Abstract:To address the limitations of general large language models （LLMs） in the field of civil engineering due to a lack of specialized knowledge， this study proposes a large knowledge model specifically designed for civil engineering， named CivilGPT. The development of CivilGPT follows a multi-step technical approach， including data preprocessing， the construction of a domain-specific knowledge graph， the generation and optimization of automated datasets， staged pre-training and fine-tuning， and alignment with engineering tasks to ensure that the model can accurately express and reason within the field of civil engineering. Additionally， this study introduces a standardized evaluation framework， Civil-Bench， based on civil engineering qualification exams. Civil-Bench encompasses 13 categories of professional engineering exam questions， including 14，823 objective questions and 269 subjective questions. Testing across 15 domestic and international language models demonstrates that CivilGPT exhibits significant advantages in civil engineering knowledge comprehension， reasoning ability， and solving complex problems. The outcomes of this research lay a technical foundation for the intelligent advancement of the civil engineering field and provide valuable insights for the development of models in other specialized domains.

Abstract:Strengthening of steel structures is usually carried out with loads on the original structure （i.e.， under preload）. In order to clarify the influence of the preload level of the structure on the strengthening effect of steel members with bolted strengthening， 168 finite element models of bi-directional eccentrically compressed H-shaped steel members with bolted strengthening are calculated and analyzed. Parameters affecting the strengthening effect are investigated， including the preload level， slenderness ratio and eccentricity. It is shown that， for the strengthening of eccentrically compressed members， the preload has an unfavorable effect on the strengthening effect， weakening the strengthening effect within 14%， the slenderness ratio also affects the strengthening effect， whereas the eccentricity has an insignificant effect on the strengthening effect. The proposed formula for predicting the strengthening effect by considering the slenderness ratio and the preload level can be used to calculate the ultimate bearing capacity of bi-directional eccentrically compressed H-shaped steel columns with bolted strengthening under preload.

Abstract:Precast concrete segmental girder bridges play a vital role in urban infrastructure construction. To address the problem of reinforcement discontinuity at joints in conventional designs， this study proposes a new joint scheme with steel rebars through joints， ensuring continuity of longitudinal reinforcement. The primary objective is to improve the cracking resistance and long-term durability of segmental girders. A series of load tests on simply supported segmental beams， combined with numerical analysis， were conducted to compare the performance of the proposed joint with that of conventional joints. Key parameters such as cracking load， ultimate load capacity， deformation behavior， and failure modes were systematically evaluated. The results demonstrate that incorporating steel rebars through joints significantly enhances the cracking resistance of segmental beams. The proposed two-dimensional finite element model effectively simulates the entire loading process of segmental girders with both continuous and conventional joints. In addition， a simplified formula for predicting ultimate load capacity is developed， offering improved accuracy in structural performance evaluation. These findings suggest that the use of steel rebars through joints provides a promising design approach to improve the engineering performance and practical applicability of precast concrete segmental girders.

Abstract:In order to revise the deformation performance index system of steel reinforced concrete（SRC） shear walls of Guangdong provincial standard Specification for Performance-based Seismic Design of Reinforced Concrete Building Structure（DBJ/T 15-151-2019）， 12 steel reinforced concrete （SRC ） shear walls were designed and tested under low-cyclic loading based on current standard. Seismic behaviors， such as failure pattern， energy dissipation capacity， deformation capacity and displacement ductility were investigated. According to the test results， two performance state division methods based on skeleton curve and failure phenomena were proposed respectively and a comparison between the two methods was made. The deformation performance index limits obtained by these two methods were consistent and could be characterized each other， which verified the rationality of the performance state division method based on skeleton curves. Based on this， the relationship among deformation， extent of damage and bearing capacity of the component was established. Based on the low-cyclic loading test results of 104 specimens， including 12 specimens in this paper and 92 specimens collected from literature， failure mode classification was revised and the regression formulas for the key performance points （θ₁， θ₅， θ₆） of SRC shear wall were established. Finally， the established regression formulas for the key performance points were adjusted by exceedance probability based on ASCE/SEI 41－13， and the deformation performance index limits of SRC shear walls were revised.

Abstract:A prefabricated facade integrated enclosure system based on steel structure building is illustrated，which adopts an integrated design method by taking both the appearance effect of the external facade and the internal decoration materials into consideration.The system is highly composite， integrated and intensive， reflecting the green concept and technical aesthetics. Through cross-subjects design， co-research and demonstration of several professional institutions， typical parts 1：1 solid model proofing， physical testing and analysis methods， this study gives the specific implementation and node construction， and physical verification.

Abstract:To solve the research difficulties regarding the temperature field and mechanical properties degradation of lead rubber bearings under heating effects of the lead core， an evaluation method for heating effect of lead core was proposed by physical tests and numerical simulations. The law of temperature field transfer under free boundary and quasi-real boundary was studied in physical tests. The temperature variation of the lead core and mechanical degradation of bearings under heating effects were predicted by the numerical simulation under real boundary conditions. The results show that the influence of the first shape factor on the heating and cooling rate was significant. The insulation effect of the covering layer was remarkable， resulting in a slow cooling rate. The boundary consisting of steel plate and concrete block has a great influence on the temperature field of the bearing， and emphasis should be placed on simulation of the boundary. The numerical simulation reproduces the physical test process， and the simulation of the bearing and the steel-concrete boundary was experimentally verified. Finally， the temperature variation of the lead core and mechanical degradation of the bearing with the real boundary was predictable. The proposed method has strong applicability， which can provide a reference for evaluating heating effects of lead core in various types of rubber bearing.

Abstract:To address the issue of the structural performance evaluation of shield tunnels give less consideration to the correlation of indexes， the association rule theory optimization analytic network process （ANP） which combined with fuzzy mathematics is introduced and the evaluation model is established. The data of each index were collected and processed， and Apriori algorithm was used to mine the association rules to obtain the association relationships and the strength of association among the indexes， according to which the judgment matrix of ANP was adjusted and the structural performance evaluation of the shield tunnel was carried out. Through the correlation analysis of the evolution of the indexes， the optimization of the maintenance strategy of the shield tunnel is proposed. From the theoretical point of view， the proposed method is more in line with the actual working conditions， and the engineering application is conducted by taking the south tunnel of East Yan'an Road as an example， and the results show that： there is only a 3.9% difference between the proposed evaluation method and the results obtained by the fuzzy hierarchical analysis method （FAHP）， which proves the feasibility of this method； The results of association rule mining indicate that the over-limit cumulative settlement is the source factor of structural performance degradation， which should be realized through limiting excessive settlement and index evolutionary antecedent to control the structural performance degradation. The research results can provide technical support for the safe service of shield tunnels.

Abstract:This paper focuses on summarizing the historical context， the latest advances and future trends of block theory and methods. Firstly， the essential concept has evolved from the traditional pyramid， the centroid sliding pyramid and the generalized joint pyramid to the integrated entrance angle based on contact theory. Secondly， the whole-space stereographic projection method， the vector method， the stereo analytical method and the whole-space block cutting and contact method have accordingly been proposed. Thirdly， the basic assumption of translational failure of rigid bodies is still retained， while the other two basic assumptions， i.e.， infinite straight persistent structural plane and geometric constraints of face–to–face contacts， have been weakened. Lastly， stability analysis of exposed convex blocks under the geometric constraints of face to face contacts has been extended to progressive failure analysis of whole-spatial arbitrarily-shaped blocks under the geometric constraints of all general contacts. Moreover， the authors present the contact theory and method based on the integrated entrance angle for block removability analysis to further promote the substantial development of block theory.

Abstract:Non-signalized intersections serve as critical nodes in urban mobility systems. In the absence of traffic signal control， these intersections often experience mixed flows of motorized and non-motorized vehicles， increasing the likelihood of vehicle–bicycle conflicts. Such conflicts frequently lead to abrupt braking and evasive maneuvers by drivers， ultimately reducing the efficiency of urban mobility. Therefore， it is essential to analyze the conflict risks at non-signalized intersections to inform intersection design improvements and enhance overall urban mobility performance. This study proposes a trajectory extraction framework based on computer vision techniques to analyze traffic interactions at non-signalized intersections. Utilizing over 2，000 minutes of video footage， we examine various traffic behaviors and conflict processes among different types of road users. A random-parameter ordered probit model is employed to capture the coupling relationships between driver behavior， environmental factors， and the severity of traffic conflicts. The results reveal that the sequence of intrusion during conflict events—combined with the type of road user involved—is highly correlated with the severity of the conflict. Notably， the involvement of electric bicycles significantly increases the risk level. The proposed trajectory extraction framework offers a valuable tool for researchers studying traffic conflict dynamics， while the findings provide theoretical guidance for enhancing urban traffic efficiency and informing practical traffic management strategies.

Abstract:With express and local trains in mixed operation mode， appropriate overtaking maneuvers can enhance the line carrying capacity of suburban rail transit systems； however， not all stations are equipped with overtaking facilities. To quantitatively analyze the relationship between line carrying capacity and multiple factors including station layout combinations， a line carrying capacity calculation method is proposed. Based on predefined parameters—station layouts， train stopping plans， train running times in sections， trains interval times， and express/local train scheduling priorities—and with the conditions for overtaking clearly defined， the method sequentially introduces express/local trains， calculates arrival/departure or passing times of each train at each station， dynamically adjusts timetables of overtaken trains， and iterates until line carrying capacity saturation is achieved. Case studies with a fixed number of intermediate stations equipped with receiving-departure tracks demonstrate that station layout combinations universally affect line carrying capacity under varying express/local train ratios， but their impacts differ in magnitude. The results emphasize the necessity of incorporating station layout constraints into line carrying capacity calculations.

Abstract:This study aims at the difficult problem of calculating and comparing the total road travel time costs of rail transit network planning schemes. Focusing on the differential road OD pairs of the comparison schemes， we conduct traffic surveys on the predicted urban road section through which the differential OD pairs pass， and revise the travel time functions of the road sections so that the travel time of the differential OD pairs is accurately calculated. Given the interactive nature of vehicle travel， a computational model based on Wardrop First Principle was constructed using the surveyed road sections to directly estimate the impact times of differential OD pairs on the road network. By calculating and comparing the sum of differential OD pair travel time and impact time between the comparison schemes， an accurate comparison of total road network travel time was completed.We selected two rail transit schemes for a particular city to verify this method， the results showed that the comparative analysis of this method is reliable and easy to implement in practical work.

Abstract:Road traffic pollution has become a major source of pollution in cities， and street-facing communities are the most vulnerable to traffic pollution. This paper aims to use unmanned aerial vehicle（UAV） to monitor the three-dimensional distribution of air pollutants in urban street-facing communities， summarize their three-dimensional distribution rules， and explain the causes of three-dimensional distribution of pollutants in urban street-facing communities by analyzing different traffic conditions， different built environments and different meteorological conditions. This paper finds that large diesel vehicles have a great impact on the emission intensity of road pollution sources， and 12% of large diesel vehicles can contribute 55% of PM₁ emissions. At the same time， according to the different land use conditions on both sides of the road， below 90m altitude， the horizontal impact range of PM₁ is 100m to 350m， and within the height of 90m to 300m， the horizontal influence range of PM₁ can reach more than 500m.

Abstract:The microfluidic model addresses the challenge of understanding soil-groundwater dynamics by enabling real-time observation of transport， transformation， and adsorption processes at the microscopic scale. This model offers crucial theoretical and technical support for groundwater pollution management. Emphasized how innovative microfluidic devices can provide insights into soil-groundwater geophysical-chemical processes at the microscopic level， covering aspects like transport， transformation， adsorption， desorption， leaching， and precipitation. The article discusses limitations in microfluidic technology， such as scale and complexity， and anticipates future research directions， including integrating mainstream and novel characterization techniques to enhance fidelity and versatility. Microfluidic technology advancing microscopic， controllable， and in situ studies of multimedia interfacial processes in soil-groundwater contaminants.

Abstract:This study focuses on the operational principles of pre-coated dynamic membranes， exploring the influence of various factors， including the support membrane pore size and material， membrane-forming material properties and particle size， as well as filtration pressure and crossflow velocity on filtration performance. The study reviews recent advancements in the application of pre-coated dynamic membranes in surface water， industrial wastewater， and municipal wastewater treatment. Furthermore， the economic viability of pre-coated dynamic membrane technology is evaluated. The study also highlights the challenges faced in actual engineering applications and outlines future research directions， aiming to offer a theoretical foundation and technical support for pre-coated dynamic membrane applications in water treatment.

Abstract:This paper considers the pricing of zero-coupon bonds with an upper bound on the dividends on the firm’s assets in the presence of a possible credit rating migration of the bond issuer. Firms were classified into two credit rating statuses， high and low based on its assets， and the change in the value of the firms’ assets satisfied the geometric Brownian motion with the dividend rate， and the change in the value of the firms’ assets satisfied the ordinary geometric Brownian motion when the number of cumulative dividends reached the expected value， taking into account that there was a plan for expected dividends on the firms’ assets. By calculating the value of corporate bonds under this model， the risk of credit rating migration was then assessed， while hedging was used to derive the conditions at the credit rating migration boundary. An innovative risk assessment model including credit rating migration was established， and then the eigenline implicit format difference method was used to solve it， and the numerical solution of the company’s asset value was obtained and its financial significance was analyzed.

Abstract:This research examines the valuation of companies incorporating dividend payments and default risk under the Merton jump-diffusion framework. By integrating Girsanov’s measure transformation theory with probabilistic methodologies， we derive analytical solutions for three pivotal metrics： the present enterprise value， survival probability， and discounted dividend value. Numerical experiments are conducted to validate the analytical solutions， with results benchmarked against Monte Carlo simulations， confirming both the accuracy and computational efficiency of the proposed approach. Furthermore， the methodology demonstrates extensibility to the pricing of exotic structured derivatives， particularly double-barrier snowball options and shark-fin instruments requiring dual-threshold modeling.

Abstract:Integrating innovative data preprocessing methods and machine learning algorithms， an intelligent prediction model is constructed based on the Breast cancer Wisconsin diagnostic dataset. Firstly， the feature recursive elimination method based on light gradient boosting machine （LightGBM） model is used for feature selection. Secondly， the integrated sampling combined with adaptive synthetic sampling （ADASYN） oversampling and one-sided selection （OSS） undersampling is used to deal with data imbalance， and a balanced training data set is obtained. Finally， with multilayer perception （MLP）， LightGBM and categorical boosting （CatBoost） as the base learner and logistic regression model as the meta-learner， an intelligent diagnosis model based on Stacking ensembled learning is constructed. It is evaluated by 5 folds cross-validation and classification prediction indicators such as accuracy， sensitivity， and area under receiver operating characteristic curve. The experimental results show that the proposed model can achieve a prediction accuracy of 98.2%， and has stable and excellent classification prediction performance， which can provide strong decision support for clinical diagnosis of breast cancer.

Abstract:Aiming at the defense attacker game with limited resources， a multi-period game theoretic model based on reinforcement learning for resource allocation is proposed. The defender allocates resources in deploying false targets and strengthening the genuine one within multiple periods. Multiple attackers， on the other hand， distribute resources in identifying false targets and attack the genuine one among multiple targets. In each period， each player bases their decision on the expected utility of the genuine target as the reward. The Q-learning method， one of the reinforcement learning algorithms， is adopted in the game theoretic model， exploring the best resource allocation strategy over the entire planning horizon. An illustrative example was studied to demonstrate the effectiveness of the proposed model and algorithm， which can support the decision making in the resource allocation problems.