Abstract:Based on the structural characteristics of vehicles， a dynamical model of suspension frame and suspension magnet chain structure of electromagnetic suspension （EMS） high-speed maglev vehicle was established， and the suspension control algorithm based on state observer was introduced to simulate the active suspension control. Combining with the finite element model reflecting the unique structural and dynamical characteristics of maglev turnout， a numerical analysis model of maglev vehicle-turnout system was established. It is shown that simulation results are close to the in-situ measurements. On this basis， the dynamic response analysis of vehicle-turnout in single- and multi-car formation at low， medium and high-speed running through the turnout was carried out. Both numerical and experimental results show that defining the parameter matching relation of the vehicle-turnout-suspension control system is the premise and key to reveal the interaction mechanism between the maglev vehicle and turnout.

Abstract:The dynamic mechanism and vibration response of a high-speed maglev vehicle-guideway coupling system were studied. Firstly， various coupling relationships in high-speed maglev systems were modeled. Next， the vehicle-guideway coupling mechanisms under dynamic and static conditions were analyzed respectively. Then， the finite element model of high-speed maglev vehicle-guideway coupling dynamics was established. Finally， the dynamic response of the guideway beam and the functional components along with the vehicle-guideway resonance speed was analyzed when the maglev train passed the bridge at high speed. The results show that the vehicle-guideway interaction frequency changes with running speed， and the vehicle-guideway vibration increases significantly at resonance speeds for a long time. If the dynamic amplification factor is less than 1.2， the minimum base frequency ratio is 1.16. If the dynamic amplification factor is less than 1.3， the minimum base frequency ratio is 1.04.

Abstract:Taking the high-speed maglev train as the research object， the vehicle-guideway coupling vibration was discussed under the condition of elastic track. The minimum suspension element model of the actual vehicle-guideway coupling was established， and the mechanism analysis of the stability and coupling vibration of the open-loop system was conducted. In order to make up for the lack of effective information utilization of proportional-integral-differential （PID） controller， a full-state feedback controller was designed based on pre-compensation. By analyzing the effect of the feedback gain matrix on the system tracking performance and vibration suppression capability through simulation， an idea was proposed to appropriately reduce the suspension clearance in exchange for the improvement of vibration suppression capability. Further， the influence law of the feedback coefficient on the performance of the suspension system was analyzed， and on this basis an optimized pre-compensation full-state feedback control framework was obtained. Through experiments， the full-state feedback control was demonstrated based on pre-compensation， which can effectively suppress the elastic coupling vibration of the guideway.

Abstract:In order to ensure the safe and reliable operation of maglev trains， the suspension control of suspension systems under the condition of parameter perturbation was studied. Firstly， the basic suspension unit of a electromagnetic suspension （EMS） maglev train was modeled and the current control model was given. Then， the reinforcement learning environment and the soft actor-critic （SAC）agent were established for the suspension system， and a reward function and an anti-suspension-contact strategy were designed for accelerating the training process. Finally， a corresponding reinforcement learning method was proposed. Compared with the conventional proportional-integral-derivative （PID） control method， the proposed method has faster dynamic response and better tracking accuracy under the condition of 50% loss of coil turns and the pole area change of the magnet.

Abstract:In order to solve the performance degradation problem of magnetic levitation system of hign-speed maglev train during long-term operation， the complex dynamic characteristics of magnetic levitation system was analyzed to build a plug and play modular control and optimization architecture based on Youla parametric form， and an online optimization algorithm based on residual was designed. Simulation results show that the designed control and optimization architecture and online optimization algorithm improve the robustness and adaptability of the magnetic levitation system for unknown disturbance effectively.

Abstract:An electromagnet suspension gap prediction method was proposed based on long short-term memory （LSTM） neural network for high-speed maglev trains. Firstly， considering the aerodynamic load during the operation of high-speed maglev train， a train simulation model was established and the dynamic response of the train was calculated. Secondly， the LSTM neural network was established by PyCharm and the prediction model of the electromagnet suspension gap for high-speed maglev trains was built by taking the simulation calculation results as the sample set. Finally， the accuracy of the proposed electromagnet suspension gap prediction algorithm was verified by judging the calculation results and evaluation indexes of the prediction model.

Abstract:To study the deformation characteristics of pebble gravel-crushed aggregate composite cushion under earthquake loading， a series of large-scale shear tests were carried out， and the effects of cushion material， normal stress， shear amplitude， particle breakage， number of shear cycles， and shear direction were discussed. The experimental results reveal that the differences in physical characteristics such as particle shape and porosity between the pebble gravel and the crushed aggregate will affect the cushion’s deformation characteristics. Due to the interaction between the cushion layers， the horizontal and cumulative normal deformations of the pebble gravel-crushed aggregate composite cushion are larger than that of the single-layer cushion.

Abstract:A stochastic inversion method for in-service underground structure load was proposed. Firstly， based on spline function， the disorderly distributed load was parameterized into a set of interpolation unknowns. Secondly， on Bayesian framework， the posterior probability density function （PDF） of the unknowns was built by incorporating the measured deformation data. Lastly， the full Bayesian inference of the corresponding unknown load was carried out based on an efficient sampling method of DREAM（differential evolution adaptive metropolis）. Testing results from a field case indicated that the expectation load obtained from the proposed method fits well with the actual recorded pressures while the ill-conditioning is encountered by traditional deterministic inversion method. Apart from the expectation load， complete PDFs of the inversion load are obtained， which presents the natural advantage of the proposed method to deal with non-uniqueness.

Abstract:To investigate the mechanical properties， failure mechanism， and the influence of the connection on the overall performance of the column-supported modular steel frame under vertical loading， the experimental study on specimens of the module-to-module joints and two-story-single-bay modular steel frame was conducted. The rotational stiffness of the module-to-module joints， the overall stability bearing capacity and the failure mode of the modular steel frame were obtained. Finally， the overall response of the modular steel frame and the joint behavior were also carefully observed， and the theoretical calculation model and numerical analysis model of the stability bearing capacity of the structure were verified. The module-to-module joint is classified as the semi-rigid type. Therefore， the effects of semi-rigid joints should be adequately taken into consideration in design. The failure mode of the modular steel frame under vertical loading is the overall instability of the upper and lower modular corner columns. The substructure model can preferably calculate the overall stability bearing capacity.

Abstract:From the aspects of state equation， continuity equation， momentum equation and other numerical techniques，the numerical techniques and recent progress of smoothed particle hydrodynamics （SPH） on water-gas two-phase flow are reviewed. Discussions on the techniques are also presented. Finally， future research prospects are put forward.

Abstract:An optional phase optimization framework for transit signal priority （TSP） was proposed and a real-time TSP algorithm was developed. Dual-ring phase structure was improved by adding optional transit phases for multiple phase settings to integrate phase repetition， reverse and insertion and other functions. Hence， the green times，cycle length and phase sequence can be optimized simultaneously. A mixed integer linear programming signal control optimization model was built， using the deviation of background signal timings as vehicle loss function， which was linearly designed and computationally cost-effective. Results of a case study show that compared with the active TSP algorithm， the proposed algorithm reduces transit delay and stops by 33.1% and 15.1% respectively， and vehicle delay and stops by 17.7% and 12.6% respectively.

Abstract:Based on license plate recognition （LPR） data and connected vehicle trajectory data， an ensemble learning method was deployed to estimate the intersection queue length. By analyzing the applicability and accuracy of different queue length estimation methods， the random forest method was applied to design a base ensemble learner and formulize the nonlinear mapping relationship among the LPR data， connected vehicle trajectory data， estimation results of the existing queue length methods and real queue length values. Simulation results show that the proposed method overperforms the existing queue length methods， with a mean absolute error of 1.3 m?cycle^-1?lane^-1 and a mean absolute percent error of 1.4%.

Abstract:City water-industry along Yangtze River has important impact on the river’s great protection. First of all， based on the present development situation of Great Protection of Yangtze River， as well as the current development status on the social third-party service in the field of city environmental protection and urban engineering construction， the demand for social third-party how to explore and innovate a new water-controlling mode under the new developing stage， is analyzed. And the theoretical basis on the new mode， i.e. city water-housekeeper， for social third-party participating in the construction of city water system along Yangtze River is studied. Then， the concept as well as connotation and basic features for city water-housekeeper are put forward and clarified. And basic element and framework system for the mode of city water-housekeeper are also established. Finally， application conditions as well as implementation road map for the mode of city water-housekeeper within Great Protection of Yangtze River is put forward， and its application progress in recent years as well as further practical promotion in the future concerning the mode is discussed.

Abstract:After constructing the index system of urban waterlogging risk assessment， the idea of game theory （GT） was used to combine the subjective weight obtained by the analytic hierarchy process （AHP） and the objective weight obtained by the improved entropy value method （IEVM）. The quantitative evaluation framework was established by combining the grey relation analysis （GRA） with the technique for order preference by similarity to ideal solution （TOPSIS）， and finally the location of the storage tanks was determined by the relative closeness of each node. It is shown that the coefficient of variation of the comprehensive value from GRA-TOPSIS is 0.628， higher than that from GRA or TOPSIS alone， which is more conducive to the identification of waterlogging risk at each node. In addition， the method can eliminate the risk of urban waterlogging and improve the standard of urban waterlogging prevention.

Abstract:A mathematical model of environmental responses for electrical-mechanical cable subsurface mooring （EMC-SSM） was established to solve the problem of the significant dynamic response of EMC-SSM under extreme sea conditions. The effectiveness of the snubber hose in matigating the environmental response of EMC-SSM and the effect of different configurations of the snubber hose on the mitigation were studied by calculating the tensile response of a specific type of EMC-SSM under sea conditions with a 100-year reoccurrence period. The results show that the dynamic response of EMC-SSM is inhibited with an appropriate length of snubber hose. However，the dynamic response of EMC-SSM is enhanced when the length of snubber hose is inappropriate. Finaly， the configuration principle of the snubber hose and the method for determining the shortest effective length were also presented.

Abstract:The unsteady flow simulation method was used to study the aerodynamic performance of the Formula SAE racing car during cornering. By using the unsteady flow method， the flow speed was set as a field function varying with time to accurately simulate the cornering situation， and the simulation results were compared with the steady cornering simulation. In conjunction with simulation results， the local flow field of diffuser and rear wing was quantitatively studied in road test using surface pressure measurement. The results show that， the front wheels and wings have a great effect on the downstream bottom airflow during unsteady cornering， and the unsteady aerodynamic robustness of the diffuser is the key point of the racing car aerodynamic design.

Abstract:The data envelopment analysis （DEA） based on undesirable outputs was used to evaluate the sustainable innovation efficiency of 30 provinces in China， and the factors affecting the sustainable innovation efficiency in China were studied through Tobit regression. The results indicate that the sustainable innovation efficiency in China is low but rise gradually. In addition， efficiency values vary widely among regions. The efficiency value of eastern China is higher than that of central and western China. The significance of factors affecting the sustainable innovation efficiency of eastern， central and western China is quite different， reflecting the heterogeneity of regional innovation systems in different regions.

Abstract:In view of the dilemma of “numerous cases and few people” in the enforcement department， a risk assessment method was proposed based on case retrieval and combined weighting in an environment of information uncertainty. Firstly， a natural person-based risk assessment index system was constructed. Secondly， a hybrid attribute distance measure and similarity calculation was introduced to take into account the ambiguity and reliability of case information， and the combined weights of attributes were optimized using the deviation minimization method. Then， a dynamic similarity threshold determination method was proposed to balance subjectivity and objectivity， and a case-like case database was constructed to estimate the enforcement risk of the target case based on the execution results of similar cases. Finally， the feasibility and validity of the assessment model were verified through a case study and comparative analysis.