1、_ 6$(7HFKQLFDO6WDQGDUGV%RDUG5XOHVSURYLGHWKDW7KLVUHSRUWLVSXEOLVKHGE6$(WRDGYDQFHWKHVWDWHRIWHFKQLFDO and engineering sciences. The use of this report is entirely voluntary, and its applicability and suitability for any particular use, including any patent infringement arising tKHUHIURPLVWKHVROHUHVSRQVL
2、ELOLWRIWKHXVHU SAE reviews each technical report at least every five years at which time it may be revised, reaffirmed, stabilized, or cancelled. SAE invites your written comments and suggestions. Copyright 2014 SAE International All rights reserved. No part of this publication may be reproduced, st
3、ored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE. TO PLACE A DOCUMENT ORDER: Tel: 877-606-7323 (inside USA and Canada) Tel: +1 724-776-4970 (outside USA) Fax: 724-776-079
4、0 Email: CustomerServicesae.org SAE WEB ADDRESS: http:/www.sae.org SAE values your input. To provide feedback on this Technical Report, please visit http:/www.sae.org/technical/standards/J2951_201401 SURFACE VEHICLE RECOMMENDED PRACTICE J2951 JAN2014 Issued 2011-11 Revised 2014-01 Superseding J2951
5、NOV2011 Drive Quality Evaluation for Chassis Dynamometer Testing RATIONALE To provide standardized metrics for evaluating drive quality on emissions and fuel economy tests. This document has been revised to include a new drive rating metric and typical driver capability ranges. FOREWORD It is genera
6、lly recognized that the manner in which a vehicle is driven during a chassis dynamometer test can impact emissions and fuel economy results. The speed vs. time tolerances used to validate a test do limit this impact, but even within these constraints drive-related effects can be significant contribu
7、tors to test variability. This document provides drive quality metrics intended to enable improved monitoring and characterization of driver-related variability. TABLE OF CONTENTS 1. SCOPE 2 2. REFERENCES 2 2.1 Applicable Documents 2 2.1.1 SAE Publications . 3 2.2 CFR Publications 3 3. DEFINITIONS .
8、 3 4. DRIVING SCHEDULES 6 4.1 UDDS 6 4.2 HFEDS 6 4.3 US06 . 6 4.4 SC03 . 6 4.5 Certification and Test Procedure 40 CFR Part 600 Fuel Economy of Motor Vehicles 3. DEFINITIONS 3.1 ETW CLASS (EQUIVALENT TEST WEIGHT) Test mass dictated by U.S. Code of Federal Regulations that is assigned to represent a
9、class of test vehicles (40 CFR 86.129-80). ETW is a weight class, and is not necessarily equal to the as-tested weight of a vehicle. 3.2 DYNAMOMETER SET INERTIA (MSET) The setting that specifies the inertia that is to be simulated by the dynamometer. The MSETequals ETW for regulatory testing using 2
10、WD chassis dynamometers. For testing on a 4WD chassis dynamometer, MSETequals 98.5% of ETW. 3.3 EFFECTIVE TEST MASS (ME) Effective Test Mass (ME) is the sum of 1) the dyno-simulated inertia (MSET) and 2) the effective inertia of the vehicle components (e.g. wheels, axles) that are rotated on the dyn
11、amometer. This value describes the total inertial load acting on the vehicle system, and is required to calculate the inertial component of cycle energy. For light-duty vehicles the effective inertia of the rotating components, per axle, may be estimated by taking 1.5% of the ETW. However vehicles w
12、ith other than single, normal-sized wheels, such as dual-wheel trucks, may require specific estimation or determination of the effective mass of the rotating drivetrain components. Using 1.5% of ETW per axle, and the defintion of MSETabove, gives the following equation for determining the effective
13、test mass for both 2WD and 4WD dynamometer testing: ETW1.015 ME (Eq. 1)3.4 Dyno Target Coefficients: Fx(F0, F1and F2) Target coefficients describe the total force (tire, drivetrain and aerodynamic drag) acting on a vehicle during an on-road coastdown. These coefficients are developed from track data
14、 (and/or equivalent analytical methodology), corrected to standard conditions, and possibly adjusted to account for differences between vehicle weight as tested on the track and weight represented by an ETW class assigned for dynamometer testing. SAE INTERNATIONAL J2951 Revised JAN2014 Page 4 of 27
15、3.5 SIMULATION MODE The operating mode where the dynamometer simulates the vehicle inertia and road load commanded by the dynamometer set inertia (MSET) and Dyno Set coefficients (Dx), respectively, so that a vehicle driven on the dynamometer operates as it would on the road. 3.6 VEHICLE SPEED (V) 3
16、.6.1 Roll Speed (VROLL) The inferred vehicle speed as measured by the dynamometer. Roll encoder speed sampled at 10Hz shall be used as the roll speed, and shall be the same or equivalent speed signal that is used to determine conformance with the speed vs. time tolerance in 40 CFR Part 86.115-78 App
17、endix 1. 3.6.2 Scheduled Speed (VSCHED) The target vehicle speed as specified by the speed vs. time requirements for a drive schedule. Scheduled speed is defined by a smooth trace drawn through the specified speed versus time relationship (40CFR 600.10978). A linear interpolation between the 1Hz spe
18、ed points given in the CFR shall be used to produce the 10Hz scheduled speed trace (VSCHED). 3.6.3 Vehicle Speed Driven (VD) The vehicle speed derived from the roll speed data (VROLL) for the purposes of calculating the drive metrics described in this document. The driven vehicle speed is used for t
19、he calculation of driven cycle energy and is calculated by taking a 0.5 s, double moving average of the 10Hz roll speed signal. After performing the double moving average, all values less than or equal to 0.03 m/s are set to zero. This is done to reduce the impact of noise in the roll speed signal o
20、n the results. A moving average was chosen over other filter types as it provides the best compromise between smoothing the time series and preserving the response time of the signal. The subscript “D“ will be used to refer to quantities calculated from the driven vehicle speed. 3.6.4 Vehicle Speed
21、Target (VT) The target vehicle speed, calculated in the same manner as VDbut using the scheduled speed (VSCHED) instead of the roll speed (VROLL). The target vehicle speed is used for the calculation of target cycle energy. The subscript “T“ will be used to refer to quantities calculated from the ta
22、rget vehicle speed. 3.7 SAMPLING PERIOD (t) The time between successive samples of VROLLand VSCHED. The calculations in this document require a sampling frequency of 10 Hz, which corresponds to a sampling period of 0.1 s. Higher sampling frequencies may be used if 10 Hz is not possible, however the
23、data must first be downsampled to 10 Hz, using good engineering judgement to ensure representative results, in order to maintain compatibility with the finite-difference calculations defined in Section 5.1. 3.8 VEHICLE ACCELERATION (a) The acceleration of the vehicle calculated as the time-rate-of-c
24、hange of the vehicle speed (V). The specifics of this calculation are detailed in Section 5.1.2. SAE INTERNATIONAL J2951 Revised JAN2014 Page 5 of 27 3.9 ROAD LOAD FORCE (FRL) The combination of intrinsic and dyno-simulated forces opposing the vehicles motion on the dynamometer that are intended to
25、duplicate the internal and external vehicle parasitic forces the engine must work against while driving on the road. These forces are primarily comprised of aerodynamic drag, driveline parasitic losses and tire rolling resistance. The road load force is calculated using vehicle speed (V) and the Dyn
26、o TargetCoefficients. 2V2FV1F0F FRL (Eq. 2)For testing performed at 20 F (-7 C), the road load force should be approximated as 1.10 x FRLunless road load coefficients derived at 20F are used. 3.10 INERTIAL FORCE (FI) The combination of intrinsic and dyno-simulated forces opposing the vehicles motion
27、 on the dynamometer that represents the effect of its mass and the rotational inertia of its driveline components while driving on the road. Inertial force is calculated using the effective test mass (ME) and vehicle acceleration (a). a M FEI (Eq. 3) Substituting Eq. 1 (See Section 3.3) for MEgives
28、a ETW1.015 FI (Eq. 4)3.11 ENGINE FORCE (FENG) The sum of the inertial force (FI) and the road load force (FRL). The engine force represents the sum of all the forces that oppose the vehicles motion while driving on the dyno. It is equal to the sum of the road load force and inertial force when this
29、sum is positive, and zero when this sum is negative. (A negative sum of FRLand FI is interpreted as braking, and not considered “engine“ force.) IRLENGFFF (Eq. 5)The term “engine“ is a general reference to the power-generating system of the vehicle, and its use is not restricted to systems that use
30、an internal combustion engine. In hybrid electric or battery-electric vehicles FENGmay represent, in part or in whole, work that is done by an electric motor. 3.12 DISTANCE INCREMENT (d) The incremental distance traveled by the vehicle during each sampled data point, calculated from the 10 Hz vehicl
31、e speed (V) and the sampling period (t). 3.13 ACCUMULATED DISTANCE (D) The total distance traveled by the vehicle, calculated by summing the distance increments (d) over the test cycle. 3.14 ENGINE WORK INCREMENT (w) The incremental work done by the vehicle during each sampled data point, calculated
32、 by multiplying the engine force (FENG) by the distance increment (d) for each 10 Hz sampled data point. SAE INTERNATIONAL J2951 Revised JAN2014 Page 6 of 27 3.15 CYCLE ENERGY (CE) The net energy a vehicle must provide in order to drive a test cycle on a chassis dynamometer. Cycle energy is calculat
33、ed by summing the engine work increments (w) over the test cycle. By definition, the engine work increments are always positive (since FENG VRQHJDWLYHZRUNLVHFOXGHGIURPWKHFFOHHQHUJVXPPDWLRQ. The exclusion of negative work is appropriate since it is associated primarily with braking events and represe
34、nts energy that is not recovered unless the vehicle is equipped with a regenerative braking system (alternative equations for the summation of work for vehicles equipped with regenerative braking may be considered in future revisions of this document). Note that the engine work increment (w) may sti
35、ll be positive even during decelerations. This is because even though the inertial force is negative during decelerations, engine force (FENG) will still be positive if the magnitude of the road load force is greater than the magnitude of the inertial force. In this case the engine is still doing wo
36、rk, but the output from the engine is insufficient to maintain the vehicles speed. 4. DRIVING SCHEDULES There are five driving schedules referenced in this document which are required by the EPA and the California Air Resources Board during emissions and fuel economy certification of vehicles with i
37、nternal combustion engines. They are the Urban Dynamometer Driving Schedule (UDDS), the “Cold“ UDDS, the Highway Fuel Economy Driving Schedule (HFEDS), the US06 Driving Schedule (US06), and the SC03 Driving Schedule (SC03). 4.1 UDDS The Urban Dynamometer Driving Schedule is defined in 40 CFR Part 86
38、, Appendix 1. It has a duration of 22 min, 52 s. It is used to represent vehicle city driving. 4.2 HFEDS The Highway Fuel Economy Driving Schedule is defined in 40 CFR Part 600, Appendix 1. It has a duration of 12 min, 45 s. It is used to represent vehicle highway driving. The Highway Fuel Economy T
39、est (HFET) consists of two HFEDS cycles. 4.3 US06 The US06 Driving Schedule is defined in 40 CFR Part 86, Appendix 1. It has a duration of 10 min. It is used to represent vehicles driving at high speeds and with aggressive accelerations. Dynamometer load reduction for low-powered vehicles may be use
40、d in accordance with 40 CFR Part 86.108 00(b)(2)(ii). The US06 cycle is subdivided into a “City“ test (0-130 s and 495-600 s) and a “Highway“ test (130-495 s) for the purposes of 5-cycle fuel economy labeling 40 CFR 86.159-08 Exhaust emission test procedures for US06 emissions. NOTE: If dynamometer
41、load reduction for low-powered vehicles is utilized, the energy-related evaluations outlined in this Recommended Practice are not valid. 4.4 SC03 The SC03 Driving Schedule is defined in 40 CFR Part 86, Appendix 1. It has a duration of 10 min. It is used to represent vehicle operation with air condit
42、ioning. 4.5 5. : 6 ? 5 ; 6. : ? 5 ; . 5 6 ? A5 6(Eq. D1) Where there are k drivers, nkis the number of samples for each driver, and skis the standard deviation of each drive metric for each driver. The data was collected before any limits or requirements for the drive ratings were defined. To use th
43、e data collected to determine a range of reasonable ratings, two values were calculated for EER, ASCR, and RMSSE which are shown below in Table D1. TABLE C1 - RANGES FOR DRIVE METRICS FROM DRIVER VARIABILITY STUDY The first value shown for each metric is the Maximum Capability Range. This value is t
44、he largest pooled standard GHYLDWLRQUHSRUWHGEDQODEIRUWKHJLYHQFFOH7KHYDOXHUHSRUWHGLV1VRDVWRLQFOXGHRIGULYHV,WLVEHOLHYH d that without any requirements in place, drivers should typically be able to meet the Maximum Capability Range. To develop a target level for drive metrics, the Average Capability Ra
45、nge is also shown. This value is the average of all of the pooled standard deviations reported by all labs and gives equal weighting to the value reported by each lab. As with WKH0DLPXP % Convert driven speed profile from mph to m/s. Vsched = Vs*0.44704; % Convert scheduled speed profile from mph to
46、 m/s. % Dyno Target Coefficients F0=ABCs(1)*4.448; % lbf to N F1=ABCs(2)*9.9504; % lbf/mph to N/(m/s) F2=ABCs(3)*22.25839; % lbf/(mph2) to N/(m/s)2) % Equivalent Test Weight ETW=Mass*0.4536; % Conversion from lb to kg % Effective Test mass Me=1.015*ETW; % SPEED CALCULATIONS / FILTERING 1 Vd_tmp=0; V
47、d_tmp(2)=0; Vt_tmp=0; Vt_tmp(2)=0; for i=3:length(time)-2, Vd_tmp(i)=1/5*(Vroll(i-2)+Vroll(i-1)+Vroll(i)+Vroll(i+1)+Vroll(i+2); Vt_tmp(i)=1/5*(Vsched(i-2)+Vsched(i-1)+Vsched(i)+Vsched(i+1). +Vsched(i+2); endVd_tmp(length(time)-1)=0; Vd_tmp(length(time)=0); Vt_tmp(length(time)-1)=0; Vt_tmp(length(time)=0; clear i Vd=Vd_tmp; Vt=Vt_tmp; clear Vt_tmp Vd_tmp % SPEED CALCULATIONS / FILTERING 2 Vd_tmp=0; Vd_tmp(2)=0; Vt_tmp=0; Vt_tmp(2)=0; for i=3:length(time)-2, Vd_tmp(i)=1/5*(Vd(i-2)+Vd
