A discrete-event model to simulate the effect of truck bunching due to payload variance on cycle time,hauled mine materials and fuel consumption

Data collected from truck payload management systems at various surface mines shows that the payload variance is significant and must be considered in analysing the mine productivity,energy consumption,greenhouse gas emissions and associated cost.Payload variance causes significant differences in gross vehicle weights.Heavily loaded trucks travel slower up ramps than lightly loaded trucks.Faster trucks are slowed by the presence of slower trucks,resulting in‘bunching’,production losses and increasing fuel consumptions.This paper simulates the truck bunching phenomena in large surface mines to improve truck and shovel systems’efficiency and minimise fuel consumption.The study concentrated on completing a practical simulation model based on a discrete event method which is most commonly used in this field of research in other industries.The simulation model has been validated by a dataset collected from a large surface mine in Arizona state,USA.The results have shown that there is a good agreement between the actual and estimated values of investigated parameters.     

A comprehensive investigation of loading variance influence on fuel consumption and gas emissions in mine haulage operation

The data collected from haul truck payload management systems at various surface mines show that the payload variance is significant and must be considered in analysing the mine productivity, diesel energy consumption, greenhouse gas emissions and associated costs. The aim of this study is to determine the energy and cost saving opportunities for truck haulage operations associated with the payload variance in surface mines. The results indicate that there is a non-linear relationship between the payload variance and the fuel consumption, greenhouse gas emissions and associated costs. A correlation model, which is independent of haul road conditions, has been developed between the payload variance and the cost saving using the data from an Australian surface coal mine. The results of analysis for this particular mine show that a significant saving of fuel and greenhouse gas emissions costs is possible if the standard deviation of payload is reduced from the maximum to minimum value.     

混合动力汽车经济型巡航的车速规划策略

选取功率分流式混合动力汽车为对象,以燃油消耗最小为目标开展巡航场景下的经济车速规划研究. 结合车辆动能管理与等效燃油最小化策略（ECMS）,提出增强型等效燃油最小化策略（R-ECMS）. 运用极小值原理推导油电等效系数,建立动能与电能间的等效关系;结合电能与燃油之间的等效关系,将车辆动能变化和电能消耗统一转化成燃油消耗. 为了兼顾电池SOC平衡以及车辆通行速度,采取非支配排序遗传算法优化R-ECMS权重系数中的参数. 仿真结果表明,与传统能量管理策略ECMS相比,R-ECMS可以降低8.06%的燃油消耗. 与采用最优算法的动态规划策略相比,R-ECMS能在实现次优的优化效果的同时大幅降低计算时间. 同时,与ECMS相比,R-ECMS在其他仿真场景下能实现6.94%的节油率,具有较好的泛化性能和应用前景.     

Energy saving performance improvement of intelligent connected PHEVs via NN-based lane change decision

Plug-in hybrid electric vehicles(PHEVs) unite the advantages of the engine and electric motor which could provide great potential in saving energy. However, the fuel economy performance of the PHEVs is highly associated with the driving condition, especially for parallel PHEVs because they could not decouple the engine work status from the driving condition. Meanwhile, fuel economy performance is not only a longitudinal issue but also related to lane selection. Lane selection is an important driving behavior and the algorithm of lane selection is necessary for the development progress of intelligent connected vehicles. Energy consumption cost is an important part of the vehicle's using consumption cost. Therefore, lane selection strategies must consider this point.With the development of intelligent connected vehicle technology, such as V2X(Vehicle to Everything), the potential of energy consumption performance of intelligent connected PHEVs could be improved by taking environment information from V2X and smart sensors into lane selection. In this paper, a neural network(NN) based method is proposed to predict the future status of the local vehicle using the information from V2X, and then another network is used to estimate the future energy consumption of each lane. The lane selection is decided on energy consumption estimation. Lastly, the effectiveness of the method is validated by simulation using Matlab combined with SUMO(Simulation of Urban MObility).     

复杂曲面铣削加工参数双神经网络优化方法研究

针对复杂曲面加工效率低、能耗高、表面质量难控制的问题,以及加工参数和目标之间关系确定的难题,建立了考虑复杂曲面特征的双神经网络优化方法。首先,用曲率表示复杂曲面加工复杂度来描述曲面特征,以曲面加工复杂度、主轴转速、进刀量、进给速度和路径间距为设计变量,以加工时间、能量消耗和表面粗糙度为目标函数,建立了复杂曲面加工参数的优化数学模型;其次,采用BP神经网络以黑箱法建立加工参数与优化目标的非线性关系,结合ALM神经网络方法对加工参数进行了优化。该方法解决了复杂曲面加工参数的优化问题,对提高复杂曲面加工效率和质量有一定的理论指导作用。     

不同方法构建建筑能耗模型比较

为了能够准确预测建筑能耗,以人工神经网络中的前馈神经网络和物理原理建立的建筑模型作为能耗方案进行预测分析,并建立了从外部和内部获得热量的系统方程.以办公楼为例,对比两模型对能耗预测结果的准确性,并且输入实际参数值,将计算结果与实际值进行对比分析.采用EnergyPlus进行了参数分析,以评估不同参数对预测结果的影响.结果表明,两种模型均适用于能耗预测,内部负荷对能耗预测的影响更为显著.     

考虑电池寿命的插电式混合动力汽车能量管理优化

考虑电池寿命对插电式混合动力汽车全寿命周期成本的影响,以综合燃油消耗和电池寿命衰减最小为目标开展电池充放电功率的多目标优化研究. 引入权重系数将多目标优化问题转化为单目标优化问题,采用动态规划（DP）算法求解实现全局最优,并根据优化结果选择最优权重系数. 为了解决动态规划算法运算速度慢、须预知工况的缺陷,以最优权重系数的优化结果训练神经网络控制器并将其应用于控制策略中. 仿真结果表明,与以油耗为单一目标的优化相比,多目标优化可使电池寿命衰减减少13.5%,而燃油消耗仅增加0.5%,在保证燃油经济性的同时有效减少电池寿命的衰减程度;基于神经网络的控制策略有效克服了动态规划算法的缺点并能达到与其相近的运算效果,具有较好的应用前景.     

基于IABC-Elman神经网络的电厂耗煤量短期预测

针对电厂耗煤量具有不确定性的特点及传统Elman神经网络利用梯度下降训练网络参数易陷于局部最优的缺点,基于人工蜂群(ABC)算法,提出了一种改进蜜源更新方式和跟随蜂选择引领蜂方式的改进ABC优化算法,结合进煤量、存煤量和发电量,建立了Elman神经网络电厂耗煤量短期预测模型(IABC-Elman)。实际算例表明,基于IABC-Elman电厂耗煤量短期预测模型结果能达到耗煤量短期预测的标准,与传统神经网络相比具有更高的预测精度。     

GPS and Google Earth based 3D assisted driving system for trucks in surface mines

In order to reduce the number of surface mining accidents related to low visibility conditions and blind spots of trucks and to provide 3D information for truck drivers and real time monitored truck information for the remote dispatcher, a 3D assisted driving system (3D-ADS) based on the GPS, mesh-wireless networks and the Google-Earth engine as the graphic interface and mine-mapping server, was developed at Virginia Tech. The research results indicate that this 3D-ADS system has the potential to increase reliability and reduce uncertainty in open pit mining operations by customizing the local 3D digital mining map, constructing 3D truck models, tracking vehicles in real time using a 3D interface and indicating available escape routes for driver safety.     

神经网络用于油田地面集输管道结垢预测

利用典型的误差反传神经网络理论,对油田地面集输管道结垢进行预测和评判,避开了各种因素对其结垢影响规律的难题,准确地预测和评判地面集输管道的结垢情况。应用人工神经网络分析某油田地面集输系统管道的结垢情况后表明,人工神经网络无需建立数学模型,学习过程通过自动调节神经元之间的连接权值完成,在选取有代表性的训练样本情况下,人工神经网络能够成功地预测和评判地面集输管道的结垢情况。     

Zigbee/Google Earth based assisted driving system in mining

The Assisted Driving System (ADS) for haul trucks operating in surface mining and construction sites is to reduce accidents related to low visibility conditions. This system is based on the GPS, Zigbee, and the Google-Earth engine as the graphic interface and mine-mapping server. The system has the capability to pin-point and track vehicles in real time using a 3D interface, which is based on user-based AutoCAD mine maps using the Google-Earth graphics interface. All equipped vehicles are shown in a 3D mine map stored in a local server through a wireless network. When low visibility conditions are present, the system indicates available exit/escape routes for driver safety. The ADS potentially increases reliability and reduces uncertainty in open pit mining operations.     

Investigation of the optimum differential gear ratio for real driving cycles by experiment design and genetic algorithm

Experiment statistical method and genetic algorithms based optimization method are used to obtain the optimum differential gear ratio for heavy truck that provides best fuel consumption when changing the working condition that affects its torque and speed range. The aim of the study is to obtain the optimum differential gear ratio with fast and accurate optimization calculation without affecting drivability characteristics of the vehicle according to certain driving cycles that represent the new working conditions of the truck. The study is carried on a mining dump truck YT3621 with 9 forward shift manual transmission. Two loading conditions,no load and 40 t,and four on road real driving cycles have been discussed. The truck powertrain is modeled using GT-drive,and DOE-post processing tool of the GT-suite is used for DOE analysis and genetic algorithm optimization.     

Hybrid dynamic model of polymer electrolyte membrane fuel cell stack using variable neural network

The polymer electrolyte membrane(PEM) fuel cell has been regarded as a potential alternative power source,and a model is necessary for its design,control and power management.A hybrid dynamic model of PEM fuel cell,which combines the advantages of mechanism model and black-box model,is proposed in this paper.To improve the performance,the static neural network and variable neural network are used to build the black-box model.The static neural network can significantly improve the static performance of the hybrid model,and the variable neural network makes the hybrid dynamic model predict the real PEM fuel cell behavior with required accuracy.Finally,the hybrid dynamic model is validated with a 500 W PEM fuel cell.The static and transient experiment results show that the hybrid dynamic model can predict the behavior of the fuel cell stack accurately and therefore can be effectively utilized in practical application.     

基于LM-BP神经网络的建筑能耗在线预测

由于神经网络不需要建立复杂的数学模型,因此基于BP神经网络的建筑能耗预测引起广泛关注.但标准BP神经网络收敛速度慢,不适于建筑能耗在线预测,因此采用了LevenbergMarquardts算法对标准BP神经网络加以改进,并将其应用于某建筑未来24小时的整点电量负荷预测.实验结果表明,改进的神经网络明显提高了训练速度,为建筑短期负荷的在线预测提供了一种方法.     

电动汽车锂电池SOC预测方法研究

为有效地对电动汽车锂电池荷电分布状态(SOC)进行预测,在分析与电池剩余电量相关因数,对动力电池组进行不同工况充放电试验的基础上,建立电池组的神经网络仿真模型。提出基于BP神经网络的电池剩余电量预测模型,利用模型可逼近任何多输入输出参数函数的性能,系统通过样本训练达到较好的仿真结果。与实验结果对比,仿真结果与实验基本吻合,验证了该方法的正确性.     

基于城市道路工况的汽车燃油消耗模型

应用车载油耗实时测试系统,在长春市不同路段、不同时段进行了实时油耗测试实验。根据实验结果分别建立了平峰期和高峰期的平均速度油耗模型、不同道路类型下的速度油耗模型以及不同速度、不同加速度下的瞬时油耗模型。通过其他道路实验对模型进行了验证,结果表明,这些模型能够较好地反映实际道路上车辆燃油消耗的变化情况,可以用来预测车辆在城市道路行驶时燃油消耗。     

矿用电动轮汽车日常维修成本数学模型分析

以某大型露天矿山同年投入使用的7辆154t630E型矿用电动轮汽车为研究对象,详细介绍了电动轮汽车日常维修成本构成,提出了一种科学估算电动轮汽车日常维修成本的新方法,建立数学模型,分析维修成本的规律,以降低维修成本,从而为国内矿用电动轮汽车日常维修成本评估提供了重要依据．     

G104国道卡车荷载概率特性分析

为了解决我国一级公路卡车荷载概率分布特性的问题,以G104国道289 d的动态称重数据为基础,采用多峰正态分布模型及多重指数分布模型,分析包括车质量、轴质量、轴距车速等参数在内的卡车荷载概率特性,得到各车辆荷载参数的概率分布类型及分布参数,并确定了G104国道代表性卡车车型.结果发现:采用多峰正态分布可以准确描述除卡车车头时距之外的其他车辆荷载参数分布,三重指数分布可准确描述卡车车头时距概率分布.G104国道卡车主要包括9种典型卡车车型,其中以2轴卡车为主,半挂卡车所占比例相对较小,半挂卡车显著车型为6轴V8卡车,且该车型平均装载质量、最大车质量均明显高于其他车型.     

公路桥梁车辆荷载的双峰分布概率模型

为了解决我国一级公路卡车荷载概率分布特性的问题,以G104国道289 d的动态称重数据为基础,采用多峰正态分布模型及多重指数分布模型,分析包括车质量、轴质量、轴距车速等参数在内的卡车荷载概率特性,得到各车辆荷载参数的概率分布类型及分布参数,并确定了G104国道代表性卡车车型.结果发现：采用多峰正态分布可以准确描述除卡车车头时距之外的其他车辆荷载参数分布,三重指数分布可准确描述卡车车头时距概率分布.G104国道卡车主要包括9种典型卡车车型,其中以2轴卡车为主,半挂卡车所占比例相对较小,半挂卡车显著车型为6轴V8卡车,且该车型平均装载质量、最大车质量均明显高于其他车型.

     

具有高区分度的视频火焰检测方法

为了在视频监控系统中准确地判断火焰区域并预测火灾的发生,提出一种新的基于人工神经网络的视频火焰检测方法.该方法在分析火焰的运动和三维颜色特征的基础上,分别通过傅里叶变换和圆形度分析、角点检测的方法研究火焰的闪烁频率、几何形状对应的时空域特征,采用获得的各类特征构成概率向量作为人工神经网络分类模型的输入,输出表示火灾发生的概率.在保持检测准确率的同时,该方法通过实验选择最优的参数组合解决神经网络容易陷入局部极值及收敛慢的问题.该方法可以区分大空间（隧道、仓库、博物馆等建筑物）中闪烁的车灯和真实火焰,能够避免在实际的视频监控系统应用中将闪烁车灯误判为火焰,有效减少环境光对检测结果的影响,降低火灾火焰的误报率.实验结果表明,采用该方法在保持检测实时性的同时,能够达到96%的检测正确率.