Signal Hardware-In-the-Loop simulation of a Hybrid locomotive

Transcription

1 International Workshop on HIL simulation HIL 16 summer school Lille, 1-2 September Signal Hardware-In-the-Loop simulation of a Hybrid locomotive Dr. T. Letrouvé 1,3, Dr. J. Pouget 1,3, Dr. W. Lhomme 2,3, Prof. A. Bouscayrol 2,3 1 SNCF Innovation & Research 2 Univ. Lille1, L2EP 3 MEGEVH network Lille, France 1 international Workshop on HIL'16 - Summer school, Lille, France September 1 st and 2 nd

2 Actual approach Customer needs Sizing and power flow studies Prototype elaboration Prototype development Prototype behavior validation Feasability response Direct validation of component behavior in real time on vehicle, Components homologation (hardware), All physics phenomenon are taken into account. 2 international Workshop on HIL'16 - Summer school, Lille, France

3 Actual approach Customer needs Time to market Sizing and power flow studies Prototype elaboration Prototype development Prototype behavior validation x2 x3 Cumulative dev. cost Feasability response Sizing Prototype Complex to develop, more than one prototype needed, Time needed, The development Cost is exponential with the time on prototype, A lot of resources are needed (prototype, staffs and tracks availability), unrepeatable tests, Safety and fault tolerance tests are made on-line. 3 international Workshop on HIL'16 - Summer school, Lille, France

4 Actual approach Customer needs Time to market Sizing and power flow studies Prototype elaboration Prototype development Prototype behavior validation Cumulative dev. cost Feasability response Sizing Prototype Adding intermediary steps: Reduce the time on prototype (reduce the cost and time of development), Virtual homologation of some subsystems ahead of time. Objective : Adding intermediaries steps 4 international Workshop on HIL'16 - Summer school, Lille, France

5 Outline HIL simulation interest and structuring problematic Signal HIL simulation of PLATHEE locomotive Conclusion and outlooks 5 international Workshop on HIL'16 - Summer school, Lille, France

6 Sommaire Hardware-In-the-Loop simulation : hybrid locomotive Energy Storage System behavior tests HIL simulation interest and structuring problematic

7 Adding simulation Customer needs Sizing and power flow studies Simulation Time to market library development using EMR (common tool, formalism unification, easy to share, ) With Simulation Actual approach Prototype elaboration Mise Prototype au point development du prototype Prototype behavior validation Cumulative dev. cost Feasability response Sizing Sim. Proto Prototype debugging ahead of time (ex. control), Easy to use, Quick development, Availability, Repeatable tests. 7 international Workshop on HIL'16 - Summer school, Lille, France

8 Adding simulation Customer needs Time to market Avec With Simulation Approche Actual approach initiale Sizing and power flow studies Simulation Prototype elaboration Mise Prototype au point development du prototype Prototype behavior validation Cumulative dev. cost Feasability response Sizing Sim. Proto Models validity range, virtual homologation complicated, real time portability of the control? 8 international Workshop on HIL'16 - Summer school, Lille, France

9 Adding simulation Customer needs Time to market Avec With Simulation Approche Actual approach initiale Sizing and power flow studies Simulation Prototype elaboration Mise Prototype au point development du prototype Prototype behavior validation Cumulative dev. cost Feasability response Sizing Sim. Proto Solution: «SuperModel» development? How many development time? All interaction are they taken into account? 9 international Workshop on HIL'16 - Summer school, Lille, France

10 Power HIL simulation: Battery test example No more model dependency : real power part + - I N T E R F A C E voltage current Upstream system Control & Energy management 1 international Workshop on HIL'16 - Summer school, Lille, France

11 Signal HIL simulation: Energy management test Signal HIL Validate new close control or energy management system in real time in interaction with emulated power systems, Validate interactions between EMS and other control unit (sensors, etc.), Validate step time and final control unit computation time, Validate on-line EMS behaviors, Test EMS and close control in fault tolerance modes. Battery + PE voltage current SIGNAL INTERFACE Upstream system 11 international Workshop on HIL'16 - Summer school, Lille, France

12 HIL simulation benefits Customer needs Time to market With simulation Actual approach With HIL Sizing and power flow studies Simulation Control development Simulation HIL Prototype behavior validation RT control validation and power subsystems validation Cumul dev. cost Feasability response Siz. Sim. HIL Proto Reduce development time, Financial benefits (mobilization et component deterioration), Safe et quality tests (fault tolerance tests), Repeatable tests, Virtual homologation available. 12 international Workshop on HIL'16 - Summer school, Lille, France

13 Sommaire Hardware-In-the-Loop simulation : hybrid locomotive Energy Storage System behavior tests Signal HIL simulation of PLATHEE locomotive

14 BB635 Diesel Electric locomotive DC Bus DCM 2*1 kw Traction Wh. Generator Aux. Mth 65 kw SM Drawbacks : - No energetic storage for traction or auxiliaries, - Oversized generator for shunting operation, - Diesel engine is uninterrupted (Auxilairies, etc.), 14 international Workshop on HIL'16 - Summer school, Lille, France - Diesel engine is not always in its maximal efficiency point.

15 PLATHEE Hybrid locomotive SC Storage system 8x2 Scaps (6,94 kwh) DC Bus DCM 2*1 kw Traction Wh. Bat 4x29 NiCd batt.(194 kwh) Generator Aux. Mth 215 kw SM Energy management strategy 15 international Workshop on HIL'16 - Summer school, Lille, France

16 PLATHEE energetic benefits validation 16 international Workshop on HIL'16 - Summer school, Lille, France

17 PLATHEE energetic benefits validation 1 v train [km/h] Different accelerations Traffic differences 5 PLATHEE Diesel-electric Time [s] Kilometer [km] Time [s] 16 5 Complexity 1 15 to make2 exactly the 25 same3 trip international Workshop on HIL'16 - Summer school, Lille, France

18 6 Consumption [L] 51,9 L 4 28,9 % 36,89 L 2 PLATHEE Diesel-electric Time [s] 1 SoCscap [%] SoCbat [%] Not the samesoc Difficut to conclude about the real benefits international Workshop on HIL'16 - Summer school, Lille, France Time [s] Time [s]

19 PLATHEE Hybrid locomotive - Energetic benefits are difficult to achieve directly on prototype, - EMS development on prototype is expensive. Objectives : Develop a real time test bench to test Energy management strategies using on-experimental results validated models Analog communication SC FC Bat. Gene. Driving CAN network EMS EMS 19 international Workshop on HIL'16 - Summer school, Lille, France

20 Structuring needs problematic of a Power HILs Power adaptations? (Signal) + - I N T E R F A C E Measures (Power) (Signal) Voltage Current Upstream system Studied system (Power) Control & Energy management Interface (Power and Signal) Control (signal) Emulated models (signal) Can 2 international EMR helps Workshop on to HIL'16 structure - Summer school, Lille, the France different parts of a Thursday, HIL simulation? September 1 st

21 EMR and Control development SCs u sc tot i s i ch i s 8 v ch m ch. i sc DC u C 2 u C i hach m ch u hach i arm i arm e mach C mcc Ω red C roue Ω roue Brak. F roue v F train freins v train F tract v train 2 F tot v train v train F res Env Bat u bt tot i s i ch i s 4 v ch m ch. i bt DC i tot i ch i aux Aux F freins-ref ICE C mth Ω arb Ω arb C ms i ms u hach_ref C roue_ref F roue_ref F tract_ref F tot_ref i arm_ref C mcc_ref 2 C mth_ref Ω arb_ref C ms_ref Generator Strategy Driving Strategy k D v ch_ref 4 P ge_ref SC 1 6 Scaps (6,94 kwh) DC Bus DCM 2*1 kw Wh. i s_ref i ch_ref i bt DC_ref Bat 1 16 NiCd batt. (194 kwh) Aux. v ch_ref 8 21 international Workshop on HIL'16 - Summer school, Lille, France i s_ref i ch_ref i sc DC_ref EMS ICE SM 215 kw Energy managment strategy

22 EMR and Control development RTS 1 SCs u sc tot i s v ch m ch. i ch i s 8 i sc DC u C 2 u C i hach m ch u hach i arm i arm e mach C mcc Ω red C roue Ω roue Brak. F roue v F train freins v train F tract v train 2 F tot v train v train F res Env Bat u bt tot i s i ch i s 4 v ch m ch. i bt DC i tot i ch i aux Aux F freins-ref ICE C mth Ω arb Ω arb C ms i ms Analog communication u hach_ref C roue_ref F roue_ref F tract_ref F tot_ref i arm_ref C mcc_ref 2 C mth_ref Ω arb_ref C ms_ref Generator Strategy Driving Strategy k D P ge_ref v ch_ref i s_ref i ch_ref i bt DC_ref v ch_ref 8 22 international Workshop on HIL'16 - Summer school, Lille, France RTS 2 i s_ref 4 i ch_ref i sc DC_ref EMS CAN Original EMS PLATHEE Obj. Model Validation 1 2 USB-CAN NI 8473 New EMS elaboration EMS RTS 3

23 Experimental setup and results 1 Same EMS than Prototype : Models validation v loc [km/h] P tra ct [kw] P au x [kw] P gs [kw] Cons [L] 23 international Workshop on HIL'16 - Summer school, Lille, France t(s) 5 P bat [kw] SoC bat [%] P sc [kw] SoC sc [%] u dc [V] (i) SOC BT [%] SOC SC [%] FUEL [L] EXP. MEAS. S-HIL RESULTS ERROR [%]

24 Experimental setup and results 2 New EMS and studies outlooks : Fair strategies comparison (maintenance, component behavior, fuel, ) v loc [km/h] P tract [kw] P gs [kw] Cons [L] 24 international Workshop on HIL'16 (e) - Summer school, Lille, France (b) (d) (a) P aux [kw] (c) t(s) P bat [A] SoC sbt [%] P sc [A] SoC sc [%] u dc [V]

25 Sommaire Hardware-In-the-Loop simulation : hybrid locomotive Energy Storage System behavior tests Conclusion and outlooks

26 Conclusion & outlooks EMR as a structuring tool : - Such a complex system and experiment needs methodology : EMR, - Integral causality allows to use description and control part in real time. Signal HIL simulation - Reduce the cost and time to market of new developments, - Do not required any prototype component, - Quick development and adjustment possibility, - Model validation in real time using original EMS and experimental measurements: EXPERIMENTAL MEASURES S-HIL RESULTS ERROR [%] SOC BAT [%] SOC SC [%] FUEL [L] Outlooks - fault tolerance tests and control robustness, - Power HIL simulation, - Full scale implementation on prototype. 26 international Workshop on HIL'16 - Summer school, Lille, France PLATHEE Supercapacitors PLATHEE NiCd Batteries

27 International Workshop on HIL simulation HIL 16 summer school Lille, 1-2 September Thanks for your attention! For more questions, comments or needs : Tony.Letrouve@sncf.fr Lille, France 27 international Workshop on HIL'16 - Summer school, Lille, France September 1 st and 2 nd