Popularization of various portable mobile devices with positioning function produces massive spatial-temporal trajectory data of moving objects, and the huge data scale has brought severe challenges to trajectory data management and analysis. Therefore, a spatial-temporal trajectory data compression algorithm based on smart phone sensors is proposed. The algorithm recognizes the turning behavior and speed change behavior of the vehicle by monitoring and analyzing the data change law of the linear acceleration sensor and direction sensor built in the smartphone, and requests GPS sensor positioning to record the corresponding trajectory feature points according to the pattern recognition result, so as to realize real-time online compression of vehicle trajectory. The proposed algorithm is compared with the representative trajectory compression algorithms characterized by feature point extraction. The results indicate that it is slightly weaker than the representative trajectory compression algorithms in compression accuracy, its spatial-temporal distance deviation is 0.4 meters more than that of the online NOPW (normalopening window) algorithm on average, and its spatial distance deviation is 0.6 meters more that of the online NOPW algorithm on average. The real-time performance of the proposed algorithm is strong, and the feature points can be obtained in the current second, the calculation efficiency of the proposed algorithm is high, and the calculation time consumption is reduced by about 30% compared with the DP (douglas-peucker) algorithm, which reduces the amount of network transmission data; It only requests positioning and sampling at key feature points, the compression results is able to adapt to changes in road conditions to some extent, thus it reduces the storage space of the mobile phone, and decreases the power consumption of the mobile phone.

In order to solve the difficulty of acquiring tagged corpus, a Chinese word sense disambiguation method is proposed on the basis of semi-supervised learning convolutional neural networks (CNN). Firstly, the word, part of speech and semantic category are extracted as discriminative features, which are acquired from 2 word units on the both left and right adjacent to ambiguous word. Word vector tool is used to denote discriminative features as vector. Secondly, tagged corpus is preprocessed to obtain initialized clustering centers and thresholds. At the same time, it is used to train convolutional neural networks. The optimized CNN is applied for determining the semantic categories of ambiguous words in the untagged corpus. Corpus with high confidence that meets threshold conditions is selected into the training corpus. The above process is repeated until the training corpus is no longer expanded. In the last, SemEval-2007: Task#5 is used as the tagged corpus, and the unannotated corpus from Harbin Institute of Technology is used as the untagged corpus. Experimental results show that the proposed method improve disambiguation accuracy of CNN by 3.1%.

Existing data fusion visualization methods have high requirements on data accuracy, complex matching process, redundant organization mode of traditional point cloud, poor dynamics for complex data and low index efficiency, and thus it is difficult to efficiently visualize multi-source data in network environment. In view of the above problems, a fast fusion visualization method of panorama and point cloud is proposed for network lightweight application. Key technologies are discussed such as two-dimensional image mapping, fast matching of three-dimensional point cloud data, optimized organization of irregular octree-based point cloud and dynamic scheduling at multiple levels of detail (LOD). A fusion-scene cross-modal interaction mechanism is designed to realize fast fusion visualization of panorama and point cloud data. Finally, a prototype system is constructed and experiments are carried out. The results show that the method reduces the time in the matching and fusion of panoramic and point cloud, and the number of frames is stable above 40 frames/s, which can support the efficient visualization and interactive analysis of fusion scenes.

Testing model is an important basis to create the test cases of safety critical function of high-speed railway signal system. To solve the problem that the characteristics of signal system are not fully represented in the modeling process of safety critical function test of high-speed railway signal system, the modeling theory of timed finite state machine (TFSM) and test case generation method are proposed. The characteristics in the test modeling of high-speed railway signal system is analyzed, and the modeling requirements are put forward. Then, based on the theory of finite state machine, a modeling method of TFSM is proposed by utilizing functional logic and clock constraints, and its formal definition is established with Z notation. Further, the TFSM is transformed into timed automata, which can prove the consistency between them, and test cases are automatically generated with timed automata based testing theory. The switch conversion function of the computer interlocking system is used as an example to build the testing model of TFSM and generate the test cases. The result shows that compared with the test cases generated by timed automata, in terms of functional logic, the test cases generated by TFSM can fully cover those generated by timed automata, and add 16 more test cases with clock constraints, showing that TFSM can meet the testing modeling requirements for the safety critical function of high-speed railway signal system.

Due to the rapid development of electric vehicles (EVs), the number of EVs has surged. Connecting EVs to micro-grid in charge-discharge scheduling become an effective way to reduce the negative impact of large-scale EVs on the power grid. Thus, given the source-charge characteristics of the EVs, with the access of EVs to DC side in micro AC/DC hybrid power grid, the two-phase robust scheduling model is built to settle the uncertainty in micro-source output and load of the microgrid system and find most economical solution in extreme situations. The model uses a box uncertainty set to describe the uncertainty, and flexibly adjusts the conservatism of the model through uncertain budget. Based on the operation constraints of each unit in the system, the objective function of minimum cost is established, and the model is transformed into a mixed integer linear programming model by the strong duality theory and BIG-M method. Finally, the model is iterated to obtain the optimal solution through the column and constraint generation algorithm. The simulation results of an example simulation show that reasonably utilizing the source-load characteristics of EVs can effectively reduce the microgrid daily operation cost. In the case of 50 EVs in parallel operation, the operation cost in the randomly charging mode is higher than that in the orderly charging and discharging mode by 1069.7 Yuan. In addition, under the restriction of power commutation, with the increase in the number of EV access, the operation cost also shows a trend of declining and then rising. In terms of real-time adjustment costs, the robust scheduling model has favorable economic benefits.

To study the DC operation parameters of the 10 kV AC cable in different laying modes and DC operation topologies, according to the example of the three-core cross-linked polyethylene (XLPE) cable widely used in 10 kV AC distribution network, the temperature field, flow field and electric field coupling simulation models are established through finite element simulation software. The temperature field, the flow field, the steady electric field and the transient electric field are analyzed respectively, in the cases of the cable laid in the soil, pipeline and trench (three laying modes) and operating in the bipole, monopole and triple (three-wire bipole structure-high voltage direct current, TWBS-HVDC) DC operation modes (three DC operation topologies). The results show that for the same laying mode, the TWBS-HVDC has the highest DC ampacity, while the monopole mode has the lowest DC ampacity. The DC voltage levels of the 10 kV AC cable in the three laying modes and three DC operation topologies are 10 kV and leave a certain voltage margin. The cable in the trench and monopole mode has the highest maximum DC transmission power, i.e., 13.2 MW, whereas the cable in the pipeline and bipole mode has the lowest maximum DC transmission power i.e., 8.7 MW. The maximum transmission power of the AC cable will have a significant increase after the cable is converted into DC operation.

Empty wagon allocation is the key link in railway transportation. The scheme of empty wagon allocation should have robustness to avoid the influence of uncertain factors. Based on the fixed technical operation time of stations and travel time between stations, the connection time relationships between arrival and departure trainsin supply stations, and between departure trainsin supply stations and in demand stations are identified. To maximize the revenue of empty wagon allocation, an empty wagon allocation model involving vehicle type substitution is established. On this basis, the fluctuation rate is introduced to describe the uncertainty due to the technical operation time of stations and travel time between stations, and the lower limit of fluctuation is set to adjust the model robustness. A robust optimization model is established for empty wagon allocation under uncertain conditions. Combined with model properties, a fast algorithm for solving the robust optimization model is designed in terms of the change of traffic flow relationship, and the nonlinear problem is transformed into a robust equivalent model that is easy to solve. The results show the empty wagon allocation scheme can accurately obtain the empty wagon flow and vehicle type substitution, and the fluctuation rate and lower limit of uncertain factors affect the revenue of the scheme. Under absolute robustness, compared with the certain model, the revenue of the scheme is decreased by 16.2%, 12.1% and 28.1%, respectively, due to the uncertain factors, including the travel time between stations, technical operation time of supply stations and technical operation time of demand stations.

Reasonable shunting operation plans can greatly improve the operation efficiency and maintenance capacity of the electric multiple unit (EMU) depots. On the premise of determinate layout of tracks and each function sections, the effect of EMU marshalling status and occupation on the shunting operartion plan is analyzed. With the objective of reducing track occupancy times during shunting operations, and taking EMU operation time, shunting operation sequence and track type as constraints, an optimization model for shunting operation plan considering train position occupancy is established, a simulated annealing algorithm based on the feasible route generation and the priority exchange for shunting operations is proposed. Finally, an example of shunting operation plan at the EMU depot is used to verify the effectiveness of the model and algorithm. The results show that the proposed approach yields a shunting operation plan in a short time and with a small number of track occupancy. The number of EMU transfers is reduced to 34, and 80% of the EMUs have realized the quick storage operation.

For automatic and safe parking of buses in public transportation environment, the safety constraints such as the degree of fitness with the platform geometry, the influence of vehicle occupancy, and the driving continuity are explored. The nonlinear constraint optimization function is used to calculate the continuous curvature of the arrival trajectory. A trajectory planning model for bus autonomous parking is built, and the Sigmoid function, cubic spline interpolation, arc-tangent function and improved arc-tangent function are used to simulate and analyze the parking trajectory of the bus under the restriction of the preceding vehicle. The experimental results show that the average disparity of 0.41 of the improved arc-tangent curve is much smaller than that of Sigmoid curve, cubic spline interpolation, and arc-tangent curve, which are 1.49, 1.051, and 0.52, respectively. Therefore, when there is vehicle occupancy in station, for the path planning of bus autonomous parking, the improved arc-tangent function is better than the other three methods. It has smoother trajectory curve, which is convenient for passengers to get on and off, and improves the vehicle operation efficiency .

In order to analyze the causes of rail corrugation in curve section of heavy haul railway, a refined wheel-rail three-dimensional contact model is established based on the friction self-excited vibration theory. The influences of different stiffness, friction coefficient and superelevation on the unstable friction self-excited vibration of wheel-rail system were discussed, and the internal causes of single rail corrugation were revealed, and the transmission and evolution process of single rail corrugation were analyzed by means of the explicit dynamic model. The results show that the mismatch between superelevation and actual running speed is the main cause of unilateral rail corrugation. The corrugation of the inner rail will lead to the instability of the wheel/rail system, and the vibration will be transmitted to the outer rail, which will induce corrugation of the rails on both sides of the small radius curve section. By properly improving the vertical and lateral stiffness of fasteners and controlling the wheel-rail friction coefficient below 0.4, the tendency of unstable vibration of wheel-rail system can be effectively reduced, and the development of ripples can be restrained.

To study the characteristics of wheelset elastic vibration and their effect on the dynamic performance of heavy-haul freight wagon, a railway wagon with a 30 t axle load is considered as the research object in this work. The running stability and the curving performance of vehicle are investigated in a comparative manner. First, the mathematical modeling method of elastic body based on multi-body dynamics is presented. Then, a finite element model of wheelset’s flexible body is established, and the elastic vibration modes are analyzed. The flexible wheelset is integrated in to the multi-rigid-body system, which then forms the rigid-flexible coupling dynamic model of heavy-haul wagon. Finally, for the multi-rigid body and rigid-flexible coupling modeling methods used for the wagon, the differences in wheelset vibration responses, running stability, and curving performance characteristics determined by these methods for a heavy-haul wagon were compared when using the main-line track irregularity with short-wave irregularity as the excitation source. The research results indicate that the deformation of elastic wheelset can alleviate the wheel-rail impact and also weaken the wheel-rail rigid restraining effect when compared with rigid wheelset. Therefore, the vibration amplitude for the flexible wheelset is reduced and the vehicle’s nonlinear critical speed is decreases by 9%. In sharp curves, the wheel-rail lateral force of flexible wheelset is reduced by 13.7%. The influence of wheelset’s elastic vibration on the dynamic performance of the heavy-haul wagon thus can not be ignored.

The effect of skew tail-wind on driving safety cannot be ignored. To investigate the effect of skew tail-wind, the aerodynamic characteristics of the vehicle under the action of skew tail-wind were tested by using a wind tunnel test device of the moving vehicle model, aiming at the vehicle and bridge models with the scale ratio of 1/20. The effects of wind speed, wind direction and wind barrier on the aerodynamic characteristic of moving vehicle were discussed. The results show that the five-component coefficients of moving vehicle well agree with different wind speeds. The lateral drag coefficient, lift coefficient and nodding moment coefficient decrease with the increase of the yaw angle. The skew tail-wind has a significantly effect on the lift coefficients when the yaw angle is small. The wind barrier makes the aerodynamic characteristics of the vehicle close to 0, but it obviously transforms the change law of the aerodynamic coefficient of the vehicle with the yaw angle. After the wind barrier is installed, the change rate of drag coefficient of vehicle is affected by the yaw angle, vehicle speed and wind speed.

The effect of the wheel load could cause the self-vibration effect of high frequency for the track slab. To analyze the fatigue characteristic of CRTS Ⅱ track slab under high-frequency load and the effect of the self-vibration of the track slab on its fatigue life, the fatigue characteristics of the track slab under the influence of this self-vibration during the interval in the wheelset was explored on the basis of existing fatigue damage criteria. The fatigue life of the track slab was predicted with reference to the effect of the de-bonding length, and the results obtained were compared with the results obtained when only the number of load actions was considered. Results show that, the possibility of damage to the track slab caused by the train wheel load is reduced when the track structure is intact. The bottom of the track slab would crack before resonance triggering occurred if the train speed was 360 km/h. When the track structure is intact, the self-vibration effect in the track slab caused by the wheel load has its greatest impact on the fatigue damage to the track slab, and the wheel load produces approximately 1.8 times the equivalent fatigue load to the track slab. The effect of the self-vibration effect in the track slab on the fatigue damage could be ignored when the slab de-bonding length is more than twice the distance between sleepers. When de-bonding length of the track slab is more than 3.2 times the distance between sleepers, it becomes difficult for the ballastless track on-site to maintain its expected 60-year service life.

Aiming at investigating the correlation between the parameters of a vehicle’s secondary suspension structure and the uneven wear of the running wheel tires, the goal of optimization was to improve the uneven wear performance of the running wheel tires and the design variables are the vehicle’s secondary suspension parameters. An indirect method of evaluating the running wheel wear was established from the analysis of a finite element model and by combining with evaluation indexes of the running wheel wear characteristics. This method minimizes the running wheel tire wear by exploring the effects of the slip angle and roll angle of the running wheel tires and their combination on the wear and uneven wear of the running wheel tires, and by optimizing the vehicle’s secondary suspension parameters. The results show that before optimization, the slip angles of the front and rear running wheels of the bogie (where the former bogie is taken as an example) are 0.5°, 0.3°, −0.4°, and −0.2°, respectively, and that the slip angles of the rear running wheel tires are 0.2°, 0.2°, −0.2°, and −0.2°, respectively. Among the vehicle structural parameters, the lateral stiffness of the secondary suspension and the vertical stiffness of the secondary suspension strongly affect the uneven wear of the running wheel tires, which can be reduced to a certain extent by selecting appropriate parameter values.

To explore the failure mechanism of Jinsha River Bridge at Zhubalong, a 1∶20 scaled numerical simulation model of the bridge was built with the measured field data in flood discharge, while the influence of air trapped in the diaphragm chamber at the bottom of T-beam is analyzed. The characteristics of flood force, flow field and failure mechanism of superstructure of the simply supported T-beam bridge under the effect of flood are studied in depth. Results show that when the submergence ratio is less than 1.342, the horizontal force coefficient increase with the ratio; when it is beyond 1.342, the coefficient decreases with the ratio. Before the backwater crosses the guardrail, the vertical force and overturning moment coefficient gradually increase with the submergence ratio. When it crosses the guardrail, the vertical force and overturning moment coefficient decrease sharply. When the submergence ratio is close to 1.000, the risk of the superstructure failure is the greatest. During the failure process of the bridge, horizontal movements occur first, and then there is a high probability that floating and overturning occur. The beam movement leads to the destruction of the link stopper, and part of the beam overturns in the falling. Solid railings contribute up to 31% of the horizontal force, and the contribution of entrapped air to the vertical force is up to 18%. Solid railings and entrapped air are important causes for flood damage or accelerating the bridge damage.

To deeply explore the seismic damage mechanism and evaluate the ductile capacity of hollow RC columns, square and rectangular hollow specimens with different aspect ratios, longitudinal rebar ratios, and stirrup ratios were tested under quasi-static loading. The crack distribution and damage state were observed. The hysteretic behavior, curvature and displacement ductility were analyzed. According to the experimental data in literature, applicability of the existing plastic hinge formulas to estimate the tip displacement capacity of hollow columns was discussed. The results show that all hollow specimens suffer flexural failure with ductility factor over 5.0, resulting in excellent seismic performance. The shear performance of hollow columns is weaker than that of solid members with the same external size and aspect ratio. The lateral bearing capacity and ultimate displacement can be properly improved by increasing longitudinal rebar ratio. However, the longitudinal steel and stirrup amount exert vague effects on the displacement ductility factor of hollow columns with low axial load ratio. The plastic hinge length of hollow columns increases with increased aspect ratio, longitudinal rebar strength/diameter, or axial load ratio, but decreases with the increment in concrete strength. The influence of stirrup on the plastic hinge is not obvious. The relative errors between the test and calculation of Mander, Sun and JRA models are less than 5%. Mander supplies the best result, which can be used to evaluate the equivalent plastic hinge length of a hollow column. The Paulay-Priestley model, adopted in many codes, overestimates the length of the plastic hinge of hollow members, leading to unsafe seismic design.

The influence of axial force and shear effect on the longitudinal deformation of shield tunnel structures under different loads is investigated by Fourier series method. Taking into account the foundation reaction caused by shear deformation, the governing differential equation of bending deformation is derived, and a formula is proposed to calculate the shear deformation. The correctness of the Fourier series solution is verified by comparison with analytical solutions. Through comparative calculation, the main work is focused on addressing the influence of cross section, end support, type of load, ratio of length to height, and elastic foundation on the shear deformation of shield tunnel, the influence of shear rigidity on bending deformation, shear deformation and internal forces, and the influence of axial load exerted by installation of segment rings on bending deformation and bending moment. Results show that the shear deformation of shield tunnel accounts for more than 20% of the total deformation, which is caused by the form of annular section and low shear rigidity of shield tunnel. When shear deformation is counted, the whole deformation gets larger, but the bending deformation and bending moment are smaller than those without consideration of shear deformation. As shear rigidity decreases, shear deformation increases, and its proportion to the entire deformation also increases. In our case studies, when the shear rigidity decreases from 8×10⁶ to 1×10⁶ kN, the whole deformation goes up by 15.7%; the bending deformation and bending moment go down by 11.7% and 17.1%, respectively; the shear deformation increases 5.77 times; and the percentage of shear deformation to the whole deformation increases from 4.64% to 27.17%. Besides, the bending deformation and bending moment are also increased by installation-caused axial load, but the influence range is limited, often less than 2% in deflection and less than 3% in bending moment.

Landslide disasters occur frequently in loess area. Landslides, especially super deep landslides, have a significant impact on the stress and deformation of existing tunnel structures. The deformation characteristics and mechanical responses of tunnel-landslide systems are extremely complex and have been the focus of academic and engineering researchers. Based on a railway tunnel project in a super deep landslide geological disaster, a three-dimensional numerical model of the "super deep loess slope-sliding zone-tunnel" system was established using FLAC3D. The local strength reduction method based on displacement mutation was used to simulate the critical state of slope instability, and variation laws of the mechanical and deformation characteristics of the existing lining structure induced by landslide were analyzed for cases of different relative positions between sliding zones and tunnels. In addition, combined with field measurements and structural damage conditions, the causes of engineering accidents were preliminarily analyzed. Numerical simulations indicate that when the sliding zone is above the tunnel, the surrounding rock exerts a significant pulling effect on the tunnel, and the maximum horizontal displacement of 27.83 mm occurs at the wall foot; when the sliding zone is below the tunnel, the tunnel structure has an obvious overall lateral translation, resulting in a maximum horizontal displacement of 185.61 mm at the wall foot. The most dangerous case is when the tunnel is above the sliding surface. The field measurements indicate that the tunnel has an overall lateral translation perpendicular to the longitudinal axis. The maximum translation (105.35 mm) at the wall foot is smaller than the case when the sliding zone locates below the tunnel. Results also show that the selected tunnel project was built in a loess (silt)-bedrock landslide, which is still in a creeping state and has not yet reached the sliding critical state.

The rock mass in a fault zone is a special type of rock mass that has attracted wide attention in the field of civil engineering. Triaxial tests are conducted on the rock mass of a concealed fault zone of a hydropower project on the Nujiang river to obtain the mechanical parameters of the rock mass in the concealed fault zone under confining pressure. Moreover, a systematic method (rock mass sampling in a fault zone, pretreatment of the rock mass prior to test, application of initial confining pressure, triaxial test, and evaluation of mechanical parameters) is summarized. This method avoids the effects of the disintegration of the fault zone and the hydraulic fluid on the samples when the triaxial test is conducted. In this way, the obtained mechanical parameters will be closer to the actual mechanical parameters of the rock mass in the fault zone in a natural state. The results show that the deformation modulus of five samples range from 449 MPa to 921 MPa, with the average value being 665 MPa. The cohesion ranges from 0.04 MPa to 0.22 MPa, with the average value being 0.13 MPa and the standard value being 0.12 MPa at a guarantee rate of 80%. The internal friction coefficient ranges from 0.32 to 0.61, with the average value being 0.46 and the standard value being 0.35 at a guarantee rate of 80%. In addition, the test values of cohesion are more discrete than those of the internal friction coefficient, indicating that the cohesion of the fault zone rock mass is more deeply affected by the material structure, fault gouge content, and cementation degree.

Peak ground-motion acceleration (PGA) directly reflects ground shaking intensity with the advantages in conceptual clarity and engineering application. Prediction of PGA, along with site condition correction, needs to be handled in site-specific seismic design. In this work, 32 stations and the earthquake data recorded are collected from the KiK-net strong motion array, Japan, so as to propose a simplified method to predict PGA amplification factors (f_PGA) corrected by site condition. Linear and quadratic empirical formulae of the f_PGA possibility model parameters with respect to the combinations of site characteristic parameters are obtained via regression analysis. Using the f_PGA model, ground surface PGA values, corrected by site conditions, can be predicted under different exceedance probability levels. Data analysis indicates that f_PGA is variable but can be simulated by a log-normally distributed function, of which mean and standard deviation are less correlated with a single site characteristic parameter but have good correlation with the linear combinations of the site characteristic parameters. The reasonable agreement between the predictions and records testifies the feasibility of the proposed method.

Non-parametric spectral inversion is the key technology to study the seismic source, propagation path and site effects. The reliability of spectral inversion was responsible for exactly understanding seismic source physics and effectively predicting seismic hazards. In this work, the non-parametric spectral inversion of ground motions during the 2016—2017 central Italy seismic sequence was used to systematically evaluate how the reliability of spectral inversion was dominantly controlled by several decisive factors, i.e., constraints for trade-offs between source and site terms, selection of one- or two-step non-parametric inversion methods, and potential effects of trade-offs among the parameters describing source spectra on their estimations. Results show the crucial effects of reference site on the inversion reliability. Multiple trials for the source spectra and source parameters were necessary for selecting the appropriate reference sites, which were preliminarily regarded as rock sites based on the surface geological information, the horizontal-to-vertical spectral ratios, etc. The usage of either one- or two-step method has obvious effects on high-frequency propagation path attenuation at far fields. However, negligible effects occurred on source parameter estimates. The one-step method was preferred because the path attenuation given by two-step method could be interfered by the site term. In order to eliminate the effects of trade-offs between corner frequency (f_c) and high-frequency decay parameter, low- and intermediate-frequency source spectra were suggested to estimate the moment magnitude and corner frequency. It was found that trade-offs between the moment magnitude and corner frequency have weak effects on their estimations for the cases of f_c > 0.5Hz. Predefined the moment magnitude can be included for the reliable corner frequency estimates for those large events with small corner frequency.

In order to study the mechanism of plant roots in preventing slope soil cracking, a self-developed uniaxial tensile test device was used to quantitatively study the strengthening effect of roots on the tensile strength of soil and analyze the tensile failure mechanism of root-soil composite. The test device consists of a loading module, a digital control module, a data acquisition module, and a sample making mold, which can accurately obtain the entire process displacement-tensile stress relationship curve and the tensile strength of test material. A series of direct tensile tests were carried out on the root-soil complex of the shrub plant Lespedeza bicolor under different root contents by the developed tensile device. The results show that the displacement-tensile stress curve of pure soil is unimodal, while the curve of root-soil complex is bimodal. The tensile strength of root-soil composite increases nonlinearly with the increase of root content, which is 28.01%−142.15% higher than that of plain soil. The tensile strength of root-soil complex can be estimated by the calculation model proposed in this paper. When the root content is 1−3, the average error is 12.12%. The tensile failure process of the root-soil complex can be divided into four phases: stress increase phase, soil failure phase, secondary increase stage, and root slip phase. The root system mainly contributes to the first and third stages.

Geogrid has been widely adopted in ballasted railways to improve the bearing capacity of ballast and its resistance to lateral deformation. The mechanical properties of geogrid-reinforced ballast aggregate are seriously affected by contaminants of coals falling from the moving hauls or clay fines from ballast layer and subgrade. A series of large-scale direct shear tests of geogrid-reinforced ballast under different normal pressures and various fouling levels were performed using the two common fouling substances of ballasted railway, i.e., clays and coals. The effects of the two fouling materials on the shear strength, peak friction angle, vertical dilatant displacement, and peak dilatant angle of the geogrid-reinforced ballast were compared and analyzed, and the mechanical mechanism that contributes to the different performances was also explored. The results showed that the presence of fouling materials decreases the shear strength and peak friction angle of geogrid-reinforced ballast, but reduces its vertical dilatant displacement and peak dilatant angle. Compared with that fouled by clay fines, the ballast fouled by coals exhibits a lower shear strength and peak friction angle, and a larger vertical dilatant displacement and peak dilatant angle, indicating increasingly adverse impacts on the mechanical performances of geogrid-reinforced ballast.

The sound absorption performance of porous asphalt concrete has an important impact on reducing tire/pavement noise. The sound absorption coefficients of porous asphalt concrete (PAC), stone matrix asphalt (SMA-13) and a dense graded asphalt concrete (AC-13) are tested adopting the standing wave ratio method at one-third octave frequencies. The effects of several concrete properties are investigated; i.e., the grade distribution type, void content of the asphalt mixture, specimen thickness and surface texture. It is found that the sound absorption coefficients of PAC with a higher void content are much larger than those of SMA-13 and AC-13, and the sound absorption spectrum first increases and then decreases with the noise frequency increasing. The linear expressions of the sound absorption coefficient with the connected void content are proposed for asphalt mixture. A higher void content and larger maximum nominal particle size result in a larger peak and average value of the sound absorption coefficient (i.e., better noise reduction performance) for the PAC, with the peak value of the sound absorption spectrum gradually moving to a higher frequency. As the specimen thickness decreases, the average value of the sound absorption coefficient decreases for the PAC, and there is no obvious change in the peak value of the sound absorption coefficient whose corresponding frequency shifts to a high frequency gradually. For SMA-13 and AC-13 with similar void percentages, the peak and average value of the sound absorption coefficient of the former are slightly larger than those of the latter. For the same PAC specimen, the average value of the sound absorption coefficient measured using a rough-surface receiving incident sound wave is about 13.9% larger than that measured using a smooth-surface receiving sound wave, which indicates that the surface texture is an important factor affecting the sound absorption performance of the PAC. In summary, increases in the void content, maximum nominal particle size and thickness contribute to improving the sound absorption performance of the PAC. The two former factors are more beneficial to the absorption of high frequency noise, while the last factor is beneficial for the low frequency.

To solve the instability problem of newly designed double-limb cold-formed C-shaped steel yurts under vertical loads such as ice and snow, the stability performance of a double-limb cold-formed C-shaped steel portal frame with an in-plane toono was investigated. The failure modes and bearing capacity of this type of portal frame were analyzed through testing and numerical simulations using three parameters: the slope of the inclined beam, the height-to-span ratio, and the toono diameter. The results show that increases in the inclined beam slope and the toono diameter can increase the load bearing capacity of the rigid frame, and an increase in the high span ratio of the rigid frame then reduces the rigid frame’s load bearing capacity. Following consideration of the influence of the inclined beam slope and the lateral displacement of the column top, the work in this paper indicates that the inclined beam slope should be designed to be within 20°, where the column effective length coefficient is calculated to be safe according to The Technical Code for Steel Structure of Light-Weight Building with Gabled Frames; this also meets the engineering design requirements.

In order to reduce the damage degree of masonry infills and improve their deformation recoverability under earthquake, based on the failure mechanism of infill walls of reinforced concrete frames, a structural measure of installing horizontal slip layers (HSLs) in masonry infills constructed by hollow bricks is put forward. Two groups of full-scale HSL-specimens were manufactured and tested under quasi-static loads to study the seismic behaviors of HSL-frame structures. A calculation method for measuring the damage degree of infills is proposed and their damage degree after adding HSLs are compared. Results show that after HSLs are added in the hollow brick infills, the diagonal bracing effect of the infill walls is weakened. With the HSLs, the failure mode of the infill walls changes from diagonal failure to sliding failure, the energy dissipation capacity of frames increases, and the damage degree of the infills and their residual deformation are decreased significantly. In our experiment, the residual deformation decreased by 27% and the failure ratio decreased by more than 93%.

To explore the failure mode and bearing capacity of the granite cladding panel under different reinforcements, flexural tests were conducted on the windward and leeward sides of 33 granite cladding panels with undercut bolt anchorage. The joints of the granite cladding panels were reinforced with metal ring, and the panel back was reinforced with glass fiber reinforced plastics (GFRP) in cross, diagonal, horizontal, and vertical paste; then the bearing capacity of the panels was analyzed.The results show that after the panel is reinforced at the joints, the failure mode of the leeward side is improved in the loading direction. Compared with the panel without joint reinforcement, its bearing capacity is increased by 0.50 times. The four GFRP reinforcement modes at the back of the panel can keep its integrity after failure emerges and also improve the bending capacity of the panel windward side in the loading direction by 2.30 times at most, verifying the feasibility of the reinforcement measure of granite cladding panels with undercut bolt anchorage for engineering applications.