Battery Management Solution

Intelligent BMS Solution using AI &
Prognostic SPA
Subrahmanyam Sista, Avinash Sista
Mahindra Satyam, Satyam Technology Center, Hyderabad, AP, INDIA
[Sista_Subrahmanyam, Avinash_Sista]@mahindrasatyam.com
Abstract--- This paper presents a Novel, Low cost and
Efficient Intelligent Battery Management Solution
(iBMS) for Electric Vehicles (EV) and Hybrid Electric
Vehicles (HEV). The solution provides a comprehensive
topology for identifying the State of Charge (SOC),
State of Health (SOH), charging and discharging
including isolation of defective identified battery cell
from healthy ones. The highly modular and scalable
solution uses Bi-directional, 4 quadrant DC-DC
converter; a non-isolated four switch topology design
for the charging/discharging and cell cut off (infected
cell), an Artificial Intelligence (AI) module using Fuzzy
Logic (FL) and Signature Pattern Analysis (SPA) for
envisaging the Battery stack health. The proposed
design offers an affordable On-Board monitoring &
diagnostics module leveraging the above intelligent
modules and Impedance Analysis. This circumvents the
need of further diagnostic tools; makes the system
highly portable, Scalable for any chemical composition
of battery cell and considerably extend the life cycle of
EV/HEV battery stacks. In this paper, we will review
some of the issues and associated solutions for battery
thermal management and what information is needed
for proper design of battery management systems. We
will discuss about the issues related to impedance
management which affects the battery life.
Keywords: Artificial Intelligence (AI), Electric
Vehicles (EV), Fuzzy Logic (FL), Hybrid Electric
Vehicle (HEV), Impedance Analysis (IA), State of
Charging (SOC), State of Health (SOH), Bi-
directional, 4 quadrant DC-DC convertor, UIS and
CIP( use in series and charge in parallel)
I. INTRODUCTION
From portable Electronics to Electric vehicles
(EVs), batteries are widely used as a main energy
source in many applications. Battery management
involves implementing functions that ensure
optimum use of the battery in a system. Examples of
such functions are proper charge handling and
protecting the battery from misuse. The basic task of
a BMS is to ensure that optimum use is made of the
energy inside the battery powering the portable
product and that the risk of damage inflicted upon the
battery is minimized. This is achieved by monitoring
and controlling the battery’s charging and
discharging process.
The purpose of the BMS is to guarantee safe and
reliable battery operation. To maintain the safety and
reliability of the battery, state monitoring and
evaluation, charge control, and cell balancing are
functionalities that have been implemented in BMS.
Similar to the engine management system in a
gasoline car, a gauge meter should be provided by the
BMS in EVs and HEVs. BMS indicators should show
the state of the safety, usage, performance, and
longevity of the battery. Due to volatility,
flammability and entropy changes, a lithium-ion
battery could ignite if misused. This is a serious
problem because an explosion could cause a fatal
accident. Moreover, over-discharge causes reduced
cell capacity due to irreversible chemical reactions.
Therefore, a BMS needs to monitor and control the
battery based on the safety circuitry incorporated
within the battery packs. Whenever any abnormal
conditions, such as over-voltage or overheating, are
detected, the BMS should notify the user and execute
the preset correction procedure. In addition to these
functions, the BMS also monitors the system
temperature to provide a better power consumption
scheme, and communicates with individual
components and operators.
Safety functions must be implemented and work
correctly to ensure that no one gets hurt and that the
BMU and battery cells are not damaged. If a battery
cell has too low a cell voltage, it may result in
voltage reversal. The consequences of reversal can be
venting or rupturing depending on battery type.
Under voltage can be prevented by measuring the
voltage. In some cases, if there are weak battery cells
in the battery pack, individual voltage measurements
may be needed to detect low voltage. The cells
should be monitored and not allowed to drop below
the lower voltage threshold, thus minimizing the risk
of voltage reversal. Cell Balancing / connect-
disconnect circuit is a critical function for high-
powered battery packs because a long series of

individual cells is only as reliable as the weakest cell.
Sometimes it may become necessary to rest a
particular cell especially in case of an over
temperature, therefore it needs to be disconnected
and re-connected again.
The use of individual cell voltage measurements
leads to a better precision in SOC and balancing,
increasing the battery lifetime and capacity. It is
important to follow the manufacturer’s suggested cell
voltage thresholds to avoid degradation of
performance and improve safety.
II. THE CURRENT STATE OF ELECTRIC
CAR BATTERY TECHNOLOGY:
The value chain of electric car battery consists of
seven steps: component production (including raw
materials), cell production, module production,
assembly of modules into the battery pack (including
an electronic control unit and a cooling system)
integration of battery pack into the vehicle and reuse
and recycling. Automobile manufacturers have
identified three types of rechargeable battery as
suitable for electric car use. Those types are lead-acid
batteries, nickel metal hydride (NiMH) batteries, and
lithium-ion (Li-ion) batteries. Lithium-ion batteries
comprise a family of battery chemistries that employ
various combinations of anode and cathode materials.
Each combination of anode and cathode has distinct
advantages and disadvantages in terms of safety,
performance, cost and other parameters. The most
prominent technologies for automotive applications
are lithium-nickel-cobalt-aluminium (NCA), lithium-
nickel-manganese-cobalt (NMC), lithium-manganese
spinel (LMO), lithium titanate (LTO) and lithium-
iron phosphate (LFP).[10]
III. PROPOSED BMS ARCHITECTURE
Our solution consists of batteries connected in
series. Battery Management Solution is divided into
the following three categories:
 State of charge
 State of health
 Thermal Management
Fig 1: Battery Management System
SOC and SOH are considered to be the
methodologies for battery evaluation.
STATE OF CHARGE (SOC):
SOC is critical, but it is not measurable
given the current onboard sensing technologies. The
ratio of the currently available capacity to the
maximum capacity can be expressed as SOC.
STATE OF HEALTH (SOH):
SOH describes the physical condition of a
battery, ranging from internal behavior, such as loss
of rated capacity, to external behavior, such as severe
conditions. Unlike SOC, there is no clear-cut
definition of SOH. A general definition of SOH is
that it reflects the health condition of a battery and its
ability to deliver specified performance compared to
a fresh battery
Figure 2 gives us an idea of how our
solution architecture looks like.

Fig 2: Architecture
In our solution we use two voltage sensors, two
current sensors, 4 temperature sensors and an
impedance sensor to measure and control the life of
the battery.
A. THERMAL MANAGEMENT
SYSTEM
The goal of a thermal management system is to
deliver a battery pack at an optimum average
temperature. LM35 is the temperature sensor used to
sense the temperature in our solution. The LM35
series are precision integrated-circuit temperature
sensors, whose output voltage is linearly proportional
to the Celsius (Centigrade) temperature. The LM35
thus has an advantage over linear temperature sensors
calibrated in ° Kelvin, as the user is not required to
subtract a large constant voltage from its output to
obtain convenient Centigrade scaling. The LM35
does not require any external calibration or trimming
to provide typical accuracies of ±1⁄4°C at room
temperature and ±3⁄4°C over a full −55 to +150°C
temperature range.
B. IMPEDANCE MEASUREMENT
Impedance is nothing but the Ohm’s Law for AC
circuits. Voltage and current are sensed and are
divided to give the impedance value. AD5934 is the
impedance sensor used to sense the impedance[3].
The AD5934 is a high precision impedance converter
system solution that combines an on-board frequency
generator with a 12-bit, 250 kSPS, analog-to-digital
converter (ADC). The frequency generator allows an
external complex impedance to be excited with a
known frequency. The response signal from the
impedance is sampled by the on-board ADC and a
discrete Fourier transform (DFT) is processed by an
on-board DSP engine. The DFT algorithm returns a
real (R) and imaginary (I) data-word at each output
frequency. Once calibrated, the magnitude of the
impedance and relative phase of the impedance at
each frequency point along the sweep is easily
calculated using the following two equations:
Magnitude = 22 IR+
Phase = tan−1 (I/R)
C. ARTIFICIAL INTELLIGENCE
APPROACH TO ESTIMATE SOC
The battery pack charge and discharge processes
are so complex that it is essential to consider many
factors such as cell voltage, current, internal
impedance and temperature gradients. The battery
pack connected in series presents a more complex
problem. Careful monitoring and control is necessary
to avoid any single cell within a Li-ion battery pack
from undergoing over-voltage or under-voltage. In a
Li-ion battery module-management system, the

individual batteries SOC must be monitored because
of overcharge and over-discharge issues. Therefore, it
is essential to have methods capable of estimating the
battery SOC.
Kalman Filter (KF) is a powerful tool for the
state estimation of systems.
To achieve the accurate estimation of SOC,
Artificial Neural Networks (ANN) and Fuzzy Logic
systems have been treated as the universal
approximates. Many techniques have been developed
to approximate the nonlinear functions for practical
applications. The BMF constructed in possesses the
property of local control and has been successfully
applied to Fuzzy Neural Control. Also, the
hybridization of fuzzy logic with ANN has been used
to improve the efficiency of function estimation.
The adaptive Neural Fuzzy method was
proposed to estimate battery residual capacity.
Although the estimation of battery residual is
accurate, the algorithm utilizes the least-square
method to identify the optimal values and hence,
learning rate is computationally expensive; much
time is wasted in training an ANN. A more practical
approach, called merged-FNN for SOC estimation is
proposed. In merged-FNN, the FNN strategy is
combined with Reduced-form Genetic Algorithm
(RGA) which performs effectively on SOC
estimation in a series-connected Li-ion battery string.
The merged-FNN achieved a faster learning rate and
lower estimation error than the traditional ANN with
a back-propagation method.
Due to the above-mentioned facts, it is not hard
to see that AI is a promising approach for fast,
precise and reliable SOC estimation. [4]
D. CHARGE ESTIMATION
ALGORITHMS
Several different techniques such as Fuzzy
Logic, Kalman Filtering, Neural Networks and
recursive, self-learning methods have been employed
to improve the accuracy of the SOC estimation as
well as the estimation of state of health (SOH).
1) FUZZY LOGIC: Fuzzy Logic is simple way to
draw definite conclusions from vague, ambiguous or
imprecise information. It resembles human decision
making with its ability to work from approximate
data to find precise solutions.
Unlike classical logic which requires a deep
understanding of a system, exact equations, and
precise numeric values, Fuzzy logic allows complex
systems to be modeled using a higher level of
abstraction originating from our knowledge and
experience. It allows expressing this knowledge with
subjective concepts such as big, small, very hot,
bright red, a long time, fast or slow. This qualitative,
linguistic representation of the expert knowledge
presents a natural rather than a numerical description
of a system and allows relatively easy algorithm
development compared to numerical systems. The
outputs can then be mapped into exact numeric
ranges to provide a characterization of the system.
Fuzzy logic is used extensively in automatic control
systems.
Using this technique we can use all the
information available to us about the performance of
a battery to derive a more accurate estimation of its
state of charge or the state of health.
2) KALMAN FILTER: The battery SOC is affected
by many simultaneous factors and is continually
changing due to the user driving pattern. The Kalman
filter is designed to strip unwanted noise out of a
stream of data. It operates by predicting the new state
and its uncertainty, then correcting this with a new
measurement. It is suitable for systems subject to
multiple inputs and is used extensively in predictive
control loops in navigation and targeting systems.
With the Kalman Filter the accuracy of the battery
SOC prediction model can be improved and
accuracies of better than 1% are claimed for such
systems.
3) NEURAL NETWORKS: A Neural Network is a
computer architecture modeled upon the human
brain's interconnected system of neurons which
mimics its information processing, memory and
learning processes. It imitates the brain's ability to
sort out patterns and learn from trial and error,
discerning and extracting the relationships that
underlie the data with which it is presented.
Each neuron in the network has one or more
inputs and produces an output; each input has a
weighting factor, which modifies the value entering
the neuron. The neuron mathematically manipulates
the inputs, and outputs the result. The neural network
is simply neurons joined together, with the output
from one neuron becoming input to others until the
final output is reached. The network learns when
examples (with known results) are presented to it; the
weighting factors are adjusted on the basis of data -

either through human intervention or by a
programmed algorithm-to bring the final output
closer to the known result. In other words, neural
networks "learn" from examples and exhibit some
capability for generalization beyond the training data.
Neural networks thus resemble the human brain in
the following two ways:
1. A neural network acquires knowledge
through learning.
2. A neural network's knowledge is stored
within inter-neuron connection strengths
known as synaptic weights.
The true power and advantage of neural
networks lies in their ability to represent both linear
and non-linear relationships and in their ability to
learn these relationships directly from the data being
modeled. Among the many applications are
predictive modelings and control systems.
Neural Network techniques are useful in
estimating battery performance which depends on
quantifying the effect of numerous parameters most
of which cannot be defined with mathematical
precision. Algorithms are refined with the aid of
experience gained from the performance of similar
batteries. [4]
E. DC-DC CONVERTER
OPERATION
A DC-DC converter allows DC voltages to be
raised or lowered similar to how a transformer does
with AC voltages. DC-DC conversion became very
popular with the advent of power transistors, and
since then many circuits have been invented. There
are countless types of DC-DC converter circuits but
they all operate on the same basic principles. The
simplest types of converter are the non-isolated buck
and boost converters. Other types include the Cuk
converter and the Buck-Boost converter, as well as
isolated designs such as the forward converter and
flyback converter.
1) POWER SWITCH CONTROL: Figure 3 shows a
simplified diagram of how the four power switches
are connected to the inductor, VIN, VOUT and GND.
Figure 4 shows the regions of operation as a
function of duty cycle D. The power switches are
properly controlled so that the transfer between
modes is continuous.
2) BUCK REGION (VIN > VOUT): Switch D is
always on and switch C is always off during buck
mode. When the error amp output voltage, VC, is
approximately above 0.1V, output A begins to
switch.
Fig 3: Representative Output Switch Configuration
Fig 4: Operation Mode vs. Vic Voltage
During the off time of switch A,
synchronous switch B turns on for the remainder of
the switching period. Switches A and B will alternate
similar to a typical synchronous buck regulator.
As the control voltage increases, the duty
cycle of switch A increases until the max duty cycle
of the converter in buck mode reaches
DMAX_BUCK, given by:
DMAX_BUCK = 100 – D4 (SW) %
Where D4 (SW) = duty cycle % of the four switch
range.
D4 (SW) = (300ns • f) • 100%
Where f = operating frequency, Hz.
Beyond this point the four switch or buck-boost
region is reached.
3) BUCK-BOOST OR FOUR SWITCH (VIN ~
VOUT): When the error amp output voltage, VC, is
above approximately 0.65V, switch pair AD remain
on for duty cycle DMAX_BUCK, and the switch pair

AC begin to phase in. As switch pair AC phases in,
switch pair BD phases out accordingly. When the VC
voltage reaches the edge of the buck-boost range,
approximately 0.7V, the AC switch pair completely
phase out the BD pair, and the boost phase begins at
duty cycle, D4 (SW).
The input voltage, VIN, where the four switch region
begins is given by:
The point at which the four switch region ends is
given by:
VIN = VOUT (1 – D) = VOUT (1 – 300ns • f) V
4) BOOST REGION (VIN < VOUT): Switch A is
always on and switch B is always off during boost
mode. When the error amp output voltage, VC, is
approximately above 0.7V, switch pair C and D will
alternately switch to provide a boosted output
voltage. This operation is typical to a synchronous
boost regulator. The maximum duty cycle of the
converter is limited to 90% typical. [5]
F. CELL CONNECTING AND
DISCONNECTING
Cell balancing is a critical function for high-
powered battery packs because a long series of
individual cells is only as reliable as the weakest cell.
It is critical that the charge level of all cells does not
stray outside the recommended SOC range. To
ensure that the batteries cannot be over-discharged
during inactivity, the BMS idle power must be
significantly less than the self-discharge of the
battery cells.
More importantly, the idle current along the
battery string must be well matched to ensure that
batteries do not become unbalanced during storage.
Once placed into operation, the batteries will
experience high charge and discharge currents. In this
situation, the charge level of each cell must be
actively balanced to derive maximum pack energy
and lifetime.
In our solution in case if a battery in the series is
not functioning properly, all the other batteries in the
series compensate the lost voltage and maintains a
constant output voltage through the 4 Quadrant DC to
DC converter. The battery is isolated from the flow
circuit.
IV. CONCLUSION
As batteries are the core energy sources in EVs
and HEVs, their performance greatly impacts the
salability of EVs. Therefore, manufacturers are
seeking for breakthroughs in both battery technology
and BMSs. Chemical reactions in the battery are
subject to operating conditions, and hence, the
degradation of a battery may vary in different
environments. Developing a comprehensive and
mature BMS is critical for manufacturers who would
like to increase the market share of their products.
The major concerns of BMSs were discussed in this
paper. They include battery state evaluation,
modeling, and cell balancing, wherein the evaluation
methodologies of battery status were viewed as the
crucial issue. A BMS framework was proposed to
deal with the deficiencies of current BMSs in both
research and commercial products. Based on previous
work, specific challenges facing BMSs and their
possible solutions were presented as a solid
foundation for future research. Due to varying
situations in real-world applications, a standard
solution was not wanted. Based on the specific
situation, different strategies should be applied to
improve and optimize the performance of BMSs in
future EVs and HEVs.

REFERENCES
[1]. Battery Management Systems in Electric and Hybrid Vehicles-Yinjiao Xing 1,
Eden W. M. Ma 1, Kwok L. Tsui 1, 2 and Michael Pecht 1, 3,*
[2].http://publications.lib.chalmers.se/records/fulltext/125124.pdf
[3]http://www.analog.com/static/importedfiles/data_sheets/AD5934.pdf
[4].http://www.mpoweruk.com/soc.htm
[5].http://cds.linear.com/docs/Datasheet/3785fc.pdf
[6]. Linden, David and Reddy, Thomas. Handbook of batteries. New York :
McGraw-Hill, 2002.
[7]. Park, Hong-Sun, et al. Two-Stage Cell Balancing Scheme for Hybrid Electric
Vehicle Lithium-Ion Battery Strings. Korea : Korea Advanced Institute of Science
and Technology, 2007.
[8]. Electric Vehicle Wikipedia Page. Available online:
http://en.wikipedia.org/wiki/Electric_vehicle
(accessed on 20 July 2011).
[9]. Claus, D. Materials and processing for lithium-ion batteries. J. Miner. Metals
Mater. Soc. 2008,
60, 43–48.
[10]. Batteries for Electric Cars; Challenges, Opportunities, and the Outlook to
2020; The Boston
Consulting Group Inc.: Boston, MA, USA, 2010; Available online:
http://www.bcg.com/ documents/file36615.pdf (accessed on 20 July 2011).
[11]. Goals for Advanced Batteries for EVs; United States Council for Automotive
Research LLC:
Southfield, MI, USA, 2009; Available online:
http://www.uscar.org/commands/files_download.
php?files_id=27 (assessed on 1 August 2011).
[12]. Stuart, T.; Fang, F.; Wang, X.P.; Ashtiani, C.; Pesaran, A. A modular battery
management system
for HEVs. Future Car Congress 2002, doi: 10.4271/2002-01-1918.

Battery Management Solution

More Related Content

What's hot (20)

Viewers also liked (18)

Similar to Battery Management Solution (20)

Battery Management Solution