Research progress in parametric modeling, intelligent optimization, scheme generation, and intelligent details of bridge structures was systematically reviewed in this paper, and the development trend of the intelligent design of bridge structures was predicted. Current research on bridge modeling mainly includes parametric building information modeling (BIM) and parametric finite-element modeling. The BIM obtained from parametric modeling was mainly used for visualization, which is difficult to convert into a finite element model (FEM) for structural analysis and optimization. The FEM obtained from parametric modeling could be used for the analysis and optimization of structures. However, the layouts of the loads and boundary conditions for a relatively complex FEM were still manually implemented. Moreover, it was difficult to accurately convert FEM into a BIM model. In research on the intelligent optimization of bridge structures, heuristic algorithms remain the dominant approach for optimization, which mainly focuses on reinforced concrete structures. Few studies have been conducted on steel and steel-concrete composite structures. In addition, most studies have focused on single-load cases. The generative approach can be used for the rapid generation of bridge schemes. However, current studies have only considered span design and bridge-type selection. There have been few research results on the intelligent detailed design of bridge structures, and the integration of intelligent detailed design and digital manufacturing has not been considered. It is foreseeable that through intelligent design technology, the main direction of bridge structural design development will be the implementation of the intelligence of scheme generation, modeling, optimization, and detailed design of bridge structures, along with the integration of digital schemes with subsequent digital manufacturing and intelligent construction.

    截至2023年底,我国公路总里程达544.1万千米,其中高速公路总里程18.4万千米。然而,我国沥青路面的设计使用寿命一般为10~15年,随着服役年限的增加,达到设计使用年限的沥青路面逐年增多。因此,在“交通强国”战略与“双碳”目标下,针对我国公路事业发展所面临的既有沥青路面存量大、老龄化严重等难题,延长既有沥青路面结构寿命,高质量保证既有路面运维水平是交通行业发展的重大需求。近年来,众多学者和科研人员在既有沥青路面结构延寿方法与技术研究方面开展了大量卓有成效的工作,取得了丰富的研究成果,为我国既有沥青路面结构延寿提供了重要理论和技术支撑。
    为充分展现我国既有沥青路面结构延寿方法与技术研究领域的最新研究成果,及时总结该领域的前沿动态,推动延寿方法和技术的不断发展与创新,《中国公路学报》编辑部联合长沙理工大学于新教授、东南大学钱振东教授、长安大学汪海年教授,共同策划了“既有沥青路面结构延寿方法与技术”专栏;并邀请东南大学罗桑教授、同济大学严宇教授、长安大学杨旭教授、长沙理工大学李盛教授、哈尔滨工业大学王大为教授、重庆交通大学何亮教授、河海大学朱浩然副教授、长沙理工大学黄拓副教授作为组稿专家,共同向该领域的知名专家学者约稿,出版本期“既有沥青路面结构延寿方法与技术”专栏。本专栏共收到论文30余篇,最终录用9篇研究内容主要集中于以下2个方面:
    (1)既有沥青路面性能演变规律研究。主要内容包括:长期服役状态下既有沥青路面结构延寿关键技术综述、基于亿次加载试验的长寿命沥青路面性能演化规律及设计体系研究、基于道路结构承载力及环境严酷程度的服役寿命评价指标研究、人工气候加速作用下胶粉再生沥青老化特征及时空演变模型、不同应力状态下沥青混合料疲劳剩余强度及损伤演化规律研究。
    (2)既有沥青路面病害成因及检测技术。主要内容包括:基于温度应力分析的半刚性基层沥青路面拱胀病害成因研究、沥青混合料隐性病害动态共面电容成像优化、基于自适应课程学习的探地雷达道路隐性病害检测增强。
    在此特向专栏组稿专家、审稿专家、论文作者的辛勤付出致谢,希望本期专栏的出版可以进一步推动既有沥青路面结构延寿设计方法及技术的不断发展与创新。《中国公路学报》将持续关注该领域的国内外最新研究进展,以期为广大专家、学者及工程技术人员提供学习、交流的平台,促进我国交通基础设施建设的高质量与可持续发展。由于水平及时间有限,专栏中的不足之处在所难免,恳请各位专家不吝指出。

To review the latest technological advancements in the detection of underwater structural defects in bridges, this study focused on innovative applications of underwater robots to improve detection accuracy and efficiency. Underwater robots can carry both non-contact detection devices, such as optical and acoustic sensors, and contact detection devices, such as ultrasonic instruments and rebound hammers, demonstrating their potential for efficient detection in complex underwater environments. This study undertook a detailed analysis of the adaptability and improvement techniques of non-contact detection methods based on optical and acoustic principles, highlighting effective approaches for enhancing image quality and detection accuracy. It also clarified the current status of underwater contact detection research and proposed a solution for the collaborative operation of underwater robots with contact detection devices. This work emphasizes that the future direction of underwater detection lies in the use of underwater robots equipped with contact-detection devices. The challenges faced by current underwater bridge structure detection technologies are summarized, and new underwater detection methods based on intelligent algorithms and multisource data fusion are proposed, offering specific directions and technical paths for future research.

Artificial intelligence (AI) technology has become a core component of national strategic science and technology. Its integration with bridge seismic engineering is emerging as a critical approach to enhancing the seismic resilience of infrastructure. Bridge seismic analysis has long faced challenges such as complex physical models and the difficulty of balancing efficiency with accuracy. Addressing these challenges, this study systematically reviews the application and innovation of traditional machine learning models, deep learning models, and next-generation AI fusion technologies in bridge seismic analysis, including: ① Intelligent synthesis and input of complex ground motions; ② Seismic capacity analysis and demand prediction; ③ Damage assessment and fragility analysis; ④ Resilience evaluation and recovery strategy optimization; ⑤ Seismic analysis of large-scale bridge networks. AI has significantly improved the efficiency and accuracy of bridge seismic analysis, opening new avenues for exploring problems involving multiple parameters and strong nonlinearity. However, existing AI models still face persistent challenges, including insufficient foundation in physical laws, weak model generalization capabilities, and difficulties in effectively integrating heterogeneous data sources. Looking ahead, AI technology will further advance the field of bridge seismic engineering through enhancing the interpretability of physical laws, developing multi-modal sensing technology, building high-fidelity databases, strengthening model generalization capabilities, and developing novel intelligent algorithms. This research facilitates a paradigm shift in bridge seismic studies, moving from reliance on manual expertise towards a deep integration of physical laws with artificial intelligence.

    截至2023年底,我国公路总里程544.1万千米,高速公路总里程18.4万千米。随着人工智能、大数据等技术的发展,如何利用新技术提高道路运维效率和质量成为我国道路建设运维领域的重要课题。为此,科技部立项国家重点研发计划项目《智慧道路建设运维关键技术(2023W1B2403500)》,旨在形成智慧道路建设运维技术标准,推动我国智慧道路高质量发展。发展道路数字化对提高道路通行效率、提升行车安全、促进经济发展具有深远的意义,智慧道路建设运维相关研究成果可为推动交通行业高质量发展,加快建设交通强国提供重要支撑。
    近年来,在智慧道路建设运维理论、方法与关键技术研究领域,众多学者和科研人员开展了大量卓有成效的工作,取得了丰富的研究成果,为我国道路建设与运维技术的提升和交通基础设施的运营安全提供了重要的理论和技术支撑。为充分展示我国智慧道路建设运维理论、方法与关键技术研究领域的最新研究成果,引领该研究领域的发展方向,推动智慧道路建设运维理论、方法与关键技术的创新,《中国公路学报》编辑部联合长沙理工大学吕松涛教授、哈尔滨工业大学王大为教授、同济大学杜豫川教授、西南交通大学艾长发教授、长安大学杨旭教授、德国亚琛工业大学刘鹏飞教授,共同策划了“智慧道路建设运维理论、方法与关键技术”专栏,并向该领域的知名专家学者约稿,出版本期专栏。专栏共收到论文27篇,最终录用9篇。研究内容集中于以下2个方面:
    (1)道路运维数字化技术。主要内容包括:智慧道路设计建造运维中的数字化技术综述、车辆荷载激励下的水泥混凝土路面振动信号时频能量解析及车速估算、高精度三维路面纹理超分辨率重构及测评方法、考虑HV-AV交互的动态换道轨迹规划方法。
    (2)道路病害智能检测研究。主要内容包括:道路窨井盖-井周路面的病害处治与智慧检测监管综述、基于YOLOM算法的路面病害轻量化检测方法、考虑完整性分割的超轻量化路面裂缝检测方法、基于SmRtRock的沥青路面隐蔽性裂缝智能识别与定位、基于CNN和尺度自适应Transformer融合网络的路面裂缝分割方法。
    在此特向专栏组稿专家、审稿专家、论文作者的辛勤付出致谢,希望本期专栏的出版可以进一步推动智慧道路建设运维理论、方法与关键技术的创新与发展。《中国公路学报》将持续关注该领域的国内外最新研究进展,以期为广大专家、学者及工程技术人员提供学习、交流的平台,促进我国道路建设运维的高质量与可持续发展。由于水平及时间有限,专栏中的不足之处在所难免,恳请各位专家不吝指出。

The intelligent cooperative vehicle infrastructure System (ICVIS) enhances existing transportation services and facilitates the development of new applications by enabling flexible, cooperative, and open communication among vehicle subsystems, roadside subsystems, personal systems, and monitoring centers. Vehicle-to-vehicle and vehicle-to-infrastructure communication significantly improve the quality and reliability of information within the cooperative system, thereby optimizing driving conditions and enhancing traffic safety and efficiency. ICVIS features a complex structure, diverse functionalities, and stringent reliability requirements. Comprehensive testing and evaluation are critical to ensure its safe and efficient operation in real-world traffic scenarios. Currently, there is no systematic framework for ICVIS testing tools and evaluation methods. Most existing approaches rely on singular evaluation techniques, and the related theories, testing tools, and technologies are still in their nascent stages, with notable limitations in both breadth and depth. In this paper, existing ICVIS evaluation methods are systematically categorized based on testing objects, tools, and data sources, starting with an overview of the ICVIS system structure and its application scenarios. First, in response to the evolution of the ICVIS system structure from a “centralized” to a “cooperative” model, the hierarchical testing concept of “device level→system level→overall level” is discussed along with its specific implementation. Next, the concept of a multi-layer domain testing toolchain is introduced, categorizing and comparing existing evaluation tools. Thereafter, the advantages and disadvantages of current testing methods are outlined based on simulation software, driving simulators, closed test sites, semi-open roads, and open roads, among other evaluation tools. Furthermore, data-driven ICVIS evaluation methods, including subjective, objective, and combined subjective-objective evaluation approaches, are explored. Finally, through the analysis of three practical ICVIS assessment cases, the application process and outcomes of these evaluation methods in real-world scenarios are demonstrated. This systematic review and synthesis of ICVIS testing and evaluation methods is intended to provide valuable insights into the development and application of ICVIS technologies.

Turning flow at signalized intersections is a fundamental parameter for urban road traffic system modeling.This is essential for traffic state prediction and the traffic management strategies.However，owing to the nonlinear characteristics of traffic flows in urban road networks，turning flow often changes randomly over time.Following the advancement of connected vehicle (CV)technology，route and turning information provided by connected taxis，connected mobility vehicles，and reserved vehicles，has enabled high-resolution turning flow predictions.A key issue that remains to be addressed is how to utilize limited CV individual data to predict accurately the future turning behavior of the mixed traffic flow.This paper decouples and analyzes the stochastic fluctuation characteristics of turning flow from the longitudinal and lateral perspectives.Longitudinally，the randomness of upstream vehicles traveling downstream at different speeds was modeled by a platoon dispersion model.Laterally，a Gaussian process method with deep kernel learning was employed to predict the randomness of turning ratio，leveraging partially observed CV turning data.The proposed model was validated in two intersection scenarios that involved Ningbo and Qinzhou in China.Results indicate that the proposed model accurately predicts high-resolution changes in turning flows.Compared with the traditional cell transmission model and a prediction method based on the CV turning ratio in the literature，the proposed model improved the prediction accuracy by more than 23.81% in the Ningbo experimental scenario and by more than 26.06% in the experimental Qinzhou scenario. Even at low CV penetration rates，the proposed model can achieve more accurate turning flow predictions，demonstrating its potential practical value.

The impact of complex wind fields on the construction quality of sea-crossing cable-stayed-suspension hybrid system bridges is significant. Digital twin technology serves as an effective approach for holistic integration and simulation of bridge construction environments and processes. This study takes a large-span cable-stayed-suspension hybrid system bridge as an engineering case. By coupling the mesoscale Weather Research and Forecasting (WRF) model with computational fluid dynamics (CFD), the influence of real bridge structures was introduced into the traditional simulation framework. High-precision wind field prediction and structural mechanical analysis methods for construction stages under complex environments were established by combining data fitting and user-defined function (UDF) boundary condition loading techniques. A digital twin scenario construction method was proposed, integrating virtual geographic environments, bridge twin models, and simulation systems. A dynamic closed-loop feedback mechanism (“data acquisition-model calibration-instruction feedback-scheme adjustment”) was developed to address the limitations of traditional static models in dynamically responding to construction changes. A four-layer virtual simulation platform for bridge construction was developed based on the Cesium engine, incorporating modules for construction progress, wind field distribution, and structural mechanical responses. Through 3D GIS-based global visualization and localized detail rendering, the platform assists engineers in decision-making optimization. Case studies demonstrate that this method dynamically maps construction processes and predicts wind-structure interactions. The findings provide strong support for advancing the intelligence and informatization of bridge construction.

Horizontal swivel construction, characterized by minimal disruption to existing railways and highways, has become a critical method for constructing long-span bridges over operational railways. However, swivel construction presents challenges, such as heavy loads, elevated overturning risks, and rapid state transitions. Therefore, effective monitoring and control are essential to ensure safety, structural stability, and precision. This study clarifies the primary and secondary objectives of swivel-monitoring feedback control and establishes their logical interconnections. The “traction-rotation” process was decomposed into five monitoring components: support system, traction system, swivel posture, structural stress, and surrounding environment. Both corresponding key and general monitoring indicators were proposed. Threshold calculation methods and state evaluation criteria were established for two typical swivel support configurations: swivel hinge-centered support and combined swivel hinge-support foot systems. A data linkage model was developed based on the physical interactions between the external environment, support system, traction system, swivel posture, and structural stress, and a dynamic feedback and control mechanism was established to ensure “mobility, stability, accuracy, and speed.” The ARIMA time-series model was incorporated to facilitate the dynamic prediction of monitoring data, allowing timely control of traction and balance systems. By integrating inertial sensing, image recognition, digital twins, and PLC wireless communication technologies, a lightweight visual monitoring feedback control system was implemented using WebGL, Python, and the B/S architecture. This system was successfully implemented during the swivel construction of a cable-stayed bridge. The results indicate that the primary objectives of monitoring and controlling rotating bridges are “mobility, stability, accuracy, and speed,” which follow a pyramid-shaped logical hierarchy, with mobility forming the foundational tier. The proposed monitoring-feedback control model for the swivel construction of bridges elucidates the interrelationships between control targets, monitoring indicators, state evaluation, and system responses. The developed lightweight, visualized swivel construction monitoring feedback control system enables real-time monitoring and regulation of traction, support, posture, and velocity states, effectively addressing challenges such as data redundancy, fragmented analysis, and state assessment and control.

Reflection cracks are a typical disease of asphalt pavement with semi-rigid base, which constantly evolve under load and environmental effects, leading to secondary pavement damage. However, as cracks develop within the pavement with small dimensions and strong concealment, it is difficult to detect and track them. Therefore, these fundamental issues are not being effectively resolved. The development of the discrete element method provides a powerful simulation tool for studying the cracking behavior of asphalt pavements, whereas the advancement in the ground penetrating radar (GPR) technology provides an effective means for detecting and tracking cracks. This study utilized a discrete element simulation method and 3D GPR detection technology to explore the expansion process of reflection cracks under vehicle loads, and the deterioration process of reflection cracks under hydraulic coupling. The 2D convex hull algorithm and the circumcircle collision detection algorithm were proposed to construct a coarse aggregate template, establishing a full-scale microscopic model of asphalt pavement that met the grading requirements. The expansion law of reflection cracks under vehicle loads was studied, and the expansion of reflection cracks towards the pavement surface was mainly divided into three stages: rapid expansion, stable expansion, and failure. The deterioration process of reflection cracks after expansion to the pavement surface under hydraulic coupling was analyzed. Compared with the situation where only the load acts, under the action of hydraulic coupling, the reflection cracks form secondary cracks and expand laterally, leading to interlayer debonding, base-layer loosening, and fragmentation. It was observed that water is the primary factor in the deterioration of reflection cracks. Based on the radar detection results, a large number of actual pavement crack core analyses were conducted, calibrating the typical morphological characteristics of cracks in six grades, from “none” to “severe.” The accuracy of the discrete element simulation results was further verified, and the stage characteristics of the entire development process of reflection cracks was revealed, from “initiation,” to “expansion,” to “deterioration.” The comprehensive presentation of the entire development process of reflection cracks, including secondary diseases, facilitates more accurate identification, evaluation, and tracking of reflection cracks and their development trends.

    截至目前，中国机动车保有量超4.6亿辆、驾驶人超5.5亿、公路通车里程达550万千米，道路交通系统体量巨大，交通安全面临着极大的压力和挑战。与此同时，自动驾驶、人工智能等新技术快速发展，也为应对交通安全问题带来新的变化和机遇。为全面推进交通安全治理工作，国务院安委会发布了《“十四五”全国道路交通安全规划》，提出“科技赋能、智慧治理”的基本原则，要求提升交通安全治理现代化、信息化、智慧化水平。“交通强国五年行动计划2023~2027”进一步强调，要健全交通运输安全生产体系，推动安全生产向事前预防转型。针对中国特有的复杂交通系统，同时结合国家战略发展方向，学术界、工业界和行业主管部门都进行了大量的实践与攻关研究，在驾驶行为、安全设施优化、安全风险评估、自动驾驶交通安全等方面取得了一系列研究成果。
    为帮助政府、研究者和工业界等及时跟进道路交通安全领域理论与技术研究的最新成果，聚焦道路交通安全研究的前沿方向，促进道路交通安全理论与技术创新，《中国公路学报》编辑部联合同济大学王雪松教授（本刊副主编）、同济大学方守恩教授、公安部道路交通安全研究中心王长君研究员、公安部道路交通管理科学研究所孙正良研究员、交通运输部公路科学研究院交通安全研究中心周荣贵研究员、东南大学刘攀教授、西南交通大学闫学东教授、武汉理工大学吴超仲教授、吉林大学李世武教授、中南大学黄合来教授、长安大学付锐教授、东北林大大学裴玉龙教授、北京工业大学赵晓华教授共同策划了“道路交通安全”专栏，并邀请武汉理工大学吕能超教授、东南大学郭延永教授、长安大学王畅教授、长安大学王秋玲副教授、清华大学裴欣副研究员、西南交通大学胥川副教授、北京工业大学李佳副教授、上海海事大学王晓梦助理教授、上海理工大学丰明洁助理教授作为组稿专家，共同向该领域的知名专家、学者约稿，出版本期“道路交通安全”专栏。
    本专栏共收到论文117篇，最终录用16篇，研究内容主要集中于以下4个方面：
    （1）自动驾驶交通安全研究。研究成果包括：基于自动驾驶安全视域的交叉口右转适驾性研究、基于混合模型的地面道路车道线识别和影响因素分析、考虑路侧感知限制的交叉口冲突风险监测效能虚拟评价方法、基于多约束自适应模型预测控制的智能车路径跟踪与稳定性集成控制、通信时延下考虑行驶状态时空价值的编队安全控制。
    （2）交通行为、心理与安全分析。研究成果包括：异常驾驶人行为识别与异常度量化研究、考虑驾驶人补偿行为特征的行车安全评价方法、基于自然驾驶试验的公路螺旋隧道驾驶人心理旋转效应分析、驾驶人认知控制能力对交通风险事件响应的影响特性、考虑客货交互压迫的小汽车换道风险预测研究、融合冲突可能性和严重性的高速公路分流区极值模型构建及应用。
    （3）事故伤害评估与防护。研究成果包括：碰撞减速工况下低龄儿童脑挫伤损伤特性试验与仿真、基于真实事故重建的头部保护测评方法评估与优化、基于人体落地机制预测的人地碰撞损伤防护方法。
    （4）交通安全设施评估与优化。研究成果包括：基于分布鲁棒优化的危化品运输事故应急救援站选址分配问题研究、汽车波形梁护栏碰撞非线性理论计算方法研究。
    在此特向专栏组稿专家、审稿专家、论文作者的辛勤付出致谢！道路交通安全基础理论与应用技术的创新发展，对中国交通强国建设具有重要支撑作用。《中国公路学报》将持续关注道路交通安全领域的最新研究进展，涵盖人、车、路、管理等多方面内容，聚焦前沿理论、关键技术和工程实践。期望通过搭建学术交流与知识共享的平台，为专家、学者及工程技术人员提供有价值的研究成果和实践经验，推动中国公路交通行业向更安全、更高效、更可持续的方向发展。由于水平及时间有限，专栏中的不足之处在所难免，恳请各位专家不吝指出。

The conventional design process of bridge structures relies heavily on the experience of designers. The structural configuration proposal is repeatedly modified, resulting in low modelling efficiency. In addition, the optimal results are hard to achieve optimality. To this end, the intelligent modelling and optimization method of half-through steel box arch bridge with composite bridge deck system was proposed in this paper. First, the structural intelligent modelling method based on the human-computer collaboration was proposed. The automated layer classification method was adopted to extract the key information such as the arch rib, K brace, bridge deck system, and bridge pier abutment in the initial condition drawings. The cross-sectional information module of the bridge components was defined. The spatial information reasoning was conducted combined with the extracted information and the length and spatial coordinates of components were determined. The dividing criteria of elements and sectional fiber was established and the number of nodes and elements was determined, realizing the automatic definition of nodes and elements. The boundary condition module was defined. According to the structural parameter module, the load input module and structural output module were defined, realizing the intelligent modelling. Based on the parametric FE model, the structural intelligent optimization method was proposed. Parameters were input with combination of the intelligent modelling technology, forming the FE model to calculate the pseudo-objective function. The standard genetic algorithm, strengthen elitist genetic algorithm, and differential evolution algorithm were employed to optimize the structural cost. The proposed intelligent FE modelling and optimization method was verified with combination of the practical engineering case. The results indicate that the parametric FE model could be established by only marking the initial condition drawings using the proposed intelligent modelling method, which can significantly improve the modelling efficiency and quality. Compared with the manual structural optimization method, the cost of structural material is reduced by about 37.9% and the optimization period is reduced by about 74% using the proposed optimization method.

Due to the construction convenience and cost-effectiveness, non-excavation grouting technology is widely employed for road defect rehabilitation. As a core component of road repair, the performance of grouting material is critical to the effectiveness of filling and reinforcement. This paper systematically reviews research progress concerning the classification, applicability, composition design, key properties, repair mechanisms, and microstructural characterization of road grouting materials. It focuses on analyzing the influence mechanisms of material composition and typical dosage ranges on performance. The potential for synergistic or inhibitory effects between mineral admixtures and chemical admixtures is highlighted, emphasizing the need for research into their interaction mechanisms to mitigate adverse effects. Furthermore, the limitations of current design methodologies prioritizing single-material performance optimization are discussed. The necessity of establishing a predictive model integrating grouting material properties, subgrade characteristics, and construction processes is underscored. This model aims to achieve optimal repair outcomes by addressing multifactor coupling issues. Additionally, the repair mechanisms of different grouting materials are analyzed. In the macroscopic perspective, void filling and cementation dominate. In the microscopic perspective, repair is achieved through reactions such as pozzolanic activity, geopolymerization, grout-soil interaction, and polymer solidification. Cement-based and geopolymer-derived grouts primarily rely on cementitious reactions and ion exchange with soil/rock particles. In contrast, polymer grouts is mainly dependent on foam compaction and cementation effects.

Lateral collaborative performance is a key indicator for evaluating the service condition of prefabricated beam bridges. The timeliness and applicability of traditional methods for obtaining lateral collaborative performance evaluation indicators can still be further improved. To address this issue, this paper proposed a dynamic characterization model for the lateral collaborative performance of prefabricated beam bridges based on the analysis of vehicle-induced response mapping relationships. This method can use the health monitoring data to realize the dynamic characterization of the changes in the lateral collaborative performance of bridges. It established a vehicle-induced response mapping model for prefabricated beam bridges using a Bayesian optimization Natural Gradient Boosting (Bo-NGBoost) model, and evaluated the location and extent of lateral connection damage based on the error distribution and magnitude of the response mapping models at different positions caused by lateral connection damage. The effectiveness of the proposed method was verified. The results showed that the Bo-NGBoost had certain advantages in terms of accuracy and robustness compared with standard NGBoost and Long Short-Term Memory (LSTM) networks. Under typical working conditions, its average coefficient of determination (R²) reached 0.986, representing improvements of 4.4% and 36.4% over standard NGBoost and LSTM, respectively. In particular, in the emergency lane region with sparse effective data, the model maintained a high R² of 0.953, whereas the R² of LSTM was only 0.374. Numerical simulations were conducted to consider various combined working conditions of different pavement roughness and traffic flow density. The R² of the deflection response was higher than 0.951, and the root mean square error (RMSE) was below 0.153 mm. In addition, by analyzing the error distribution and magnitude of the response mapping models at different positions, the location of lateral connection damage can be identified and its extent evaluated. In actual monitoring scenarios, the mapped strain responses of the model have an R² exceeding 0.981 with the true data, and an RMSE below 2.381×10^-6, with the true values generally falling within the 95% confidence interval. This indicates that the method not only accurately reflects the mapping relationship of the lateral vehicle-induced response of bridges, but also can locate and evaluate lateral connection damage, thereby providing strong support for the maintenance of prefabricated beam bridges.

To address the demand for full-time monitoring of traffic loads on long-span freeway bridges, a full-time spatiotemporal traffic load identification method was developed using computer vision and data fusion technologies. First, an integrated approach combining YOLO-v8, ByteTrack, and LPRNet algorithms was established to recognize vehicle parameters and license plate characters from on-bridge roadside camera videos. A fuzzy matching algorithm based on Levenshtein distance was employed to match vehicle license plate data from roadside camera videos with electronic toll collection (ETC) records, providing a cost-effective solution for vehicle weight identification. Second, utilizing the same field-of-view and image coordinate transformation relationships between daytime and nighttime scenes, a labeled dataset was generated based on video-identified vehicle parameters captured during the daytime. This dataset was used to train an artificial neural network to predict vehicle wheel positions at night, enabling the identification of vehicle load sequences in camera-monitored regions at night. Third, a previously developed hybrid virtual-real traffic simulation approach was used to reproduce the full-time spatiotemporal distribution of traffic loads across the entire bridge deck. Additionally, a dynamic time-warping algorithm was employed to fine-tune the vehicle trajectories to align the theoretical and measured bridge responses. Finally, the proposed method was comprehensively validated using monitoring data from a cable-stayed bridge. The results indicated that the machine vision approach achieves full-time multi-target recognition accuracy of ≥92%, full-time vehicle license plate recognition and matching accuracies of 94% on average, and nighttime vehicle axle position prediction error of 0.2 m in both the transverse and longitudinal directions. The hybrid virtual-real traffic simulation approach provides average weighted vehicle longitudinal location matching errors of 0.89 m and 0.49 m, and vehicle traffic lane matching errors of 7.7% and 5.2%, for the left and right bridge parts, respectively. The deflections calculated from the identified full-time traffic loads closely match the measured static deflections caused by traffic loadings, yielding a correlation coefficient of 0.999. This work provides a cost-effective and highly efficient methodology for full-time monitoring of spatiotemporal traffic loads on long-span freeway bridges in China.

To reveal the effect of lane-changing intentions on vehicle trajectories and ensure the efficiency and safety of highway traffic, this study constructed a vehicle trajectory-prediction model while considering lane-changing intentions. The model considers the intention of a vehicle to change lanes and integrates the attention mechanism into a convolutional bidirectional long short-term attention network. To predict the interaction between vehicles and surrounding vehicles and the chronology dependence of driving trajectory data, spatial-temporal features were extracted using a convolutional neural network and bidirectional long short-term memory network. Additionally, an attention mechanism was incorporated to facilitate the model in focusing on important time series. The heading angle was used to define the vehicle lane-changing process. Lane-changing intention labels were added to the data based on the change in the heading angle and then transformed into a one-hot vector spliced with the trajectory information as the model input. Model-validation comparison experiments were performed using the NGSIM and HighD freeway datasets. The results show that, compared with the current mainstream trajectory-prediction models, the long-time domain (3-5 s) root mean square error and the average and final displacement error metrics of CI-CBLA show different degrees of reduction, and that the optimization for the HighD dataset is more significant than that of the NGSIM dataset. Considering that the vehicle lane-changing intention effectively improves the trajectory-prediction accuracy, the CI-CBLA model performs well in freeway vehicle-trajectory prediction. To verify the generalizability of the model, the Ubiquitous Traffic Eye dataset was selected for training and validation. The results show that the model can predict the trajectory of urban roads in a short-time domain.

Accurate measurement of suspender force is important in construction control, health monitoring, damage diagnosis, reinforcement and maintenance of suspension bridges. However, at present, there is still a lack of effective methods for short suspenders in the mid span area. Therefore, A laser scanning-based method for testing the suspender forces of suspension bridges, referred to as the scanning method, is proposed in this paper. Firstly, to address the difficulty of measuring the cable shape in suspension bridges, a method of quickly capturing the geometric shape of the main cable using three-dimensional laser scanning is proposed, and a precise method for calculating the cable shape based on the scanned point cloud is given. Secondly, the formula for calculating the suspender force of the suspension bridge is derived based on the measured shape of the main cable. The suspender force is solely determined by the geometry of the main cable, its specific gravity, and the horizontal force within it, independent of factors such as stiffness and boundary conditions in the suspenders. This approach effectively addresses challenges associated with accurately measuring cable forces in short suspension bridges using frequency-based methods. Then, field tests were conducted on the Guihua Bridge in Wushan to verify the accuracy of the proposed cable force calculation method and the reliability of the proposed laser scanning cable force testing method. The test results indicate that, in terms of measurement accuracy, approximately 50% of the suspender force measurements obtained using the traditional frequency method exhibit an error exceeding 10%, whereas only 2 out of the 62 suspender force test samples in the proposed method have an error exceeding 5%; in terms of measurement efficiency, it took about 330 minutes to complete a full-bridge hoisting cable test using the frequency method, while the proposed method requires only about 70 minutes. The field test efficiency of the proposed method is nearly four times higher than that of the traditional frequency method. Finally, the proposed method was applied to the Egongyan Railway Bridge in Chongqing, where the true cable forces before the cable failure were retrospectively evaluated. The evaluation results based on the historical laser scanning data show that the cable forces of the broken cable and adjacent cables of the Egongyan Railway Bridge are within 10% deviation from the designed cable forces, which proves that cable forces are not the main cause of the bridge cable failure.

The gradation of aggregate has a significant impact on the performance of cement-stabilized crushed aggregate base. An in-depth and systematic analysis of the differences in the design theories of different gradation design theories and the influences of the strength and shrinkage characteristics of has important engineering guiding significance. In this paper, the methods of computer vision, discrete element simulation, and laboratory tests were conducted to compare and verify the differences in the optimal gradation void ratio of different gradation design theories (n method, k method, fractal theory, stepwise blending method). The influence of different maximum particle sizes (31.5, 26.5, 19.0 mm) on the strength, temperature shrinkage and dry shrinkage characteristics of the material were studied. The research results show that among the four gradation design methods, the design gradation of the k method (k=0.73) can achieve the minimum void ratio and the maximum 7-day compressive strength under the same compaction degree conditions. The void ratio of cement-stabilized crushed stone materials is linearly negatively correlated with the strength. Under the same compaction degree, the maximum particle size shows a significant negative correlation with the strength. With the increase of the maximum particle size, cement-stabilized crushed aggregate materials have better anti-dry shrinkage characteristics. Below 0 ℃, the larger the maximum particle size, the smaller the temperature shrinkage. The delay time from mixing to compaction of cement-stabilized crushed stone materials has a significant impact on its strength. Within 1 hour after mixing, the 7-day strength decreases significantly with the extension of the compaction delay time. The research results reveal the influence of the maximum particle size on the performance of cement-stabilized crushed aggregate materials, and also provide certain references for the optimization of gradation design, improvement of construction technology, and prediction of shrinkage characteristics of cement-stabilized crushed aggregate base materials with dense skeletons.

To investigate the long-term structural damage mechanism of asphalt mixtures under complex humidity conditions, this study focused on asphalt mixtures with 100% replacement of natural aggregates by steel slag. Constant humidity curing environments (60%RH, 80%RH, and 95% RH) were established. A multiscale coupling analysis was conducted by combining X-ray CT image analysis and long-term dynamic modulus testing. The results showed that the pore structure of steel slag asphalt mixtures under high humidity followed a three-stage evolution: micropore formation, development of small and medium pores, and coalescence into large pores. Valid pores dominated the volumetric expansion and performance degradation process. Significant differences were observed between mixtures with different gradations in pore evolution patterns and damage responses. In the SMA-13 gradation, strong pore coalescence formed large connected networks (average volume reached 47.12 mm³). In contrast, the AC-13 gradation showed micropore development (pore count increased by 91.4%), resulting in better structural stability. Dynamic modulus testing revealed that increased humidity and extended curing time significantly reduced the mixture stiffness. The hydration reactions of free calcium oxide (f-CaO) and free magnesium oxide (f-MgO) were the primary damage-inducing factors. In the micro-macro correlation analysis, the Mantel test was introduced to quantify the relationship between the pore structure parameter matrix and the dynamic modulus response matrix. The results confirmed that porosity and average coordination number were significantly negatively correlated with the dynamic modulus. The coupling relationship between microstructural parameters and macro performance varied with gradation. This study provides a theoretical basis and data support for optimizing the performance and promoting the efficient application of steel slag asphalt mixtures in road engineering.

In order to improve the safety of bridge erecting machine construction and reduce the operation risk, this paper constructs a bridge erecting machine operation state intelligent monitoring system based on Digital Twin (DT) and Virtual Reality (VR). The system includes four main functional modules: bridge erecting machine construction process simulation, multi-source heterogeneous data fusion, structural safety analysis and monitoring, and virtual reality panoramic roaming. Firstly, Unity 3D platform is used to create a virtual construction environment, and the simulation of key construction processes is realized through C#; subsequently, the fusion of heterogeneous data from multiple sources is carried out based on CASREL model, mapping the whole life cycle of the physical entity, and real-time monitoring of the operation status of the bridge erecting machine. The system also integrates the Collider component, which monitors the risk of collision by synchronizing reality with reality through laser range sensors and serial communication; Subsequently, a dynamic finite element analysis module is developed to combine with APDL command flow to analyze the safety of the bridge erecting machine structure; Finally, immersive panoramic roaming is realized through HTC Vive to enhance the operator's interactive experience. The construction of 32 m-span simply supported box girder of high-speed railway is taken as the test object to verify the effectiveness of the system. The results show that the key construction parameters of SLJ900/32 mobile bridge erector (such as the distance of heavy load crossing, the distance of main outrigger moving forward and the height of falling and placing box girder) are all within the safe construction range, which provides systematic engineering implementation guidance for bridge erecting machine operations. Based on the fusion of heterogeneous data from multiple sources, the system realizes dynamic finite element analysis, accurately evaluates structural mechanical properties, and applies collision detection algorithms to monitor potential risks in real time and issue timely early warnings. In addition, the virtual reality panoramic navigation function enables operators to gain real-time insight into the overall situation of the construction process. which enhances the depth of knowledge of the construction environment and decision-making ability.

Welding is a common connection method widely used in the construction of steel structure bridges. Identification and tracking of weld seam is one of key technologies for achieving automated and intelligent welding. However, Steel surfaces typically have high brightness and reflectivity, which severely interfere with seam recognition. Moreover, tack welding is usually required before welding to fix the relative positions of the workpieces in steel structure. Existing seam tracking technologies struggle to handle intermittent welding at tack points, and repeated welding may lead to numerous adverse issues. Therefore, a method for intermittent fillet seam identification and tracking based on line structured light vision is proposed. Firstly, for weld seam images containing laser stripe patterns, median filtering is employed to remove speckle noise from the weld seam images. Secondly, the center points of the laser stripes are extracted based on image grayscale features. Finally, an improved RANSAC algorithm is used to fit the extracted laser stripe center points, obtaining the feature lines and their intersections (i.e., weld seam feature points), thereby achieving precise discrimination between continuous and intermittent weld seams. Experimental results show that compared with the robot's taught path, the maximum error of the weld seam path extracted by this method is 1.56 mm, with an average processing time of 0.17 ms per frame. This study provides a feasible solution for robot jump welding of intermittent fillet welds in steel structures.

The volume expansion characteristics and water-damage risks of steel slag restrict its engineering applications as a potential substitute aggregate for asphalt pavements. To address the challenge of predicting the volume expansion and water stability of steel slag asphalt mixtures, this study established a machine-learning prediction model that incorporated multiple factors. Based on immersion expansion tests and 300 water stability tests covering variables-such as asphalt type, steel slag content, f-CaO content, gradation, and environmental conditions, a backpropagation neural network model was developed based on water-induced volume expansion and a CatBoost prediction model was optimized using Bayesian optimization and cross-validation. SHapley Additive exPlanations (SHAP) theory was employed to analyze the feature importance and parameter sensitivity that affect water stability. The results indicate that the volume expansion of the steel slag asphalt mixtures was significantly correlated with the gradation composition, f-CaO content, and immersion time. The CatBoost model achieved the highest prediction accuracy for the residual stability and tensile strength ratio (TSR) and effectively reflected the prediction error, with R² >0.997 and MSE<0.344 5. Among the material factors influencing water stability, the f-CaO content of the steel slag coarse aggregate (mean SHAP values: 2.05, 1.21, 1.17, and 4.62, 1.44, and 0.77, respectively) was the most crucial, followed by the asphalt type (0.84 and 0.82), steel slag content (0.36 and 0.32), and asphalt content (0.12 and 0.38). There was an interactive effect between the feature combinations of steel slag f-CaO content-asphalt content and f-CaO content-steel slag content on water stability. To satisfy water stability requirements, the steel slag content in the surface layer of the asphalt pavement should not exceed 75%. Additionally, the f-CaO content thresholds for steel slag with particle sizes of 2.36, 4.75, and 9.5 mm should be controlled within 2.0%, 2.25%, and 2.0%, respectively. This study provides theoretical support for controlling steel slag expansion and predicting water stability, thereby promoting the resourceful utilization of steel slag in asphalt pavements.

To improve the alignment accuracy and adjustment efficiency of the multi-round pre-assembly of long-span arch bridges, this study developed a manufacturing alignment control method based on digital pre-assembly. This method includes high-precision multistation point-cloud registration of large arch rib segments, calculation of the pre-assembly target alignment, and fine-tuning of the pre-assembly alignment. First, a rotating triangular pyramid target was developed for precise point-cloud registration, and a corresponding registration algorithm based on Random Sample Consensus was presented. Second, the sources of alignment error in the multi-round pre-assembly were analyzed, and a calculation method for determining the pre-assembly target alignment of arch rib segments based on digital pre-assembly was proposed. This yielded an optimal pre-assembly target alignment and an accurate continuation of the overall arch rib alignment. Finally, for the point-cloud model of the optimal pre-assembly target alignment, an alignment adjustment method based on the rapid extraction of geometric parameters was introduced, enabling one-time high-precision positioning of arch rib segments. The proposed method was fully applied to the arch rib segment pre-assembly of the Deyu Wujiang Bridge. The results showed that the newly developed registration device and method achieve high-precision multistation point-cloud registration of large arch rib segments, improving the registration accuracy from the millimeter level to the submillimeter level. The alignment restoration error of the common segment in adjacent rounds is the fundamental reason for distortion in the multi-round arch rib alignment continuity. Hence, to ensure the overall alignment connection accuracy in the multi-round pre-assembly, the alignment difference of the common segment in the subsequent round should remain consistent with that in the previous round. By applying the proposed pre-assembly alignment calculation and adjustment method, the arch rib pre-assembly alignment accuracy improves by 55% compared with the code limits, and the actual assembly time for the arch ribs is reduced by 7 days from the original 28 days, enhancing pre-assembly manufacturing efficiency by 25%. The manufacturing alignment control method proposed in this study, which leverages digital pre-assembly, can be extended to other bridge types that employ high-precision connecting components such as flanges or bolts between segments.

To address the common technical challenge of controlling convergence deformation at the arch feet of initial supports in soft rock tunnels, this study explores the occurrence conditions of convergence deformation at the arch feet of the initial supports in construction via the bench cut method from the perspective of structural load and constraint conditions. This method proposes the use of prestressed feet-lock cables to actively control the convergence deformation at the initial support arch feet. The supporting mechanical and deformation control effects of prestressed foot-lock cables on the arch feet of the initial supports in soft rock tunnels are studied by combining theoretical calculations and field tests. Moreover, a construction method for the prestressed feet-lock cable is proposed. The results show that the convergence deformation at the arch feet caused by the horizontal load of the surrounding rock is larger than the expansion deformation at the arch feet caused by the vertical load of the surrounding rock. This is the fundamental reason for the convergence deformation at the arch feet of the initial support. The traditional feet-lock pipe (bolt) has a limited restraint effect on the arch feet of the initial support and cannot play an axial anchoring role. The initial support has an insufficient horizontal restraining effect on the arch feet, which explains the significant convergence deformation of the arch feet. After setting up a 10 m-long end resin anchored prestressed feet-lock cable with a diameter of 21.8 mm and applying a high pretension force (i.e., a designed pretension force of at least 300 kN), the deformation rate at the arch feet of the initial support significantly decreased. The working principle is that the prestressed feet-lock cable plays a strong, fast, and active control role in the convergence deformation at the arch feet of the initial support while forming a “prestressed anchor cable retaining wall” with a steel rib and shotcrete initial support, which plays an active reinforcement role in the rock surrounding the tunnel. For the severe collapse of the tunnel face and vault after tunnel excavation, the “anchor first and then support” scheme is difficult to implement, considering its high support cost, long process time, and inability to use prestressed anchor cables in large quantities, the prestressed feet-lock cable support provides an alternative with high promotion and application value.

To address the challenges of low automation and inefficient iteration in the virtual pre-assembly of thin-walled steel beam structures, this study explored a virtual pre-assembly surface-line-point layer extraction approach and a two-stage virtual assembly technique (coarse and fine) tailored for thin-walled steel beams. Utilizing 3D laser scanning technology to capture point cloud data from the steel beams, the study implemented intelligent virtual pre-assembly through a sequence of voxel preprocessing, feature extraction, virtual coarse assembly, and virtual fine assembly. A robust method for the layer-by-layer extraction of thin-walled steel beam features was proposed, incorporating the RANSAC algorithm, alpha shape algorithm, and the getLineIntersectionPoint function. Principal component analysis (PCA) and the iterative closest point (ICP) algorithm were employed for the coarse and fine stages of virtual assembly, respectively. The results demonstrate that: ① The surface-line-point layer-by-layer extraction method efficiently extracts point clouds of complex beam assembly surfaces, meeting the accuracy requirements for the virtual pre-assembly of thin-walled steel beams in engineering applications; ② Utilizing the feature point data extracted through the surface-line-point method, the application of PCA for coarse assembly and ICP for fine assembly provides a robust and reliable assembly strategy for steel beams; and ③ The implementation of this method in a rapid reconstruction project involving multiple-main-beam I-beams demonstrates that the overall average and root mean square errors validate the capability of the method to achieve accurate, comprehensive, and high-precision virtual pre-assembly of thin-walled steel beams. The proposed method satisfies practical engineering assembly requirements and serves as a reference for the virtual pre-assembly of similar bridge types.

    桥梁，作为交通生命线网络的枢纽工程，其抗震性能的优劣直接决定着震后“生命通道”的畅通与否，关乎区域经济韧性与灾后恢复能力。然而，传统抗震分析方法长期依赖经验公式与简化模型，在精准捕捉地震动的强随机性、结构响应的强非线性以及复杂系统耦合作用等多重挑战面前，显得力有不逮。值得关注的是，近年来人工智能技术的爆发式发展，为破解这些深层次难题开辟了全新路径。通过将强大的数据驱动能力与深度的物理机理认知相融合，人工智能正推动桥梁抗震分析实现从“经验依赖”到“智能驱动”的范式跃迁——人工智能算法能够从海量数据中挖掘隐藏规律，深度学习模型可以高精度模拟复杂的非线性破坏模式与灾变过程，智能优化技术则可赋能抗震优化方案的快速生成与性能评估。
    为积极推动人工智能技术在桥梁抗震领域的深度应用与创新突破，及时反映该交叉学科的最新研究进展与发展趋势，助力智能抗震理论与技术的快速发展，《中国公路学报》编辑部联合北京工业大学韩强教授（我刊副主编）、北京工业大学温佳年高级研究员（我刊青年编委）、西南交通大学魏凯教授（我刊青年编委）、合肥工业大学钟剑副教授、同济大学王晓伟副教授共同策划了“人工智能在桥梁抗震中的应用”专栏，并邀请哈尔滨工业大学郭安薪教授、东南大学/江苏大学王景全教授（我刊副主编）、香港理工大学毕凯明教授、中南大学国巍教授、同济大学管仲国教授、西南交通大学赵灿晖教授、天津大学李宁教授、河北工业大学王东升教授、北京交通大学江辉教授、湖南大学伍隋文教授、长安大学周敉教授、防灾科技学院孙治国教授作为组稿专家，共同向该领域的知名专家学者约稿，出版本期“人工智能在桥梁抗震中的应用”专栏。
    本专栏共录用相关论文7篇，研究内容主要集中于以下3个方面：
    （1）已取得成果与应用前景分析。主要内容为：人工智能在桥梁抗震中的研究进展与展望。所报道内容系统梳理了人工智能技术在桥梁抗震五大核心场景（地震动合成、响应预测、损伤评估、韧性优化、网络级分析）中的应用。
    （2）桥梁结构响应预测与易损性分析。主要内容包括：中小跨径常规体系连续梁桥地震反应的机器学习预测及可解释性比较研究，基于LSTM的近断层桥梁地震响应和易损性快速预测方法，基于LSTM的近断层桥梁地震响应和易损性快速预测方法。所报道内容体现了人工智能技术在桥梁结构响应预测和易损性分析过程中的高效性与精确性。
    （3）桥梁韧性评估与抗震优化。主要内容包括：考虑构件修复次序的桥梁震后功能恢复模型及韧性评估，基于多目标群体智能算法的悬索桥减震优化方法研究，基于复Morlet小波梁式桥低阶振型识别新方法研究。所报道内容展现了人工智能技术在解决最优问题方面的优势。
    值此专栏出版之际，谨向为本专栏倾注心血的组稿专家、审稿专家以及所有论文作者致以最诚挚的谢意！正是诸位同仁的鼎力支持与卓越贡献，方使本期聚焦“人工智能赋能桥梁抗震”的专题得以成功呈现。我们殷切期望，本专栏的出版能够有效推动桥梁工程智能抗震新技术与新应用的蓬勃发展。《中国公路学报》将一如既往地密切关注人工智能与土木工程交叉融合领域，特别是智能抗震方向的国内外前沿动态与突破性成果，致力于为广大研究者、工程师搭建高水平的学术交流与知识共享平台，共同促进我国桥梁工程向更智能、更安全、更可持续的方向迈进。人工智能在复杂工程系统中的应用尚处于发展阶段，专栏中难免存在疏漏与不足之处，恳请各位专家、读者不吝批评指正，以帮助我们不断进步。

To evaluate the force state of cable-clamped bolts in long-distance suspension bridges accurately, a detection method based on longitudinal ultrasonic waves is proposed, considering the uneven distribution of the axial stress of the bolt. The axial stress of the bolt was simulated using a finite element software to show the pattern of the axial stress distribution of the cable-clamped bolt. Considering the uneven axial stress distribution, the relationship between the travel time of the ultrasonic longitudinal wave and bolt stress was derived based on the acoustoelastic effect, and an adaptive function of the particle swarm algorithm was established. Experiments were designed to calibrate the intrinsic parameters of the bolts. Ultrasonic longitudinal echo signals were collected, and the signal-to-noise ratio was improved using a wavelet transform. An autocorrelation algorithm was applied to calculate the travel time of the longitudinal waves. A particle swarm algorithm was used to determine the optimal values of the average axial stress and effective clamping length of the bolt to obtain the true value of the screw axial force. The results show that under different axial forces, the effective clamping length is deep into the inner side of the nut, which is approximately 7% of the thickness of the nut and is consistent with the actual force situation of the bolt. As the actual force of the bolt increases, the absolute and relative errors of the measured bolt force decrease. When the axial force is greater than 70% of the required bolt axial force, the absolute error of the detection result is less than 20 kN, and the relative error is less than 7.22%. The proposed method can effectively measure the bolt clamping length and the axial force, indicating that it has promising potential for detecting the bolt force of suspension bridge cable clamps.

Intersection optimization is a crucial measure for enhancing urban road traffic efficiency. Existing methods rely on analytical expressions of intersection performance evaluation metrics at the aggregate level, making it difficult to integrate optimization involving different design patterns. To improve the flexibility and scalability of intersection design optimization models, we propose a simulation-based coordinated optimization model for intersection layouts and signal control. The optimization problem employs an iterative process to develop a lane-based intersection optimization design model that generates decision variables for intersection geometric layout and signal control. Simulations by Simulation of Urban MObility (SUMO) were used as an evaluation model to provide performance metrics for the corresponding design solutions. During the optimization process, the decision variables were transferred to the simulation model through an interface and automatically adjusted to the geometric configuration and signal timing parameters in the simulation environment. A simulation was performed to derive an optimization objective value, which was used as a fitness function for the particle swarm optimization algorithm to update the decision variables of the current scheme. The updated variables were then returned to the intersection optimization model. Thus, the optimal allocation of spatial and temporal resources for the intersection was obtained through continuous iterations. The optimization model was decoupled from the analytical expression of the evaluation indicators by embedding the SUMO simulation module into the optimization framework. Through case studies in high- and low-traffic scenarios, the feasibility and optimization benefits of the model were compared and verified. Compared to traditional optimization algorithms, this method reduces delays by 5.7% and 21% under low- and high-traffic conditions. The experimental results show that the optimal geometric design and signal timing scheme derived from this method are more effective in improving traffic efficiency and reducing the environmental burden under high-flow conditions, and that the simulation-based optimization model for geometric design and signal control co-optimization is free from reliance on analytical models for operational evaluation, which makes it easy to extend the model to various design modes.

Uncertainty analysis of structural performance indicators based on the statistical characteristics of measured influencing parameters is crucial for the accurate evaluation of prestressed concrete structures. However, due to the complex coupling of multidimensional uncertain influencing parameters in structures, existing evaluation methods have difficulty fully revealing the propagation law between measured random distributions of influencing parameters and the uncertainty of performance indicators. To address this issue, This paper proposes a probability density estimation method for concrete structures based on uncertainty propagation theory, incorporating the measured distribution characteristics of effective prestress. Firstly, the dimensionality reduction integration method is introduced to decouple the multidimensional prestress system into a linear superposition of single-variable subsystems. Combined with the measured prestress distribution characteristics, an analytical calculation method for the first four statistical moments of performance indicators is proposed. Secondly, using the statistical moments of structural response as constraints and based on the kernel density maximum entropy principle, the number of kernel functions is incorporated into the optimization framework. An objective function that comprehensively considers maximum entropy and statistical moment errors is proposed, establishing an improved kernel density maximum entropy method for estimating the probability density of prestressed concrete responses. The results show that, compared with traditional Monte Carlo simulations, the errors of the first to fourth-order moments of the statistical moment analytical solution are 0.01%, 0.11%, 0.29%, and 0.57%, respectively, which can accurately characterize the main stochastic features of the performance indicators. The improved kernel density maximum entropy method overcomes the limitation of traditional methods that focus solely on maximum entropy constraints, significantly enhancing the fitting accuracy of performance indicator probability density. Through engineering validation, the theoretical calculation error of the mid-span camber value of a prefabricated T-beam is controlled within 5%, and the measured camber values fall within the theoretical envelope region defined by the 95% confidence interval of elastic modulus. The results demonstrate that the proposed method exhibits good agreement with experimental data, effectively quantifies the impact of prestress uncertainty on structural performance, and provides theoretical and practical support for the performance evaluation of prestressed concrete structures.

Lane detection models based on a single feature layer often have difficulty in simultaneously improving the precision of lane recognition and positioning, especially in challenging scenarios with lane occlusions and poor illumination. This severely hinders the progress of environmental understanding in assisted or autonomous driving systems. To this end, a lane detection model based on multi-feature scale-wise regression was proposed. Lane features containing global and local contextual information were obtained using a joint attention mechanism to enhance lane representation and improve model adaptability in scenarios with lane occlusions and poor illumination. The lanes were then regressed from larger to smaller-scale feature layers with line proposal information to enhance the lane positioning precision. In addition, to improve the quality and generalization of line proposals, they were extended as model-trainable parameters for optimization. Besides, a local slope angle was introduced in the IOU formula to further constrain lane shapes. Finally, a multitask learning mechanism was adopted to adaptively learn the weights of various losses to alleviate the complexity of model parameter tuning. The test results demonstrate that the proposed model can effectively detect lanes in complex road scenarios. Furthermore, the proposed model achieves F1 scores of 80.04% on the CULane dataset with a detection rate of 128 fp·s^-1, better than those of CondLaneNet, GANet, and Lane2Seq when the backbone network is DLA34.

Electric vehicle wireless charging systems on urban roads have been widely studied because of their convenience and high efficiency. An improved DD coil structure is proposed to address the issue of high-efficiency energy transmission when different types of primary-and secondary-side coils are matched and when the primary-and secondary-side coils are offset and deflected. This structure employs semi-circular coils instead of rectangular coils. It adds an orthogonal Q coil based on a semi-circular DD coil. Based on the basic structure of the improved DD coil, an embedded core structure was designed, and the improved DD coil was optimized from the perspective of a lightweight coupling mechanism. The COMSOL finite element simulation software and MATLAB were utilized to conduct joint simulations, and the interoperability and anti-bias performance of the unipolar DD, improved DD, and DDQ coils were compared and analyzed. A system model using an improved DD coil was established, and the energy transmission characteristics of the system during the rotation and migration of the secondary coil were analyzed. The results demonstrated that, compared with the traditional unipolar DD coil, the mutual inductance of the improved DD coil is increased by over 70%. Compared to the DDQ coil, especially when the secondary-side coil is circular, the mutual-inductance value of the improved DD coil is increased by more than 4%, and the material consumption is reduced by 2%. The maximum output power of the electric vehicle wireless charging system using the improved DD coil can reach 3.46 kW, with an efficiency of 94.73%. During the process of 0°-45° rotation deviation of the secondary coil, the system output power fluctuation was only 0.12%. The efficiency fluctuation was only 0.01%, indicating that the system had high energy transmission stability.

To expand the application of waste tire rubber in the road sector, this study systematically investigated the synergistic optimization mechanism of waste tire rubber modification and blending technology on the road performance of chip seals. Rubber particles were treated via oxidative amination and incorporated into the chip seal system using an equal volume replacement method. Contact angle measurements, scanning electron microscopy microscopic morphology analysis, and Fourier-transform infrared chemical characterization were employed to verify improvement in the rubber-asphalt interfacial compatibility achieved by the modification treatment. The anti-skid and anti-stripping performances of the chip seals were evaluated using a wet-track abrasion test, hand-spreading sand method, and a standard abrasion test. The vibration damping and noise reduction characteristics of rubber chip seals were characterized using indoor noise and tire vertical vibration attenuation tests. The results indicated that sequential treatment of the rubber particle surface with a 2% NaClO+2% CH₄N₂O solution effectively reduced the aggregate loss rate of the chip seal and enhanced surface roughness. With an increase in the nominal maximum size of the crushed stone and rubber particle content, the pavement texture depth increased, surface texture characteristics were enhanced, and vibration damping and noise reduction performance were improved. Based on the combined results of the wear and noise tests, the aggregate detachment rate initially increased rapidly with the number of wear cycles before gradually stabilizing. As the particle size of the rubber granules increased, the vibration-damping performance of the chip seal improved; however, the resistance to detachment decreased. For rubber particle sizes ranging from 1.18 to 2.36 mm and an incorporation rate of 40%, the aggregate detachment rate was lower than that using other incorporation rates, with a reduction of 28%-55%. Considering noise reduction, vibration damping effects, and road performance, it is recommended to use crushed stone of size 4.75-7.1 mm combined with rubber particles of size 1.18-2.36 mm for chip seals, with a rubber particle content ranging from 35% to 40%.

This study aims to establish a dynamic theoretical characteristic model for Air springs with an additional chamber (ASAC) to elucidate the mechanism of the amplitude-frequency dependency, dynamic stiffness resonance peaks and amplitude-independent fixed-point characteristics. Firstly, the variation of dynamic stiffness under different excitation amplitudes and frequencies was experimentally studied. An analytical dynamic characteristic model of ASAC was established and nondimensionalized, based on thermodynamic principles. Then, the proposed model was verified by experiments, followed by derivations of closed-form expressions for the frequency and amplitude of the amplitude-independent point and the resonance peaks. Finally, the impact of the air chamber stiffness ratio and damping coefficients on the amplitude-independent point was analyzed. The results show that the nonlinearity of dynamic characteristics is caused by the amplitude-frequency dependency of air damping; the resonance of the air mass at a specific frequency causes resonance peaks in dynamic stiffness, whose magnitude is related to the pipe length, diameter, and excitation amplitude. Besides, the frequency of the amplitude-independent point is solely dependent on the stiffness ratio of the two air chambers, while the amplitude is equal to the stiffness of the main chamber. The model can characterize amplitude and frequency dependence and is applicable to common road excitation frequencies and amplitude ranges. Some design suggestions for ASAC were also provided.

The pipe jacking method, a novel construction technique for metro cross-passages, is characterized by a short construction period and high safety, making it increasingly popular. To investigate the structural mechanical response of the main tunnel segments at the launching end during cross-passage pipe jacking, an in-situ monitoring study was conducted based on the Zhengzhou Metro Line 12 cross-passage pipe jacking project. The monitoring results were used to analyze the variation patterns and spatial distribution of strain increments in the main tunnel segments during the pipe jacking process. The findings show that, during the pre-support stage, the circumferential strain increment in the cutting ring, half-cutting ring, and adjacent rings exhibits a “vertical elliptical” distribution. As the jacking force increases, this distribution gradually shifts to a “horizontal elliptical” pattern. During the breakthrough stage at the launching end, the strain increment at the 90° position of the half-cutting ring changes significantly. The circumferential strain drops sharply from 88×10^-6 in the pre-support stage to -235×10^-6, while the longitudinal strain increases abruptly from -8.0×10^-6 to 70.0×10^-6. The rates of change are -12.9×10^-6·(100 kN)^-1 and 3.1×10^-6·(100 kN)^-1, respectively. When the jacking force reaches 5 000 kN, the circumferential additional tensile strain at the 210° and 330° positions of the cutting ring is relatively large, at 132×10^-6 and 100×10^-6, respectively, while the circumferential additional compressive strain at the 90° position of the half-cutting ring reaches a maximum value of 297×10^-6. During the construction process, the jacking force significantly affects the circumferential strain of the tunnel segments, while its impact on the longitudinal strain is relatively small. The change in circumferential strain increments in the cutting ring, half-cutting ring, and adjacent ring at the 180° position of the arch bottom lags behind the change at the 0° position of the arch crown. The circumferential strain increments at the 90° position of the half-cutting ring and the adjacent ring differ significantly, while at the 270° position, the circumferential strain increments are relatively similar. These results provide valuable guidance for the construction of subway cross-passages using the pipe jacking method.

In recent years, with the development of water conservation projects, a large number of dredged silt foundations have been generated in coastal cities in China. Dredged silt has poor engineering properties and requires treatment prior to its use in construction. The widely used traditional vacuum-preloading method has poor effectiveness in treating dredged silt foundations owing to issues such as clogging. The improved horizontal combined vertical drainage plate (PHD & PVD) method has problems such as an insufficient reinforcement effect and clogging in the last duration. To further improve the reinforcement effect of the PHD and PHD & PVD vacuum preloading method, this study proposes a new method in which air is periodically introduced during the vacuum preloading process to promote drainage. It is called the PHD and PHD & PVD vacuum preloading-aeration method. Four sets of indoor model experiments were conducted using self-made glass model boxes. Parameters such as drainage volume, pore water pressure, and surface settlement were monitored during the tests, and the soil moisture content and shear strength were tested after treatment. The effects of different ventilation durations on the reinforcement of dredged silt using the PHD and PHD & PVD vacuum preloading-aeration method were compared and analyzed. The experimental results show that, compared with the conventional PHD and PHD & PVD vacuum preloading, the intermittent ventilation PHD and PHD & PVD vacuum preloading group increased the discharge of dredged silt by 6.60% to 14.03%, the settlement increased by 9.30% to 25.66%, and the overall average shear strength of the cross plate increased by 10.08% to 29.74%. The best consolidation effect was achieved when air was introduced for 1 h/day. Compared to the conventional PHD and PHD & PVD vacuum preloading method, the clogging phenomenon near the drain can be effectively reduced, and the consolidation efficiency is improved. In addition, micro-experiments confirmed that intermittent ventilation can effectively alleviate the problem of particle accumulation around the drain and improve the reinforcement effect of dredged silt. This study verified the effectiveness of the PHD and PHD & PVD vacuum preloading-aeration method, which will aid in the future development and application of dredging silt treatment technologies.

In recent years, the amount of filter mudcake, a type of waste soil material, has increased. However, the resource utilization rate remains generally low. The massive accumulation of filter mudcakes has caused problems such as land occupation, resource waste, and ecological environmental pollution, which are not conducive to the construction of waste-free cities. This study aimed to recycle and reuse filter mudcakes as subgrade filling materials. Through a series of studies, including curing agent formulation tests, strength performance tests, dry-wet cycle resistance tests, water stability tests, and centrifuge modeling tests, the formulation of curing agents suitable for improving filter mudcake, the road performance of recycled material derived from filter mudcake, and the long-term performance maintenance of filter mudcake recycled materials for subgrade filling were gradually explored. The results show that the curing agent combination of “carbide slag + blast furnace slag + rice husk ash” demonstrates the optimal improvement effect on filter mudcake, with the optimal dosage ratio being 7%, 7%, and 4%, respectively. Using this ratio, a block-type filter mudcake recycled material was prepared. At a curing age of 14 days, the bearing ratio and unconfined compressive strength exceeded the performance requirements of traditional slag-filling materials and common subgrade fillers specified in the standards. Moreover, it met the requirements for dry-wet cycles and prolonged water immersion. Under hypergravity centrifugal conditions simulating equivalent water immersion (up to approximately 5.7 years) and considering vehicle loads on the subgrade, the strength, water absorption, and settlement of the subgrade constructed with the filter mudcake recycled material comply with relevant standard requirements, demonstrating excellent long-term durability and strong potential for engineering applications in subgrade construction.

The hierarchical control of the speed planning and energy management are effective solutions for energy conservation and emission reduction in hybrid electric vehicles (HEVs) under urban transportation connected conditions. Considering the gearshift and engine fuel consumption of the powertrain system, optimization in the speed planning level is challenging. Adding gearshift integer optimization to energy management under the conditions of speed trajectory tracking is also a challenge. Therefore, in this study, a two-level gearshift optimization control method at both the speed planning and energy management levels with different time domains is proposed for connected HEVs. The upper-level speed-planning control unit transfers the speed target to a lower-level energy-management controller. The lower-level controller further optimizes the energy allocation factor between the engine, motor, and gearshift. The simulation test results of real-time urban traffic road scenes show that, compared to traditional control methods in the upper and lower layers, the proposed two-level gearshift optimization method for speed planning and energy management can reduce fuel consumption by 17.33%.