Low NO X Diesel Engine. Steps. Christopher Sharp 2017 CLEERS Workshop October 3, 2017

Transcription

1 Low NO X Diesel Engine Program Update and Next Steps Christopher Sharp 217 CLEERS Workshop October 3, 217 1

2 CARB Low NO X Programs at SwRI Stage 1 Evaluating Technologies and Methods to Lower Nitrogen Oxide Emissions from Heavy-Duty Vehicles ( ) CNG and Diesel Initial i Technology Evaluation Primary focus on Regulatory Cycles Completed Stage 2 Heavy-Duty Low Load Emission Control ( ) Diesel Expand previous technology evaluation to low-load load and urban operating cycles Stage 3 Further Evaluation and Development of Low NOX Technologies on 217 (non-turbocompound) Engine Platform ( ) 218) Diesel Focus on both Low Load (Real world) and Regulatory cycles 2

3 Stage 1 Program Engine 214 Volvo MD13TC Euro VI A diesel engine with cooled EGR, DPF and SCR 1477 rpm rpm Representative of OEM s planned U.S. direction for future GHG standards on Tractor engines Incorporates waste heat recovery mechanical turbo-compound (TC) Tailpipe NOx, g/hp hrhr FTP RMC Average SD COV 8.5% 11% SD % Std 5.9% 4.6% Engine out NO X ~ 3 g/hp hr No tailpipe NH 3 Tailpipe N 2 O ~.5 g/hp hr CO2, g/hp hr MD13TC Baseline Vocational (FTP) 217 GHG Standards Tractor (SET) 3

4 Final ARB Stage 1 Low NO X Configuration All catalysts are coated on 13 diameter substrates SCRF is 13 X 12 on high porosity filter substrate Remaining catalysts are 13 X 6 on thin wall, low thermal mass substrates All sensors shown are production-type Modified engine calibration NO 1 X Levels with Development Aged Parts, g/hp hr Cold FTP Hot FTP Composite RMC SET Engine Out Tailpipe hours at 65 C 4

5 Model-Based SCR Controller with Mid-Bed NH 3 SCR Model T Sensor Feedback In Thermal Cell T In ṁ exh Model ṁ exh SCR Model Cell Thermal Model T In ṁ exh SCR Model Cell Thermal Model T In ṁ exh NO X T wall NO X T wall NO X T wall NO X NH 3 NO 2 / NO X Kinetic Model NH 3 NO 2 / NO X Kinetic Model NH 3 NO 2 / NO X Kinetic Model NH 3 NO 2 / NO X θ 1 θ 2 θ 3 Primary controls challenge is repeatable hot-start/warmed-up NO X <.1 g/hp-hr Separate coverage observer models for SCRF and SCR (7 cell model for each) Primary calibration parameters are controller gains and coverage targets Same calibration used for FTP, RMC-SET, CARB Idle, vocational cycles Slightly modified coverage targets for WHTC 5

6 Cold-FTP NO X and Temperature -45 seconds Final System, DevelAged Parts PNA function most important during first 9 seconds s of cold-start t Thermal management active until ~ 375 secs Mini-burner, engine calibration, elevated idle SCRF light off at 125 secs Full SCR conversion by 22 secs Development Aged parts Cold FTP result =.6 g/hp hr 6

7 Hot-FTP Thermal Management Final System, Devel Aged Parts Te emperature, de egc PNA In SCRF In SCRF In with TM SCRF Inlet T With TM ~.8 g/hp hrhr Time, sec Te emperature, de egc PNA In SCRF Out SCRF Out with TM SCRF Outlet T Time, sec These idle segments from the engine result in a small drop in SCRF and SCR catalyst temperature (down to 175C with no intervention) Small cool wave passes through the catalysts and requires some intervention This makes a difference between hot-starts <.1 g/hp-hr and >.15 g/hp-hr Engine calibration changes at idle made to help minimize, but engine alone was not enough Small amount of thermal management from the mini-burner was used as a countermeasure 7

8 Final Aging Protocol ] SCRF Inlet Temperature [ C] g/hr Soot Rate Exhaust Flow = 975 kg/hr Low Temperature Soot & Ash Accumulation Mode Time [] [s] Mode ssive Oxidation / Ash Accumulation Pa Soot & A Operation / oval Mode High Temp HC Remo Active Regenerat tion Mode Accelerated Aging g based on SwRI DAAAC protocol 29 Cummins ISX mule engine (DAAAC modified) 4-hour duration Regeneration is via in-exhaust injection upstream of PNA Final duration was 847 hours 1% FUL thermal exposure 23% FUL chemical exposure This is based on regeneration frequency of ~ 1.7% (near x2 from base engine) Resulted in 194 hours of regeneration for FUL thermal equivalent This is more than 3 Active Regeneration events 8

9 Final Aging - Issues Early PNA face coking resolved by adjusting cycle but resulted in large HC buildup that had to be baked off Regeneration process had to be adjusted to insure complete soot cleaning some early localized exotherms possible Mechanical Canning failure on PNA at 71 hours PNA mat failed Large buildup of HC and soot on PNA had to be recovered Ingestion of mat into SCRF (mal-distribution and local exotherms?) had to be mechanically removed without disturbing deep ash PNA SCRF Inlet SCRF Channels Abnormal Mt/Ah Mat/Ash Normal Ash Load 9

10 1 Final Tailpipe NO X Results.2.71 g/hp hr Tailpipe NO X, Baseline Degreened Devel Aged Final Aged Note: IRAF calculation l for NO X would add adjustment of +.4 g/hp hr to Final Aged results with Low NO X Engine (+.3 g/hp hr for Devel Aged and Degreened). Aftertreatment NO X Conversion Efficiency, % X Test Config FTP Transient Cold Hot Composite RMC SET WHTC Baseline 75% 98.5% 95% 97% 97% Devel Aged 98% 99.7% 99.5% 99.3% 99.4% Final Aged 96% 99.3% 98.8% 98.2% 98.8% 1

11 Cold-FTP Final Aged vs Devel Aged PNA Performance PNA Stored NOx, grams Devel Aged Final Controls Final Aged Inlet Temp SCRF Light Off Full SCR Conversion Temp, degc PNA Inlet NO X reduction across system components 2 secs, % Full Cycle, NO X conv % NO X conv Devel Aged Final Aged Devel Aged Final Aged PNA 44% 27% 1% 5% SCRF 64% 28% 9% 84% SCR SCR/ASC 1% 13% 8% 8% Time, sec Cold-start performance change is primarily due to loss of NO X storage capacity on PNA More NO X reaches SCRF before it reaches light-off temperature, downstream SCR still too cold to help 11

12 Hot-FTP Final Aged vs Devel Aged SCRF Performance Devel Aged Final Controls Final Aged Devel Aged Final Controls Final Aged SCRF Out Final Controls = 9%,.3 g/hp hr Final Aged = 87%,.4 g/hp hr Tailpipe Final Controls = 99.7%,.9 g/hp hr Final Aged = 99.3%,.2 g/hp hr SCRF Out NOx, g/hr Tailpipe NOx, g/hr Time, sec Time, sec Hot-start performance change due primarily to change in SCRF performance Lower NH 3 storage capacity, higher tendency towards ammonia oxidation More demand on downstream SCR catalyst Early cycle tailpipe performance still maintained but later there is more NH 3 release to downstream catalyst Small increase in late cycle NO generation due to larger amount of NH 3 to be oxidized 12

13 Final GHG Results Cycle Measured CO 2 and N 2O Emissions Overall CO 2 / Fuel Consumption Impact WHTC very similar to FTP Slight increase for Final Aged (about.3%) due to backpressure and slightly higher MB fueling to reach temperature thresholds CO 2 impact on FTP driven by low temperatures res from turbocompound Different GHG approach would require less thermal management Impact could be reduced via better packaging and integration 13

14 Example Vocational Cycle NYBCx4 EO NOx TP NOx DOC In T DPF Out T Aftertreatment Out T Speed EO NOx TP NOx PNA In T SCRF In T SCR In T Speed NOx, g/h hr or Temp, degc Speed NOx, g/h or Temp, degc Speed, rpm Time, sec Baseline Engine EO, g/hphr TP, g/hphr 4 NOx Conversion, % Fuel Rate, kg/hr Baseline % 5.3 ULN Engine % 5.3 % Change 35% 84% n/a None Time, sec 4 Ultra Low NO X Engine (Final Aged Parts) Duty cycle is average 6% of maximum engine power (not including idle segment) Test cycle starts after the idle Precondition before idle with same cycle 14

15 Next Steps - Stage 1b Background and Plan Re-aging and testing ti of new set of Stage 1 parts Answer questions about real system durability after canning failure Provide robust set of parts for Stage 2 (low load cycle) demonstration parts Test Plan Obtain new set of de-greened Stage 1 parts Repeat the 1-hour aging process Test points Initial (-hour), 333 hr, 667 hr, 1 hr (final) intermediate test points this time to allow better monitoring of change in performance during aging Also includes a re-baseline on the engine updated baseline for more robust GHG comparison (same cell, closer in time) October 217 March 218

16 Next Steps - Stage 2 Background and Objectives Next program to follow-on from current ARB Demonstration Program already awarded - $1M in additional funding Stage 2 Program focus is Low-temperature and Low Load (urban) Vocational duty cycles Key Topics Development of Low-Load Load duty cycle profiles Development of a Heavy-Duty Low Load Cycle Re-calibration of ARB Diesel Demonstration Engine to achieve low NO x on Low Load profiles What is the impact on GHG for this kind of control? Appropriate load metrics for in-use testing at Low Load Program is currently in progress focus has been on Low Load cycle development 16

17 Stage 2 Low Load Cycle Development Analysis of In-Use Vehicle Data Fleet DNA and CARB (CE-CERT) Combined Dataset: 751 unique vehicles located across the US ~6+ Gb of raw data 25 Distinct Locations 44 Unique Vocational Designations 55 Unique Fleets 17

18 Stage 2 Low Load Cycle Development Example Candidate Profile Cluster 7 Temperature e, degc Aftrtrtmnt1DslPrtcltFltrInt_ Aftrtrtmnt1PrtcltTrpOtltGasTemp EngExhaustGasTemp speed Time, sec This profile represents shows a cooling trend after warmer temperature operation, but still all cool temperatures Vehicle Spee ed, mph 18

19 Stage 2 Low Load Cycle Development Example Candidate Profile Cluster 1 partial Aftrtrtmnt1DslPrtcltFltrInt_ EngExhaustGasTemp Aftrtrtmnt1PrtcltTrpOtltGasTemp speed Temperatur e, degc Time, sec Vehicle Spee ed, mph This profile represents shows a warming trend after a period of cool operation 19

20 Stage 2 Low Load Cycle Development Status We are screening through the initial set of 12 candidate profiles identified by the NREL analysis some may have unrepresentative events (long key-off periods, etc.) we do not want some we may want another data file from the same cluster that has more complete data set Develop the final set of 1 low-load vehicle profiles This will enable us to start low-load engine development Work with these profiles to develop combined Low Load cycle Analytical tools and approach will be useful in other upcoming data analysis efforts 2

21 Next Steps Stage 3 Motivation for Stage 3 Stage 1 Concerns Stage 1 engine choice appears to be more of a corner case regarding architecture in 217+ Very low exhaust temperatures due to turbocompound no capability to bypass or work around this Engine architecture limited the ability to make engine modifications within the program scope Limited flexibility on engine-out NO X and temperature Further constrained for some AT technologies due to AFR control limitations Combination resulted in a more complicated AT system SCRF, NO X storage (PNA), large supplemental heat input (mini-burner) Large heat input resulted in substantial GHG impact Essentially eliminated i benefit of turbocompound, especially in transient cycles

22 Objectives for Stage 3 Stage 1 result appears to be closer to a worstcase example May not be representative of wider market More complicated and costly system than other potential examples Stage 1 Focus FTP Desire is to examine a platform that is more representative of broader market in 217+ Is a less complicated system feasible for the broader Stage 2 Focus Low industry? Load and Vocational Is a lower GHG impact apparent without high temperature WHR? Integrate both Regulatory and Low Load Cycles at the start Stage 1 focused on FTP / WHTC / RMC low load added later Stage 2 effort will provide realistic low load and vocational duty cycle profiles that can be used for development Stage 3 target is control on all cycles Stage 3 Focus Control On All Cycles

23 Stage 3 Overall Timeline Task ARB ARB n/a SCAQMD SCAQMD ARB MECA ARB ARB ARB Procure Engine and AT Install and Baseline Description Aftertreatment Delivery (and Supplier Tests) Engine Calibration and AT Selection Aftertreatment Integration and Emission Test Low Load Optimization Final Aging Final Calibration and Demonstration Baseline Recheck Final Reporting Sep 17 Oct 17 Nov 17 Dec 17 Jan 18 Feb 18 Mar 18 Apr 18 May 18 Jun 18 Jul 18 Aug 18 Sep 18 Oct 18 Nov 18 Dec 18 Jan 19 Feb 19 Mar 19 Apr 19 May 19 Jun Program start anticipated October 217 waiting on completion of CARB contracts to begin engine availability at SwRI in October Likely program duration is through early 219

24 Stage 3 Early Program Timing Program start anticipated October 217 Engine availability in similar timeframe Initial tasks will be engine baseline and initial engine calibration efforts There will be a gap in the program of ~ 5 months in order to allow lead time for AT hardware supply Engine will be available in test cell at SwRI Probably in the Dec 217 / Jan 218 timeframe Opportunity to use this gap to explore additional engine hardware options Sep 217 Oct 217 Nov 217 Dec 217 Jan 218 Feb 218 Mar 218 Apr 218 May 218 June 218 Expected Start ARB Baseline Install and Initial Efforts Time Gap for AT Procurement

25 Engine Hardware List for Potential Stage 3 Added Work Cylinder deactivation Turbine bypass EGR cooler bypass Exhaust manifold material/construction in conjunction with turbine bypass More complicated (still being investigated) variable speed coolant pump early exhaust valve opening (EEVO) Hardware/design/engineering provided by suppliers SwRI scope is installation, integration, controls, calibration

26 More Information California ARB website emissions/low-nox/low-nox.htm SwRI Contact Christopher h Sharp chris.sharp@swri.org 26