Abstract:In order to investigate the current development status and key issues in the field of cross-modal retrieval based on real-valued features for unlabeled datasets (hereinafter referred to as cross-modal retrieval), this paper conducts an analysis and summary of the existing literatures. Cross-modal retrieval refers to the retrieval of samples from one modality that are relevant to a given query from another modality. Firstly, using a time complexity-based classification approach, existing cross-modal retrieval methods are categorized into feature-based methods and score-based methods. Secondly, the research status of these two categories of methods is described, and the main issues in the current stage for each category are analyzed and discussed. Furthermore, two mainstream datasets and commonly used evaluation metrics for cross-modal retrieval are introduced, and the performance of the two categories of methods on public datasets is compared and analyzed. Finally, key issues to be addressed in the field of cross-modal retrieval are summarized. The research indicates that although significant progress has been made in existing cross-modal retrieval methods, there are still key issues that urgently need to be addressed. These key issues represent important directions for future development in the field of cross-modal retrieval.

Abstract:To enhance the capability of the synthetic aperture radar (SAR) system in anti-retransmission deception jamming, a technique for waveform design and optimization based on non-linear frequency modulation (NLFM) signals is proposed. This approach utilizes autonomous transmit-receive strategies to optimize waveform groups, ensuring that the waveforms emitted during agile transmission are mutually orthogonal. This enables effective suppression of retransmission deception jamming in complex environments. Firstly, the mechanism of retransmission deception jamming in SAR systems is analyzed, and the rationality and effectiveness of waveform agile transmission methods are discussed. A method for anti-jamming using orthogonal waveform design is then proposed. Secondly, NLFM signals are generated using the segmented function method, and the waveform group of NLFM signals is optimized based on the Lagrange algorithm in conjunction with the genetic algorithm. Finally, simulation experiments are conducted to verify the effectiveness of the optimized waveform set designed by the proposed method in countering retransmission deception jamming in SAR systems. The results show that when NLFM waveforms are generated by the segmented function method and optimized for orthogonality using the Lagrange genetic algorithm with appropriate jamming forwarding delays, they improve the mainlobe width and peak sidelobe ratio of the waveforms. This enhancement in orthogonality of agile waveforms leads to improved waveform quality.

Abstract:To address the issues of overfitting on new classes, limited generalization ability on old classes, and catastrophic forgetting of incremental learning-based incremental intrusion detection system (IDS) when dealing with attacks of new classes, an adversarial assistance-augmented incremental IDS is proposed. In the incremental training, the regularization of adversarial samples is leveraged to mitigate the overfitting on new classes. A dual distribution simulation buffer that stores both clean and adversarial samples of old classes is proposed to enhance the generalization ability to old classes. In addition, weighted cross-entropy loss is introduced into the training process to alleviate the catastrophic forgetting. Experimental results on the CSE-CIC-IDS2018 dataset and the UNSW-NB15 dataset show that direct participation of adversarial samples in training leads to deterioration of the recognition performance, while participation in the form of detached data distribution enhances the recognition performance of the model. The storage of adversarial samples in the buffer effectively suppresses the loss of the models generalization ability for old classes, and the adjustment of learning weights by weighted cross-entropy loss alleviates the catastrophic forgetting caused by the imbalance between the new classes and the data in the buffer. The proposed method offers a viable strategy for detecting real attacks within dynamic and complex networks, presenting substantial practical applicability.

Abstract:To investigate the fretting wear of the contact parts of electrical connectors caused by random vibrations during operation, resulting in decreased contact performance due to wear debris, stepped stress and constant stress random vibration tests were conducted. Electrical capacitance tomography technology was used to detect the wear debris between the contact parts of the electrical connectors during fretting wear. The effects of vibration period, direction, magnitude, and load current on wear debris accumulation and distribution were studied. Surface morphology and energy spectrum analysis were conducted on the specimens after testing, and fractal dimension was used to study the uniformity of debris distribution, quantifying the extent of wear on the contact surfaces. Results of the tests show that vibration period, direction, and magnitude have a positive cumulative effect on the generation and accumulation of wear debris between contact parts. The coupling effect of load current, leading to heating and stress relaxation, delays the performance degradation process. Axial vibration causes the most severe wear, with the root of the pin being the area of severe wear. The degradation patterns of the electrical connectors under stepped stress random vibration are similar to those under constant stress, with wear debris characteristic values under stepped stress showing a certain step-change feature. Significant changes occur in contact surface morphology during fretting wear of the connectors. The results of surface morphology and spectrum analysis align with the characteristics of wear debris and the variation in contact resistance.

Abstract:To optimize the energy consumption of unmanned aerial vehicle (UAV)-assisted information collection system in wireless sensor networks (WSN), we propose a joint optimization algorithm for UAV trajectory and sensor power allocation that considers inter-link interference. This approach comprehensively takes into account both UAV flight energy consumption and sensor node uplink data transmission energy consumption. Firstly, we construct a system energy minimization problem model based on practical constrains. Then, considering the non-convex nature of the multi-constraint optimization problem, we employ the block coordinate descent (BCD) method to decompose the problem of minimizing system energy consumption into two sub-problems: power allocation with a fixed UAV trajectory and UAV trajectory optimization with fixed power allocation. We apply convex approximation to transform the non-convex problem into a solvable approximate convex optimization problem by leveraging the mathematical characteristics of the sub-problems. We obtain an approximate sub-optimal solution to the original non-convex problem through alternating iterative optimization of the two sub-problems. Finally, the feasibility and effeetiveness of the proposed algorithm are verified through simulation experiments. The results show that the proposed algorithm can reduce system energy consumption by up to 22%, significantly outperforming the comparison algorithms. Furthermore, the advantages of the algorithm become more pronounced as the number of sensor nodes increases. This study provides a systematic approach for energy optimization in information collection in WSNs, effectively reducing system energy consumption by slightly increasing UAV flight energy consumption.

Abstract:In order to solve the problems that the current deep learning dehazing algorithm being ineffective in using multi-scale features when processing non-uniform foggy images, resulting in color distortion and incomplete detail recovery of the restored images, this paper proposes an image dehazing algorithm with mixed attention and multi-feature interaction. Firstly, the coding module is used to extract features at different scales. Secondly, a mixed attention module is constructed to globallay perceive the image fog weights to different fog concentrations using channel attention mechanisms. Then, a multi-feature interaction module is designed to facilitate information exchange between features of different scales, effectively use the semantic information in low-resolution features, retain the spatial details and color information of high-resolution features, and use the gated fusion module to aggregate features of different scales. Finally, the decoding module reconstructs the fused features to obtain a fog-free image. Experimental results show that the dehazing images recovered by the proposed algorithm not only have natural colors and clear details subjectively, but also outperform the existing mainstream algorithms in objective indicators. These research findings offer novel approach for both research and application of deep learning-based image dehazing.

Abstract:To address the problem of synchronizing chaotic systems with unknown parameters and complex state variables that cannot be separated, a misalignment complex correction function projection synchronization (MCCFPS) control scheme is proposed. First, an adaptive tracking controller is designed to accommodate the differences between any bounded complex chaotic systems. This controller enhances adaptive capability and adjusts convergence speed through dynamic control strength and convergence factors, and proves the convergence of the controller through Lyapunov theory, providing theoretical support for the reliability of the control scheme. Subsequently, the dynamic models of two heterogeneous complex chaotic systems are analyzed. Fuzzy entropy of the time series of chaotic and complex chaotic systems is calculated and compared to validate that complex chaotic systems have higher complexity. This highlights the superiority of their synchronization control in secure communication applications. The MCCFPS synchronization scheme is further designed. Finally, the effectiveness of the MCCFPS scheme is verified through numerical simulations. The results demonstrate that the scheme successfully achieves synchronization control of complex chaotic systems, exhibiting good synchronization effects and control performance. Overall, compared to traditional chaotic systems, this study introduces complex variables and adopts a more sophisticated synchronization control scheme, MCCFPS, to directly achieve chaotic synchronization in the complex domain, effectively enhancing system security. This research provides new ideas and approaches to the synchronization control problem of complex systems, holding significant practical value.

Abstract:With the increase of uncertainty in the construction environment of industrial internet engineering, emergency response capability has gradually become an important factor in ensuring the successful implementation of engineering projects. In order to improve the emergency response ability of industrial internet engineering supply chain, considering the interactive relationships between entities, the game relationship among emergency response entities of industrial internet engineering supply chain is studied in the construction stage of industrial internet engineering projects. A fusion evolutionary game model is established to reveal the integration mechanism of emergency response entities of industrial internet engineering supply chain. The simulation results show that the interest connection of industrial internet engineering supply chain enterprises is the key driving force for the integration of emergency response entities. Mutual benefit, reciprocity and mutual trust serve as the prerequisite basis of successful integration. When the continuous intergration profit increment and income subsidies of participating enterprises outweigh the impact of phased integration strategies adopted by other enterprises on the chain, participants will choose continuous integration strategies regardless of the initial intention of enterprises. Conversely, the evolutionary results will diverge. Understanding the mechanism of emergency response integration and incentivizing active participation of emergency response entities in integration can improve emergency response efficiency and emergency response capacities, ensuring the stability of the supply chain and facilitating the successful implementation of industrial internet project construction.

Abstract:To improve the voltage and power fluctuations arising from large spacing between the trarsmitting arrays in the dynamic wireless power transfer (DWPT) systems, this paper proposes a novel transmitting structure. Firstly, based on the bipolar pad (BPP) coil structure, the domination transmitter (Dtx) coil and the compensation (Cx) coil are connected serially to form a transmission segment. The DWPT system adopts segment control to minimize losses. Secondly, decoupling capacitors are used to decouple the Dtx coil and Cx coils, and the circuit principles of the system are analyzed. Then, the coupling structure is optimized and designed using Ansys Maxwell finite element analysis software, and the optimal switching positions are selected to minimize power fluctuations. Finally, an experimental platform is set up to validate the feasibility of the structure. The experimental results show that the power fluctuation of the new magnetic coupling structure is 3.9%, and the voltage fluctuation is 2.9%. Compared with the previous coupling structure, the power fluctuation has decreased by 15.8%, and the voltage fluctuation has decreased by 7.6%. In the transition region between adjacent Dtx coils, compared with the improved coupling structure, the power output has increased by 33.8%, with the system′s transmission efficiency stabilizing around 88%. The proposed novel transmitting structure in this paper helps reduce the fluctuations in output voltage and power during the movement process, and provides a theoretical support for the design and application of magnetic coupling structures.

Abstract:To study the fast and efficient modelling method of rock structures, calibration and 3D point cloud establishment were conducted to investigate the optimal parameters of binocular vision modelling through images correction, alignment, deskew and 3D modelling algorithms based on the LenaCV binocular system. Optimal parameters for binocular vision modeling were determined according to the characteristics of target objects suitable for point cloud modeling. Relationship between the optimal shooting distance and baseline distance for 3D point cloud modeling was discussed. The results show that the optimal grid size of the checkerboard is 10 mm. The optimal calibration distance is 25 times the baseline distance with an error limit of 0.5 pixels. Objects with rough surfaces have better 3D modelling effect than that with smooth surfaces. Objects with higher proportion of targets in the field of view show better 3D point cloud modeling than lower ones. The LenaCV binocular system with a baseline distance of 12 cm gives better results than that with a baseline distance of 6 cm in creating and outputting 3D point cloud. Based on on-site tests, the optimal calibration conditions, photographic conditions, and object characteristics for modeling were discussed, providing practical experience for the reconstruction of three-dimensional structures of rock masses and verifying the feasibility.

Abstract:In the event that the L3 autonomous driving system fails or has difficulty handling complex traffic environments, the driver is required to takeover in an emergency, which can easily lead to traffic accidents. In order to assess the takeover risk of L3 autonomous vehicles, a takeover scenario on the urban expressway was designed and driving simulation experiments were carried out. Based on the theory of driving risk field, dynamic and static risk distribution functions were used to reflect the influence of other traffic units on the takeover risk of ego vehicle. And then, vehicle performance probability factor was introduced to indicate the probability of potential traffic accidents caused by abnormal vehicle trajectories during the takeover process, as well as considering the influence of takeover response time, a risk evaluation model of autonomous driving takeover was constructed. The model parameters were calibrated on the basis of the takeover reaction time and trajectory data obtained from the driving simulation experiments and compared with the inverse time-to-collision to verify the model. The results showed that the values of takeover risk index calculated by the model from 1 s to 9 s after the driver took over were consistent with the inverse time-to-collision. However, the root mean square error of risk index during the takeover (0.059) decreased by 37% compared to the root mean square error of the inverse time-to-collision (0.093). In summary, the constructed model can effectively assess the risk of driver takeover, and the model is more accurate than the inverse time-to-collision in describing the risk.

Abstract:In order to facilitate road engineering personnel to have a more specific and in-depth understanding of roads in cold regions, and to help researchers study the occurrence and development laws of road diseases, and to improve their understanding of the temporal and spatial conditions of road structures, the researchers have conducted research on the deformation laws of road surfaces in cold northern regions. The researchers selected typical highway sections and established 10 monitoring sections to continuously monitor the on-site elevation of roadbed frost heave and thaw settlement deformation. The authors analyze the deformation characteristics and response mechanisms of road surfaces under the influence of temperature fields, and analyze the time-domain deformation characteristics, dynamic deformation characteristics, and differences of highways under natural conditions. By analyzing periodic monitoring data, the researchers constructed parameters such as vertical relative deformation, vertical relative deformation rate, and maximum multiple of deformation rate between cross-sections. They used these parameters to compare the deformation and differences between different sections, cross-sections, and monitoring points. The researchers have provided the coordination law between road surface deformation and temperature field in cold northern regions. The researchs results confirm that the road surface deformation based on temperature field in cold northern regions has typical periodic characteristics. At the same time, through comparative research, the researchers have discovered the differences in deformation characteristics in the longitudinal and transverse directions of the filling and excavation sections, the distribution pattern of deformation extremes, as well as the relative deformation patterns and deformation rate differences during the thermal expansion, frost heave, and thaw settlement stages. The researchers have found that the deformation rate of excavated sections is more than 10 times that of filled sections, and the thermal expansion rate from August to October is greater than that from October to November. The researchers analyzed the melting rate at different times and found that melting mainly occurred from March to May. The authors analyzed that the rate of frost heave is relatively stable, and also found that for fill sections, the frost heave from November to January is greater than that from January to February, while the opposite is true for excavation sections. The authors summarized the time-domain response laws and characteristics of temperature field for road surface deformation in cold regions, providing a basis for research on roadbed, pavement structure and materials, as well as road diseases in cold regions.

Abstract:The parameters of geotechnical are randomly distributed in space. To better reflect the actual engineering geological conditions, considering the uncertainty of soil in the study of foundation bearing capacity, establishing a bearing capacity prediction model holds significant engineering value. Incorporating spatial variability of geotechnical parameters based on random field theory into the study of coupling beam pile foundation, a two-dimensional random finite element model is established using numerical methods to analyze the bearing capacity of coupling beam pile foundation and pile foundation, validated against model test results. Subsequently, a convolutional neural network is employed to establish a model between the random field images of soil parameters and the ultimate bearing capacity of foundations for bearing capacity prediction, and the impact of different parameters is studied based on the prediction model. The results indicate that considering the spatial variability of soil, the foundation′s bearing capacity is in basic agreement with experimental results, with random results consistently higher than deterministic analysis. Under random conditions, both coupling beam pile and pile foundations exhibit normally distributed bearing capacities. The accuracy of the bearing capacity prediction model established using convolutional neural networks is high and can be utilized for parameter analysis. The bearing capacity of foundations increases with increasing soil parameters and decreases with increasing coefficient of variation. Under random conditions, the bearing capacity of coupling beam pile foundation is higher than that of pile foundation, effectively leveraging soil strength to withstand bearing capacity loss caused by parameter uncertainties.

Abstract:In order to accurately predict the actual consolidation rate of soil layer, based on Gibson′s large-stain consolidation theory, a model for the large-strain nonlinear consolidation of soft soils has been developed with consideration of the nonlinear compressibility and permeability characteristics and the time-dependent drainage boundary condition of soils. On this basis, explicit analytical solutions for the large strain nonlinear consolidation model under single-stage and multi-stage loading can be derived. By using the obtained analytical solutions, the influence of interface parameters on consolidation behavior are analyzed. The results indicate that the influence of external loading on the dissipation of excess pore water pressure in soil is that the excess pore water pressure dissipates faster with increasing the loading, but the value of loading has no influence on the final dissipation time of excess pore water pressure. Besides, the influence of parameter Ic(α－2) on soil consolidation behavior is that the consolidation rate changes with the change of parameter values. It is worth noting that the influence of parameter values describing nonlinear behavior of soil on consolidation behavior is delayed under the time-dependent drainage boundary. The time-dependent drainage boundary under different values of interface parameter reflects the actual permeability of soil boundary better. In practical engineering, the corresponding interface parameters can be inversely investigated according to the actual dissipation rate of excess pore water pressure, so as to accurately predict the consolidation process of soil layer.

Abstract:To investigate the mechanical performance and microscopic characteristics of Engineered Cementitious Composite (ECC) after high temperatures, material mechanical tests and microscopic observations of ECC after high temperatures were conducted. The compressive tests and the bending tests of ECC specimens were carried out after natural cooling and spray cooling methods at room temperature (25 ℃), 200 ℃, 300 ℃, 400 ℃, and 500 ℃ and analyzed the failure characteristics of ECC microstructure by scanning electron microscope, and investigated the damage mechanism of ECC after high temperatures. The results show that the concrete on the surface of ECC does not peel off after high temperatures, and there is no burst phenomenon within 500 ℃. As the temperature increases, the fiber gradually melts from the surface of the concrete to the inside, the water loss increases and the maximum burning loss rate is 13.9%. In terms of mechanical properties, the compressive strength of ECC after natural cooling decreases first then increases, and then decreases with the increase in temperature. After water spray cooling, the compressive strength decreases monotonously with the increase in temperature, and the strength decreases significantly. The flexural strength after high temperature decreases gradually with the increase in temperature, and the decrease of natural cooling is more significant than that of water cooling. Combined with the change of microstructure, when ECC undergoes high temperature, the partial melting of the fiber leads to the weakening of the bonding performance between the fiber and the matrix. With the increase in temperature, the hydration products gradually show independent dispersion. However, the secondary hydration of unhydrated particles after water spray cooling is evident, which makes the flexural strength of ECC increase by 17% compared with natural cooling. ECC has excellent thermal stability, and cooling methods influence the apparent characteristics, mechanical properties, and microscopic characteristics of ECC.

Abstract:The essence of macro rutting of asphalt pavement is the result of the evolution of the meso-structure of the mixture. To understand the development process of rutting behavior, the evolutionary mechanism of rutting in asphalt mixture was investigated based on the Jamming transition theory of particle materials. The image processing technology was used to construct a meso-structure model of the asphalt mixture. Rutting simulation tests were conducted based on the time-temperature equivalence principle. The macro and micro variations of the mixture structure during rutting deformation were analyzed. The influence of Unjamming behavior in asphalt pavement and the evolution mechanism of rutting were investigated. The results show that the rutting evolution process of the asphalt mixture can be explained well by the Jamming transition theory, in which the depth of rutting is closely related to the jamming degree of the asphalt mixture system. With the increase of loading time and loading force, the Jamming degree of the mixture particle system gradually increases, along with the increase of rut depth and area fraction, and eventually tends to a stable state. The Jammed state of the particle system is affected by the joint action of asphalt mortar and aggregate skeleton. With the increase in the test temperature, the particle displacement and displacement difference inside the specimens increased, and the Jamming blockage degree of mixture particle system decreased. The transition of the asphalt pavement particle system from a stable Jammed state to an unstable Unjammed state is the internal mechanism inducing rutting and oil flooding. This study successfully applied the Jamming theory to an amorphous system, providing a new research method for understanding the rutting evolution behavior of asphalt mixture.

Abstract:The effect of screw pile and pile side soil is relatively complex. In order to study the vibration characteristics of screw pile under vertical load, the longitudinal vibration characteristics of screw pile in hysteretic damping foundation are studied from the three-dimensional fluctuation of pile side soil, considering the layered characteristics of soil and the vertical action of pile side soil. Based on the three-dimensional wave theory, the wave equation of pile side soil is established, and the vibration response solution of screw pile under the condition of complete coupling between pile and soil is obtained by Laplace transform and modified impedance function transfer method. The theoretical calculation results are compared with the field measured curves to prove the rationality of the screw pile-soil interaction model. The results show that compared with the assumption that the soil around the pile is homogeneous, considering the layered characteristics of the soil can fully consider the reflection and refraction of the stress wave at the layered interface, which is more in line with the actual situation. In the range of effective pile length, with the increase of pile length, the side friction resistance of pile side soil to screw pile is larger, and the damping effect is more obvious. The existence of the screw thread of the screw pile has a certain vibration reduction effect on the pile body, and with the increase of the outer diameter of the screw teeth, the amplitude level and resonance frequency of the complex stiffness curve of the pile top decrease significantly. The theoretical model can better simulate the interaction mechanism between the screw pile and the layered soil, and provide theoretical support for the application of the screw pile.