SHC Swedish Centre of Excellence for Electromobility Cost effective electric machine requirements for HEV and EV Anders Grauers Associate Professor in Hybrid and Electric Vehicle Systems
SHC SHC is a national centre of excellence for research and development of electric and hybrid vehicles. It is an arena where Sweden s automotive industry, universities and government agencies meet and collaborate to generate new technology, insights and competence for the future.
Well known concept plus new technology New layout of vehicle + =? New engine requirements New uses of vehicles http://en.wikipedia.org/wiki/file:stantonbury_1907.jpg Copyright expired http://en.wikipedia.org/wiki/file:hansomcab.jpg New vehicle requirements Do the vehicle makers know enough to put the right requirements on the electric drives in vehicles?
Challenge Electric machines bring new strengths & weaknesses requires new performance/cost compromises. Powertrain designers should not need understand all complex details of electric machine design. El. machine designers should not need understand all about power train performance, control & sizing. Still they must meet halfway Describe both: Req s/values - Capability/costs Jointly discuss how to find good compromises. May lead to new EM ideas, or new PT ideas
Aim of today s workshop 1) How are vehicle requirements on EM s best described? 2) How is performance limits of the EM best described? 3) How can we start a dialogue about the best requirements compromises? What is the value of different requirements? What is the cost of changing the EM capabilities? Before starting: are these the right questions?
Limited scope There is no perfect solution to this challenge A solution always is a compromise between Accuracy OR Easy to use. Focus on the primary function of the electric drive: Torque at different speeds Our solutions shall be useful for concept studies and definition of preliminary requirements. (must be combined with more detailed investigations) Must capture dominant effects, but may miss small or second order effects.
Examples from today s PT analysis Often scales EM with only one parameter - Power! See s field weakening as a fact and not an option. Express requirements in a format reflecting what the EM can perform, rather than what the powertrain needs. OR Specifies several long driving cycles and leave it to the electric machine designers to translate them themselves. Have too simple picture of the EM s capabilities to question what they are offered. The used driving cycles are a mixture of what driver wants and compromises made for conventional power trains. Can for instance not answer which EM provides highest performance: 100 Nm and 3 s OR 70 Nm and 10 s?
Examples from today s EM spec s Thermal limitations expressed in a way which cannot be used to determine vehicle performance in an arbitrary cycle. Does not show that the same machine can have different field weakening range. Double measures on some quantities: 1sec, 10 sec and base torque are not independent. Continuous torque is often confused with base torque.
Workshop rules There are no stupid questions! We have very different backgrounds! It is OK to question and discuss but don t criticise I bring ideas to start the discussion, but am more than happy to see them replaced by better ideas. Write down all ideas!
Program Time Agenda 10.00 Welcome and Introduction of participants 10.20 Overview of the challenge, Anders Grauers, SHC 10.40 Introduction -Power train requirements on electric machines 11.00 Requirements of a high 1 sec peak torque for a mild hybrid Sören Eriksson Volvo Cars 11.20 Group discussion: How is the PT s requirements on the EM best described? 12.00 Lunch 13.00 Introduction - How to describe electric machine performance? 13.30 Groupdiscussion: How to describe EM performance such that thept designer can determine its function in an arbitrary situation? 14.30 Introduction - How do we find the best compromises? Some important compromises in the design of the EM Francisco Márquez-Fernández LTH 15.00 Coffee break 15.30 Discussion on critical requirement compromises: What is the customer value of different requirements? What does different EM performance cost? 16.00 Discussion: Which compromises are important? Which PT req s should be reduced or increased? 16.30 Summarize results and open questions. What should be further investigated and discussed? 17.00 End of Workshop
How are vehicle requirements on EM s best described?
What are good requirements? Orthogonal: Not linked No double measures Smallest set with full coverage (only critical req s) Implementation-independent Represent interface quantities which can be measured. Minimize use of internal states in PT or EM Easy to use for the EM designer. For requirement setting process: Only use hard requirements when necessary, - Rather use soft evaluation criteria's, like cost penalties.
The PT requirements on EM Peak torque vs. speed Continuous torque vs. speed Thermally critical load cycles Evaluation criteria for efficiency Which are the critical requirements unfortunately depends on the electric machines properties, like thermal time constant! I.e. difficult to reduce requirements without knowing something about the EM.
Required torque operating range Typical driving Take-off in Max gradient Driving uphill Extreme situations Topspeed Does this allow significantly different spec? Are really all points important?
Evaluation factor for efficiency It is not useful if the power train designer prescribes a certain efficiency distribution. Instead a distribution of operating points is directly useful for the EM designer to find good balance between different types of losses.
Dynamic overload difficult! EM s with the same continuous torque, but different thermal time constants, react very different! If we don t know the thermal properties of the EM we cannot say which req s are critical. We need to assume something to be able to simplify the requirements. (for example losses prop to T^2) One or two node thermal model: Models Winding + Stator core temp. Too simple for design of EM! But probably OK to estimate thermal response for simple driving scenarios. Useful for control purposes
Dynamic load difficult! We need to understand EM to find good ways of simplifying the requirements. Thermal time constants allows separating load cases into: Only depending on present state Quick dynamics (Slow dynamics) Steady state long term average
Dynamic load difficult! Quick dynamics: Temp mostly returns to average level between them
Separate test cases (from base load) Short enough to not significantly influence stator core temp. The winding temp quickly returns between them.
Identify which scenarios are critical
Identify which scenarios are critical 60 sec time constant 20 sec time constant Time constant of EM very important for quick dynamics. The order of criticality sometimes change with time constant!
Steady state (slow dynamics) Very difficult to define a base load from typical driving cycles like Artemis Cycles However, certain special cases may be the critical ones. Gradients (which can go on for many minutes or even an hour) High speed driving on motorways.?
Downsize EM changes the requirements Downsized regarding what? Peak torque - Average torque power - time constant? A smaller EM may have very different distribution of operating points. Type of PT will influence what compromises can be made in the use of the EM: EV PHEV Mild HEV Note: Some limits can be compensated for by rest of PT, but some cannot.
Some comments Handle transient load with cooling system OR with longer thermal time constant? Time constant important when writing requirements: but what are typical winding and stator time constants in different machines?
Program Time Agenda 10.00 Welcome and Introduction of participants 10.20 Overview of the challenge, Anders Grauers, SHC 10.40 Introduction -Power train requirements on electric machines 11.00 Requirements of a high 1 sec peak torque for a mild hybrid Sören Eriksson Volvo Cars 11.20 Group discussion: How is the PT s requirements on the EM best described? 12.00 Lunch 13.00 Introduction - How to describe electric machine performance? 13.30 Groupdiscussion: How to describe EM performance such that thept designer can determine its function in an arbitrary situation? 14.30 Introduction - How do we find the best compromises? Some important compromises in the design of the EM Francisco Márquez-Fernández LTH 15.00 Coffee break 15.30 Discussion on critical requirement compromises: What is the customer value of different requirements? What does different EM performance cost? 16.00 Discussion: Which compromises are important? Which PT req s should be reduced or increased? 16.30 Summarize results and open questions. What should be further investigated and discussed? 17.00 End of Workshop
How is performance limits of the EM best described?
What is a good way of specifying? A good way of specifying the performance of the electric machine shall: Be short Little risk for misinterpretation Allow the user to determine how the EM will work in any dynamic situation. Don t require the reader to understand the internal function of the EM To define transient load from a fixed base load is not good, as a driving cycle does not have any defined base load.
What are the limiting factors for EM? Machine internal Electromagnetic limits Thermal limits (of Insulation / Magnets / ) Insulation (voltage-) limits Mechanical limits (rotor stress) External limits: Power limit due to battery Limit due to dc link voltage Current limit of inverter
Proposed performance description Maximum torque speed operating area - Limited by instantaneous factors: Saturation, Inverter current, inverter voltage, battery power, Thermal limits: One-/two-node thermal model + loss maps per node OR Continuous torque-speed operating points, limited thermally by the electric machine Iso-lines for different steady state winding temp (and stator core temp) + thermal time constant for winding
Program Time Agenda 10.00 Welcome and Introduction of participants 10.20 Overview of the challenge, Anders Grauers, SHC 10.40 Introduction -Power train requirements on electric machines 11.00 Requirements of a high 1 sec peak torque for a mild hybrid Sören Eriksson Volvo Cars 11.20 Group discussion: How is the PT s requirements on the EM best described? 12.00 Lunch 13.00 Introduction - How to describe electric machine performance? 13.30 Groupdiscussion: How to describe EM performance such that thept designer can determine its function in an arbitrary situation? What limiting factors need to be included? 14.30 Introduction - How do we find the best compromises? Some important compromises in the design of the EM Francisco Márquez-Fernández LTH 15.00 Coffee break 15.30 Discussion on critical requirement compromises: What is the customer value of different requirements? What does different EM performance cost? 16.00 Discussion: Which compromises are important? Which PT req s should be reduced or increased? 16.30 Summarize results and open questions. What should be further investigated and discussed? 17.00 End of Workshop
How can we start a dialogue about the best requirements compromises? What is the customer value of different requirements? What does different EM performance cost? Which compromises are important? Which PT req s should be reduced or increased?
Some ideas Will it be cheaper to increase thermal time constant than to increase the continuous torque through better cooling? If we can reduce the most critical requirement what will then be the second most critical requirement? The answer of what it cost to change performance will depend on if for example machine size has to be constant or not.
Summary Do we have a common language which both PT and EM designers are comfortable with? Can we categorise requirements as: Must haves, the more the better, give or take Can we categorise EM performance as: Costly, For free, In conflict with xx Have we identified requirements or performances which we should investigate further?
Anders.grauers@chalmers.se