-plus-noise covariance matrix is optimized by estimating the interference steering vector and interference power. Secondly, based on subspace theory, an optimization model of steering vector constraint error is established, and the convex optimization model is solved by iterative method to obtain the optimal weight vector. The simulation results show that the proposed algorithm improves the robustness of the beamformer in the case of target vector constraint error and array error. The algorithm performs well in low-speed snapshot and its output performance is better than the existing methods.

Aimed at improving the performance of an Electro-Hydrostatic Actuator (EHA), a novel Cascade Control (CC) algorithm based on Damping-Variable Sliding Mode Control (DV-SMC) and PID is proposed in this paper. A high-order model of EHA was divided into two low-order subsystems, i.e. mechanical subsystem and hydraulic subsystem. Furthermore, double-loop PID and DV-SMC were applied for the two subsystems, respectively. The proposed method can adjust the damping ratio of the subsystem adaptively. At the beginning of sliding, small damping ratio was used, while an overdamped subsystem was obtained in the end by adaptive adjustment. Therefore, the EHA rapidness and the suppression of overshoot can be guaranteed simultaneously. Finally, simulative validation was carried out to verify the effectiveness of the proposed method, and the optimal parameters of the sliding mode surface are discussed and given.

A fault-tolerant fixed-timepath following guidance control method for the UAVs subject to thedisturbancesand actuator faults is studied. Both backstepping and fixed-time convergence techniques are employed for developing the line-of-sight path following control strategies to guarantee the convergence of the UAV to its reference trajectory in fixed time with elegant transient performance. Command filters and auxiliary systems are introduced in the guidance control algorithms design to avoid the arduous calculation of derivatives of virtual control terms in backstepping. To address turning rates constraints of the UAV, the barrier Lyapunov functions are incorporated with the control scheme to prevent the drastic change of the guidance control system states. A nonlinear fixed-time observer is designed for estimating complex unknown external disturbances and eliminating the actuator faults and the influence of external environment disturbance on following performance. Simulation results show the effectiveness and robustness of the proposed line-of-sight path following guidance control algorithm, and it has good path following fault-tolerant control performance.

Aiming at the problem of poor performance at off-design points in the current common single-point optimization design method of large aircraft wings considering multiple cruise conditions, a synthetical aeroelastic optimization framework with multiple cruise conditions is proposed, and the multi-point aeroelastic optimization of a large aircraft composite wing is studied. The laminate thickness of skin, web, flange and other composite components of the jig shape is optimized to minimize the wing structure weight using genetic algorithm, subjected to the constraints of aeroelasticity, stress/strain, strength and other conditions, and the jig shape design is carried out according to the optimization results. The lift-to-drag characteristics of the optimization results are analyzed and verified by the high-precision CFD/CSD coupling method. The results show that the multi-point aeroelastic optimization can effectively reduce the structure weight and maintain the aerodynamic performance of the pre-designed cruise configuration, thus reducing the overall fuel consumption. Furthermore, the results of multi-point optimization and single-point optimization are compared and the relationship between the considered cruise conditions number and the optimization results is analyzed. The results show that the performance of the multi-point aeroelastic optimization is better than that of the single-point aeroelastic optimization, and the overall performance increases with the increase of the number of cruise conditions considered in the optimization.

Response surface method is a widely used agent model analysis method for structural reliability analysis because of its good applicability and maneuverability. Aimed at the difficulties of the response surface method, such as balancing the efficiency and accuracy, a structural reliability calculation method based on improved weighted response surface is proposed. In the iterative process, three weighting factors, including the distance between the sample point and the design point, the absolute value of limit state function and the value of joint probability density function, are considered to weight the sample points. The quadratic polynomial response surface function without cross term is updated by weighted regression and reusing all known sample points. After the iterative convergence, the sample points with the larger weight among the existing sample points are selected to fit the quadratic polynomial response surface function with cross terms. Finally, with numerical examples and engineering cases, the feasibility of the proposed method is verified by comparing with traditional sampling methods and other response surface methods. The results show that the proposed method has higher efficiency and meanwhile guarantees accuracy.

With the development of Global Navigation Satellite System(GNSS), the prospect of GNSS has been widely recognized in the world. In particular, the positioning solutions with fast and accuracy calculation are essential for the GNSS receiver design. The most of the current satellite selection algorithms in the GNSS receiver fix the number of satellites in advance, which limits the performance of the algorithm. This paper proposes an Imperialist Competitive Algorithm (ICA) for satellite selection. In order to obtain better geometric configuration of satellite constellation, the prior information (elevation and azimuth of visible satellite) is introduced for prior constraint. The Geometric Dilution of Precision (GDOP) and number of satellites are two objectives of the optimization algorithm. Comprehensive decisions are used to quickly select satellites, making the selection of satellites more flexible, as well as reducing the computational burden of multi-constellation satellite receivers. Experiment results based on simulation and field data showed that, after priori constraints are introduced, at elevation angle 5°, the average number of satellites selected by the algorithm proposed in this paper is 51.8% of the maximum visible satellites based on simulation data and the average number of satellites is 45.4% of the maximum visible satellites based on field data. The average GDOP is decreased by 0.209 2 and 0.248 4 compared to the satellite selection without a priori constraint. At the same time, the average calculation time for once satellite selection is about 0.168 4 s and 0.303 1 s, with an improvement of 95.79% and 92.42% compared to the time consumption (i.e. 4 s) of the traversal method.

Geomagnetic matching navigation plays an important role in the field of navigation guidance. The construction accuracy of geomagnetic reference map determines the effectiveness of geomagnetic matching navigation. Aimed at the problem that the existing geomagnetic reference map construction accuracy is difficult to meet the needs of practical applications, a high-precision geomagnetic reference map construction method based on sparse representation and dictionary learning is proposed. First, the sparse dictionary is initialized using Rectangular Harmonic Analysis (RHA). Then, K-SVD is used to train the sparse dictionaries. Finally, the feature that the low-resolution and high-resolution reference maps have the same sparse coefficients is used to reconstruct the high-resolution geomagnetic reference maps. Experimental results show that the proposed method has higher construction accuracy for geomagnetic reference maps, lower requirements for training datasets, and better robustness to noise. Compared with the PSO-Kriging interpolation method, with a magnification factor of 4, the Peak Signal to Noise Ratio (PSNR) value is increased from 26.31 dB to 26.73 dB; the Structural Similarity Index (SSIM) is increased from 0.498 to 0.524; the Root Mean Square Error (RMSE) is decreased from 14.96 nT to 13.78 nT.

Aiming at the problem of multi-scale and multi-scene Synthetic Aperture Radar (SAR) ship detection, an object detector without pre-training based on EfficientDet is proposed. The existing SAR image ship detectors based on convolutional neural networks do not show excellent performance that it should have. One of the important reasons is that they depend on the pre-training model of the classification tasks, and there is no effective method to solve the difference between the SAR image and the natural scene image. Another important reason is that the information of each layer of the convolutional network is not fully utilized, the feature fusion ability is not strong enough to deal with the detection of ships in multiple scenes including sea and offshore, and especially the interference of complex offshore background cannot be ruled out. SED improves the method in these two aspects, and conducts experiments on the public SAR ship detection data set. The detection accuracy index AP of SED reaches 94.2%, which, compared with the classic deep learning detector, has exceeded the best RetineNet model by 1.3%, and achieved a balance among model size, computing power consumption and detection speed. This verifies that the model can achieve excellent performance in multi-scale SAR image ship detection in multiple scenes.

A phosphoric acid-nickel chloride reaction system was designed by utilizing the comprehensive corrosion properties of phosphoric acid and the activation effect of chloride ions. By characterizing surface potential and microscopic morphology, the nickel immersion process was analyzed at different phosphoric acid concentration and temperature. The result shows that the concentration of phosphoric acid is a key factor affecting the surface potential and morphology of the nickel-immersed coating. When the concentration of phosphoric acid is 25% and the temperature is 30℃, nickel-immersed coating with stable chemical properties, good coating and uniform grain size can be prepared. In this reaction system, the nickel-immersed coating grows in a spherical manner after nucleation. After reacting for 600 s, a nickel-immersed coating with a thickness of about 1 μm is obtained, and its surface potential can reach about -0.51 V.

With the consumption of modern industry, the limited land resources have been unable to support the sustainable development of human society in the future, so more and more attention has been paid to the exploitation of marine resources. Marine riser is a key component in deep-sea mining system, which connects mining ship on the sea surface and the mining car on the seabed. In order to avoid the long and soft lifting mine pipeline in the deep-sea operation to touch the seabed or tie up and other phenomena, which will affect the operation efficiency, buoyancy modules are usually used to lift the riser. Considering several design parameters of the buoyancy lifting modules layout, including the total lifting buoyancy, horizontal tension, number of buoyancy segments and the distribution length of buoyancy segments, the static configuration of the marine riser is evaluated by finite element method, and the general influence law of these parameters on the structure configuration is explored, so as to propose a reasonable buoyancy distribution scheme.

For the problems of less onboard fuel pump fault data source, low diagnosis efficiency, high maintenance costs, and lack of effective fault characteristics, we use vibration signals and pressure signals collected from onboard fuel transfer system experimental platform, and put forward an onboard fuel pump fault diagnosis method based on Empirical Mode Decomposition (EMD) and Support Vector Machine (SVM). First, EMD is used to extract values of vibration signals energy as characteristic parameters at different frequency bands, and fault characteristic vectors are constructed by combining with the mean value of port pressure signals. Then, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Salp Swarm Algorithm (SSA) and Grid Search (GS) algorithm are used to optimize the penalty parameters c and Radial Basis Function (RBF) parameters g of SVM, and the optimized SVM diagnostic performance is evaluated. Finally, SVM, Extreme Learning Machine (ELM) and BP neural network are used as classifiers, and the diagnostic performance of the three classifiers is evaluated. The results show that the fault diagnosis rates of the SVM using the three-population intelligent optimization algorithm can reach 100%, none of them fall into the local optimal solution during the optimization process, and the optimization time is equal. Among them, the training time of GA is the shortest, so GA can be used to optimize the SVM parameters. When GA_SVM is used as the fault classifier, the time is shorter and the fault diagnosis rate is higher. Therefore, the GA_SVM classification model can be used to realize the efficient fault diagnosis of airborne fuel pump.

There is a large random error in the angular rate measurement signal of Magnetically Suspended Control & Sensing Gyroscope (MSCSG), which is not conducive to improving the sensitivity accuracy of MSCSG. In this paper, taking the principle prototype of MSCSG as the research object, the Allan variance analysis method is proposed to analyze the random error of the measured data of MSCSG. According to the principle of MSCSG angular velocity sensitivity, the measurement formula of MSCSG rotor deflection angular velocity is derived. Five typical random error coefficients are calculated by Allan variance analysis and least square fitting. The calculation results show that, among MSCSG random errors, zero deviation instability, rate following and rate slope are the main components, while quantization noise and angle random walk error account for the smaller proportion. According to this, the directivity of MSCSG error source is analyzed, and the method of restraining and compensating random error is given, which lays a theoretical foundation for improving the sensitivity accuracy of MSCSG.

In order to solve the problems in mission planning, such as the low deployment efficiency of relay Unmanned Aerial Vehicle (UAV) and the deployment scheme cannot meet the minimum number requirements, a fast relay UAV deployment strategy is proposed. First, according to the task requirements of the least relay nodes, a deployment model based on the least relay nodes is established. Then, the search mode of the depth-first search algorithm is optimized, and the fast search of feasible links between nodes is realized. Finally, the Rapid Depth-First Search (RDFS) algorithm is introduced into the Artificial Bee Colony (ABC) algorithm to solve the deployment scheme of the least relay nodes. The simulation results show that under the same task scale, the solution speed of this strategy is about 53.56% higher than that before improvement, and the number of deployed relay UAVs is reduced by about 11.88% compared with the existing methods.

Aimed at the problem that when the Sparrow Search Algorithm (SSA) is close to the global optimum, the population diversity decreases, and it is easy to fall into the local optimal solution. A Chaotic Sparrow Search Optimization Algorithm (CSSOA) is proposed. Firstly, the population was initialized by improving the Tent chaotic sequence, the quality of the initial solution was improved, and the global search ability of the algorithm was strengthened. Secondly, the method of Gaussian mutation was introduced to strengthen the local search ability and improve the search accuracy. At the same time, a Tent chaotic sequence was generated based on the search stagnation solution, and this chaotic sequence was used to chaotically disturb some individuals who were partially trapped in the local optimum, prompting the algorithm to jump out of the limit and continue the search. Finally, through simulations of 12 benchmark functions, the results show that the proposed algorithm can overcome the shortcomings of SSA being easily trapped in local optimum, and improve the search accuracy, convergence speed and stability of the algorithm. Meanwhile, CSSOA is applied to the simple image segmentation problem, which verifies the feasibility of applying CSSOA to practical engineering problems.