Modeling and predicting low-speed vehicle emissions as a function of driving kinematics

An instantaneous emission model was developed to model and predict the real driving emissions of the low-speed vehicles. The emission database used in the model was measured by using portable emission measurement system (PEMS) under actual traffic conditions in the rural area, and the characteristics of the emission data were determined in relation to the driving kinematics (speed and acceleration) of the low-speed vehicle. The input of the emission model is driving cycle, and the model requires instantaneous vehicle speed and acceleration levels as input variables and uses them to interpolate the pollutant emission rate maps to calculate the transient pollutant emission rates, which will be accumulated to calculate the total emissions released during the whole driving cycle. And the vehicle fuel consumption was determined through the carbon balance method. The model predicted the emissions and fuel consumption of an in-use low-speed vehicle type model, which agreed well with the measured data.     

基于车载诊断系统的轻型乘用车实际道路油耗特征分析

近年来的一系列研究表明,轻型车实际道路油耗与型式认证油耗间的差异在不断增大,之前的实际道路油耗研究大多基于用户上报数据或车载排放测试系统(PEMS).车载诊断系统(OBD)是一种监测发动机及排放控制系统实时工作状态的仪器.本研究通过OBD解码器采集瞬时发动机进气量计算机动车瞬态油耗并通过2辆轻型汽油车的实验室台架测试证明该方法具有很好的准确性(差异在±5%以内).本研究进一步在北京选取了7辆轻型汽油车,利用OBD解码器开展实际道路油耗测试,并研究运行工况及车辆载重对其实际道路油耗的影响.实际道路结果显示测试车辆在典型工况下的油耗比型式认证油耗高23.1%~46.4%,平均差异为29.2%.通过微观运行模态方法将实际道路油耗修正到法规测试循环(NEDC工况)的交通特征下,道路油耗仍然比型式认证油耗高18.7%±7.1%.研究表明,平均速度与实际道路油耗具有很强的相关性.实际道路油耗在30 km·h-1以下时会随速度降低而显著上升.此外,研究还发现如果车辆载重增加260 kg,实际道路油耗将上升6.2%±2.2%.     

面向排放量化的低速区间机动车比功率分布特性与模型

随着排放建模方法从基于行驶周期和平均速度演变为基于VSP(Vehicle Specific Power)参数,利用VSP分布刻画交通状态成为最新的研究需求.近期研究中,针对城市快速路上大于20 km·h-1的速度区间建立了基于平均行程速度的VSP分布数学模型,却未对低速区间的VSP分布特征作深入研究.基于北京快速路大量逐秒浮动车数据,研究0～20 km·h-1的VSP分布与平均行程速度的关系.通过分析大量逐秒浮动数据的VSP分布与平均速度间关系,发现VSP分布与平均行程速度具有规律性:各VSP分布的峰值出现在VSP Bin=0处,且随速度的增加单调递减.因此,针对VSP分布的正、负区间以及VSP Bin=0处分别建立数学模型,并利用该模型进行机动车油耗/排放测算.对比分析油耗/排放的预测值和实测值得出,所建立的VSP分布模型可以有效用于机动车油耗/排放测算.     

Carbonyl emissions from heavy-duty diesel vehicle exhaust in China and the contribution to ozone formation potential

Fifteen heavy-duty diesel vehicles were tested on chassis dynamometer by using typical heavy duty driving cycle and fuel economy cycle. The air from the exhaust was sampled by 2,4- dinitrophenyhydrazine cartridge and 23 carbonyl compounds were analyzed by high performance liquid chromatography. The average emission factor of carbonyls was 97.2 mg/km, higher than that of light-duty diesel vehicles and gasoline-powered vehicles. Formaldehyde, acetaldehyde, acetone and propionaidehyde were the species with the highest emission factors. Main influencing factors for carbonyl emissions were vehicle type, average speed and regulated emission standard, and the impact of vehicle loading was not evident in this study. National emission of carbonyls from diesel vehicles exhaust was calculated for China, 2011, based on both vehicle miles traveled and fuel consumption. Carbonyl emission of diesel vehicle was estimated to be 45.8 Gg, and was comparable to gasolinepowered vehicles （58.4 Gg）. The emissions of formaldehyde, acetaldehyde and acetone were 12.6, 6.9, 3.8 Gg, respectively. The ozone formation potential of carbonyls from diesel vehicles exhaust was 537 mg O3/km, higher than 497 mg O3/km of none-methane hydrocarbons emitted from diesel vehicles.     

Real-world vehicle emission factors in Chinese metropolis city--Beijing

The dynamometer tests with different driving cycles and the real-world tests are presented. Results indicated the pollutants emission factors and fuel consumption factor with ECE15 EUDC driving cycle usually take the lowest value and with real wodd driving cycle occur the highest value, and different driving cycles will lead to significantly different vehicle emission factors with the same vehicle. Relative to the ECE15 EUDC driving cycle, the increasing rate of pollutant emission factors of CO, NOx and HC are - 0.42--2.99, - 0.32--0.81 and - 0.11--11 with FTP75 testing, 0.11--1.29, - 0.77--0.64 and 0.47--10.50 with Beijing 1997 testing and 0.25--1.83, 0.09--0.75 and - 0.58--1.50 with real wodd testing. Compared to the carburetor vehicles, the retrofit and MPI TWC vehicles‘ pollution emission factors decrease with different degree. The retrofit vehicle (Santana) will reduce 4.44%--58.44% CO, - 4.95%--36.79% NOx, -32.32%--33.89% HC, and -9.39%--14.29% fuel consumption, and especially that the MPI TWC vehicle will decrease CO by 82.48%--91.76%, NOx by 44.87%--92.79%, HC by 90.00%--93.89% and fuel consumption by 5.44%--10.55%. Vehicles can cause pollution at a very high rate when operated in high power modes; however, they may not often operate in these high power modes. In analyzing vehicle emissions, it describes the fraction of time that vehicles operate in various power modes. In Beijing, vehicles spend 90% of their operation in low power modes or decelerating.     

基于远程数据的重型柴油车实际道路碳排放测算准确度

利用车载排放测试系统(PEMS),对3辆加装有车载诊断系统(OBD)远程监控设备的重型柴油车进行实际道路排放测试,以获取车辆实际道路基于电子控制器单元(ECU)的OBD远程油耗数据,分析碳排放结果的准确性及其影响因素.研究发现,基于碳平衡法获取的ECU油耗换算CO₂与车载排放测量系统(PEMS)直接测量的CO₂排放结果存在差异,二者偏差平均为2.06%.对影响ECU碳排放计算的关键因素(ECU油耗量和ECU行驶距离)的准确性进行了分析,发现ECU油耗的偏差均在3%以内,而ECU测量距离的平均偏差为2.41%.ECU油耗和行驶距离的准确度会受到车速、加速度和整个行程动态的影响.当车辆低速行驶时,其准确度较低.     

Intensity and daily pattern of passenger vehicle use by region and class in China: estimation and implications for energy use and electrification

Given the explosive growth of the passenger vehicle market and energy demands in China, research on vehicle-use intensity and driver-travel patterns is critical for better assessing travel demand and its implications for alternative fuel vehicles, energy security, and environmental policies. This study attempts to estimate annual vehicle kilometers traveled (AVKT) per privately-owned passenger vehicle and their daily distance patterns by vehicle class and geographic region. The data sample from a survey consists of 169,292 privately owned passenger vehicles, made by 177 car manufacturers during 2003–2018, running in 82 cities from 27 provincial regions. The log-transformed average AVKT is estimated to be 12,377 km with 95% probability ranging from 5490 to 28,579 km. The investigation reveals that vehicles from South China have the highest AVKT at 13,320 km. Generally, vehicles in small cities have higher AVKT than in big cities, except AVKT of tier 1 cities being higher. Another trend is that more expensive or larger vehicles tend to be driven more. A model is fitted for estimating AVKT based on region, city type, automaker, price range, and certain vehicle features including class and age. Data of daily commuting distances in recent years are also analyzed. The average daily commuting distances typically range from 21 to 28 km. Using the validated Gamma distribution method, daily distance distributions are specified for different regions. It is found that 99% of the daily driving distance is no more than 88.0–112.0 km, depending on region. Utility factors of plug-in electric vehicles are also estimated to be much higher than those based on driving data in the USA. These findings suggest global mitigations strategies on vehicle fuel use, electrification, and greenhouse gases should consider vehicle-use intensity at the regional level.

     

同型规范变换的不同预测模型具有的兼容性和等效性

传统的不同预测变量的预测模型之间不具有兼容性和等效性,而同型规范变换和误差修正相结合的不同变量的预测模型的预测相对误差与预测对象的维数、样本数及预测模型类型皆无关,仅与预测变量的数据特性、相似样本的模型输出值及其相对误差和相似度有关,因而同型规范变换的不同预测变量的预测模型之间具有兼容性和等效性.其重要意义在于：只要对任意一个预测变量建立了基于规范变换的某种预测模型,就可以将此预测模型直接用于具有同型规范变换的其他预测变量的预测;若再将其与误差修正法相结合,还可以极大地提高模型的预测精度,获得与实际值很接近的预测结果.依据受3个因子影响的灞河口COD_Mn指数数据、受4个因子影响的伊犁河雅马渡站年径流量数据和牡丹江市TSP年均值的时序数据,分别建立具有同型规范变换（n_j=2）的3个不同预测变量的3种智能预测模型和一元线性回归预测模型,并验证了3个不同预测变量的预测模型之间的兼容性和等效性.对同一个预测样本,用同型规范变换和误差修正相结合的不同预测变量的预测模型的实际预测值及其预测相对误差绝对值不仅差异甚微,而且预测值与实际值非常接近,其预测的相对误差绝对值平均值几乎全都小于3%,最大相对误差绝对值均小于5%,小于或远小于20种传统预测模型和方法预测的相应误差.     

多情景视角下中国能源消费和碳达峰路径

准确预测能源消费及碳排放量对于科学有序落实我国"2030年前碳达峰,2060年前碳中和"目标有重要现实意义.提出了一种融合位置扰动和模拟退火的改进粒子群算法（IPSO）优化基于两层"分解-集成"策略的预测方法：首先利用趋势分解（TD）将原始能源消费时序分解成趋势项和非趋势项,继而使用经验模态分解（EMD）将非趋势项分解成若干本征模态函数（IMFs）和一个残差项,然后对上述趋势项、IMFs和残差项分别建模预测,利用IPSO优化多元线性回归模型（MLR）预测趋势项,采用长短期记忆神经网络（LSTM）预测非趋势项的本征模态函数IMFs和残差子序列,最后通过相加集成求取最终能源消费预测值.实证分析表明,基于TD-EMD两层"分解-集成"策略的IPSO-MLR-LSTM模型融合了TD、EMD、IPSO和LSTM的优点,更全面地捕捉了趋势项和非趋势项演化规律,提升了预测性能,将其应用于能源消费领域是可行且有效的.最后,预测了在不同能源消费结构、经济增长、人口数量、能源效率和人均生活能源消费水平情景下的我国2021~2035年能源消费和碳排放量,并给出相关政策建议.     

神经网络在海水腐蚀预测中的应用

根据我国材料自然环境腐蚀网站长期以来积累的海水腐蚀数据,采用BP人工神经网络算法,建立了碳钢及低合金钢的海水腐蚀预测模型.该模型以合金成分、环境因素为输入参数,以平均腐蚀速率为输出参数.以碳钢、低合金钢的17种钢种在青岛、厦门、榆林海水腐蚀试验站16年腐蚀数据建模.选定A3钢与10CrCuSiV在以上三地16年的腐蚀数据为验证样本.结果表明该网络具有良好的预测精度,能够正确反映海水环境腐蚀性因素及金属材料腐蚀暴露时间与其腐蚀速率的关系,用于碳钢及低合金钢在海洋全浸环境中的腐蚀预测.     

基于wavelet-SVM的PM10浓度时序数据预测

太原是以煤炭为主要能源的重工业城市,PM_(10)(particulate matter)是太原市的主要大气污染物,因此研究其变化趋势,并给出污染物浓度预测结果,为相关部门进行大气污染防治,为突发污染事件应急提供理论支持是一项非常重要的工作.支持向量机(support vector machine,SVM)应用于PM_(10)污染物浓度时序数据预测时,表现出良好的泛化能力.在预测模型建立过程中通常选择历史数据作为学习模型的输入特征,然而这样的数据表示形式,结构单一,信息表达不完备,在很大程度上将影响预测模型的泛化能力.本文以山西省太原市城区4个监测站点的PM_(10)日浓度数据为研究数据,通过小波变换(wavelet transform)将一维输入数据转化为由低频信息和高频信息构成的高维数据,并以该数据为输入数据建立wavelet-SVM预测模型.结果表明,相较于传统SVM模型预测,wavelet-SVM模型预测结果具有更高的精度,尤其能更加准确捕捉到PM_(10)浓度突变点,为大气污染预警提供有效信息支持,并且wavelet-SVM模型对于PM_(10)浓度时序数据变化趋势的预测精度有明显提升,能更好地预测PM_(10)浓度变化趋势,揭示PM_(10)浓度时序数据内在规律.     

中国城市居民生活能源碳排放的时空格局及影响因素分析

随着经济社会的发展与人民生活水平的提高,生活用能强度逐年增大,城市居民生活能源碳排放日益成为碳排放的新增长点.本文采用Theil指数、空间自相关分析了我国城市居民生活能源碳排放的时空格局演变特征,并利用STIRPAT模型分析了影响城市居民生活能源碳排放的主要因素.结果表明:12001—2012年我国城市居民生活能源碳排放总量及人均生活能源碳排放量均呈增长趋势,其年增长率分别为9.69%、3.29%;2八大经济区域间城市居民人均生活能源碳排放的差异是构成我国城市居民人均生活能源碳排放总体差异的主要原因,其对总差异的贡献率达到了57.90%;3我国城市居民人均生活能源碳排放具有显著的空间正相关性,2001—2012年间城市居民人均生活能源碳排放的"冷点"区变化较为稳定,主要分布在东部和南部经济区,而"热点"区主要分布在西北、东北和黄河中游经济区;4城市人口规模、城市居民可支配收入、城市居民生活消费支出、城市居民年龄结构均对城市居民生活能源碳排放量具有加剧作用,而城市居民能源消费结构具有减缓作用,且北方城市居民生活能源碳排放量明显高于南方;5现有样本数据支持环境Kuznets曲线假说,即随着经济的发展,城市居民生活能源碳排放量存在转折点.     

珠江三角洲典型城市机动车尾气排放的影响因素分析

机动车技术水平和道路路况是影响机动车尾气排放的两大因素。本文以中山、佛山和广州三个城市为代表城市,首先采用问卷调查和实时采集交通流视频的方式收集三个城市的机动车技术水平信息和道路路况,并分析了上述三个城市的燃油类型和累计行驶里程和交通流的分布特征。分析结果衰明,备城市的机动车技术水平特征及差距,其中广州的出租车、公交车已规模推广环保型燃料,各类车型燃料使用比例不同使得尾气成分有一定差异;广州公交车累计行驶里程和年均行驶里程分布较靠后,公交系统的使用率较高;出租车年均行驶里程低于佛山、中山。在交通流分布上,广州总交通量大于中山;广州各级道路交通量呈梯度递减分布,小时交通量差距较大;中山仅国道承担较大范围的运输,其他各级道路分布形状统一,非工作类出行比例低。     

大气腐蚀数据降维最优维度研究

目的确定金属大气腐蚀数据降维的最优维度。方法分别采用PCA、MDS、Isomap和LLE四种方法对大气腐蚀数据进行降维处理,并采用集成学习算法建立预测模型。针对不同的降维方法和近邻点个数计算,使用MAPE(Mean Absolute Percentage Error,相对百分误差绝对值的平均值)对预测结果进行评价,将最佳预测速率所对应的维度作为最优维度。结果不同的降维方法和近邻参数作用下,最优维度从2维到7维不等。流形学习方法对大气腐蚀数据进行降维的MAPE均小于线性降维方法。结论适合每种降维方法的最优维度可能是不同的,通过对MAPE进行对比,得到大气腐蚀数据在各种降维方法的最优维度。     

城市汽车年检碳排放数据挖掘及其应用研究

基于城市汽车年检中CO₂的过程动态检测数据,对小型汽油客车CO₂的排放因子和排放水平信息进行了挖掘,并与文献研究结果进行了对比.结果表明,从城市汽车年检数据中挖掘的CO₂排放结果可以作为研究城市机动车碳排放的重要参考依据;合理有效地利用汽车年检数据可以为城市车辆尾气排放的精准分级管控、城市交通运输碳排放达峰的量化分析,以及城市交通源的污染物和CO₂协同减排提供重要的数据支撑;从国Ⅰ到国Ⅴ不同排放阶段汽油车的常规污染物CO、NO_x和HC的排放水平下降非常明显,而对应的CO₂排放水平差异不大;CO₂排放因子随累计行驶里程、车龄、基准质量和排量的变化关系反映出,如果需要削减城市汽车碳排放水平,应鼓励使用基准质量小或者排量小的车辆,淘汰高油耗高排放的老旧车辆,鼓励公共绿色出行而降低单车活动水平,增加纯电动车辆优化车队能源结构.     

基于精细化年均行驶里程建立机动车排放清单

基于环保检测数据，提出“里程-车龄”曲线用以获取满足“车辆类型-燃料种类-排放标准”三级分类的精细化年均行驶里程.使用《道路机动车大气污染物排放清单编制技术指南（试行）》推荐值、车辆类型均值、“里程-车龄”曲线3种方式获取年均行驶里程并分别建立排放清单，发现年均行驶里程的本地化与精细化可以极大降低行驶里程不确定性对排放清单准确性的影响.采用精细化年均行驶里程，计算得到青岛市2017年机动车CO、VOCs、NO_x、PM₁₀、NH₃和SO₂的排放量分别为7.07，1.14，2.84，0.10，0.08和0.08万t.分析排放构成可知，老旧车淘汰在当前仍可作为青岛市机动车排放治理的有效举措.结合路网信息与交通数据，得到0.01°×0.01°高时空分辨率网格化排放清单.结果表明，青岛市机动车排放分布在不同时段变化明显.以NO_x为例，排放的早晚高峰分别出现在8：00与17：00，占到了全天总排放的8.17%和7.53%.同时，排放分布存在着空间异质性，排放从城市中心至边缘呈逐渐降低趋势，沿高速路呈明显带状分布.     

Real-world fuel efficiency and exhaust emissions of light-duty diesel vehicles and their correlation with road conditions

The real-world fuel efficiency and exhaust emission profiles of CO, HC and NO_x for light-duty diesel vehicles were investigated. Using a portable emissions measurement system, 16 diesel taxies were tested on different roads in Macao and the data were normalized with the vehicle specific power bin method. The 11 Toyota Corolla diesel taxies have very good fuel economy of (5.9 ± 0.6) L/100 km, while other five diesel taxies showed relatively high values at (8.5 ± 1.7) L/100 km due to the variation in transmission systems and emission control strategies. Compared to similar Corolla gasoline models, the diesel cars confirmed an advantage of ca. 20% higher fuel efficiency. HC and CO emissions of all the 16 taxies are quite low, with the average at (0.05 ± 0.02) g/km and (0.38 ± 0.15) g/km, respectively. The average NO_x emission factor of the 11 Corolla taxies is (0.56 ± 0.17) g/km, about three times higher than their gasoline counterparts. Two of the three Hyundai Sonata taxies, configured with exhaust gas recirculation (EGR) + diesel oxidation catalyst (DOC) emission control strategies, indicated significantly higher NO₂ emissions and NO₂/NO_x ratios than other diesel taxies and consequently trigger a concern of possibly adverse impacts on ozone pollution in urban areas with this technology combination. A clear and similar pattern for fuel consumption and for each of the three gaseous pollutant emissions with various road conditions was identified. To save energy and mitigate CO₂ emissions as well as other gaseous pollutant emissions in urban area, traffic planning also needs improvement.     

怠速起停对汽油车油耗及颗粒数量排放的影响

以一辆配置怠速起停系统的国V缸内直喷汽油车为研究对象,使用底盘测功机试验系统、全流稀释采样系统和固态颗粒计数系统,试验研究怠速起停对缸内直喷汽油车油耗与颗粒数量排放的影响,并分析起动温度、试验循环等因素的影响.结果表明,NEDC循环车辆冷机起动时,车辆怠速起停系统开启的百公里油耗降低了5.1%,颗粒数量排放升高了16.7%;车辆热机起动,怠速起停开启产生的百公里油耗降幅增大到7.3%,颗粒数量排放升幅减小至9.3%;WLTC循环热机起动,车辆怠速起停系统开启的百公里油耗降幅减少到1.7%,颗粒数量排放升幅减少到6.2%.怠速起停有利于降低汽车的百公里油耗,但不利于缸内直喷汽油车颗粒数量排放的控制.     

Variability of fuel consumption and CO2 emissions of a gasoline passenger car under multiple in-laboratory and on-road testing conditions

An increasing divergence regarding fuel consumption(and/or CO₂ emissions) between realworld and type-approval values for light-duty gasoline vehicles(LDGVs) has posed severe challenges to mitigating greenhouse gases(GHGs) and achieving carbon emissions peak and neutrality. To address this divergence issue, laboratory test cycles with more real-featured and transient traffic patterns have been developed recently, for example, the China Lightduty Vehicle Test Cycle for Passenger cars(CL...     

Influence of driving cycles on exhaust emissions and fuel consumption of gasoline passenger car in Bangkok

The influence of different driving cycles on their exhaust emissions and fuel consumption rate of gasoline passenger car was investigated in Bangkok based on the actual measurements obtained from a test vehicle driving on a standard chassis dynamometer. A newly established Bangkok driving cycle (BDC) and the European driving cycle (EDC) which is presently adopted as the legislative cycle for testing automobiles registered in Thailand were used. The newly developed BDC is constructed using the driving characteristic data obtained from the real on-road driving tests along selected traffic routes. A method for selecting appropriate road routes for real driving tests is also introduced. Variations of keyed driving parameters of BDC with different driving cycles were discussed. The results showed that the HC and CO emission factors of BDC are almost two and four times greater than those of EDC, respectively. Although the difference in the NOx emission factor is small, the value from BDC is still greater than that of EDC by 10%. Under BDC, the test vehicle consumes fuel about 25% more than it does under EDC. All these differences are mainly attributed to the greater proportion of idle periods and higher fluctuations of vehicle speed in the BDC cycle. This result indicated that the exhausted emissions and fuel consumption of vehicles obtained from tests under the legislative modal-type driving cycle (EDC) are significantly different from those actually produced under real traffic conditions especially during peak periods.