人工智能算法在铁道车辆动力学仿真中的应用进展

梳理了人工智能算法在铁道车辆系统动力学仿真中的应用实例和国内外相关文献,概述了铁道车辆动力学仿真中常用的机器学习和深度学习算法,归纳和评述了2种学习算法在铁道车辆系统动力学建模与仿真中的应用分类;从铁道车辆系统动力学建模、动力学性能预测与动力学性能优化等方面入手,详细讨论了人工智能算法应用在力元建模和仿真、轨道不平顺预测、运行平稳性预测、噪声预测、侧风安全性预测、运行安全性预测、悬挂优化、轮轨匹配优化、结构优化以及主动与半主动控制等领域的优势和局限性,指出了现阶段人工智能算法在动力学仿真应用中主要面临的训练样本缺乏、泛化能力不够、可解释性欠缺等问题;展望了今后人工智能算法和车辆系统动力学交叉研究的发展方向和重点研究内容。研究结果表明：融合经典力学和人工智能算法结合的混合建模理论可作为之后的重点研究方向;人工智能算法对解决随机动力学中的随机不确定性,提高随机动力学的性能具有较大的应用潜力;通过人工智能算法与优化算法相结合来实现动力学性能优化,可充分发挥人工智能算法的优势。     

A virtual environment for the formulation of policy packages

The interdependence and complexity of socio-technical systems and availability of a wide variety of policy measures to address policy problems make the process of policy formulation difficult. In order to formulate sustainable and efficient transport policies, development of new tools and techniques is necessary. One of the approaches gaining ground is policy packaging, which shifts focus from implementation of individual policy measures to implementation of combinations of measures with the aim of increasing efficiency and effectiveness of policy interventions by increasing synergies and reducing potential contradictions among policy measures. In this paper, we describe the development of a virtual environment for the exploration and analysis of different configurations of policy measures in order to build policy packages. By developing systematic approaches it is possible to examine more alternatives at a greater depth, decrease the time required for the overall analysis, provide real-time assessment and feedback on the effect of changes in the configurations, and ultimately form more effective policies. The results from this research demonstrate the usefulness of computational approaches in addressing the complexity inherent in the formulation of policy packages. This new approach has been applied to the formulation of policies to advance sustainable transportation.     

Truck parking in urban areas: Application of choice modelling within traffic microsimulation

Urban truck parking policies include time restrictions, pricing policies, space management and enforcement. This paper develops a method for investigating the potential impact of truck parking policy in urban areas. An econometric parking choice model is developed that accounts for parking type and location. A traffic simulation module is developed that incorporates the parking choice model to select suitable parking facilities/locations. The models are demonstrated to evaluate the impact of dedicating on-street parking in a busy street system in the Toronto CBD. The results of the study show lower mean searching time for freight vehicles when some streets are reserved for freight parking, accompanied by higher search and walking times for passenger vehicles.     

考虑船体变形耦合作用的船舶推进系统建模与控制

提出考虑船体变形耦合作用的船舶推进系统仿真建模设计方法与系统框架,通过船体变形与动力装置振动烈度间的量化关系体现耦合作用对模型的影响;研究适合推进系统的先进控制技术,通过对简化后的船舶推进系统不确定模型进行仿真试验,结果表明,具有强自适应能力的神经网络控制和模糊控制技术能够满足系统对不确定性及时变非线性控制的要求,具有良好的控制效果。     

Application of a validated primary production model (BLOOM) as a screening tool for marine, coastal and transitional waters

In order to manage aquatic systems, it is necessary to apply methods relating the environmental variables and system-state parameters with external factors that affect the system. External factors can be natural (i.e. the movement of water) or partly-anthropogenic (i.e. nutrient loads). In addition to the national authorities, who have been implementing environmental policies for several decades, the EU is presently implementing the Water Framework Directive (WFD) aimed at establishing a new set of standards for the ecological and water quality of water systems. Among these are the phytoplankton biomass and composition. Phytoplankton affects turbidity, oxygen depletion, total productivity of the system and the occurrence of (harmful) algal blooms. A range of methods is available to relate phytoplankton to the controlling environmental conditions. Among these are statistical relations for instance of the Vollenweider type as well as deterministic simulation models. At the end of the 1970s, a generic deterministic phytoplankton module called BLOOM was developed, which has since been applied to a wide range of fresh water and marine systems. Here we test the applicability of this model as a screening tool for coastal waters. We conclude that the model is able to reproduce observed chlorophyll levels adequately under a wide range of conditions. Subsequently the model is applied to demonstrate the potential impacts of reductions in nitrogen, phosphorus or both nutrients simultaneously. Depending on which factors are initially controlling, the impacts of these reductions vary considerably both between locations and during the season. While this type of application lacks explicit relations between nutrient concentrations and external loadings, it does consider a number of relevant conditions in a consistent way and requires remarkably little data and effort. It is therefore a valuable screening tool.     

Vehicle dynamics control of an electric-all-wheel-drive hybrid electric vehicle using tyre force optimisation and allocation

ABSTRACT

Hybrid Electric Vehicles (HEV) offer improved fuel efficiency compared to conventional vehicles at the expense of adding complexity and at times, reduced total power. As a result, HEV generally lack the dynamic performance that customers enjoy. To address this issue, the paper presents a HEV with electric All-Wheel-Drive capabilities via the use of torque vectoring electric rear axle drive (TVeRAD) to power the rear axle. The addition of TVeRAD to a front wheel drive HEV improves the total power output. To improve the handling characteristics of the vehicle, the TVeRAD provides torque vectoring at the rear axle. A bond graph model of the drivetrain is developed and used in co-simulation with CarSim. The paper proposes a control system which utilises control allocation to optimise tyre forces. The proposed control system is tested in the simulation environment with a high fidelity CarSim vehicle model. Simulation results show the control system is able to maximise vehicle longitudinal performance while avoiding tyre saturation on low mu surfaces. More importantly, the control system is able to track the desired yaw moment request on a high speed double lane change manoeuvre through the use of the TVeRAD to improve the handling characteristic of the vehicle.     

Optimal tyre usage for a Formula One car

Variations in track temperature, surface conditions and layout have led tyre manufacturers to produce a range of rubber compounds for race events. Each compound has unique friction and durability characteristics. Efficient tyre management over a full race distance is a crucial component of a competitive race strategy. A minimum lap time optimal control calculation and a thermodynamic tyre wear model are used to establish optimal tyre warming and tyre usage strategies. Lap time sensitivities demonstrate that relatively small changes in control strategy can lead to significant reductions in the associated wear metrics. The illustrated methodology shows how vehicle setup parameters can be optimised for minimum tyre usage.     

Data-driven train set crash dynamics simulation

Traditional finite element (FE) methods are arguably expensive in computation/simulation of the train crash. High computational cost limits their direct applications in investigating dynamic behaviours of an entire train set for crashworthiness design and structural optimisation. On the contrary, multi-body modelling is widely used because of its low computational cost with the trade-off in accuracy. In this study, a data-driven train crash modelling method is proposed to improve the performance of a multi-body dynamics simulation of train set crash without increasing the computational burden. This is achieved by the parallel random forest algorithm, which is a machine learning approach that extracts useful patterns of force–displacement curves and predicts a force–displacement relation in a given collision condition from a collection of offline FE simulation data on various collision conditions, namely different crash velocities in our analysis. Using the FE simulation results as a benchmark, we compared our method with traditional multi-body modelling methods and the result shows that our data-driven method improves the accuracy over traditional multi-body models in train crash simulation and runs at the same level of efficiency.     

Modelling and validation of coupled high-speed maglev train-and-viaduct systems considering support flexibility

This study presents a more realistic modelling of the maglev-based high-speed railway line in Shanghai, China. Focus is placed on an accurate simulation of the two subsystems: the train subsystem including the magnets and the viaduct subsystem including the modular function units of the rails. The electromagnet force–air gap model with a proportional-derivative (PD) controller is adopted to simulate the interaction between the maglev train via its electromagnets and the viaduct via its modular function units. The flexibilities of the rails, girders, piers and associated elastic bearings are all considered in the modelling of the viaduct subsystem to investigate their effects on an interaction between the two subsystems. By applying the proposed model to the Shanghai maglev line, the essential characteristics of the coupled system can be duly captured. The accuracy and effectiveness of the proposed approach are then validated by comparing the computed dynamic responses and frequencies with the measurement results. It is confirmed that the proposed modelling with a detailed simulation of the magnets and modular function units can duly account for the dynamic interaction between the train and viaduct systems. Moreover, the effects of the inclusion of the flexibilities of the rails, girders and elastic supports to the response of the coupled system are respectively investigated, the results of which prove that their involvements are essential to the accurate prediction of the response of the coupled maglev train–viaduct system.     

Finite element based meta-modeling of ship-ice interaction at shoulder and midship areas for ship performance simulation

The climate change has made the transit through Arctic area more feasible, which demands reliable methods to evaluate ship performance. Ship performance in ice is a cross-scale problem, where the desired output such as ship speed lies in larger scale while the actual ship-ice interaction happens in smaller scale. Due to insufficient knowledge in ice mechanics and the demand for computational efficiency, existing approaches for modelling ship-ice interaction from ship performance perspective are mostly either (semi-) empirical, or simplified analytical, with reduced dimensions and extensively simplified mechanics. This paper presents a novel approach to model ship-ice interaction, which maintains the accuracy of the modelling with Finite Element Method (FEM) in ship-ice interaction scale, while being computationally very cheap, therefore is capable to be applied in ship scale simulations. The ice failure is firstly qualitatively investigated through full-scale and model-scale observations, as well as a numerical simulation with Extended Finite Element Method (XFEM). The model is then simplified and executed by Abaqus to automatically run a large database. A neural network is used to fit the results to get a simulation-free tool for ship-ice interaction calculation. Finally, the uncertainty in the results due to an important assumption is quantified. The results show that the obtained neural network fits the database with excellent performance. Therefore, it can be applied in ship scale simulations with improved accuracy compared to empirical or analytical approaches.