基于移动导航数据的信号配时反推

在交通管理和评价时,信号配时对监测评价路口运行状态,评价路口配时方案至关重要.但是,大范围的实时信号配时方案的获取尚缺乏简明有效的途径.本文提出两种基于移动导航数据计算固定配时路口信号配时的方法.第一种方法是在不考虑驾驶员驾驶行为差异性时,得到路口红灯和车均延误的关系模型,从而计算某相位的红灯时长.另外一种方法是基于车辆通过停止线的时间,结合本文提出的上升梯度法,得到某阶段红灯时长.本文通过实际的路口案例计算,将预测结果和已知路口的信号配时比较,表明此方法计算得到的红绿灯时长准确度较高,为后续进行路口运行状态和通行能力研究提供了数据支持.     

载人潜水器耐压球壳的多目标优化设计

采用基于响应面(RSM)近似模型和遗传算法(GA)对某深海载人潜水器耐压球壳进行多目标优化设计。首先建立深海载人潜水器耐压球壳的结构优化模型,通过试验设计(DOE)获得设计目标的响应特性,拟合得到响应面近似模型,最后采用Pareto遗传算法对响应面模型进行多目标优化求解,得到潜水器耐压球壳的优化设计方案。     

���ڸ��������ݵ��Ŷӳ��ȼ�ⷽ���о�

排队长度作为评价信号交叉口运行效率的一个重要指标,能有效反映交叉口处的运行状况.传统排队检测模型大多基于线圈检测器,且模型假设过于理想化,本文提出一种面向低采样率浮动车数据、具有良好数据驱动性的信控交叉口在线排队长度检测方法,方法关键在于利用队尾浮动车位置估算最大排队长度.检测过程采用固定时间间隔,主要步骤包括地图匹配、等距划分交叉口进口道并统计停车点数量、判定队尾浮动车的位置、修正得到最大排队长度估计值.实测数据表明,此方法的精度与浮动车比率有直接的关系,在浮动车比率较高的许多主干路交叉口,精度可以达到理想效果,30 m以内的平均绝对误差对高峰期的排队检测依旧具有很大价值.     

An inverse finite element method for the analysis of VIV data

This paper presents an algorithm for the frequency domain solution of dynamic linear “inverse” problems, that is for the processing of measurement data (strain, acceleration etc.) acquired on a mechanical structure, in order to estimate the loads acting on the structure and its corresponding response. The problem is formulated as a constrained (force equilibrium) optimization (small deviations from measurements, small loads) problem, which is transformed into an unconstrained problem, then into differential equations. The algorithm is applied to the estimation of hydrodynamic forces induced by the shedding of vortices from an offshore oil riser.     

A Novel Approach to Vehicle Dynamics using the Theory of a Cosserat Point and its Application to Collision Analyses of Platooning Vehicles

A fully three-dimensional, computationally inexpensive vehicular model is presented. In contrast to traditional rigid body models, the vehicle's sprung mass is modeled as a (nonlinearly) deformable body. The formulation of the equations of motion is based on a continuum theory known as the theory of a Cosserat point. These equations largely preserve the relative simplicity of rigid body dynamics but incorporate deformable features. The ease of computer implementation permits the simultaneous simulation of vehicle and collision dynamics of multiple vehicles and highway objects. In this paper, the theory of a Cosserat point is summarized and its general application to vehicle and collision dynamics is illustrated. A three-dimensional collision algorithm is discussed with emphasis on small closing velocities (negligible residual crush, elastic rebound). The novel model is compared to standard procedures.     

Contents Index Volume 35 and 36

Active safety systems would benefit from tyre force and friction potential information. Different sensor concepts, including, among others, the EU–funded Apollo–project developed tyre sensor based on optical position detection, are being studied. The sensor can measure tyre carcass deflections with respect to the rim. The carcass deflections can be used to calculate tyre forces and they may be exploited in the estimation of friction potential. The waveforms of the sensor signal are illustrated. The vertical and lateral force estimations are presented with unavoidable compensation parts. The tyre sensor measurements were compared to the measurement–vehicle results and good correlations achieved. Continuing activities are concerned with the estimation of friction potential and the detection of aquaplaning.     

Limit braking of a high-performance motorcycle

A mathematical-model-based study of the limit braking of a high-performance motorcycle and rider is described. Front and rear brakes are operable independently. A dry road and high friction are presumed, such that full braking of the front wheel would lead to an overturn or ‘stoppie’ in colloquial parlance. Effective braking needs to maintain some loading on the rear wheel. A planar but otherwise detailed system model is set up and braking strategies for front and rear are devised. Parameters of the braking control schemes are derived with the help of an optimisation process, minimising the final speed in braking from high speed over a fixed time interval. Simulation results are examined critically and the strategy is developed until efficient use of the friction available is made. The nature of optimal braking events is demonstrated. The influences of slipper-clutch torque setting and the rear-tyre target load chosen are shown.     

A model-based traction control strategy non-reliant on wheel slip information

A traction control system (TCS) for two-wheel-drive vehicles can conveniently be realised by means of slip control. Such a TCS is modified in this paper in order to be applicable to four-wheel-drive vehicles and anti-lock braking systems, where slip information is not readily available. A reference vehicle model is used to estimate the vehicle velocity. The reference model is excited by a saw-tooth signal in order to adapt the slip for maximum tyre traction performance. The model-based TCS is made robust to vehicle modelling errors by extending it with (i) a superimposed loop of tyre static curve gradient control or (ii) a robust switching controller based on a bi-directional saw-tooth excitation signal. The proposed traction control strategies are verified by experiments and computer simulations.     

Observer-based clutch disengagement control during gear shift process of automated manual transmission

A clutch disengagement strategy is proposed for the shift control of automated manual transmissions. The control strategy is based on a drive shaft torque observer. With the estimated drive shaft torque, the clutch can be disengaged as fast as possible without large driveline oscillations, which contributes to the reduction of total shift time and shift shock. The proposed control strategy is tested on a complete powertrain simulation model. It is verified that the system is robust to the variations of driving conditions, such as vehicle mass and road grade. It is also demonstrated that the revised system with switched gain can provide satisfactory performance even under large estimation error of the engine torque.     

用神经网络分析估计斜拉桥的施工控制参数

结合一座三塔四跨预应力混凝土斜拉桥施工过程,进行用神经网络分析估计斜拉桥施工控制参数的研究。用神经网络分析估计斜拉桥施工控制参数的过程主要包括建立神经网络、计算训练样本、训练神经网络和网络仿真四部分。运用神经网络仿真计算进行斜拉桥施工监控的具体方法是,先计算当前施工状态的索力和标高,再预测下一节段的索力和标高。经过往复循环,逐一进行节段预测调整,从而指导斜拉桥顺利施工。通过对比该斜拉桥合拢后部分节段索力及主梁控制点标高的计算值和实测值可知,用神经网络分析进行斜拉桥施工监控的效果达到设计的理想桥梁线型和索力要求。