Module 6

1) What is power follower strategy for energy management in hybrids? Is

this a rule based strategy or an optimization based strategy?

Power follower control

 In this strategy the ICE is the primary source of power and the EM is used
to provide additional power when needed by the vehicle.
 Care is always taken to maintain the SOC of batteries within safe limits.
 The rule base that is generally used is :
 Below a certain minimum vehicle speed, only
the EM is used.
 If the demanded power is > the maximum power
that the ICE can produce at its operating speed,
the EM is used to produce excess power.
 The EM charges the batteries by regenerative
braking.
 The ICE shuts off when the power demand falls
below a limit at the operating speed. This is done
to prevent inefficient operation of ICE.
 This is a very simple and effective strategy but the major disadvantage is
that the efficiency of the entire drive train is not optimized.

Findings based on analysis of power flow in HEV drive train is utilized to design
deterministic rules not based on optimization. These rules are generally
implemented using lookup tables to split requested power between the ICE and
EM.

2) List four examples of rule based strategies that can be applied to energy
management in hybrid vehicles?
The most commonly used strategies are:
 Power follower control
 Modified power follower
 State Machine based controller
 Fuggy Logic (FLC) Based Controllers for HEVs

Downloaded from Ktunotes.in

Power follower control
 In this strategy the ICE is the primary source of power and the EM is used
to provide additional power when needed by the vehicle.
 Care is always taken to maintain the SOC of batteries within safe limits.
 The rule base that is generally used is :
 Below a certain minimum vehicle speed, only
the EM is used.
 If the demanded power is > the maximum power
that the ICE can produce at its operating speed,
the EM is used to produce excess power.
 The EM charges the batteries by regenerative
braking.
 The ICE shuts off when the power demand falls
below a limit at the operating speed. This is done
to prevent inefficient operation of ICE.
 This is a very simple and effective strategy but the major disadvantage is
that the efficiency of the entire drive train is not optimized.

Modified power follower

 In order to improve the power follower controller a cost function is
introduced.
 The role of this cost function is to strike a balance between fuel
consumption and emissions at all operating points of HEV.

State Machine Based controller

 The state machine dictates the operating mode of the HEV such:
 ENGINE (ICE propelling the vehicle)
 BOOSTING (both ICE and EM propelling
the vehicle)
 CHARGING (ICE propelling the vehicle
and charging the battery)
 The transition between the operating modes is decided based on:
 The change in driver demand.
 A change in vehicle operating condition.

Downloaded from Ktunotes.in

  A system or a subsystem fault.

Fuggy Logic (FLC) Based Controllers for HEVs

 Looking into a hybrid drive train as a multi-domain, nonlinear and time
varying plant, fuzzy logic seems to be the most logical approach to the
problem.
 Instead of using deterministic rules, the decision making property of fuzzy
logic can be adopted to realize a real time and sub-optimal power split.
 Fuzzy logic is an extension of the conventional rule-based methods and has
following advantages over them:
 Robustness: It is inherently robust because it
does not require precise, noise free inputs and the
output is a smooth function despite a wide range
of input variations.
 Adaptation: Since FLC processes user defined
rules governing the system, it can be modified
easily to improve or drastically alter system
performance.
 Flexibility: FLC is not limited to a few feedback
inputs and one or two outputs and it is not
necessary to measure or compute rate-of-change
of parameters.

3) Classify and explain the basic principles of rule based energy management
system?
The Rule Based strategies consist of following subcategories:
Deterministic Control:
The deterministic controllers are subdivided into
 Power follower
 State Machine
 Thermostat Control

Fuzzy based:

Downloaded from Ktunotes.in

The fuzzy based control strategies are of three types
 Predictive
 Adaptive
 Conventional

Rule Based Control Method

Basic Principles of Rule Based Control Methods:
 Rule based control strategies can cope with the various operating modes of
HEV.
 The rule based strategies are developed using engineering insight and
intuition, analysis of the ICE efficiency charts shown in following Figure
2 and the analysis of electrical component efficiency charts.

Fig .1 Efficiency map of ICE

Downloaded from Ktunotes.in

 Fig .2 Efficiency map of ICE showing upper and lower
boundaries for EM operations
 An example of developing rule based strategy can be explained using the
ICE efficiency map shown in Figure 2
 The lines, which are drawn using engineering insight and intuition, divide
the map into 3 regions: A, B, and C. The rules for operation of ICE in these
three regions are :
 In the region A only EM is used because
in this region the fuel efficiency of the ICE
is poor .
 In region B only ICE is used since this the
region of high fuel efficiency .
 In region C both ICE and EM are used.

Power follower control

 In this strategy the ICE is the primary source of power and the EM is used
to provide additional power when needed by the vehicle.
 Care is always taken to maintain the SOC of batteries within safe limits.
 The rule base that is generally used is :
 Below a certain minimum vehicle speed, only
the EM is used.
 If the demanded power is > the maximum power
that the ICE can produce at its operating speed,
the EM is used to produce excess power.
 The EM charges the batteries by regenerative
braking.

Downloaded from Ktunotes.in

  The ICE shuts off when the power demand falls
below a limit at the operating speed. This is done
to prevent inefficient operation of ICE.
 This is a very simple and effective strategy but the major disadvantage is
that the efficiency of the entire drive train is not optimized.

Modified power follower

 In order to improve the power follower controller a cost function is
introduced.
 The role of this cost function is to strike a balance between fuel
consumption and emissions at all operating points of HEV.

State Machine Based controller

 The state machine dictates the operating mode of the HEV such:
 ENGINE (ICE propelling the vehicle)
 BOOSTING (both ICE and EM propelling
the vehicle)
 CHARGING (ICE propelling the vehicle
and charging the battery)
 The transition between the operating modes is decided based on:
 The change in driver demand.
 A change in vehicle operating condition.
 A system or a subsystem fault.

Fuggy Logic (FLC) Based Controllers for HEVs

 Looking into a hybrid drive train as a multi-domain, nonlinear and time
varying plant, fuzzy logic seems to be the most logical approach to the
problem.
 Instead of using deterministic rules, the decision making property of fuzzy
logic can be adopted to realize a real time and sub-optimal power split.
 Fuzzy logic is an extension of the conventional rule-based methods and has
following advantages over them:
 Robustness: It is inherently robust because it
does not require precise, noise free inputs and the
output is a smooth function despite a wide range
of input variations.

Downloaded from Ktunotes.in

  Adaptation: Since FLC processes user defined
rules governing the system, it can be modified
easily to improve or drastically alter system
performance.
 Flexibility: FLC is not limited to a few feedback
inputs and one or two outputs and it is not
necessary to measure or compute rate-of-change
of parameters.

4) What is a Controller Area Network?

CAN (Control Area Network)
 In vehicle and outside of the vehicle communications are fast growing
technology areas in automotive applications.

 Automotive communications networks have typically used LANs (local

area networks) of the controller area network (CAN) variety.
 CAN networks are converting over to protocols that are more fault tolerant
and have guaranteed communications times and no issues with message
latency as has been common for multiple access collision detection
(MA/CD) protocols in the past.
 Time triggered protocols (TTP) have been proposed for many years and
are now beginning to enter the automotive arena as a new protocol derived
from a TTP/C basis known as Flexray.

IN VEHICLE NETWORKS
 During most of the early years of electronic engine control and for
hybrid propulsion technologies some form of CAN communications
b/w the various subsystems has been in use.
 Early SAE (Society of Automotive Engineers) standards categorised in-
vehicle networks according to data handling speed.
 Setting a standard based on data handling speed in effect determined
the types of devices that could be served and the types of data
communication protocols that could be applied.
 Early hybrid propulsion communications architectures relied on class B
and some provision was made for class C, CAN, where very fast data
exchange was necessary, such as in the power train controller.

Downloaded from Ktunotes.in

  Class A is a low speed protocol used primarily for in-vehicle body
electrical functions such as power seat and power window controls.
 Protocols on class A networks include local interconnect network (LIN)
and time triggered protocol/A(TTP/A).
 Class B n/ws are designed for data sharing between devices in the hope
of minimizing if not eliminating redundant sensors – for example,
instrumentation, speed-control functions, emissions systems and others.
 CAN-class C network protocol was developed for real time control
applications; hence the reference to power train control functions
 Vehicle dynamics control also included in class C protocol.

Table 1 . The various in-vehicle network classes with typical in vehicle

applications served.

 The CAN is a fast, high rate network enabling communication between

ECUs.
 In CAN most data can be updated every 10ms and the data is checked to
assure data reliability.

Downloaded from Ktunotes.in

 Fig. 3 A typical CAN network in an HEV
Control Variables
 The control variables connect various ECUs each other, fall in one of the
3 categories:
 Mechanical
 Electrical
 Discrete
 The control variables falling in the mechanical category are:
 All the variable related to ICE
 Gear ratio
 Rpm of each rotational component
 The control variables falling in the electrical category are:
 Currents in the batteries, inverters and EM

Downloaded from Ktunotes.in

  Voltages across EM, inverters and battery
terminals
 EM torque

5) What are the typical objectives a fuzzy logic based energy management
control strategy addresses, and what inputs are mainly employed in the
strategy? (5M) [APJKTU: DEC 2018]
What are the advantages of fuzzy logic-based energy management control
strategy in hybrid vehicles? [Sep 2020 [5M]]?
What are the advantages of fuzzy logic based energy management control
strategy in hybrid vehicles? [May 2019 [4M]]
Answer

Looking into a hybrid drive train as a multi-domain, nonlinear and time varying
plant, fuzzy logic seems to be the most logical approach to the problem.

Instead of using deterministic rules, the decision making property of fuzzy logic
can be adopted to realize a real time and sub-optimal power split.

Fuzzy logic is an extension of the conventional rule-based methods and has

following advantages over them:
i. Robustness: It is inherently robust because it does not require precise, noise
free inputs and the output is a smooth function despite a wide range of input
variations.
ii. Adaptation: Since FLC processes user defined rules governing the system, it
can be modified easily to improve or drastically alter system performance.
iii. Flexibility: FLC is not limited to a few feedback inputs and one or two outputs
and it is not necessary to measure or compute rate-of-change of parameters.

How to Use Fuzzy Logic in Design of Controllers

The energy management and control strategy using FLC performs following
actions:
i. Maximizes fuel economy, minimize emissions and distribute the driver’s
request for power between two sources: ICE and Motor.
ii. Maximize fuel economy at any point in operation, that is, provide dynamic or
instantaneous optimization.
iii. maximize some other attributes such as acceleration of vehicle

Downloaded from Ktunotes.in

The four components of FLC are:
i. Fuzzification, which is the change from crisp values to fuzzy values.

(For ICE speed, the crisp value may be 2000rpm and a possible rule for ICE
speed may be “If ICE rpm is too low, then inject more fuel”. The fuzzy value
associated with the “If X, then Z” statement would be for X<2000 rpm.)
ii. Rule base has a collection of rules: several hundred rules may be developed
and applied.
iii. Inference applies the defined rules to the inputs.
iv. De-fuzzification transforms the results of the inference process to crisp
outputs.

Fuzzy Strategy
The FLC, explained in this section, satisfies the following objectives:
a. minimize NOx emissions
b. sustain battery SOC
c. achieve desired torque requested by the driver

The inputs to this FLC are:

i. Acceleration pedal stroke (Acc)
ii. EM speed ( Ѡem )

 The configuration of the drive train: An induction motor (IM) used in the
drive train and the IM is directly coupled to the ICE.
 Since the IM is directly coupled to the diesel ICE, it will be in the field
weakening region in most of the ICE operating, the generating torque
decreases as the ICE speed increases.
 Hence, it is required to describe the required torque as a ratio defined as K
to the rated torque at a rotational speed.
 The positive K means that the IM acts as a powering source and negative
K means that the IM acts as a generator.
 Once K is determined, the torque command becomes

Torque Command= rated torque at K X a rotational speed

Some basic principles of generating the torque command from the acceleration
pedal stroke and ICE rotational speed in the HEV can be described as follows:

Downloaded from Ktunotes.in

Low ICE Speed:
When the ICE rotational speed is low, it generates pollutant emissions with
low efficiency.

Hence, in this operating condition, the torque assistant control by the IM

should be performed. Assistant torque is commanded to increase in proportion to
the acceleration pedal stroke as in the conventional ICE vehicle, that is:
K acceleration pedal stroke

Medium ICE Speed:

When the diesel ICE’s speed is medium, it can supply sufficient torque to the
hybrid drive train.
Hence, battery recharging control is performed instead of torque assistance
control when acceleration pedal stroke is below some extent.
The torque assistance of the IM should be should be achieved to satisfy the
driver’s acceleration need if the acceleration pedal is pressed beyond some
extent.
Since, the diesel ICE torque is subjected to saturation beyond 80% of the pedal
acceleration pedal stroke; the motoring action of the IM is made to begin from
that point.

High ICE Speed:

When the diesel ICE speed is high, the torque assistance control is performed
as that of the medium speed range.
In battery recharging control the IM’s output power is kept constant. Now
since the ICE can produce more power than in the medium speed range, the factor
should be made to be negatively greater in order to supply more power to the
batteries than that of the medium speed range.
As the speed increases, the ratio of the power capability of the ICE to that of
the IM increases.
Hence, it is beneficial to recharge the battery at high speed, rather than at
medium speed.

Implementation of Fuzzy Logic

Based on the three principled of operation of ICE, discussed above, the fuzzy
rule base can be developed.
The development of the FLC is described in the following subsections.

Input / Output Membership Functions for Fuzzy Logic

For the considered example, there are two input variables namely:
i. The acceleration pedal stroke Acc

Downloaded from Ktunotes.in

ii. The IM rotational speed Ѡrpm

6) State and explain the optimal control problem associated with hybrid
vehicles?
The optimal control of HEVs (Hybrid Electric Vehicles) is an important topic not
only because it is useful forPower management control but also indispensable for
the optimal design of HEVs. Different vehicle systems can be compared to each
other only when the controllers guarantee the optimality for each deployed
system.

Technically, we can obtain optimal control trajectories if the whole driving-cycle

information is given prior, and if we have determinate performance indexes, such
as fuel consumption, exhaust emission, or acceleration performance. Under those
circumstances, the Dynamic Programming (DP) approach guarantees the global
optimal results and had been investigated in several works.
The results obtained through DP are unbeatable but, unfortunately, cannot be
implemented directly. Instead, a post-processing step is required by using rule
extraction, e.g., through Neural Networks, which approximates the results of the
optimal control pattern.
Even with this post-processing step, these strategies cannot cover all driving
conditions. Hence, the real-time controller based on DP is effective only for the
driving cycle that is used for rule extraction. To remedy this problem, stochastic
dynamic programming and driving pattern detection with multiple driving cycles
had been suggested as possible solutions.

Another approach based on optimal control theory which, basically, realizes the
minimization of the Hamiltonian has been applied to the optimal control problem
for HEVs. The Equivalent Consumption Minimization Strategy (ECMS), which
really began from the heuristic concept that electric energy could be equivalent
to fuel usage, was introduced

The optimal control based on Pontryagin’s minimum principle (PMP) is simple

enough to be implemented in real-time applications because it is based on
instantaneous optimization.

Downloaded from Ktunotes.in

7) With the help of block diagram explain the battery management
supporting system of hybrid vehicle? [Dec 2019-5M]

Downloaded from Ktunotes.in

Battery ECU or Battery Management System:
The battery ECU or the battery management system (BMS) monitors and
measures temperature and assures cooling is adequate.
The BMS avoids the stress of heat and over-temperature and the effects of
excessive charging or discharging are eliminated or lessened.
The BMS is essentially for long battery life and optimum fuel efficiency.

8) With the help of block diagram explain the hierarchical power and data
transmission network of hybrid vehicles? [Dec 2019-5M]
 Modern vehicles, and particularly hybrid vehicles, are evolving to a
four layer power and data communications architecture.
 The four layer model is based on broad classes of functionality,
data transmission speed and data protocols associated with broad
classes of speeds.
 The four layer model includes: power distribution in an open
architecture sense of generation, storage and distribution systems;
safety and mobility including air bag deployment control to power
train and x-by-wire controls; body layer including interior and
exterior lighting, customer amenities and instrumentation; and
infotainment layer including high end audio products, TV, wireless

Downloaded from Ktunotes.in

 phone, voice actuated systems, navigation aides, and other additions
in a scaleable and flexible architecture.
 In-vehicle architectures, besides being open, are hierarchical by
design. A hierarchical architecture adheres to strict protocols of top-
down command flow and upward flow of data.
 Lateral data flow is allowed, but lateral command flow is prohibited.
 This is the architecture of choice for hybrid propulsion in which a
high level vehicle system controller is coordinating its various
subsystems as shown in Figure.

 The communications networks serving the full vehicle will be of different

classes and share data via gateways.
 For example, outside of the power train, chassis and energy system
functions illustrated in Figure (above) there would be additional networks
for safety and mobility, body control and infotainment as shown in
following Figure, and perhaps others interconnected via gateways.
 In Figure the power and data architecture is shown with class A/B
network for body and energy management and a Class C network for power
train and hybrid system control as well as safety and mobility plus
infotainment.
 This high speed network can be upgraded to include a separate, higher
speed, dedicated network to serve the infotainment functions as these
become more pervasive in the automobile.

Downloaded from Ktunotes.in

9) Why an energy management control system is required in an HEV? Do
you think an elaborate energy management system similar to that applied to
a hybrid vehicle, is required in an electric vehicle? Explain. [May 2019 [5M]]
The major functions of the control system in HEV and EV are:
1. To maximize the fuel efficiency
2. To minimize the exhaust emissions

The minor functions of the control system are component monitoring and
protection such as:
1. Battery state of charge(SOC) monitoring
2. Battery temperature monitoring
3. EM overheating
4. ICE overheating

The control system in HEV requires additional control of ICE parameters during
the various operational modes such as:
1. Ignition timing
2. Tuned intake manifold
3. Camshaft angle for exhaust valves
4. Camshaft angle for intake valves
5. Fuel injector settings such as:
a. Timing of injection
b. Fuel flow rate

Downloaded from Ktunotes.in

An elaborate energy management system as seen in hybrid vehicles includes the
operation and control of IC components during each operational modes, which
isn’t necessary in electric vehicles, which focuses on the control of EM and its
associated electrical components during the different operational modes.

10) Draw the block diagram of a general Fuzzy Logic Controller (FLC) and
show the core components of the FLC and the inputs and outputs relevant
to a hybrid electric vehicle control. [May 2019 [5M]]

Downloaded from Ktunotes.in

The core components of FLC are:
1. Fuzzification, which is the change from crisp values to fuzzy values. (For
ICE speed, the crisp value may be 2000rpm and a possible rule for ICE
speed may be “If ICE rpm is too low, then inject more fuel”. The fuzzy
value associated with the “If X, then Z” statement would be for X<2000
rpm.)
2. Rule base has a collection of rules: several hundred rules may be developed
and applied.
3. Inference applies the defined rules to the inputs.
4. Defuzzification transforms the results of the inference process to crisp
outputs.
The inputs relevant to hybrid electric vehicle control are: Acceleration pedal
stroke and speed of electric motor. The output is the required torque defined as K
to the rated torque at a rotational speed.

11) Explain fuzzy logic implementation of energy management system in a

parallel HEV with induction motor and ICE with an objective of reduction
in environmental pollution with the help of a block diagram. [May 2019
[5M]]
For parallel hybrid electric vehicle control, the inputs to the FLC are:

Downloaded from Ktunotes.in

 1. Accleration pedal stroke (Acc)
2. EM speed (em)

The configuration of the drivetrain is shown by:

The configuration of the drive train: An induction motor (IM) used in the drive
train and the IM is directly coupled to the ICE.
Since the IM is directly coupled to the diesel ICE, it will be in the field weakening
region in most of the ICE operating, the generating torque decreases as the ICE
speed increases.
Hence, it is required to describe the required torque as a ratio defined as K to the
rated torque at a rotational speed.
The positive K means that the IM acts as a powering source and negative K means
that the IM acts as a generator.
Once K is determined, the torque command becomes: Torque Command= K x
rated torque at a rotational speed.
The torque command is generated from the acceleration pedal stroke and ICE
rotational speed depending on low, medium and high ICE speed.
For low IC speed,
When the ICE rotational speed is low, it generates pollutant emissions with
lowefficiency. Hence, in this operating condition, the torque assistant control by
the IM should be performed. Assistant torque is commanded to increase in

Downloaded from Ktunotes.in

proportion to the acceleration pedal stroke as in the conventional ICE vehicle,
that is: K acceleration pedal stroke
For medium IC speed:
When the diesel ICE’s speed is medium, it can supply sufficient torque to
the hybrid drive train. Hence, battery recharging control is performed instead of
torque assistance control when acceleration pedal stroke is below some extent.
The torque assistance of the IM should be should be achieved to satisfy the
driver’s acceleration need if the acceleration pedal is pressed beyond some extent.
Since, the diesel ICE torque is subjected to saturation beyond 80% of the pedal
acceleration pedal stroke; the motoring action of the IM is made to begin from
that point.
For high IC speed:
When the diesel ICE speed is high, the torque assistance control is
performed as that of the medium speed range. In battery recharging control the
IM’s output power is kept constant. Now since the ICE can produce more power
than in the medium speed range, the factor should be made to be negatively
greater in order to supply more power to the batteries than that of the medium
speed range. As the speed increases, the ratio of the power capability of the ICE
to that of the IM increases. Hence, it is beneficial to recharge the battery at high
speed, rather than at medium speed

The input/output membership functions for fuzzy logic are:

1. The acceleration pedal stroke Acc
2. The IM rotational speed  rpm

The ranges of input variables are:

1. The Acc is set to zero when the driver does not press the acceleration pedal
at all and set to a 100 when the driver presses the acceleration pedal
completely
2. The rpm  rpm can vary from the diesel ICE’s idling speed to its maximum
speed.

The output is the normalized ratio of the torque command to rated torque at a
speed.
The inputs and the outputs are normalized between zero and one.

Downloaded from Ktunotes.in