基于神经网络的车用汽油机过渡工况空燃比辨识

以HL495Q电喷汽油机为研究对象,提出了一种基于BP神经网络的空燃比辨识方法,比较了不同拓扑结构的神经网络对空燃比辨识精度的影响,得到了一种最优的空燃比模型。试验结果表明,空燃比模型能高精度地逼近空燃比的实际动态过程,模型的平均相对误差小于2%。     

面向交通枢纽的车辆三维真实感实时仿真

为分析和解决城市交通拥挤问题并提高城市道路利用率提供可行的途径,提出了一种面向交通枢纽的车辆运行仿真方法,通过场景、道路与车辆的三维动态建模,实现交通枢纽交通状况的实时真实感仿真.首先,提出了基于道路关键点连接网络模型表示交通枢纽的通行道路.其次,基于粒子系统实现车辆的动态运行实时仿真,并采用基于空间剖分的车辆碰撞检测方法对车辆运动控制算法进行了优化.最终,通过对路段的动态观测和反馈机制实现车辆行驶路线的规划和调度.实验结果表明,本文提出的方法可以生动直观地呈现实际路面的交通状况,并且能以较为流畅的帧速率实现交通场景的动态仿真.     

Asynchronous n-step Q-learning adaptive traffic signal control

Ensuring transportation systems are efficient is a priority for modern society. Intersection traffic signal control can be modeled as a sequential decision-making problem. To learn how to make the best decisions, we apply reinforcement learning techniques with function approximation to train an adaptive traffic signal controller. We use the asynchronous n-step Q-learning algorithm with a two hidden layer artificial neural network as our reinforcement learning agent. A dynamic, stochastic rush hour simulation is developed to test the agent’s performance. Compared against traditional loop detector actuated and linear Q-learning traffic signal control methods, our reinforcement learning model develops a superior control policy, reducing mean total delay by up 40% without compromising throughput. However, we find our proposed model slightly increases delay for left turning vehicles compared to the actuated controller, as a consequence of the reward function, highlighting the need for an appropriate reward function which truly develops the desired policy.     

快速二阶BP神经网络在公路交通量预测中的应用

在二阶BP神经网络基础上加以改进，提出一种快速二阶BP神经网络，并将把该方法成功地用于公路交通量的预测中，通过与其它方法的比较分析，得出快速二阶BP神经网络预测方法加快了收敛速度，提高了结果的准确度，为科学地预测公路交通量提供了有力依据。     

人工神经网络在路堤沉降预测中的应用

以人工神经网络法为主，研究了多因素条件下的沉降预测问题，同时与浅岗法和S型曲线法这2种在近几年推广应用的预测方法进行了对比。结果表明，3种方法预测的最终沉降大体相近，它们之间的区别在于人工神经网络法预测的沉降较大（同时更接近实测值）；S型曲线法较小；浅岗法居中。由于神经网络是用实测数据直接建模，少了人为干扰因素，并且偏大的数值对工程来说是偏于安全的，所以选用人工神经网络预测沉降比较适宜。     

大型机电设备仿真建模的智能方法研究

本文提出了一种基于神经网络的大型机电设备仿真建模智能方法，详细讨论并解决了该法实施过程中的若干关键技术问题。通过对某船用蒸气动力装置简易训练模拟器进行仿真建模研究，表明该法对大型机电设备的仿真建模非常有效。     

水下分布式远程感知技术的发展

目前,水下分布式远程感知技术已经具备了快速发展的基本条件,备受各主要海军国家的重视,预期在未来5~10年内将会有重大进展。但水下分布式远程感知技术的发展还面临着从基础、理论到实践的众多课题,包括系统构成和运用的概念,分布式传感器阵对目标的探测、分类和识别,固定、可部署传感器和移动节点构成的传感器场的配置,水下分布式网络的通信和控制,以及系统的部署和保障等。本文讨论水下分布式远程感知系统的基本组成,实现系统功能所面临的主要挑战,需要关注的主要关键技术,并提出相应的看法和建议。     

A New Wireless Packet Scheduling Algorithm Based on EDF and Its Application in Multi-bottleneck Networks

Most of current wireless packet scheduling algorithms aim at resource allocation as fairly as possible or maximizing throughput. This paper proposed a new packet scheduling algorithm that aims at satisfying delay requirement and is the improvement of earliest due first (EDF) algorithm in wired networks. The main idea is to classify the packets based on their delay bound, scheduling the most "urgent" class of user and the users that have the best channel condition with higher priority. This algorithm can easily integrate with common buffer management algorithms, when buffer management algorithm cannot accept new arrival packets, try to modify scheduling policy. Packet scheduling algorithms in multiple bottleneck wireless networks were also discussed. A new variable multi-hop factor was defined to estimate the congestion situation (including channel condition) of future hops. Multi-hop factor can be integrated into packet scheduling algorithms as assistant and supplement to improve its performance in multi-bottleneck wireless networks.     

Competition and disruption in a dynamic urban supply chain

Rapid changes and complexities in business environments have stressed the importance of interactions between partners and competitors, leading supply chains to become the most important element of contemporary business environments. There is a concomitant need for foresight in describing supply chain performance in all operating environments, including those involving punctuated disruptions. Furthermore, the urban metropolis is now widely recognized to be an environment which is especially vulnerable to supply chain disruptions and for which integrated supply chain decisions can produce very substantial net benefits. Accordingly, this paper presents a dynamic supply chain network model formulated as a differential variational inequality; the model is fashioned to allow consideration of supply chain disruption threats to producers, freight carriers, and retail enterprises. The DVI is solved using a fixed-point algorithm, and a simple numerical example, introduced to illustrate how the impacts of supply chain disruptions may be quantified, is presented.     

区域综合运输通道抗毁性分析

区域运输通道是区域运输线路的骨干.通道抗毁性分析对发现通道网络薄弱环节、优化通道网络规划、保障运输安全等具有重要意义.以成渝经济区为例,抽取其运输通道网络,构建L空间下的复杂网络模型,构建基于节点度、节点运输重要度和节点介数的攻击策略,以网络全局效率和最大连通子图节点比为测度,借助Matlab和Pajek软件进行抗毁性分析仿真.结果表明：成渝经济区运输通道平均路径长度较大,连通性有待提高,在进行通道中公路、铁路网络规划时应注意联络线的布设,重庆、成都等应作为关键节点重点建设和保护.