改进的BP神经网络在路基沉降预测中的应用

针对传统BP神经网络存在的缺点,提出基于遗传优化的变梯度反向传播的BP神经网络预测方法,采用遗传算法优化BP神经网络的初始权重,建立路基沉降预测模型。该模型可克服BP神经网络模型存在的收敛速度慢、易陷入局部极小点等缺点。结合现场实测数据,将该优化模型与指数曲线模型、双曲线模型、灰色预测模型和传统BP神经网络预测模型对比,结果表明改进的BP神经网络在路基沉降预测中精度最高,适宜于广泛推广应用。     

神经网络参数识别法在重庆石板坡大桥中的应用

BP神经网络法的自适应学习能力、非线性映射能力、鲁棒性和容错能力以及快速收敛能力可有效解决连续刚构桥施工控制中参数估计的核心问题,通过实例证明,其参数估计结果与实测数据吻合性较好,识别精度较高,有相当的实践意义.尤其是对于必须考虑非线性影响、不确定系统的控制等问题,如果经典算法识别精度低,可考虑采用非经典神经网络算法进行重要参数的识别.     

基于神经网络的交叉口交通流量预测方法研究

主要介绍利用BP神经网络模型来预测交叉口交通流量的方法。在概述了神经网络的工作原理和BP网络模型的设计之后,通过举例更进一步说明此方法的可行性和预测效果。     

Evaluation of energy consumption of vehicles in EU Trans-European Transport Network

The paper presents the results of field tests evaluating energy consumption in the vehicles of Trans-European Transport Network (TEN-T) of selected EU countries: Poland, Germany and France. The energy consumption of vehicles in a highway system was assessed based on the telemetry analysis systems for traction parameters, tachograph record of digital speed waveform and their statistical analysis. The empirical cumulative distribution functions of speed transitions (acceleration, deceleration) were used to determine the kinetic energy losses of the vehicle (fuel consumption). To assess the statistical significance of differences between cumulative distribution functions the Smirnov–Kolmogorov test was used.     

Transit network design based on travel time reliability

This paper presents a transit network optimization method, in which travel time reliability on road is considered. A robust optimization model, taking into account the stochastic travel time, is formulated to satisfy the demand of passengers and provide reliable transit service. The optimization model aims to maximize the efficiency of passenger trips in the optimized transit network. Tabu search algorithm is defined and implemented to solve the problem. Then, transit network optimization method proposed in this paper is tested with two numerical examples: a simple route and a medium-size network. The results show the proposed method can effectively improve the reliability of a transit network and reduce the travel time of passengers in general.     

Accounting for dynamic speed limit control in a stochastic traffic environment: A reinforcement learning approach

This paper proposes a novel dynamic speed limit control model accounting for uncertain traffic demand and supply in a stochastic traffic network. First, a link based dynamic network loading model is developed to simulate the traffic flow propagation allowing the change of speed limits. Shockwave propagation is well defined and captured by checking the difference between the queue forming end and the dissipation end. Second, the dynamic speed limit problem is formulated as a Markov Decision Process (MDP) problem and solved by a real time control mechanism. The speed limit controller is modeled as an intelligent agent interacting with the stochastic network environment stochastic network environment to assign time dependent link based speed limits. Based on different metrics, e.g. total network throughput, delay time, vehicular emissions are optimized in the modeling framework, the optimal speed limit scheme is obtained by applying the R-Markov Average Reward Technique (R-MART) based reinforcement learning algorithm. A case study of the Sioux Falls network is constructed to test the performance of the model. Results show that the total travel time and emissions (in terms of CO) are reduced by around 18% and 20% compared with the base case of non-speed limit control.     

A behavioural comparison of route choice on metro networks: Time,transfers, crowding,topology and socio-demographics

Understanding travellers’ behaviour is key element in transportation planning. This article presents a route choice model for metro networks that considers different time components as well as variables related to the transferring experience, train crowding, network topology and socio-demographic characteristics. The route choice model is applied to the London Underground and Santiago Metro networks, to make a comparison of the decision making process of the users on both cities. As all the variables are statistically significant, it is possible to affirm that public transport users take into account a wide variety of elements when choosing routes. While in London the travellers prefer to spend time walking, in Santiago is preferable to spend time waiting. Santiago Metro users are more willing to travel in crowded trains than London Underground users. Both user groups have a similar dispreference to transfers after controlling for the time spent on transfer, but different attitudes to ascending and descending transfers. Topological factors presented on a distorted Metro map are more important than actual topology to passengers’ route choice decisions.     

Traffic volume forecasting based on radial basis function neural network with the consideration of traffic flows at the adjacent intersections

The forecasting of short-term traffic flow is one of the key issues in the field of dynamic traffic control and management. Because of the uncertainty and nonlinearity, short-term traffic flow forecasting could be a challenging task. Artificial Neural Network (ANN) could be a good solution to this issue as it is possible to obtain a higher forecasting accuracy within relatively short time through this tool. Traditional methods for traffic flow forecasting generally based on a separated single point. However, it is found that traffic flows from adjacent intersections show a similar trend. It indicates that the vehicle accumulation and dissipation influence the traffic volumes of the adjacent intersections. This paper presents a novel method, which considers the travel flows of the adjacent intersections when forecasting the one of the middle. Computational experiments show that the proposed model is both effective and practical.     

A Bayesian approach to detect pedestrian destination-sequences from WiFi signatures

In this paper, we propose a methodology to use the communication network infrastructure, in particular WiFi traces, to detect the sequence of activity episodes visited by pedestrians. Due to the poor quality of WiFi localization, a probabilistic method is proposed that infers activity-episode locations based on WiFi traces and calculates the likelihood of observing these traces in the pedestrian network, taking into account prior knowledge. The output of the method consists of candidates of activity-episodes sequences associated with the likelihood to be the true one. The methodology is validated on traces generated by a known sequence of activities, while the performance being evaluated on a set of anonymous users. Results show that it is possible to predict the number of episodes and the activity-episodes locations and durations, by merging information about the activity locations on the map, WiFi measurements and prior information about schedules and the attractivity in pedestrian infrastructure. The ambiguity of each activity episode in the sequence is explicitly measured.     

Co-design of traffic network topology and control measures

The two main directions to improve traffic flows in networks involve changing the network topology and introducing new traffic control measures. In this paper, we consider a co-design approach to apply these two methods jointly to improve the interaction between different methods and to get a better overall performance. We aim at finding the optimal network topology and the optimal parameters of traffic control laws at the same time by solving a co-optimization problem. However, such an optimization problem is usually highly non-linear and non-convex, and it possibly involves a mixed-integer form. Therefore, we discuss four different solution frameworks that can be used for solving the co-optimization problem, according to different requirements on the computational complexity and speed. A simulation-based study is implemented on the Singapore freeway network to illustrate the co-design approach and to compare the four different solution frameworks.