Durham E-Theses
`
`S01ne drive train control problems zn hybrid z. c
`engine/battery electric vehicles
`
`Masding, Philip Wilson
`
`How to cite:
`
`:\lasding, Philip Wilson (1988) Some drive train control problems in hybrid i.c engine/battery electric
`vehiclesi Durham theses 1 Durham University. Available at Durham E-Theses Online:
`http:// etheses. dur. ac. uk/ 6408 /
`
`Use policy
`
`The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or
`charge, for personal research or study, educational, or not-for-profit purposes provided that:
`
`a full bibliographic reference is made to the original source
`
`• a link is made to the mctadata record in Durham E-Theses
`
`the full-teA-t is not changed in any way
`
`The full-text must not be sold in any format or medium without the formal permission of the copyright holders.
`
`Please consult lhe full Durham E-Theses policy for further details.
`
`BMW1048
`Page 1 of 254
`
`

`

`Academll' Support Office, Durham Cnivcrsity, l"niversity Office, Ole\ Elvct, Durham J)Hl 3HP
`e-mail: e-theses.aclmin-'.91dur.a,c.uk Tel: +44 0191 334 6107
`http:/ /ethescs.dur.acuk
`
`2
`
`BMW1048
`Page 2 of 254
`
`

`

`Some Drive 'I'rrulil CoR11t1ml Piroblem§ l[n
`
`Hybrid i.c Engine/Battery Electric Vehicle§
`
`The copyright of this thesis rests with the author.
`
`No quotation from it should be published without
`
`his prior written consent and information derived
`
`from it should be acknowledged.
`
`by Philip Wilson Masding B.Sc.
`
`A Thesis Submitted for the Degree 9f Doctor of Philosophy
`
`School of Engineering and Applied Science
`
`University of Durhain
`
`1988
`
`-- 2 NOV 1989
`
`BMW1048
`Page 3 of 254
`
`

`

`Declaration
`
`None of the work contained in this thesis has been previously submitted
`
`for a degree in this or any other university. It is not part of a joint resea,rch
`
`project.
`
`Copyright
`
`The copyright of this thesis rests with the author. No quotation
`
`from it should be published without his prior written consent and information
`
`derived from it should be acknowledged.
`
`BMW1048
`Page 4 of 254
`
`

`

`Acknowledgements
`
`I would like to acknowledge the tremendous support and encouragement I
`
`have had from my supervisor, Dr. J.R. Bumby, throughout this research. I
`
`am also grateful to Mr N. Herron and Mr K. McGee for their assistance in
`
`maintaining and building the rig and to Mr N. Herron in particular who designed
`
`the mechanical system for the gearbox. Finally I would like to thank Mr C. Dart
`
`for his help with electrical work on the rig.
`I am also grateful to the Ford Motor Company, Lucas Chloride E.V. Systems
`
`Ltd., and Shell Research Ltd. for the provision of equipment.
`
`BMW1048
`Page 5 of 254
`
`

`

`Abstiract
`
`This thesis describes the development of a microprocessor based control
`
`system for a parallel hybrid petrol/electric vehicle. All the fundamental
`
`systems needed to produce an operational vehicle have been developed and
`
`tested using a full sized experimental rig in the laboratory.
`
`The work begins with a review of the history of hybrid vehicles,
`
`placing emphasis on the ability of the petrol electric design to considerably
`
`reduce the consumption of oil based fuels, by transferring some of the load
`
`to the broad base of fuels used to generate electricity.
`
`Efficient operation of a hybrid depends on the correct scheduling of
`
`load between engme and motor, and correct. choice of gear ratio. To make
`
`this possible torque control systems using indirect measurements provided
`
`by cheap sensors, have been developed. Design of the control systems is
`
`based on a theoretical analysis of both the engine and the motor. Prior to
`
`final controller design, using the pole placement method, the transfer functions
`
`arising from the theory are identified using a digital model reference technique.
`
`The resulting closed loop systems exhibit well tuned behaviour which agrees
`
`well with simulation.
`
`To complete the component control structure, a pneumatic actuation
`
`system was added to a 'manual gearbox' bringing it under complete computer
`
`control. All aspects of component control have been brought together so
`
`that an operator can drive the system through simulated cycles. Transitions
`
`between modes of operation during a cycle are presently based on speed, but
`
`the software is structured so that efficiency based strategies may be readily
`
`incorporated in future.
`
`Consistent control over cycles has been ensured by the development
`
`of a computer speed controller, which takes the ·place of an operator. This
`
`system demonstrates satisfactory transition between all operating modes.
`
`ii
`
`BMW1048
`Page 6 of 254
`
`

`

`CONTlENT§
`
`List of Symbols
`Lisl of Fignrt"s
`
`CHAPTER 1: INTRODUCTION
`
`lPage
`
`Vll
`
`X
`
`1
`
`1.1 Possible Improvements or Alternatives to I. C. Engine Vehicles 5
`
`1.1.1 Novel Engines
`
`1.1.2 The Electric Vehicle
`
`1.2 The History of Hybrid LC. Engine/Electric Passenger Cars
`
`1.3 The Context of the Present Work
`
`CHAPTER 2: THE LABORATORY TEST SYSTEM
`
`2.1 Load Emulation
`
`2.2 The Hybrid Drive System
`
`2.3 Computer Control and Signal Monitoring
`
`2.3.1 Motorola M68000 System
`
`2.3.2 Duet 16 Personal Computer
`
`2.4 Instrnmentation
`
`2.5 Software System
`
`2.5.1 System Initialisation
`
`2.5.2 Active Rig Control
`
`2.5.3 Data Logging
`
`2.5.4 Data Transfer
`
`2.5.5 Data Analysis
`
`2. 6 Discussion
`2.7 Softw<Lrc Documeni.ct-1-lon
`
`6
`
`7
`
`9
`
`17
`
`32
`
`32
`
`34
`
`36
`
`36
`
`38
`
`38
`
`40
`
`40
`
`43
`
`44
`
`45
`
`46
`
`4 7
`
`50
`
`CHAPTER 3: PHYSICAL ANALYSIS OF THE ENGINE AND MOTOR 57
`
`3.1 Analysis of the Electric Traction System
`
`3.1.1 D.C. Machine Analysis
`
`3.1.2 Motor Torque Equation
`
`3.1.3 The Effects of the Power Electronic Controller
`
`59
`
`59
`
`61
`
`62
`
`l1l
`
`BMW1048
`Page 7 of 254
`
`

`

`3.2 Physical Analysis of the LC. Engine
`
`3.2.1 Inlet Manifold Pressure Variations
`
`3.2.2 Transfer Function for Engine Torque
`
`3.2.3 Throttle Servo-System
`
`3.2.4 Engine Speed on No-Load
`
`3.3.5 Transfer Function for Engine No-Load Speed
`
`CHAPTER 4: EXPERIMENTAL IDENTIFICATION RESULTS
`
`4.1 Calibration of Indirect Torque Measurements
`
`4.1.1 Engine Torque Model Calibration
`
`4.1.2 Engine Torque Model Testing
`
`4.1.3 Motor Torque Model Calibration and Testing
`
`66
`
`66
`
`69
`
`71
`
`71
`
`72
`
`79
`
`80
`
`80
`
`81
`
`82
`
`4. 2 Gain Functions for the Three Operating Modes of the Motor 83
`
`4.2.1 Field Boost Mode
`
`4.2.2 Full Field Mode
`
`4.2.3 Field Weakening Mode
`
`4.3 Transfer Function Identification
`
`4.4 Transfer Function Identification for the Motor
`
`4.4.1 The Closed Loop Transfer Functions for Current
`
`4.4.2 Direct Identification of the Motor Torque
`
`Transfer Function
`
`4.5 Manifold filling Delay and Engine Torque Variations
`
`4.6 Engine No-Load Speed Transfer Function
`
`4.7 Discussion
`
`CHA'.PTER 5: CONTROLLER DESIGN
`
`5.1 Controller Design Method
`
`5.2 Control Algorithm
`
`5.3 Design of Individual Controllers
`
`5.3.1 Engine Torque
`
`5.3.2 Electric Motor Torque Control
`
`5.3.3 Motor Torque Control Test Results
`
`5.3.4 Mode Determination
`
`iv
`
`84
`
`85
`
`86
`
`88
`
`92
`
`92
`
`93
`
`97
`
`99
`
`100
`
`118
`
`118
`
`121
`
`123
`
`123
`
`126
`
`127
`
`128
`
`BMW1048
`Page 8 of 254
`
`

`

`5.3.5 Engine Speed Synchronisation
`
`5.3.6 Engine Starting and Load Transfer
`
`5.4 Quantisation Errors and Noise
`
`5.5 Model Reference Controller Design
`
`5.5.l Application of the Model Reference Technique
`
`to the Motor
`
`5.6 Discussion
`
`CHAPTER 6: THE AUTOMATED GEAR CHANGING SYSTEM
`
`6.1 The Case for a Variable Transmission in Road Vehicles
`
`6.2 Variable Transmission Systems
`
`6.3 Transmission System Hardware on the Hybrid Vehicle Rig
`
`6.4 Interface Circuitry
`
`6.5 Software Control
`
`6.5.1 Stage 1: Shift into Neutral
`6.5.2 Stage 2: Speed Matching
`6.5.3 Stage 3: Engaging the New Gear
`6.5.4 Error Handling
`
`6.5.5 Location on Power Up
`
`6.6 Res11lts
`6. 7 Discussion
`
`CHAPTER 7: INTEGRATED DRNE TRAIN CONTROL AND
`
`DRIVE CYCLE TESTING
`
`7.1 Component Sequencing Control
`
`7.2 A Speed Based Mode Controller
`
`7 .3 Ori ve Cycle Testing
`
`7.3.1 Speed Conversions
`
`7.4 Complete Computer Control
`
`7.4.1 The Speed Control Loop
`
`7.5 Automated Cycle Control Using the
`
`Speed Based Mode Controller
`
`7.5.1 Preliminary All Electric Performance
`
`V
`
`130
`
`131
`
`133
`
`136
`
`139
`
`140
`
`159
`
`159
`
`162
`
`166
`
`168
`
`168
`169
`
`171
`172
`172
`
`174
`
`174
`
`175
`
`186
`187
`189
`190
`
`192
`193
`
`194
`
`197
`197
`
`BMW1048
`Page 9 of 254
`
`

`

`7.5.2 Calibration of the Expert System
`7.5.3 All Electric Performance with Expert Control
`
`7.5.4 Effect of Mode Transitions and Gear Changes
`
`7. 6 Discussion
`
`CHAPTER 8: CONCLUSIONS AND PROPOSALS
`
`FOR FUTURE WORK
`
`8.1 General Conclusions
`
`8.2 Proposals for Future Work
`
`REFERENCES
`
`199
`200
`
`200
`
`202
`
`216
`216
`222
`
`226
`
`vi
`
`BMW1048
`Page 10 of 254
`
`

`

`lList of §ymbols
`
`a
`
`Cr
`
`!,
`
`g
`
`G.(s)
`
`g,( w')
`g,(z)
`
`'J
`
`J
`
`K
`
`K,
`
`k '
`
`Zero of gc( w').
`
`Coefficient of drag
`
`Coefficient of rolling resistance
`
`Back e.m.f. voltage, V
`
`Counter value from flywheel speed probe
`
`Number of teeth on the flywheel speed probe gear
`
`Gain of gc(w').
`
`Armature closed loop transfer function
`
`P+l Controller in w'-plane form.
`
`Bilinear discretisation of %( w').
`
`Final drive ratio
`
`Field closed loop transfer function
`
`Armature current, A
`
`Field current, A
`
`Mean field current during a transient, A
`
`Flywheel inertia, kgm2
`
`Equivalent inertia of vehicle mass M, kgm2
`
`Constant relating dynamometer speed to load
`
`Scaling factor, flywheel count to roadspeed
`
`Back e.m.f. constant (= 2
`
`5~r)
`~
`
`gaT,/2
`
`Manifold filling delay gain
`
`Gain of armature controller
`
`Power stroke delay gain
`
`Kr
`
`Motor torque constant
`
`Gain of engine torque/speed transfer function
`
`Manifold pressure/throttle gain
`
`Vil
`
`BMW1048
`Page 11 of 254
`
`

`

`Armature inductance, H
`
`Field inductance, H
`
`Mass air flow into inlet manifold, kg/ s
`
`Mass charge flowing out of inlet manifold, kg/s
`
`Vehicle mass, kg
`
`Mass of gaseous mixture m the inlet manifold, kg
`
`Engine or motor speed, r.p.m.
`
`Speed at which linearisation takes place, r.p.m.
`
`Adaptive gam matrix
`
`Engine power, kW
`
`Inlet manifold depression, mbar
`
`M
`
`N
`
`No
`p
`
`Pm(o)(N)
`
`Manifold depression at no load and N r.p.m., mbar
`
`R.
`
`Rt
`
`tr
`
`T,
`
`Armature resistance, !l
`
`Field resistance, !l
`
`Vehicle wheel radius, m
`
`Motor torque, Nm
`
`Torque in gearbox output shaft, Nm
`
`Engine torque, Nm
`
`Vehicle aerodynamic and rolling resistance loss torque, Nm
`
`Controller design criteria: rise time, sec
`
`Control system sampling period, sec
`
`Armature voltage, V
`
`Field voltage, V
`
`Adaptive error at j'h sampling point
`
`Field current set-point function
`
`Plant model coefficient vector
`
`Armature current set-point function
`
`Signifies small change in variable
`
`Vlll
`
`BMW1048
`Page 12 of 254
`
`

`

`Controller design criteria damping factor
`
`Engine throttle position, 0.9' steps
`
`Demand throttle position, 0.9' steps
`
`Electric motor accelerator/ brake demand
`
`Mean demand valne during transient
`
`Inlet manifold filling time constant, sec
`
`Engine inertia time constant, sec
`
`Field flnx, Wb
`
`Plant model input/output vector
`
`y
`
`Cont.roller t.orque demand
`
`lX
`
`BMW1048
`Page 13 of 254
`
`

`

`List of Figures
`
`Fig.
`
`l.l ScriC's Hybrid Electric Vehicle Drive Tn-i.in
`
`Fig.
`
`l.2 Pil.rillkl Hyhrid Eleci ric V,,hiclc DriYe Train
`
`Fig.
`
`I. :J Typical Engine Efficifl1cy Map for a
`
`'>Ok\\' I.C. Engine
`
`Pag<'
`
`28
`
`28
`
`29
`
`Fig.
`
`1.4 Influence of Weighting F,ccfor on I.he Performance of the Hybrid
`
`Vehicle
`
`.JO
`
`Fig. 1.'> llse of t.he Engine Over an llrban Driving C:yde for an Optimally
`
`Conf.rolled Hybrid Vehicle
`
`Fig. 1.6 Possible Hybrid Vehicle Control System
`
`Fig. 2.1 Test Beel Layout.
`
`Fig. 2.2 Pc,wer Loading Provided by the Dynarnomct.er
`
`Fig. 2.:3 Test Bed Computing and lnt.erface Equipment.
`
`Fig. 2A Software St.ructure
`
`Fig. 3.1 Elements of t.he Electric Traction Drive
`
`Fig.
`
`:J.2 Bk1ck Diagram of Elect.ric Motor Equal.ions
`
`Fig. 3.3 Variation of Field Current. With Operat.ing Mode
`
`'
`
`Fig. 3.'I Block Diagram for Mot.or Current. Cont.ml
`
`Fig. 3.-'> Engine Physical Processes
`
`Fig. 3.6 Block Diagram for t.he Thrott.le Servo-System
`
`Fig. 4.1 J.C. Engine Power C'h,cract.eristics
`
`30
`
`31
`
`53
`
`54
`.ss
`
`-., , ..
`
`15
`
`76
`
`77
`
`78
`
`103
`
`Fig. 4.2. LC. Engine Power C'haracteri:d.ics: Varia.t.ion of Regression Con-
`
`st.ant.s with Speed
`
`103
`
`Fig. 4.:J Comparison of Torq11c Transdnn~r l\:Ie.asure1nent.s wit.h J.C. Engine
`
`Torque Prt"didion
`
`104
`
`Fig. 4.4 The Effecl of t.he Engine Cooling Fan on Torque M_easurement.s 104
`
`Fig. 4.-5 \'ariation of Mulor Torque Const.ant. wit.h Fidel Current
`
`10.S
`
`X
`
`BMW1048
`Page 14 of 254
`
`

`

`Fig
`
`-1.6 Verifi\ation of the lndirf'ct. Torque fvlt-~o.:rnremcnt.s for t.hc l\t[ot.or Hl.5
`
`Fig .. :J..l(a.) Varia.Lio11 uf" F'it'ld Current. wit.h Input. De1nand in t.he Field Boo::;t.
`
`l\fod,-·
`
`lflfi
`
`Fig. 4.,(b) Variation of Arnlidun• Current wit.h lnput Demand in the Field
`
`lfoos\. Mode
`
`l IJ(j
`
`Fig. 4.8 Variation of Current. with lnp,L1- Demand ill t.he Full Field Mode
`
`107
`
`Fig.
`
`4.9(a) Variation of Field Curren!. wit.h
`
`lnpnt. Demand m
`
`t.he Field
`
`Weakening Mode
`
`107
`
`Fig. 4.9(h) Variation of Armature Current. wit.h Input. Demand m t.he Field
`
`\Veakening :Mode
`
`Fig. 4.10 Adaptive ldenlificat.ion System
`
`108
`
`109
`
`Fig. 4.11 ~lodel ldent-ificalion for the Closed Loop Armature Current Control
`
`Response G, ( c)
`
`110
`
`Fig. 4.12 Compct.rison of Torque Transfer Functions with Experi111ent.al Da.t.a
`
`for t.he Field Boost .. Mode
`
`110
`
`Fig. 4. n Comparisons of Gains for the Manifold Filling J:/elay and Steady
`
`St.ate Data Relating Manifold Depression to Tlnot.1-le Opening
`
`Fig. 4.14 Block Diagram of t.he Linearised Engine Model
`Fig. 4.15 Engine Behaviour al 1 noo r.p.nL
`
`Fig. 4.,16 Engine Behaviour at. 3000 r.p.m.
`
`111
`
`112
`
`113
`
`113
`
`Fig. 4.17 \"erifirnt.iou of Transfer Functions for Engine Speed on No-Load 114
`
`.Fig.
`
`4.1~ Predicted and Mea.sured Torque and Current. Responses for a
`
`Transient in the Field Boost- Mode
`
`114
`
`Fig.
`
`4.J\I Predict.eel and Measured Torque and Curren!. Responses for a
`
`Transient in ihe Full Field Mode
`
`11.5
`
`Fig.
`
`4.20 Predicted and Measured Torque and Current Responses for a
`
`Transient in t.he Field VVeakf.•ning J\lode
`
`11.5
`
`XI
`
`BMW1048
`Page 15 of 254
`
`

`

`Fig. 4.21 Moicor Torque Moclcl: Test. for tile Direct Identification m the Field
`
`Boos1 Mode
`
`[ l(j
`
`Fig 4.2:2 Mot.or Torqnc Muriel: Tes! for the Direcl ldent.ifirni.ion irr the Full
`
`Field Mode
`
`116
`
`Fig. 4.:2::1 M.n1or Torque Morici: TPsf for tlw Direct. ldentifiration in the Field
`
`Weakening Mode
`
`117
`
`Fig.
`
`-5.J I a) llnrnmpensat.ed Lorns for Control of Engine Torque at. 2000 r.p.m.
`
`MS
`
`Fig.
`
`-5. l ( b) Compensated Locus for Control of Engine Torque at. 2000 r.p.m.
`
`145
`
`Fig. 5.2(a) Simulated and Experiment.al Performance of t.he Engine Torque
`
`Control System at. 2000 r.p.m.
`
`146
`
`Fig.
`
`.5.2(h) Simulated and Experiment.al Performance of t-he Engine Torque
`
`C'outrol System at. 3000 r.p.m.
`
`146
`
`Fig.
`
`,1.:3 Engi11t> Torque Control Sy~t.en1: Comparison of. Indirect. and Direct.
`
`Torque i\.lec1.surements During a Large Sl:ep Dist.urba11ce
`
`Fig. 5.4 !\lot.or Torqne Control System
`
`.,_
`
`141
`
`148
`
`Fig.
`
`-5.-5( a) Unrnrnpensa.t.ecl Lorns for Motor Torque Control in the Field
`
`Boost !\-lode
`
`149
`
`Fig. 5.-S(b) Cornpensat.ed Locus for Mot.or Torque Control in the Field Boost.
`
`Mode
`
`149
`
`Fig.
`
`-5.6(a) :Motor Torque Control Tesl 111
`
`t-he Field Boost II-Ioele at. Design
`
`Condit.ions
`
`150
`
`Fig. 5.6(b) Motor Torque Gont.rol Test m the Fidd Boost. lllode Away From
`
`Design Co11dilion::;
`
`Fig.
`
`-5.7 !\lot.or Torque Control Test 111 1-he Full Field Mode
`
`1-50
`
`J.51
`
`Fig.
`
`.S.ii(a) Mot.or Torque Control %st. in I.he Field Weakening Mode at
`
`Design Conditions
`
`1.52
`
`XII
`
`BMW1048
`Page 16 of 254
`
`

`

`Fig.
`
`-5.o(b) Mot.or Tonp.1e C:011trol Test in the Field Weakening Mori,· Away
`
`From Design Condit.ions
`
`I '•)
`
`-)~
`
`Fig .. :"dl rviot.or Torque Cout.rol Den1011sLra.1.i11g Sctl.isfn.ct.ory Transition Between
`
`01wraJing Modes
`
`1.53
`
`Fig. S.10 rvlot.or Torq1.1t' Control: Elfeci. of Using Field VVe,1.kening Gains in
`
`the Field Boost Mode
`
`Fig.
`
`-5.11 Engine Speed Control Block Diagram
`
`153
`
`154
`
`Fig . . 5.12(a) Uncompensated Lams for C'ontrol of Engin,- Speed on No-Load
`
`15-5
`
`Fig.
`
`.'i.12(b) Compensated Locus for Control of Engine Speed on No-Load
`
`l.'i.5
`
`Fig. 5.13 Step Test for the Engine Speed Control System
`
`Fig.
`
`.'i.14 Ana.lysis of t.he Engine Starling and Load Transfer Ptocess
`
`Fig.
`
`.'i.15 Hill Climbing Control Design Process
`
`].'i6
`
`l (16
`
`1-'> 7
`
`Fig.
`
`.5.16 Locus of Control Parameters During a Hill Climb Design Procedure
`
`1-56
`
`Fig.
`
`-5.17 Comparison of Reference Model Performance and,. Control System
`
`Designed by the Hill Climb Method
`
`158
`
`Fig. 6.1 Road Load and Motor Operating Curves for All Electric Vehicles
`
`178
`
`Fig.
`
`6.2 Tractive Effort. and Road Load Curves for All Elect.ric Vehicles
`
`Showing t.he Effect. of Gear R at.io
`
`Fig. 6.3 The C:earch a.nge Ivlechanis1n
`
`Fig. 6.4 The Pneumati, -circuit
`
`Fig. 6.-5 Regions Covered by the Posltion Sensors
`
`Fig. 6.6 Gear Change Algorithm Flo\'cchar\
`
`179
`
`180
`
`181
`
`182
`
`183
`
`Fig. 6.7(a) Torq1w and Speed Profiles Dming a Down Change from Third to
`
`Second
`
`184
`
`Xlll
`
`BMW1048
`Page 17 of 254
`
`

`

`Fig. 6.7(h) Torq11e itml Speed Profiles During an Up C:ha.nge from Secom! t.o
`
`Thi rel
`
`Fig. 6.8(a.) SI icge Ti rni ngB
`
`for a Dnv,n1 Cli<Lnge frnn 1 Thirrl
`
`tn Sf'C()Ild
`
`Fig. 6.x(bJ Stage Tin1i11gs
`
`for ,-\,1)
`
`lip Change from Second
`
`t.o Third
`
`Fig. 7.1 Compki.e VelH·ick
`
`C 01 np(JJ 1 eu t Cout.rol Sy:;1.{'111
`
`Fig. 7.2 Flowchart for Comp011ent Sequencing Logic
`
`Fig. 7.:3 ECEl.5 Cycle wit.h Manual Accelerat.or and Brake Control
`
`Fig. 7.4 Block Diagram for A ut.on1id.ic Cycle Speed Control
`
`184
`
`l 8'>
`
`18.5
`
`20,1
`
`20.5
`
`206
`
`207
`
`Fig. 7 . .5 ]dentifica.t.ion Experiment for the Flywheel and Dynamorneter
`
`208
`
`Fig. 7.6 CompenScded Locus for the Flywheel and Dynarnomet.er
`
`Fig. 7.7 Step Test for t.he Cycle Speed Control Syst.ern
`
`Fig. 7 .8( a) ECE ).5 Cycle TT sing the l.l nn10dified Speed Controller
`
`Fig. 7.8(b) Variation in Motor Torque
`
`Fig. 7.9 Flywheel Inertia. Calibration
`
`Fig. 7.lO(a) ECEl.o Cycle Using I.he Expert Cont.rol System
`
`20'1
`
`20Y
`
`210
`
`2111
`
`211
`
`212
`
`Fig. 7.JO(b) Variation in Motor TorljUe Showing lmrnecliat.e Drop at Break
`
`Point.s
`
`212
`
`Fig.
`
`7.ll(a) ECEJ.5 Cycle Using Speed Bitscd
`
`lvlode a.ucl Gear Shifting
`
`Stral.egy
`
`Fig. 7.ll(b) Variation Ill Mot.or Torque
`
`Fig. 7.ll(r) Variat.ion 111 Engine Torque
`
`Fig. 7.12(a) ECEl-5 Cycle Using Battery Recharge lvlode
`
`Fig. 7.12(b) Variation in Engine and l\1olor Torque
`
`213
`
`21:J
`
`214
`
`2]0
`
`21 'o
`
`XIV
`
`BMW1048
`Page 18 of 254
`
`

`

`CHAPTER 1
`
`INTRODUCTION
`
`Conventional internal combustion (i.e.) engine vehicles, based on the
`
`Otto and Diesel cycles, have dominated road vehicle transport for the greater
`
`part of this century. Among their many advantages are unlimited range, ease
`
`of refuelling, good power characteristics and the low cost materials required for
`
`their construction [JPL, 1975 pp 3-2l[Unnewehr and Nasar, 1982 pp 214-217]
`
`Despite these facts conventional vehicles bring with them problems of air
`
`pollution and inefficient use of dwindling fuel resources, which at the present
`
`rates may be depleted early in the twenty-first century [Foley, 1976]. Although
`
`diminishing energy resources, and in particular depletion of fossil fuels, poses
`
`a problem to many areas of technology it is particularly pronounced in the
`
`transport sector where the choice of fuels is most restricted. This dependence
`
`on one type of fuel
`
`is
`
`illustrated by examining figures for
`
`total energy
`
`consumption in Britain over recent years. Energy consumption in Britain
`
`peaked in about 1973 [Bumby and Clarke, 1982] subsequently falling by 11 %
`
`in the years up _to 1981 as a result of price rises, conservation measures and
`
`recession. Price rises were particularly great for oil based fuels and many
`
`non transport users switched to gas or coal over · this time period. As a
`
`result of this switch over amongst more flexible users, the transport sector has
`
`become responsible for using a progressively greater proportion of the total
`
`amount of oil used in the U .K. Over the period 1973-1981 this proportion
`
`rose dramatically from 29% to 4 7%. Further analysis of the transport sector
`
`itself shows that most fuel is used by passenger cars. In 1978 such vehicles
`
`accounted for 63% of all fuel used in vehicles, whilst the public transport
`
`sector for example, including taxis, coaches and buses used only 4.4% of the
`
`total [Dept. of Energy, 1978]. Faced with this heavy dependence on oil,
`
`inspite of its likely exhaustion in the relatively near term, there has been
`
`19
`
`BMW1048
`Page 19 of 254
`
`

`

`increasing interest m possible alternative vehicle propulsion systems over the
`
`last twenty years. Amongst the many possible alternative power plants are
`
`the Stirling engine, the Brayton engine (gas turbine) the all electric vehicle
`
`and the hybrid vehicle [JPL, 1975]. Of these alternatives the hybrid power
`
`plant is the subject of the present study. A hybrid power plant maybe defined
`
`as one that is powered by two or more energy sources and as such there are
`
`a multitude of possible combinations. Examples of some which have been
`
`investigated are:
`
`Heat engine/battery electric
`
`Flywheel/battery electric
`
`Heat engine/flywheel/battery electric
`
`Pneumatic/battery electric
`
`Battery /battery
`
`The purpose of this study is to consider the control problems relating to the
`
`heat engine/battery electric power train in particular. As many authors have
`
`pointed out [Mitcham and Bumby, 1977l[Unnewehr and Nassar, 1982] using
`
`two power plants adds considerably to the complexity and cost of the whole
`
`vehicle and hence the power train must offer other compensating factors
`
`such as lower running costs, reduced pollution and less noise in sensitive
`
`areas. Of the alternatives to conventional i.e.
`
`vehicles mentioned earlier,
`
`the electric vehicle would seem to offer all these advantages without the
`
`added complexity of a duel power source. Unfortunately, due to current
`
`battery technology, electric vehicles are basically 'energy deficient' [Unnewehr
`
`and Nassar, 1982}. Despite the fact that many types of advanced batteries
`
`have been the subject of much research and development, many people still
`
`feel that for the present, at least, the only practical storage battery for
`
`use in elect-ric vehicles is the lead/ acid type [van Niekerk et al, 1980l[Dell,
`
`1984}, which has an energy density of 40 Wh/kg as opposed to the energy
`
`2
`
`BMW1048
`Page 20 of 254
`
`

`

`density of petroleum which is 12300 Wh/kg [Unnewehr and Nassar, 1982J.
`
`In addition the energy density of lead/ acid batteries falls at high discharge
`
`rates which always occur in vehicle applications during acceleration. As a
`
`consequence of these factors even advanced electric Vehicles such as the ETV-
`
`1, developed for the U.S. Department of Energy by Chrysler and General
`
`Electric, suffer from severe range limitations varying in this case from 160 km
`
`at 60 km/h down to 95 km at 90 km/h even under favourable steady
`
`cruise conditions (Kurtz et al, 1981]. One way of reducing the severe range
`
`restrictions of the straightforward all electric vehicle is to combine it with
`
`some secondary energy source in a hybrid design. As can be seen in the
`
`list of hybrid drives above, all incorporate an electrical system. Of these
`
`systems the flywheel/battery electric, the pneumatic battery /electric and the
`
`battery /battery can be considered as primarily electric since the secondary
`
`energy storag_e device has very limited capacity. Nevertheless this secondary
`
`device can make a substantial improvement to the vehicle range simply by
`
`providing energy for short bursts of acceleration.
`
`For example flywheel
`
`systems, despite their modest energy storage capacity, can easily provide
`
`enough energy to accelerate a vehicle up to 30 mph once, thus protecting
`
`the batteries from the damaging high discharge rates which normally occur
`
`during such operation.
`
`In one such study simulations showed that electric
`
`vehicle range could be extended by 80% by using a flywheel of about 0.5 kWh
`
`capacity [Burrows et al, 1980].
`
`On the other hand the hybrid heat engine/ battery electric is by no
`
`means primarily electric, with the power and stored energy capacity of the
`
`engine typically exceeding that of the electrical system. In this case the motor
`
`takes on a secondary role in shielding the engine from operating conditions
`
`which are particularly unfavourable to it. Typically this means that such
`
`a vehicle will operate electrically when the total power requirement is low.
`
`3
`
`BMW1048
`Page 21 of 254
`
`

`

`Since low loading is typical of urban driving this vehicle has the potential to
`
`reduce pollution and noise in the urban environment whilst at the same time
`
`still offering the extended range capability characteristic of the engine.
`
`In
`
`general this system seeks to emphasise the particular strengths of each power
`
`source whilst playing down the disadvantages.
`
`When considering any type of hybrid electric vehicle two basic me(cid:173)
`
`chanical configurations form the basis of nearly all designs. These are the
`
`series and parallel configurations, which are illustrated by figures 1.1 and
`
`1.2 respectively.
`
`In the parallel hybrid both the engme and the motor are
`
`connected directly to the roadwheels. Provision is usually made for them
`
`to operate either independently or jointly in powering the vehicle.
`
`In the
`
`series hybrid however, only the motor is connected to the road wheels and
`
`must therefore be sized to meet all vehicle acceleration and top speed power
`
`requirements. In this case the engine supplies power indirectly via the gen(cid:173)
`
`erator battery combination. Although the series system has been considered
`
`for large bus applications [Brusaglino, 19821[Bader, 198ll[Roan, 1976], on the
`
`grounds that it gives greater flexibilty in positioning components and ease
`
`of controlling the i.e. engine, the multiple energy c;,nversions invariably lead
`
`to a low overall efficiency. Consequently almost all designs for passenger car
`
`applications have tended to favour the parallel configuration. As seen previ(cid:173)
`
`ously this latter application is most important because passenger cars offer the
`
`greatest potential fuel savings in the country as a whole. Clearly if hybrid
`
`vehicles could offer significant fuel savings in this transport sector, the benefit
`
`to society would be substantial. Before considering what potential the hybrid
`
`heat engine/electric has for making fuel savings in this sector it is important
`
`.to consider what it is competing against in terms of likely improvements to
`
`conventional vehicles and other alternative vehicle power systems.
`
`4
`
`BMW1048
`Page 22 of 254
`
`

`

`1.1 Possible limp1mvements or Alternatives to LC. Engine Vehicles
`
`It is important to recognise that any alternative to i.e. engine vehicles
`
`1s competing against a mature, established technology backed up by a large
`
`industrial research base. As a result conventional vehicles have shown a
`
`steady improvement over the years and seem likely to continue to do so. It 1s
`
`therefore pointless to compare possible economic and environmental advantages
`
`of a new vehicle technology against the i.e. engine vehicles which are being
`
`built today. A more realistic comparison is to judge the new technology
`
`against an estimate of what conventional vehicles will achieve 10-15 years
`
`from now. This 10-15 year period will then allow a reasonable time span for
`
`the new technology to be put into production on a worthwhile scale. Some
`
`of the likely improvements in i.e. engine design which might be made over
`
`the next few years are discussed by Blackmore and Thomas [Blackmore and
`
`Thomas, 1979j. This review suggests that major economy improvements could
`
`be achieved by using higher compression ratios coupled with leaner mixtures
`
`thus saving 20% of fuel use over all operating conditions. Lean mixtures
`
`must be used in conjunction with the higher compression ratios to avoid
`
`problems with knocking. Other savings can be achieved by improvements to
`
`cold starting ( saving 25% on short trips ), power modulation by variable
`
`mixture strength as opposed to the inefficient throttling method ( saving 20%
`
`on average ) and finally up to 10% can be saved by improving fuel and
`
`lubricants. Apart from changes to the engine itself further improvements
`
`can be made by altering vehicle aerodynamics and designing better tyres and
`
`gearboxes.
`
`Electronic engme control systems also offer substantial improvements
`
`m fuel economy. One major application for electronics is in optimising
`
`ignition timing. Standard mechanical distributors produce a crnde variation
`
`in spark timing on the basis of engine speed, as sensed by a centrifugal weight
`
`5
`
`BMW1048
`Page 23 of 254
`
`

`

`system, and engine load, as determined indirectly by a vacuum diaphragm
`
`lever system connected to the inlet manifold. Load and speed are in fact
`
`the correct indicators needed to optimise ignition timing, but the mechanical
`
`system can never re-produce the complex function relating them to optimum
`
`spark advance.
`
`In contrast, in an electronic system, it is possible to store
`
`the best spark advance as a function of manifold depression and speed in a
`
`two dimensional look-up table. As described by Main [Main, 1986] Ford have
`
`produced a microprocessor system which not only controls spark timing but
`
`also controls a continuously variable transmission system (CVT).
`
`Unlike some proposed vehicle systems the hybrid heat engine/battery
`
`electric can incorporate all the changes mentioned above in its own engine
`
`and so its potential advantages remain undiminished.
`
`1.1.l Novel Engines
`
`Apart from all electric and hybrid vehicles which may generally be
`
`considered as the electric alternative, whether partial or complete, there are
`
`two other major possibilities; alternative fuels and novel engines. Leading
`
`examples of alternative fuels are those based on agricultural products such as
`
`alcohol made from sugar. This has been successfully applied in Brazil with 95%
`
`alcohol by volume being used in cars with modified carburetters. According
`
`to the Jet Propulsion Laboratory ( JPL) [JPL, 1975] lesser concentrations, up
`
`to 28%, maybe used in ordinary cars without any such modifications. A
`
`second alternative is to produce synthetic petroleum from coal as has already
`
`been done in S. Africa. Neither of these alternatives are without drawbacks
`
`though, alcohol for example is particularly prone to absorb water in storage
`
`and the petrol from coal process is undesirable on both environmental and
`
`efficiency grounds [Mitcham and Bumby, 1977].
`
`Amongst the most likely novel engines are the gas turbine, based on
`
`6
`
`BMW1048
`Page 24 of 254
`
`

`

`the Joule Brayton cycle, and the Stirling engme. Gas turbines are already
`
`highly successful as aero-engines but need much development for small scale
`
`use m cars.
`
`In addition they use more expensive materials in order to
`
`withstand the high temperatures associated with a continuous combustion
`
`process [Bumby and Clarke, 1982]. External combustion engines such as the
`
`Stirling design promise a broader choice of fuels and the ability to optimise
`
`combustion conditions for better efficiency and reduced emissions. Despite
`
`improvements to heat exchangers however the Stirling engine still suffers from
`
`large thermal inertia.
`
`In their study of alternative vehicle power systems, JPL concluded
`
`that the Stirling engine and the gas turbine might achieve gains of 20% and
`
`30% respectively when compared with the conventional vehicles of the time
`
`(1975 ). When such improvements are viewed in the light of the predicted
`
`improvements to the conventional car itself, these figures are not outstanding,
`
`particularly when it is considered that in all cases the alternative engine costs
`
`more to produce.
`
`1.1.2 The Electric Vehicle
`
`Although of somewhat limited range the all electric vehicle must
`
`not be dismissed prematurely as a likely replacement means of transport,
`
`particularly in certain applications. Fleet trials of electric vehicles have shown
`
`them to be very useful in applications with a well