Using the carbon fiber-reinforced composite tank can remarkably reduce the weight of the launch vehicle.However，the analysis method for cryogenic tanks subjected to mechanical-thermal loads remains to be studied，especially to accurately consider the microthermal stress produced between fiber and matrix in the cryogenic environment.A representative volume model containing multiple fibers was adopted，combined with the matrix and fiber failure criteria，to establish a microscopic stress field and failure prediction model.The k-means clustering method was used for dimensionality reduction calculation，and an efficient and high-fidelity trans-scale analysis method for composite tanks was proposed.The results of illustrative examples show that the proposed method can accurately predict the elastic constants and failure strength of the composite single-layer plate according to the thermal and mechanical constants of fiber and matrix.The leakage failure process of a composite tank subjected to mechanical-thermal loads was simulated，and the critical load and failure state were given.

Based on the data of A-share listed enterprises in Chinese equipment manufacturing industry from 2011 to 2021，the mechanism of digital transformation and dynamic capability on enterprise innovation performance were empirically tested.The results show that digital transformation can promote the innovation performance of equipment manufacturing enterprises by enhancing dynamic capabilities，and organizational inertia has a different degree of moderating effect between digital transformation and dynamic capabilities.The analysis of enterprise heterogeneity shows that the impact of digital transformation on enterprise innovation performance is significantly different between state-owned enterprises and non-state-owned enterprises，as well as among the eastern，central and western regions.The research conclusions remedy the theoretical gap between dynamic capabilities and organizational inertia，and provide practical guidance for the internal mechanism of digital transformation to promote enterprise innovation performance.

To study the evacuation characteristics in the Airbus 330 cabin crew under different hatch opening conditions，the Pathfinder software was used to simulate the evacuation characteristics.The evacuation characteristics of personnel were explored in four situations： all cabin doors opened， emergency doors could not be opened， front cabin doors could not be opened， and rear cabin doors could not be opened.The results show that the simulation time for evacuation is 63.3 s when all hatches are open，while 118.3 s when the emergency hatch cannot be opened.The per-unit-time flow of people at No.3，No.4，No.7，and No.8 hatches decreases by 0.09 people，0.2 people 0.09 people，0.04 people respectively.It takes 64s to evacuate people when the front cabin door could not be opened.The number of people evacuated from No.3，No.4，No.5，and No.6 doors are 47，48，52，and 50 respectively，indicating that the failure to open the front cabin door does not have no significant impact on the evacuation time.When the rear cabin door could not be opened，the evacuation time is 112.3s.The time interval from the first person to the last person from the No.3，No.4，No.5 and No.6 doors is 67.7 s，67.6 s，109.9 s and 107.1 s，respectively.This indicates that more passengers in the economy class choose to escape from the hatches No.5 and No.6，which increases the evacuation time.The research results provide a reference for understanding the evacuation characteristics of personnel under different hatch opening conditions.

The unmanned aerial vehicle with folding wing can serve in multiple mission profiles，the flight environment is complex and susceptible to low-altitude gusts. A new parameterized aeroelastic model with non-intrusive characteristics was developed to quickly calculate discrete gust response cha-racteristics of Z-shaped folding wing at different folding angles. Firstly，structural and aerodynamic enhancement design was made to the existing Z-shaped folding wing. Secondly，based on the Nastran input file for the flutter calculation of the folded wing in the extended state，the parameterized aeroelastic model under any folding angle was reconstructed. Finally，the effects of parameters such as folding angle，flight speed，gust velocity amplitude and hinge damping between different wing segments on the gust response characteristics of the folding wing were examined. The results show that increasing folding angles can effectively alleviate the gust responses， the increase in flight speed and gust velocity amplitude significantly contributes to the responses of wing tip acceleration and wing root bending moment， the presence of hinge-damping parameters can alleviate the acceleration response of the wing tip，but it also increases the bending moment response of the wing root.

The strength degradation law of composite materials is very important for studying the structural fatigue life. However， some parameters involved in existing residual strength models are confirmed using residual strength test data， so the cost of model construction is high. By exploring the relationship between fatigue life and residual strength of composite materials， taking the cumulative distribution function of fatigue life as the starting point， a damage degree with clear physical meaning was constructed，and then an improved residual strength model was proposed. The improved residual strength model didn’t need residual strength test data but only initial static strength and fatigue life data. A fatigue life prediction model called S-N-φ model was constructed based on the improved residual strength model， which can consider the influence of initial static strength on fatigue life. The results show that the prediction accuracy of the improved residual strength model is satisfactory， and the S-N-φ model has better prediction accuracy than the classical S-N curve model.

Human action recognition is a key technology for understanding pedestrian intentions from video captured by unmanned aerial vehicles （UAV）.However，UAV platforms have limited computing power，and existing action recognition methods are inefficient.A lightweight spatial grouping attention graph convolutional network （SGA-GCN） was proposed to reduce network depth to improve the efficiency and ensure the accuracy of action recognition.In order to capture body parts that represent global motion，spatial grouping attention was introduced to enhance local features with high similarity to global features.Moreover，since it was impossible to effectively distinguish actions with similar motion trajectories solely based on joint and skeletal features，a high-order feature encoding of skeletal angles was constructed to capture changes in angles between limb joints that better reflected subtle motion differences and improved feature representation capabilities.Finally，to address the low frame rate issue in UAV aerial video，a linear interpolation scheme based on inter-frame differences was proposed to increase sample information quantity.Experimental results demonstrate that compared to the existing state-of-the-art （SOTA） methods，the proposed approach achieves better performance in terms of recognition rate，parameter quantity，training time and execution time on the UAV-Human dataset.

Based on the NNW-PHengLEI software，for surfaces with uniformly distributed rough elements，rough surface boundary condition corrections and transition momentum thickness Reynolds numbers were introduced into the k-ω SST turbulence model and the γ-Re_θt transition model respectively.At the same time，some rough surface settings were added to achieve local/global rough surface boundary layer flow simulation capability.Test results show that when an airfoil is in a fully turbulent state，rough surfaces can reduce the stall angle of attack by 4°，and an equivalent roughness height of 1×10^-3 of chord length can cause a decrease in maximum lift coefficient by approximately 36%.Additionally，rough surfaces can advance the boundary layer transition，leading to an increase in skin friction drag.The computational analysis proves that the rough surface boundary condition proposed by Hellsten et al.can reflect the roughness effect more accurately than the dissipation rate boundary condition.The computational aerodynamic forces are consistent with the experimental data，and the flow field simulation results conform to the flow characteristics of rough surfaces.In addition，Hellsten’s model makes it possible to ensure stable coupling simulation with the transition model and its rough surface correlation function.The simulation results for the transition onset location and its trend with roughness level and freestream turbulent intensity are consistent with the experimental data.Still，the transition length needs further verification and correction through more rigorous experiments.

The changes in stiffness and damping of the bolted joint surface can cause the changes of dynamic characteristics of the whole bolted joint structure，so it is of great practical value to accurately obtain the dynamic parameters of the joint surface in engineering.Based on the bolted structure，an improved equivalent model of the stiffness of the joint surface with uneven distribution in the range of bolt preload was proposed，and the distribution of different stiffness matrix elements in the finite element modeling was analyzed.The stiffness parameter identification of bolted joint surface was carried out by combining experiment and finite element analysis method.The results show that the increase of the number of stiffness matrix elements can improve the accuracy of natural frequency solutions to some extent.At the same time，considering the uneven stiffness distribution of the joint surface in the range of bolt preload，the equivalent modeling accuracy of the bolted joint structure can be effectively improved.

To address the drawback of traditional painting trajectory planning algorithm requiring extensive searching for nearest point pairs，a initial trajectory point extraction algorithm based on the point cloud thin-slicing technology was proposed.This algorithm determined slice thickness by calcula-ting the density of the point cloud model and designated points on the slice as initial trajectory points， thus improving operational efficiency by avoiding point pair searches. To tackle issues such as insufficient trajectory precision and excessive curvature caused by traditional linear interpolation algorithm， a trajectory fitting algorithm was introduced. This algorithm firstly maped trajectory points in three-dimensional space to a two-dimensional coordinate system and utilized the isolation forest algorithm to eliminate outlier trajectory points，followed by combining curve fitting methods for trajectory planning. Experimental results demonstrate that the trajectories planned by this algorithm exhibit higher precision and smoothness， while also enhancing the overall efficiency of the algorithm.

In order to carry out fault diagnosis with only health status data，an optimized swin transformer deep neural network architecture was constructed to extract and reconstruct health data features，which proposed an unsupervised learning method for rolling bearing fault diagnosis. Compared with autoencoder，depth encoder，convolutional autoencoder，and sparse autoencoder，the accuracy is 98.62%，76.46%，68.69%，77.69%，68.00%，respectively，which is more than 20% higher than the accuracy of comparison network.

As a navigation solution with higher accuracy and better robustness compared to single navigation systems，GNSS/INS integrated navigation has been widely applied in various carriers.Aiming at the problem of the satellite navigation signal interruption caused by environmental obstruction or electromagnetic interference to reduce the accuracy of integrated positioning，GNSS/INS integrated navigation positioning method by a fully connected neural network （FCNN）was proposed.This method consisted of training and prediction modules.Under normal GNSS signal conditions，the me-thod utilized the position and velocity information calculated by INS and the position and velocity information output by the integrated navigation to train the FCNN model.When the GNSS signals were interrupted or fail，the pre-trained FCNN model was used to predict the navigation solutions.Experimental data was employed to validate the proposed method.The results indicate that the GNSS/INS based on FCNN integrated navigation method proposed in this study effectively suppresses the divergence of single INS positioning errors，thereby improving the accuracy and availability of GNSS/INS integra-ted positioning results when the GNSS signals are interrupted.

Based on the transformer architecture，a knowledge tracing prediction model for learning trajectory was proposed， which solved the following problems in the field of knowledge tracing using the transformer architecture： the model lacked the learning of knowledge point information； the attention scores in the self-attention mechanism showed a long-tail distribution and required square computatio-nal overhead； the prediction strategy of the model lacked consideration of learners’ability. In the data preprocessing stage， LTKT used the knowledge integration mechanism in the field of education to integrate multiple knowledge points involved in the subject， and the integrated knowledge formed was used as input to the model along with other learning trajectory information； LTKT introduced a sparse self-attention mechanism according to the characteristics of the long-tail distribution of attention scores into the encoder and decoder structure， and embedded a position encoding containing absolute distance and relative distance in it， so that the deep attention mechanism could also learn the position relationship between topics. In the prediction strategy， LTKT used the bilinear layer to fuse the learning ability features extracted by the learning ability extraction module and the output of the decoder to comprehensively predict the student's answer performance at the next moment. Experiments were carried out on two real large public datasets， and compared with other excellent models. The results show that LTKT has significantly improved the AUC.

In the context of the dual-carbon goal，the digital transformation of manufacturing enterprises is of great significance in promoting green transformation and achieving a balance between economic and environmental benefits.A-share listed companies in the manufacturing industry from 2011 to 2021 were taken as the research sample and empirically examined the impact of digital transformation on ESG responsibility performance of manufacturing enterprises.The results show that digital transformation positively promotes ESG responsibility performance of manufacturing enterprises and the conclusions are still valid after robustness test and endogeneity test.The mechanism test shows that digital transformation can promote the ESG responsibility performance of manufacturing enterprises by alleviating their financing constraints.The moderating effect shows that digital inclusive finance can increase the influence of digital transformation on the improvement of ESG responsibility performance of manufacturing enterprises.The heterogeneity analysis show that digital transformation has a more significant effect on ESG performance in manufacturing enterprises with high investments in innovation resources and in low-carbon pilot cities.The study conclusions provide countermeasures and suggestions for enhancing ESG responsibility performance from the government and enterprise levels.

The damage mechanism of fiber reinforced resin matrix composites is complex.To ensure long-term stable application，advanced health monitoring technology must be used to monitor the damage.A sensor based on carbon nanopaper can sensitively monitor resistance changes and impact damage on carbon fiber reinforced polymer （CFRP）composite.A damage monitoring system based on an artificial neural network （ANN） deep-learning algorithm was designed.Through data analysis，the system could effectively monitor the occurrence and location of CFRP damage for a long time，and the damage location accuracy was as high as 92%.It can be inferred that the damage monitoring system can evaluate the health status of composite materials.

A humidity influence of air thermophysical properties on the turbofan engine performance model was considered to study the influence of atmospheric humidity on turbofan engine performance. The influences of atmospheric temperature and humidity on engine thrust， exhaust temperature and fuel flow were calculated and analyzed， and the numerical humidity correction curves were compared with the experimental curves. The results show that the engine thrust， exhaust temperature and fuel flow gradually decrease as the relative humidity increases at the same atmospheric temperature and the influence of relative humidity is more significant when the atmospheric temperature is higher.The humidity correction coefficients of engine thrust，exhaust temperature and fuel flow linearly correlate with the humidity ratio，consistent with experimental results. But when the atmospheric temperature is below 0 ℃， the influence of humidity on engine performance may not be considered in practical applications.

Addressing the issue that the current energy method cannot accurately describe the time-varying meshing stiffness of the gears with tooth flank spalling，resulting in the imperfect dynamic modelling method for the fault of gear transmission system，an improved energy method was proposed.Considering various types of tooth frank spalling，a time-varying meshing stiffness formula that consi-dered the tooth frank spalling was established.The established time varying meshing stiffness was introduced into a six-degree-of-freedom transmission system，creating a dynamic model of the gear transmission system that could simulate the tooth frank spalling.The Newmark-β method was used for solving，the numerical simulation results were compared and analyzed with the experimental data to verify the accuracy of the established model.Based on this model，the influence of the main parameters of tooth frank spalling on the time-varying stiffness，the corresponding stiffness variation law and the frequency characteristics of the dynamic response were investigated.The results show that the secondary frequency peak with a spacing of 10 Hz occurs on both sides of the characteristic frequency of the gear transmission system with tooth frank spalling.

According to the CCAR-33R2 standard 33.78（a）， a test method of the engine hailstone ingestion was studied and verified to investigate the hailstone ingestion capability of one turbofan engine. The test process was arranged； the test conditions were determined； the engine and test bench were analyzed； the protection， photography， and projection devices were designed； and the producing method of hailstone was studied. The hailstone was projected to the engine by the projection device. At the same time， a target calibration test and an air gun test were designed before the actual test. The target calibration test ensured the speed and accuracy of the projection， while the air gun test tested the influence of the projection device on the engine intake. Furthermore， a high-speed photography test was conducted to verify the speed and impact accuracy of the hailstone. The engine thrust decreased for a short time during the process of hailstone ingestion and then resumed stable operation. After hailstone ingestion， the exhaust temperature of the engine decreased by about 2%～3%， the rotor difference changed by about -0.6%～0.2%， the thrust decreased by about 0.1%～1.6%， and other parameters did not have significant change. The impact response of the fan was collected through the time-domain signal of the low-pressure measuring point. After disassembling the engine， the deformation of the intake edge of 5 fan blades was checked， no block falling occurred， and no structural damage that may endanger flight safety was found， which verified the airworthiness conformance of the turbofan engine for hailstone ingestion.

Rocket image target recognition and tracking is an important application field of image target recognition，which is an important support for rocket test launch and flight control，and has great significance for rocket target tracking and attitude analysis and control.Image tracking of rocket target in ascending stage is an important stage of rocket flight measurement and control，but at present，the video image tracking of rocket in ascending stage mainly relies on manual operation of pinion controller to achieve rocket tracking in the image.The tracking image has some phenomena such as tracking lag and picture shaking，and the tracking effect is greatly affected by human factors.Combining full convolution theory with deep learning method，a method of rocket image target recognition and tracking based on full convolution deep neural network was proposed.Images of rocket launch and ascending flight were collected as samples，and a full convolutional network model was constructed and trained.An end-to-end semantic segmentation method was adopted to realize semantic judgment of rocket targets at pixel level on the basis of deep classification network，with good recognition rate and robustness.Based on the recognition of the rocket target，the PTZ control model was established，the high-quality image of the rocket ascent stage was obtained through the intelligent control of the PTZ，and the tracking of the rocket target was realized.

In order to solve the problem that the classical K-means clustering algorithm reguired users to know the number of clusters in advance and the clustering results were sensitive to initialization of the algorithm， a comprehensive scheme was proposed to improve the random initial partitioning of K-means algorithm and visually determine the number of clusters. Firstly， the data was standardized to make it obey normal distribution， and the most important features were extracted by principal component analysis to achieve dimensionality reduction of high-dimensional data. Then， the farthest centroid selection and min-max distance rule were used to modify the random initialization of K-means algorithm to avoid empty clusters and ensure data separability. Based on these， the statistical empirical rule was used to estimate the range of the number of clusters， and the optimal number of clusters was assessed by searching the elbow of sum-of-squared-error curve within this range. Finally， by calculating and comparing the silhouette coefficients of each cluster， the clustering quality of the algorithm was evaluated， thereby ultimately determining the inherent number of clusters in the data. The simulation results show that the proposed scheme can not only visually determine the potential number of clusters in the data， but also provide an effective method for high-dimensional data analysis in the era of big data.

In engine performance analysis，the ratio of the cooling air flow of high-pressure turbine guide vane to compressor inlet flow is a fixed value.However，the actual bleed air flow of the turbine is affected by some factors，such as the pressure difference between the inlet and outlet of the cooling air flow path，flow area and flow resistance，resulting in deviations between the current simulation accuracy and actual performance of the engine.To further improve the simulation accuracy of engine performance，a modeling method based on the flow characteristics of high-pressure turbine guide vane was proposed.The computational model of the core engine performance was improved based on this method，and the performance parameters of one core engine were calculated with this model.The numerical results show that，for this core engine the actual cooling air flow of turbines decreases after considering the flow characteristics of high-pressure turbine guide vane，and the cooling air flow is more sensitive to the change of the total pressure recovery coefficient of the cooling air flow path.In the performance calculation of the core engine，after using the cooling air flow correction of high-pressure turbine guide vane，the total temperature at the outlet of the combustion chamber decreases by 1% to 2%，the temperature ratio of the core engine increases by 0.2% to 0.45%，the unit cycle power increases by 0.24% to 0.48%，and the pressure ratio of the core engine and the fuel consumption rate per unit cycle power change relatively little.

The carbon emissions from the power industry remain a central focus of China’s current carbon reduction efforts.The production-related and consumption-related carbon emissions from electricity across 30 provinces in China were analysed and clarified the scale and pathways of spatial and sectoral transfers of carbon emissions between these provinces.The primary analysis was conducted using the multi-regional input-output （MRIO） model.The results indicate that carbon transfer is the primary driver of regional disparities in carbon emissions related to both production-related and consumption-related.Overall，the trend revealed that carbon emissions are being transferred from economically developed provinces to less developed provinces with surplus electricity supply.At the departmental level，the majority of power transfers are caused by electricity demand from the construction and service industries，which accounts for 57.89% of carbon emissions.It is effectively identified that the spatiotemporal characteristics and transfers of carbon emissions from electricity across various provinces，providing a scientific basis and theoretical foundation for the development of carbon reduction programs for the power industry at the provincial level.These efforts will contribute to achieving the goals of carbon peaking and carbon neutrality.

Taking a six degrees of freedom desktop upper limb rehabilitation robot （DULRR） as the research object，it was observed that traditional position control cannot meet the needs of patient rehabilitation training and may lead to secondary injuries during the rehabilitation process.To address this issue，a position closed-loop adaptive compliance control method was proposed.Firstly，based on the kinematic model of DULRR，a position controller based on fuzzy PID was constructed.Then，utilizing the impedance model’s ability to convert force signals into velocity and position signals，an adaptive compliance controller based on pressure sensors was proposed.Combined with the proposed fuzzy PID controller，a complete DULRR passive rehabilitation training control method was formed.Finally，the superiority of the adaptive compliance control method based on position closed-loop was verified through simulation analysis and prototype experiments.The experimental results show that compared with traditional PID controllers，fuzzy PID in the DULRR system has shorter response time and smaller steady-state error，demonstrating better trajectory tracking ability.Meanwhile，the controller exhibits good flexibility，meeting the needs of early passive rehabilitation training for patients and avoiding secondary injuries during the rehabilitation process.

In view of the problems of difficulty in extracting gear fault features and low diagnosis accuracy under noisy environment，a gear fault diagnosis method was proposed，which combined VMD parameter optimization based on comprehensive evaluation indicators，KPCA feature fusion and BP network.Firstly，in order to effectively evaluate the IMF components after VMD decomposition and avoid the problem of manually setting relevant parameters for VMD，a comprehensive evaluation index based on envelope entropy and kurtosis was designed to establish a fitness function for VMD parameter optimization and screen the optimal IMF components.Secondly，after performing VMD decomposition according to the optimal parameters，a multi domain feature set was extracted from the optimal IMF component，and then the KPCA model was used to fuse its features.Finally，fault diagnosis was performed using the BP network model.The experiment shows that under the same experimental conditions，compared with other traditional methods，this method improves the recognition rate of gear faults，with an accuracy of up to 98%，proving the effectiveness of this method.

In the complex and changing market environment， high-end equipment manufacturing industry must enhance technical strength and improve innovation effectiveness through collaborative innovation， of which partner selection is a key step to achieve innovation. Firstly， by analyzing and summarizing relevant literatures on partner selection evaluation indicators， and combining with the colla-borative innovation development needs of high-end equipment manufacturing industry，the collaborative innovation partner selection index system of high-end equipment manufacturing industry was estab-lished from seven dimensions： complementarity， compatibility， innovation resources， innovation ability，innovation environment， reputation and trust as well as technical level. Then，considering the complexity of the high-end equipment manufacturing industry and evaluation fuzziness，a group decision-making model of EDAS was constructed based on possibility under the probabilistic hesitancy fuzzy environment，and applied it to the collaborative innovation partner selection under indicator system. Finally，a numerical example was used to verify the research method，which provides a certain reference for the collaborative innovation partner selection of high-end equipment manufacturing industry.

In order to solve the flight stability issue of small-sized single-person rotorcraft，considering the small-sized and heavy load of the entire aircraft，a stability-enhancing structure scheme was proposed and modal simulation analysis was conducted. Firstly，the center of gravity was determined to ensure that the design met the requirements of flight stability.Secondly， a stability-enhancing structure scheme was proposed to address stability issues such as sway and vibration.Finally，the damper scheme was determined through cockpit dynamics simulation with different damping values of damper and the mode simulation analysis was completed.The results show that the target aircraft can achieve stable flight and meet the practical application requirements through reasonable structural design and flight control strategies while maintaining a small size.

To solve the problems of the large amount of assembly information and complex assembly process in modern aircraft component assembly，an online visual assembly model based on digital twin and other related concepts was proposed.Comprehensively considering physical data and process parameters such as the dimensional accuracy of parts，assemblies，and segments involved before and after the assembly process of aircraft components，an information integration platform was built combined with Oracle database technology，which realized the data interaction between physical space and twin space in the form of mapping tables，visually managed various information in the assembly process.The model improves assembly efficiency，at the same time helps staff make scientific decisions.

In response to the complexities of fault signal transmission path，instability and difficulties in extracting fault feature for aircraft engine main bearing，a fault recognition method was proposed based on the fusion of time-domain feature parameters，frequency-domain feature parameters and intrinsic mode function （IMF） energy moment feature parameters for dimensionality reduction.Firstly，60 groups of bearing rolling element fault，inner ring fault，outer ring fault and bearing without fault data were selected respectively then time-domain，frequency-domain and energy moment features were extracted from these instances.Addressing the issue of high dimensionality，extensive data and redundant information of the fusion vector composed of three parameters，principal component analysis （PCA） was employed to reduce the dimensionality of these data and effective principal components were extracted based on cumulative contribution rates of principal components.Finally，the dimensionality reduction feature vectors were input into the support vector machine （SVM） for pattern recognition to diagnose the types of bearing faults.The results demonstrate that compared to models employing single feature parameters，this method effectively extracts fault feature vectors from complex signals.Subsequently，it accurately identifies and classifies fault types using these feature vectors，achieving a fault recognition rate of 98.75%.

In order to explore the changes in the flow and heat transfer characteristics of the internal cold channel in turbine blades，the effects of rotation numbers 0，0.03，0.06，0.09 and 0.3 on the flow resistance and heat transfer capacity of constricted and expanded serpentine channels with inlet Reynolds number Re=3×10⁴ were investigated.The results show that the Coriolis force and centrifugal force together push the fluid in the center of the channel to the pressure surface，and then the upper wall of the channel is returned to form a vortex with the central fluid，thereby changing the flow structure in the channel.With the increase of the rotation number，the channel resistance coefficient shows a trend of first increasing and then decreasing.When the rotation number rises from 0 to 0.3，the channel resistance coefficient decreases by 53.79%.Rotation destroys the boundary layers of the suction surface and the pressure surface by strengthening the fluid disturbance to enhance the overall heat transfer capacity of the channel.When the rotation number increases from 0 to 0.3，the local heat transfer capacity of the first and second processes increases significantly higher than that of other processes，and increases by 68.49% compared with the relative stationary channel.The overall heat transfer capacity of the constricted and expanded serpentine channel is increased by 21.18% compared with the stationary state.

In the flight test measurements of civil aircraft noise airworthiness certification，the influence of meteorology is crucial to the final test results.In order to ensure that applicants correctly measure meteorological parameters，correct noise data for meteorological effects and obtain accurate test results，technical index requirements was put forward for test equipment used by ground weather stations，vehicle-mounted weather stations and meteorological aircraft.The review guidance requirements were given for measurement methods，test flight methods，data validity，etc.The atmospheric stratification method and the quadratic interpolation method of the atmospheric sound attenuation coefficient used to correct the noise measurement data were proposed，analyzed，and verified.In the process of noise airworthiness measurement data correction，the noise results can meet the baseline conditions and airworthiness requirements by calculating the atmospheric sound attenuation coefficient and correcting the measurement data.The meteorological data test flight，measurement，analysis，and data correction methods proposed improve the accuracy and reliability of the final noise results and ensure the comparability and reliability of noise test results under different meteorological conditions.

Due to its powerful information storage and programming capabilities， DNA is regarded as the most promising new material for information storage and processing. There is an urgent need to develop novel DNA manipulation techniques and information processing strategies to achieve nanoscale information storage and processing in non-traditional media， thereby providing the tools for construc-ting a new generation of intelligent molecular information processing systems. Based on the basic mechanism of signal regulation and communication in organisms， relevant research in recent years from the perspective of DNA molecular allostery was summarized. Firstly， relevant research on DNA circuits was summarized， including DNA strand displacement technology and the application of protein enzymes. Secondly， the important research results of intelligent molecular information processing from two aspects of information processing and computing platform were analyzed and summarized， inclu-ding DNA neural networks， analog biochemical reaction systems with learning ability， and computing frameworks based on DNA origami. By sorting out these studies， the key problems faced in current intelligent molecular information processing research were analyzed and potential solutions were proposed. Finally， the future directions and applications of this field were summarized and prospected.It is believed that molecular information processing technology will bring more efficient and intelligent information processing methods to human society.

As one of key tasks in the field of remote sensing image processing，object detection has always been a research hotspot.Although significant progresses have been made in this field，the deep learning methods still face significant challenges in dealing with scale changes and complex backgrounds in remote sensing images，which limits the further improvement of detection accuracy to some extent.To address this issue，an innovative object detection method for remote sensing images was proposed，which integrated a saliency guided image adaptive fusion module and improved Faster RCNN to enhance the accuracy of object detection.Firstly，in the image preprocessing stage，a saliency guided image adaptive fusion module was proposed，which effectively integrated the semantic information of the image and shallow fine-grained details，allowing the model to prioritize the object region while minimizing background interference.Secondly，after introducing MobileNetV3 as the feature extractor of Faster RCNN，an attention enhanced feature pyramid network was proposed，which combined attention with upsampling to further enhance target features and output high-quality feature maps，effectively improving the extraction effect of multi-dimensional features and providing more accurate and rich feature information for subsequent object detection tasks.Furthermore，a multi-scale region proposal network was designed，which can more accurately capture the features of objects of different sizes and shapes，thereby enhancing the expression ability of features and effectively improving the detection accuracy of targets.Finally，experiments on the DIOR and ROSD datasets demonstrated that the proposed network model exhibits higher detection accuracy compared to other advanced methods，fully demonstrating its superiority and effectiveness.

In order to investigate the influence of root slot on the static cascade performance of transo-nic compressor， the high-load diffusion cascade model NACA65-K48 was used as the basis and the influences of root slot with different width and height on the aerodynamic performance and flow structure of the compressor cascade were studied through numerical simulation. The results show that the root slot can effectively weaken the vortex strength on the suction surface， suppress the corner separation along both the pitch and blade height directions， and reduce the flow loss of the cascade. The inhibitory effect firstly increases and then decreases with the increase of the slot width， and it shows a similar trend with the rise in the height. When the width is 12% chord and the height is 5% blade height， the root slot structure of the blade has the most significant effect on improving the flow within the slot， reducing the cascade flow loss by 11.19%.

To improve the surface quality of TC4 alloy prepared by selective laser melting，the ultrasonic assisted magnetic abrasive finishing technology for polishing was adopted.Simulated the size and distribution of magnetic flux density under different magnetic pole dimensions and finishing areas.The cavitation bubble in the composite field force was analysed，the magnetic field on the cavitation effect on the influence mechanism was studied.The changes in surface morphology of the specimen before and after applying ultrasonic vibration were analyzed.The influence of amplitude，finishing rotate speed and finishing time on surface roughness were studied.The results show that as the diameter and height of the magnetic poles increase，the magnetic induction intensity increases and the distribution of magnetic flux density gradually changes from uneven to uniform.The finishing effect is better at the position of 80 mm，with an average magnetic field force of 17.12 mN and a surface roughness of 0.673 μm.After applying ultrasonic vibration，the original surface bumps and defects are completely removed，the surface exhibites fine scratches and the surface roughness is reduced to 0.243 μm.Compared with the magnetic abrasive finishing process，the surface roughness is reduced by 60%.The suppression effect of the magnetic field on the cavitation effect reduces the energy released when the cavitation bubble collap-ses， thus reducing the damage caused by the cavitation effect on the surface.

Based on imprinting theory and upper echelons theory，took 2015—2021 Shanghai and Shenzhen A-share military-civilian integration listed enterprises as research samples，empirical regression methods were used to verify the impact of the CEO’s IT background on the innovation performance of military-civilian integration enterprises as well as the intrinsic mechanism.The results show that CEO’s IT background has a significant positive effect on the innovation performance of military-civilian integration enterprises.CEO’s IT background has a significant positive effect on digital transformation.Digital transformation plays a partial mediating role in the promotion process of CEO’s IT background to the innovation performance of military-civilian integration enterprises.Government subsidy plays a positive moderating role.After the robustness and endogeneity tests，the above findings remain robust.Further heterogeneity analysis reveals that the promotion effect of CEO’s IT background on the innovation performance of military-civilian integration enterprises is more significant in civil-involved military enterprises and the eastern region.

Under the cellular vehicle-to-everything （C-V2X） communication technology framework，the accuracy of basic safety messages （BSM） is crucial for ensuring road traffic safety.However，BSM data is susceptible to non-malicious factors such as sensor faults or environmental disturbances，leading to data anomalies that may misguide driving decisions.In response to this issue，two-phase learning strategy for correcting anomalies in BSM was proposed.In the first phase，an unsupervised hybrid generative model was used to learn the behavior patterns and distribution characteristics of normal BSM data and a memory module was introduced to construct a fine-grained prototype repository in the feature space for enhancing the model’s understanding of the diversity of normal behavior patterns.In the second phase，based on the network parameters obtained in the first phase，a self-supervised learning strategy was employed for data correction.Results show that the proposed solution exhibits good correction capability and significantly reduces the error in BSM.

Fuzzy support vector machine is a classification algorithm that combines support vector machine and fuzzy theory.The existing fuzzy support vector machine algorithms can overcome the impact of noise data to some extent，but they have cost sensitivity，leading to inaccurate estimation of the prior distribution of data.A new fuzzy support vector machine algorithm was proposed.When designing the fuzzy membership function of samples，this algorithm better captures the distribution information of data by using the outlier factor constructed by the similarity of sampling points and their neighborhood density.The optimized model is validated using UCI datasets，which proves its good performance.

A method of expanding and improving the hierarchical control structure model （HCSM） of system-theoretic process analysis （STPA） using functional attribute（FA）and directional interaction tag （DIT） was proposed. Based on this method， the hierarchical functional control structure and interaction model （HFCSIM） of the system and essential improvement to STPA was obtained. Through this modification， issues such as the lack of specific methods and forms follow， incomplete interaction information between components， excessive reliance on “Brainstorming” and the difficulty in ensuring model consistency could be solved， and the systematicness， completeness and correctness of the analysis results could be fundamentally ensured. Finally， the effectiveness of the modified method was validated by taking the aircraft wheel braking system as an example.

In order to meet the jointing requirements of large-scale components，TA15 wire was used as raw material to join TA15 base metal prepared by laser deposition manufacturing （LDM） and hot isostatic pressing （HIP） with tungsten inert gas （TIG）.The results show that there are differences in the microstructure of different regions of the jointing specimens.The microstructure of the HIP base metal is equiaxed grains，and the internal structure of the grain is bimodal structure.The heat affected zone A （HAZ-A） of HIP base metal presents equiaxed grains.The TIG jointing zone presents β columnar grains，and the intracrystalline is fine basketweave structure.The LDM zone presents β columnar grains，and the basketweave structure inside the grains is coarser than that in the TIG jointing zone.The grain shape of heat affected zone B （HAZ-B） of LDM base metal is still columnar grains.The tensile strength and elongation of the jointed specimens at room temperature are 1 046.3±13.7 MPa and 6.2%± 0.6 % respectively.Compared with the base metal，the tensile strength of the jointed specimens increases and the plasticity decreases.The fracture position of the jointing specimens is HAZ-B，and the fracture mechanism belongs to semi-cleavage and semi-ductile fracture.

There are a large number of thin-walled components in aerospace products，which deform before the assembly process due to manufacturing errors and components stress release.If all the dimensions or tolerances of the assembled components can still be interchanged within the limit size，it often requires too high requirements for the components，the production cycle is also too long，and the manufacturing cost of the components was also too high.In response to the uncertainty of product components deformation， the modeling method of measured data was conducted，which effectively integrated technologies such as components geometry，material elasticity，physical data，virtual assembly and simulation analysis.Combined with components data collection before assembly，dynamic simulation analysis was conducted on the assembly process of component deformation bodies.So that out of tolerance deformation components can still be assembled normally.Furthermore，by combining big data and artificial intelligence technology，an equivalent model and application software for the assemblability of on-site deformable bodies are provided，realizing the rapid determination of whether deformed components can be assembled based on measured data of components on the production site，ensuring the continuous production of the production line.

Composite material tanks are prone to microcracks at cryogenic temperature conditions.Accurately predicting the initiation and evolution of structural microcracks at cryogenic temperatures is crucial for the design and optimization of tank structures.An embedded multi-fiber representative volume element model was established to study the influence of cryogenic temperature conditions on the initiation and evolution of microcracks in composite laminates that considered the in-situ effect.This model was consisted of two 0° adjacent constraint layers and a 90° intermediate layer.Under transverse tensile loads and cryogenic temperature conditions，the microcrack initiation strain and tensile stress in the structure were obtained，and the initiation and evolution of microcracks were studied.The results show that cryogenic temperatures significantly influence the initiation location and evolution path of microcracks in structures. Microcracks are prone to occur in structures at cryogenic temperatures. The interface debonding is more likely to occur in regions with denser fiber arrangements at cryogenic temperatures. The thermal stress results in variations in microcrack initiation location between ordinary temperature and cryogenic temperature conditions. Notably，the structural damage evolution pattern in the intermediate layer remains consistent under both ordinary and cryogenic temperatures.