高速公路动态交通流Elman神经网络模型

为了提高高速公路交通流建模的精度,分析了离散的高速公路动态交通流数学模型,基于Elman网络原理,建立了回归神经网络交通流模型。回归神经网络的输入层、上下文层、隐含层和输出层的节点数目分别选为8、30、30和2,采用Levenberg-Marquardt算法对回归神经网络进行训练,并对一条5路段的高速公路进行仿真。结果表明:回归神经网络平均相对误差为8.683 7×10-5,最大相对误差为4.237 1×10-4,与BP神经网络和RBF神经网络相比较,Elman回归神经网络能更好地逼近交通流数学模型,真实地描述交通流基本特性,能准确地建立动态交通流模型,适应交通状况的变化。     

基于串联排队网络的三峡-葛洲坝水利枢纽联合调度模型

为了提高三峡-葛洲坝水利枢纽的整体通过能力,分析了三峡-葛洲坝水利枢纽联合调度的实际需求,建立了三峡-葛洲坝水利枢纽联合调度数学模型,考虑了闸室面积利用率最大、整体待闸时间最小两个目标函数和船舶编排过程中的八个约束,应用串联排队网络理论求解模型。算法将申报船舶按照航向分成四个船舶序列,动态计算每艘船舶的权重,兼顾船舶长度与宽度优先,待闸时间约束、葛洲坝船闸通航条件限制和任务均衡的要求,循环排船,逐步优化。应用结果表明应用该数学模型和编排算法编制一个计划期的两坝五闸计划仅需2 min,编排时间短,葛洲坝2#船闸的闸室面积利用率高于70%,并且客船和旅游观光船均排在前面的闸次中,说明客船的待闸时间约束是满足的,并且在航向上是上下航向交替运行,没有出现倒闸情况,编制的计划满足实际调度需要。     

基于船舶供电网络的模拟系统设计

本文主要介绍了基于船舶供电网络的模拟系统的主要功能和设计方法,该模拟系统采用集中式管理分布式控制的硬件架构,可模拟在各种不同的电网结构和不同的运行工况下,电网中各设备的状态和参数,并通讯至能量管理系统,能便捷的验证能量管理系统功能的有效性,提高软件开发人员的故障效率。     

Spatiotemporal traffic forecasting: review and proposed directions

This paper systematically reviews studies that forecast short-term traffic conditions using spatial dependence between links. We extract and synthesise 130 research papers, considering two perspectives: (1) methodological framework and (2) methods for capturing spatial information. Spatial information boosts the accuracy of prediction, particularly in congested traffic regimes and for longer horizons. Machine learning methods, which have attracted more attention in recent years, outperform the naïve statistical methods such as historical average and exponential smoothing. However, there is no guarantee of superiority when machine learning methods are compared with advanced statistical methods such as spatiotemporal autoregressive integrated moving average. As for the spatial dependency detection, a large gulf exists between the realistic spatial dependence of traffic links on a real network and the studied networks as follows: (1) studies capture spatial dependency of either adjacent or distant upstream and downstream links with the study link, (2) the spatially relevant links are selected either by prejudgment or by correlation-coefficient analysis, and (3) studies develop forecasting methods in a corridor test sample, where all links are connected sequentially together, assume a similarity between the behaviour of both parallel and adjacent links, and overlook the competitive nature of traffic links.     

Locating rental stations and bikeways in a public bike system

To determine the spatial distribution of rental stations and bikeways in a public bike system, this paper proposes a facility location and network design model. The model is developed as a multi-objective programing problem that considers four objectives (minimizing cyclist risk, maximizing cyclist comfort, minimizing adverse impacts on traffic and maximizing service coverage) and multiple constraints (monetary budget, network connectivity, station spacing, bikeway types, station number and value ranges of decision variables). The ε-constraint method solves the programing problem for the public bike system in Daan District, Taipei City, Taiwan. The nine non-dominated alternatives generated are all markedly better than existing locations of rental stations and bikeways. Scenario analysis results indicate that increasing the construction budget for bikeways significantly improves cyclist safety and comfort whilst increasing the adverse impact on traffic. Planners can use this model to develop public bike systems that spatially integrate rental stations and bikeway networks.     

Emission rates of intermodal rail/road and road-only transportation in Europe: A comprehensive simulation study

Intermodal rail/road transportation combines advantages of both modes of transport and is often seen as an effective approach for reducing the environmental impact of freight transportation. This is because it is often expected that rail transportation emits less greenhouse gases than road transportation. However, the actual emissions of both modes of transport depend on various factors like vehicle type, traction type, fuel emission factors, payload utilization, slope profile or traffic conditions. Still, comprehensive experimental results for estimating emission rates from heavy and voluminous goods in large-scale transportation systems are hardly available so far. This study describes an intermodal rail/road network model that covers the majority of European countries. Using this network model, we estimate emission rates with a mesoscopic model within and between the considered countries by conducting a large-scale simulation of road-only transports and intermodal transports. We show that there are high variations of emission rates for both road-only transportation and intermodal rail/road transportation over the different transport relations in Europe. We found that intermodal routing is more eco-friendly than road-only routing for more than 90% of the simulated shipments. Again, this value varies strongly among country pairs.     

Formulating cargo inventory costs for liner shipping network design

This study examines how to incorporate the inventory costs of containerized cargoes into existing liner service planning models such that the designed networks could be improved while not causing extra modeling/computational burden. Two approaches are compared: (i) not considering the inventory costs at all and (ii) incorporating the inventory costs associated with onboard time and those related to transshipment by assuming a fixed connection time. The two models are compared with the ideal model capturing the exact inventory costs on a route choice problem and a capacity planning problem based on extensive randomly generated and practical numerical experiments. The results show that: first, ignoring the inventory costs in service planning models may lead to network design with much higher costs (poor network design decisions); second, in service planning models assuming weekly frequency, the inventory costs associated with onboard time could be formulated exactly, and those related to the connection time of weekly services could be approximated by assuming fixed connection time of 3.5 days for ports with 1 day’s minimum connection time and 4.5 days for ports with 2 days’ minimum connection time.     

Large-scale evacuation network optimization: a bi-level control method with uncertain arterial demand

To improve the efficiency of large-scale evacuations, a network aggregation method and a bi-level optimization control method are proposed in this paper. The network aggregation method indicates the uncertain evacuation demand on the arterial sub-network and balances accuracy and efficiency by refining local road sub-networks. The bi-level optimization control method is developed to reconfigure the aggregated network from both supply and demand sides with contraflow and conflict elimination. The main purpose of this control method is to make the arterial sub-network to be served without congestion and interruption. Then, a corresponding bi-objective network flow model is presented in a static manner for an oversaturated network, and a Genetic Algorithm-based solution method is used to solve the evacuation problem. The numerical results from optimizing a city-scale evacuation network for a super typhoon justify the validity and usefulness of the network aggregation and optimization control methods.     

Vulnerability analysis for large-scale and congested road networks with demand uncertainty

To assess the vulnerability of congested road networks, the commonly used full network scan approach is to evaluate all possible scenarios of link closure using a form of traffic assignment. This approach can be computationally burdensome and may not be viable for identifying the most critical links in large-scale networks. In this study, an “impact area” vulnerability analysis approach is proposed to evaluate the consequences of a link closure within its impact area instead of the whole network. The proposed approach can significantly reduce the search space for determining the most critical links in large-scale networks. In addition, a new vulnerability index is introduced to examine properly the consequences of a link closure. The effects of demand uncertainty and heterogeneous travellers’ risk-taking behaviour are explicitly considered. Numerical results for two different road networks show that in practice the proposed approach is more efficient than traditional full scan approach for identifying the same set of critical links. Numerical results also demonstrate that both stochastic demand and travellers’ risk-taking behaviour have significant impacts on network vulnerability analysis, especially under high network congestion and large demand variations. Ignoring their impacts can underestimate the consequences of link closures and misidentify the most critical links.     

基于BP人工神经网络的滑坡预测分析

文章以汶川地震引发的滑坡为研究对象,以震中距、地震烈度、坡度、前缘高程、坡高和岩性等影响坡体稳定性的因素为切入点,利用BP人工神经网络对实际坡体的稳定性进行了预测分析。结果表明,BP人工神经网络方法能有效预测坡体的稳定情况。