Description:TECHNICAL FIELD
[001] The present disclosure, in general, relates to the field of vehicular maintenance. More
particularly, the present disclosure relates to adaptive performance prognostics of
powertrain components in a vehicle.
BACKGROUND
[002] The following description includes information that may be useful in understanding the
present invention. It is not an admission that any of the information provided herein is
prior art or relevant to the presently claimed invention, or that any publication
specifically or implicitly referenced is prior art.
[003] Conventionally, to facilitate regular maintenance of a vehicle and its components,
automobile OEMs (original equipment manufacturers) provide a standard periodic
maintenance schedule. The standard maintenance schedule is generally designed based
on the distance that the vehicle has travelled or age of the vehicle. However, this
approach has certain limitations as it does not account for the actual usage of the
components, the driving behavior of different users and ambient conditions during
vehicle usage etc. For instance, the standard maintenance schedule may require an
engine belt of a vehicle to be replaced between 60,000 to 1,50,000 miles. However, this
does not take into consideration the actual usage scenarios. That is, a vehicle that is
usually driven in regions of hot weather may experience the wear and tear of the engine
belt much before the set standard maintenance schedule and hence, a delay in replacing
the engine belt may lead to engine failure. Similarly, a vehicle that is usually driven in
regions of normal weather may not experience a fast wear and tear of the engine belt
and hence, untimely replacing it may lead to discard of usable remaining life. In another
instance, a vehicle that is usually driven in harsh terrains may experience wear and tear
of its components much earlier than a vehicle that is driven in well paved terrains.
Therefore, it is imperative to have in a place a maintenance schedule that is not fixed
and takes into account actual condition/performance of a component based upon its
usage, driving and ambient conditions etc.
[004] To overcome such limitations, various OEMs have started utilizing machine learning
techniques to perform prognostics of vehicles. However, the conventional machine
learning techniques’-based prognostics approach suffers from various drawbacks. For
2
instance, said approach provides only a binary output, i.e., it only states whether
components are OK or NOT OK but doesn't mention about the amount of component
life left. Further, said approach requires a lot of “NOT OK” instances to train the model.
But since, the “NOT OK” instances are very less in real world usage scenario for certain
components, the training of the model is not robust and hence, the predicted output is
not accurate.
[005] Hence, there is a need for a method and system for vehicle component prognosis that
overcomes the above-mentioned limitations.
SUMMARY
[006] The present disclosure overcomes one or more shortcomings of the prior art and
provides additional advantages. Embodiments and aspects of the disclosure described
in detail herein are considered a part of the claimed disclosure.
[007] In one non-limiting embodiment of the present disclosure, a method for adaptive
prognostics of performance of one or more components associated with powertrain in
a vehicle is disclosed. The method comprises receiving, in real-time during each driving
cycle, telematics data associated with each of the one or more components. The
telematics data comprising one or more parameters associated with a component being
evaluated, attributes corresponding to the component being evaluated and feature(s)
data associated with the attributes. The method further comprises populating, at the end
of each driving cycle, a matrix consisting of plurality of predefined bins with data points
received for each component separately. In an aspect, each bin indicates a component
operation time in each bin during said driving cycle. Further, each of the plurality of
predefined bins are aligned into one or more priority groups. Next, the method discloses
comparing data density value of each populated bin with a corresponding pre-calibrated
data density value. Moving on, the method discloses activating one or more populated
bins when the data density value in a populated bin is found greater than the precalibrated
data density value. In addition, the method comprises recording one or more
attributes corresponding to the one or more activated bins and normalizing the one or
more attributes. Further, the method discloses comparing one or more normalized
attributes with corresponding one or more defined reference attributes to identify each
of the one or more activated bins as healthy or unhealthy. The method further comprises
3
calculating, from a total number of activated bins in each priority group, a number of
unhealthy bins in each priority group and comparing the calculated number of
unhealthy bins against a predefined unhealthy bin weightage of each priority group.
Further, the method comprises identifying a health status for each priority group based
on the comparison. Finally, the method further comprises providing a final performance
prognostic judgement for the component being evaluated based on the identification.
[008] In another non-limiting embodiment of the present disclosure, the plurality of
predefined bins are aligned into one or more priority groups based on an operational
behavior or stability of the component being evaluated in each bin of the plurality of
predefined bins. Further, when two or more priority groups are identified as unhealthy,
the final prognostic performance judgement is based on a higher priority group among
the two or more priority groups. Furthermore, providing the final performance
prognostic judgement comprises providing a remedial action suggestion to either one
of a user-authorized mobile widget or OEM service center.
[009] In yet another non-limiting embodiment of the present disclosure, the plurality of
predefined bins are created by dividing a component operating zone for a given driving
cycle using the one or more parameters associated with the component, into a plurality
of equally sized grids. In an aspect, each grid represents a bin, and number of the
plurality of predefined bins varies with respect to a type of the component being
evaluated.
[010] In yet another non-limiting embodiment of the present disclosure, normalizing the one
or more attribute values further comprises normalizing each of the one or more attribute
values based on a sensitivity factor corresponding to each of the one or more attribute
values and a standard value associated with a feature associated with the one or more
attribute values.
[011] In yet another non-limiting embodiment of the present disclosure, the method further
comprises defining the one or more defined reference attribute values by setting one or
more nominal reference attribute values based on a specification or predefined initial
operation of the component being evaluated. For defining the one or more defined
reference attributes, the method further comprises modulating the set one or more
nominal reference attribute values based on the one or more normalized attribute values
4
and a predefined calibration factor. Furthermore, the one or more defined reference
attributes are unique for each of the plurality of predefined bins.
[012] In yet another non-limiting embodiment of the present disclosure, a system for adaptive
prognostics of performance of one or more components associated with powertrain in
a vehicle is disclosed. The system comprises an I/O interface, a memory, and at least
one processor operatively coupled to the I/O interface and the memory. The at least one
processor is configured to receive, in real-time during each driving cycle, from a
telematics device, telematics data associated with each of the one or more components.
The telematics data comprising one or more parameters associated with a component
being evaluated, attributes corresponding to the component being evaluated, and
feature(s) data associated with the attributes. The at least one processor is further
configured to populate, at the end of each driving cycle, a matrix consisting of plurality
of predefined bins with data points received for each component separately. In one
aspect, each bin indicates a component operation time in each bin during the driving
cycle. Further, each of the plurality of predefined bins are aligned into one or more
priority groups. Next, the at least one processor is configured to compare data density
value of each populated bin with a corresponding pre-calibrated data density value.
Moving on, the at least one processor is configured to activate one or more populated
bins when the data density value in a populated bin is found greater than the precalibrated
data density value. In addition, the at least one processor is configured to
record one or more attributes corresponding to the one or more activated bins and
normalizing the one or more attributes. Further, the at least one processor is configured
to compare one or more normalized attributes with corresponding one or more defined
reference attributes to identify each of the one or more activated bins as healthy or
unhealthy. The at least one processor is further configured to calculate, from a total
number of activated bins in each priority group, a number of unhealthy bins in each
priority group and compare the calculated number of unhealthy bins against a
predefined unhealthy bin weightage of each priority group. Further, the at least one
processor is configured to identify a health status of each priority group based on the
comparison. Finally, the at least one processor is configured to provide final
performance prognostics judgement for the component being evaluated based on the
identification.
5
[013] In yet another non-limiting embodiment of the present disclosure, the at least one
processor is configured to align the plurality of predefined bins into one or more priority
groups based on an operational behavior or stability of the component being evaluated
in each bin of the plurality of predefined bins. Further, when two or more priority
groups are identified as unhealthy, the final prognostic performance judgement is based
on a higher priority group among the two or more priority groups. Furthermore, to
provide the final performance prognostic judgement, the at least one processor is further
configured to provide a remedial action suggestion to either one of a user-authorized
mobile widget or OEM service center.
[014] In yet another non-limiting embodiment of the present disclosure, to create the plurality
of predefined bins, the at least one processor is configured to divide a component
operating zone for a given driving cycle using the one or more parameters into a
plurality of equally sized grids. In one aspect, each grid represents a bin, and a number
of the plurality of predefined bins varies with respect to a type of the component being
evaluated.
[015] In yet another non-limiting embodiment of the present disclosure, to normalize the one
or more attribute values, the at least one processor is further configured to normalize
each of the one or more attribute values based on a sensitivity factor corresponding to
each of the one or more attribute values and a standard value associated with a feature
associated with the one or more attribute values.
[016] In yet another non-limiting embodiment of the present disclosure, the at least one
processor is configured to define the one or more defined reference attribute values. In
one aspect, to define the one or more defined reference attributes the at least one
processor is configured to set one or more nominal reference attribute values based on
a specification or predefined initial operation of the component being evaluated. The at
least one processor further modulates the set one or more nominal reference attribute
values based on the one or more normalized attribute values and a predefined
calibration factor. Furthermore, the one or more defined reference attributes are unique
for each of the plurality of predefined bins.
[017] The foregoing summary is illustrative only and is not intended to be in any way limiting.
In addition to the illustrative aspects, embodiments, and features described above,
6
further aspects, embodiments, and features will become apparent by reference to the
drawings and the following detailed description.
BRIEF DESCRIPTION OF DRAWINGS
[018] The features, nature, and advantages of the present disclosure will become more
apparent from the detailed description set forth below when taken in conjunction with
the drawings in which like reference characters identify correspondingly throughout.
Some embodiments of system and/or methods in accordance with embodiments of the
present subject matter are now described, by way of example only, and with reference
to the accompanying Figs., in which:
[019] Figure 1 represents an exemplary environment 100 for adaptive prognostics of
performance of powertrain components in a vehicle, in accordance with embodiments
of the present disclosure,
[020] Figure 2 represents an exemplary block diagram 200 of a system for adaptive
prognostics of performance of powertrain components in a vehicle, in accordance with
embodiments of the present disclosure,
[021] Figure 3 depicts an exemplary plot 300 representing division of engine or engine
component operating zone into a plurality of bins, in accordance with embodiments of
the present disclosure,
[022] Figure 4 depicts a plot 400 representing normalization of an attribute, in accordance
with embodiments of the present disclosure,
[023] Figure 5 depicts a plot 500 for defining reference attribute, in accordance with
embodiments of the present disclosure, and
[024] Figure 6 represents an exemplary flowchart of a method 600 for adaptive prognostics
of performance powertrain components in a vehicle, in accordance with embodiments
of the present disclosure.
[025] It should be appreciated by those skilled in the art that any block diagrams herein
represent conceptual views of illustrative systems embodying the principles of the
7
present subject matter. Similarly, it will be appreciated that any flow charts, flow
diagrams, state transition diagrams, pseudo code, and the like represent various
processes which may be substantially represented in a computer readable medium and
executed by a computer or processor, whether or not such computer or processor is
explicitly shown.
DETAILED DESCRIPTION
[026] The foregoing has broadly outlined the features and technical advantages of the present
disclosure in order that the detailed description of the disclosure that follows may be
better understood. It should be appreciated by those skilled in the art that the conception
and specific embodiment disclosed may be readily utilized as a basis for modifying or
designing other structures for carrying out the same purposes of the present disclosure.
[027] The novel features which are believed to be characteristic of the disclosure, both as to
its organization and method of operation, together with further objects and advantages
will be better understood from the following description when considered in connection
with the accompanying figures. It is to be expressly understood, however, that each of
the figures is provided for the purpose of illustration and description only and is not
intended as a definition of the limits of the present disclosure.
[028] Regular inspection and maintenance of vehicle components, especially those associated
with the powertrain are very crucial to the overall well-being of the vehicle and the
user. Conventionally, maintenance of such components is facilitated based on a
standard periodic maintenance schedule provided by the OEMs that is generally
designed based on the distance that the vehicle has travelled or age of the vehicle.
However, this approach does not account for the actual usage of the components, the
driving behavior of different users and ambient conditions during vehicle usage etc. and
may therefore lead to premature replacement of healthy components causing discard of
usable remaining life of the components or delayed replacement of unhealthy
components causing sudden component failure. Further, other approaches based on
Machine Learning also face certain limitations such as not being able to predict an
actual remaining life of a component and inaccurate predictions on the health status of
components due to inefficient training and lack of unhealthy data to train the model.
8
[029] In order to overcome the above-mentioned challenges, the present disclosure provides
a system and method for adaptive prognostics of performance of engine and powertrain
components in a vehicle considering the actual usage of the components, the driving
behavior of user and the ambient conditions. A detailed explanation of the proposed
solution is disclosed in the forthcoming paragraphs.
[030] Figure 1 depicts an exemplary environment 100 for adaptive prognostics of
performance of powertrain components in a vehicle, in accordance with an embodiment
of the present disclosure. The exemplary environment 100 depicts one or more
components 102a, 102b …. 102n such as injector, throttle, etc. associated with the
powertrain of a vehicle (not shown). It may be understood by a skilled person that the
powertrain of a vehicle comprises essential elements required to thrust the vehicle in
motion, including but not limited to engine, transmission, driveshaft, axles and
differential. The exemplary environment 100 further depicts a telematics device 104
that is configured to capture telematics data associated with each of the one or more
components 102a, 102b … 102n. Further, the exemplary environment 100 depicts a
system 106 configured to receive, in real-time, during a driving cycle of the vehicle,
the telematics data from the telematics device 104 and process the telematics data to
provide a final performance prognostics judgement of each of the one or more
components 102a, 102b ... 102n of powertrain of the vehicle. In one exemplary aspect,
the final performance prognostics judgement of each of the one or more components
102a, 102b ... 102n of powertrain of the vehicle may be shared with a user through a
user-authorized mobile widget 108. In another exemplary aspect, the final performance
prognostics judgement of each of the one or more components 102a, 102b ... 102n may
be transmitted to OEM’s service center 110 that may further notify the user of a required
action on any of the one or more components 102a, 102b ... 102n. A detailed
explanation of the exemplary environment 100 is provided in the forthcoming
paragraphs in conjunction with Figures 2, 3, 4 and 5.
[031] Figure 2 represents an exemplary block diagram 200 of a system 202 for adaptive
prognostics of performance of powertrain components in a vehicle (not shown), in
accordance with an embodiment of the present disclosure. The system 202 may
comprise an I/O interface 204, at least one processor 206, a memory 208 but not limited
thereto. The at least one processor 206 may be operatively coupled to the user interface
9
204 and the memory 208. Further, it may be noted by a skilled person that the system
202 may be a part of an Electronic Control Unit (ECU) of the vehicle or may be
implemented independently by placing it outside the ECU, for example in a server or
cloud.
[032] In one non-limiting embodiment, the at least one processor 206 may be implemented
as one or more microprocessors, microcomputers, microcontrollers, digital signal
processors, central processing units, state machines, logic circuitries, and/or any
devices that manipulate signals based on operational instructions. Among other
capabilities, the at least one processor 206 may be configured to fetch and execute
computer-readable instructions stored in the memory 208. The I/O interface 204 may
include a variety of software and hardware interfaces, for example, a web interface, a
graphical user interface, and the like.
[033] Now, referring to figure 1 and as described in the preceding paragraphs, the system 106
may receive in real-time during a driving cycle (or a trip), the telematics data for each
of the one or more components 102a, 102b ... 102n from the telematics device 104.
Now, in accordance with the embodiment of the system 202 as depicted in figure 2, the
telematics data for each of the one or more components 102a, 102b ... 102n may be
received by the at least one processor 206 through the I/O interface 204.
[034] Further, the telematics data may comprise data associated with one or more parameters
of a component being evaluated, obtained throughout the driving cycle. In one
exemplary aspect, considering the component being evaluated to be the engine of the
vehicle, then the one or more parameters may comprise engine speed and calculated
load. In another aspect, considering the component being evaluated to be the throttle,
then the one or more parameters may comprise engine speed and throttle angle.
Similarly, the one or more parameters may vary depending upon the type of component
of the powertrain. Additionally, the telematics data may also comprise attributes
corresponding to each of the one or more components 102a, 102b ... 102n and feature(s)
data associated with the attributes. In an exemplary embodiment, if one of the
components whose performance is to be evaluated is the injector component, then the
corresponding attribute may be fuel trim or fuel density and the feature(s) data may
comprise ambient temperature as one of the features influencing the attribute fuel trim
or fuel density. Therefore, it may be understood by a skilled person that an attribute is
10
associated with a component being evaluated and the features may be comprehended
as one or more factors influencing the attribute.
[035] Once the driving cycle ends, the at least one processor 206 may populate one or more
matrices, each matrix consisting of plurality of predefined bins with data points
received for each component separately. The same has been illustrated in Figure 3,
considering the one component to be evaluated as engine component (such as injector).
The data density in each bin represents component operation time in each bin with
respect to a total component operation time during the driving cycle. For instance, bin
X illustrated in figure 3 depicts a data density of 9.45% implying that for a total
component (or engine) operation duration, the component (or engine) spends 9.45% of
its total time in bin X where the engine speed is around 2500-3000 RPM, and the
calculated load is between 30-40 units. Further, the number of bins are pre-calibrated
by OEMs for different components, by dividing a component operating zone using the
one or more parameters into a plurality of equally sized grids (or bins). In simpler terms,
the number of bins in a matrix for a component may remain fixed. However, the
number of bins created may vary depending upon the type of component being
evaluated. Further, the pre-calibrated bins may be assigned an identification number
that remains constant throughout the vehicle life.
[036] Furthermore, the plurality of predefined bins may be aligned into one or more priority
groups. In one non-limiting embodiment, the priority groups may comprise a Priority
group – 1 (having the highest priority), a Priority group – 2 and a Priority group – 3
(having the lowest priority). Each priority group may be associated with a
corresponding remedial action that may be required to be taken on the component being
evaluated. For instance, the priority group – 1 may be associated with a “Need to
replace” action, the priority group – 2 may be associated with a “Need to repair” action,
and the priority group – 3 may be associated with a “Need to inspect” action. However,
it must be appreciated by a skilled person that a number of priority groups and the
association of a priority group with a remedial action may be selected in another
suitable manner other than the one disclosed herein.
[037] Further, the alignment of the plurality of predefined bins into the priority groups may
be based upon an operational behavior or stability of the component being evaluated in
said bin. For instance, the bins in which the operational behavior of the component is
11
highly erratic may be assigned to the priority group - 3 that may be highly sensitive to
any changes in the performance. Similarly, the bins in which the operational behavior
is fairly steady may be assigned to priority group – 1 that may have low sensitivity to
any changes in performance. In an additional embodiment, the bins are aligned into a
priority group at an initial drive cycle of the vehicle and remains same throughout the
life of the vehicle. Further, the importance and usage of aligning the plurality of
predefined bins into one or more priority groups has been explained in detail in the
forthcoming paragraphs.
[038] Upon populating the plurality of predefined bins, the at least one processor 206 may
compare the data density in each populated bin with a corresponding pre-calibrated data
density value to judge whether the component being evaluated has truly operated in a
particular bin or has spent at least a pre-fixed minimum amount of time in a particular
bin. For instance, in one driving cycle, considering there are around 1 lakh data points
that are to be populated in the plurality of predefined bins. Upon populating, it may be
observed that there exist certain bins where the number of data points is very less, say
less than 500 indicating that the component spends a very minimal amount of time in
those particular bins. Such bins may not prove to be useful for further analysis and
hence, the at least one processor 206 may discard such bins from further processing
based on the comparison. In one non-limiting embodiment, the at least one processor
206 may activate only those bins for further processing for which the data density value
is found to be greater than the corresponding pre-calibrated data density value as
depicted in Table 1, as an exemplary embodiment. Further, it may be noted by a skilled
person that the pre-calibrated data density value may be defined based on a knowledge
of the control system of the vehicle and may either be same for every bin or may vary
from one bin to another.
Bin No
Bin Data Density
(%)
Calibrated data
density (%)
Comparison
Result
BIN 1 0.00 0.5 Discard
…. …. …. ….
BIN 10 0.01 0.5 Discard
BIN 11 5 0.5 Activate
12
BIN 12 1 0.6 Activate
BIN 13 5 0.5 Activate
BIN 14 2 0.8 Activate
…. …. …. ….
BIN 120 5 0.5 Activate
Table - 1
[039] The variation of the pre-calibrated data density value from one bin to another may be
understood by way of an example of automatic or hybrid vehicles. For example, in an
automatic or hybrid vehicle, the engine stops every time the vehicle halts at a traffic
signal and is thus, cranked from 0-1000 RPM every time the engine starts again and
therefore, the data density in the bins corresponding to 0-1000 RPM may be very high.
However, such data points may not be useful for further analysis and hence, the
predefined calibrated data density value for such bins may be set higher than the precalibrated
data density value for other bins.
[040] Moving on, for each of the one or more activated bins, the at least one processor 206
may record one or more attributes corresponding to the one or more populated bins and
normalize the recorded one or more attributes. In one non-limiting embodiment, the
one or more attributes are normalized by taking into consideration the effect of a feature
on the one or more attributes. For instance, considering the example from preceding
paragraphs, the attribute fuel trim or fuel density is influenced by the feature ambient
temperature, and hence, the one or more attributes recorded may exhibit variation
because of fluctuations in the ambient temperature during the driving cycle. Thus, to
normalize the one or more attributes, the at least one processor 206 may first calculate
a sensitivity factor of each attribute, say ‘a’, for the feature, say ‘b’, by using the
equation (1)–
?????????????????????? ???????????? = ??/??……...(1)
[041] Further, the at least processor 206 may normalize the one or more attribute based on
the corresponding sensitivity factor and a predefined standard value for the feature as
illustrated in Figure 4. In view of the example described above, the standard value for
the feature ambient temperature may be considered as 25oC and the fuel trim attribute
13
for each activated bin may be normalized considering ambient temperature as 25oC as
also depicted in Table 2.
Bin No
Bin Data
Density
(%)
Calibrated
data density
(%)
Comparison
Result
Attribute Normalized
Attribute
BIN 1 0.00 0.5 Discard - -
…. …. …. …. …. ….
BIN 10 0.01 0.5 Discard - -
BIN 11 5 0.5 Activate 1 1.02
BIN 12 1 0.6 Activate 1 1.02
BIN 13 5 0.5 Activate 1 1.02
BIN 14 2 0.8 Activate 1.01 1.1
…. … …. …. …. ….
BIN
120
5 0.5 Activate 1 1.02
Table - 2
[042] Next, the at least one processor 206 may compare the one or more normalized attributes
with corresponding one or more defined reference attributes to identify each of the one
or more activated bins as healthy or unhealthy, as illustrated in Table 3 below. In one
non-limiting embodiment, the one or more defined reference attributes are generated
by auto-learning by the at least processor 206. Further, to define/auto-learn the one or
more defined reference attributes for each bin, the at least one processor 206 may set
one or more nominal reference attribute values in accordance with the specification or
predefined initial operation of the component being evaluated. The at least one
processor 206 may further modulate the set one or more nominal reference attribute
values based on the one or more normalized attributes and a predefined calibration
factor for a predefined engine operation duration threshold based on the equation (2)
?????????????????? ?????????????????? = ?????????????? ?????????????????? + (???????????????????? ?????????????????????????????????? ??????????????????)
?????????????????????? ???????????? … (2)
[043] Further, it may be noted by a skilled person that the one or more defined reference
attributes for each bin are independent or unique. This helps to account for any kind of
parts variation which may impact components performance to be evaluated.
14
[044] An exemplary embodiment illustrating the defining of the one or more reference
attributes is illustrated in Figure 5 for the predefined engine operation duration
threshold of 100hrs, where the one or more reference attributes are frozen after 100hrs
of engine ON-off cycles.
Bin
No
Bin
Data
Density
(%)
Calibrated
data
density
(%)
Comparison
Result
Attribute Normalized
Attribute
Reference
Attribute
Status
BIN
1
0.00 0.5 Discard - - - -
…. …. …. …. …. …. …. ….
BIN
10
0.01 0.5 Discard - - - -
BIN
11
5 0.5 Activate 1 1.02 1.05 Healthy
BIN
12
1 0.6 Activate 1 1.02 1.08 Healthy
BIN
13
5 0.5 Activate 1 1.02 1.05 Healthy
BIN
14
2 0.8 Activate 1.01 1.1 1.09 Unhealt
hy
…. … …. …. …. …. …. ….
BIN
120
5 0.5 Activate 1 1.02 1.08 Healthy
Table - 3
[045] Now, to compare the one or more normalized attributes with corresponding one or more
defined reference attributes, the at least one processor 206 may employ any
conventionally known comparison logic, including but not limited to, greater than-less
than comparison and other suitable comparison logics. Based on the comparison, the at
least one processor 206 may identify a bin as healthy if the result of comparison is a
success and may identify a bin as unhealthy if the result of comparison fails.
15
[046] The identification of bins as healthy or unhealthy provides a preliminary indication of
performance status of a component. However, to avoid any false detections by the
system 202, another level of processing is performed by the at least one processor 206
so as to provide a final performance evaluation of the component being evaluated.
[047] To achieve this, the at least one processor 206 may first identify the health status of the
one or more bins in each priority group. For this, the at least one processor 206 may
utilize a pre-defined unhealthy bin weightage for each priority group. In one nonlimiting
embodiment, the unhealthy bin weightage may denote a maximum allowable
unhealthy bins for each priority group. The at least one processor 206 may then
calculate, from a total number of activated bins in each priority group, a number of
unhealthy bins for each priority group and compare, for each priority group, the number
of identified unhealthy bins with the corresponding predefined unhealthy bin weightage
and provide a health status for each priority group. An illustrative example for the same
is depicted in table 4.
Priority BIN Unhealthy
bin
Weightage
Unhealthy
bin %
Health
status
Performance
Evaluation
1
[Need to
Replace]
BIN 30 to
BIN 60 10% 8% Healthy
Need to Repair 2
[Need to
Repair]
BIN 61 to
BIN 100 30% 35% Unhealthy
3
[Need to
Inspect]
BIN 101-
BIN 120 BIN 1 - 50% 60% Unhealthy
BIN 29
Table - 4
[048] As illustrated in table – 4, the unhealthy bin weightage for priority group – 1 is 10%,
i.e., from the 30 bins in the priority group – 1, if only 20 bins are activated, the
remaining 10 bins will be discarded from further analysis, and the at least processor
206 would identify the priority group -1 as unhealthy if the number of unhealthy bins
in priority group – 1 is greater than 2 (10% of 20). However, in the example depicted
above, the bins in priority group – 1 are identified as healthy as the number (or
16
percentage) of unhealthy bins is less than the unhealthy bin weightage of 10%.
Similarly, the at least one processor 206 may identify a health status for each priority
group.
[049] Based on the identified health status of each priority group, the at least one processor
206 may provide a final performance prognostic judgement for the component being
evaluated. Now, in accordance with the example depicted in table 4, the bins belonging
to priority groups 2 and 3 are identified as unhealthy. To provide the final performance
prognostic judgement, the at least one processor 206 may provide a remedial action
suggestion associated with the higher priority group, that is, in said example, the
performance prognostic judgement for the component being evaluated is “Need to
repair” as associated with the priority group – 2. Therefore, based on the health status
of bins in each priority group and a degree of priority of each priority group, the at least
one processor 206 may provide a final performance prognostic judgement.
[050] Further, in one non-limiting embodiment, the at least one processor 206 may utilize the
I/O interface 204 to provide the final performance prognostic judgement of the
component being evaluated either to the user-authorized mobile widget 108 or may
transmit the final performance OEM’s service center 110 that may further notify the
user. In an exemplary embodiment, the user-authorized mobile widget 108 may be a
dedicated application designed to provide the final performance prognostic judgement
and thereby alert the user of the action to be taken for the one or more components
(102a, 102b…102n).
[051] In this manner, the system 106, 202 may provide a comprehensive prognosis of
performance of the one or more components 102a, 102b, 102c taking into account the
actual usage of the components, ambient conditions and other external factors.
[052] Figure 6 illustrates a flowchart 600 of an exemplary method for dynamically evaluating
performance of engine and powertrain components in a vehicle in accordance with an
embodiment of the present disclosure. The method 600 may also be described in the
general context of computer executable instructions. Generally, computer executable
instructions may include routines, programs, objects, components, data structures,
procedures, modules, and functions, which perform specific functions or implement
specific abstract data types.
17
[053] The order in which the method 600 is described is not intended to be construed as a
limitation, and any number of the described method blocks may be combined in any
order to implement the method. Additionally, individual blocks may be deleted from
the methods without departing from the spirit and scope of the subject matter described.
[054] At step 602, the method 600 may include receiving, in real-time during each driving
cycle, telematics data associated with each of the one or more components (102a, 102b
… 102n). In one non-limiting embodiment, for receiving the telematics data, the at least
one processor 206 may be used in conjunction with the I/O interface 204. Further, in
one implementation, receiving telematics data may comprise receiving one or more
parameters associated with a component being evaluated, attributes associated with the
component being evaluated and feature(s) data associated with the attributes. In an
exemplary embodiment, if one of the components whose performance is to be evaluated
is the injector component, then the corresponding attribute may be fuel trim or fuel
density and the feature data may comprise ambient temperature as one of the features
influencing the attribute fuel trim or fuel density.
[055] At step 604, the method 600 may include populating, at the end of each driving cycle,
one or more matrices consisting of plurality of predefined bins with data points received
for each component separately, indicating a component operation time in each bin
during the driving cycle. In one non-limiting embodiment, the at least one processor
206 may be used for said populating. Further, in one aspect, the number of bins may be
pre-calibrated by OEMs for different components, by dividing a component operating
zone for any given driving cycle using the one or more parameters into a plurality of
equally sized grids (or bins). Further, the plurality of predefined bins may be aligned
into one or more priority groups at an initial drive cycle of the vehicle and remains same
throughout the life of the vehicle. In one aspect, the priority groups may comprise a
Priority group – 1 (having the highest priority), a Priority group – 2 and a Priority group
– 3 (having the lowest priority) and each priority group may be associated with a
corresponding remedial action that may be required to be taken on the component being
evaluated.
[056] At step 606, the method 600 may include comparing data density value of each
populated bin with a corresponding pre-calibrated data density value. In one non-
18
limiting embodiment, for said comparing, the at least one processor 206 may be used
to judge whether the component being evaluated has truly operated in a particular bin
or has spent at least a pre-fixed minimum amount of time in a particular bin.
[057] At step 608, the method 600 may include activating one or more populated bins when
the data density value in a populated bin is found greater than the pre-calibrated data
density value. In one non-limiting embodiment, for said activating, the at least one
processor 206 may be used.
[058] At step 610, the method 600 may include recording one or more attributes
corresponding to the one or more activated bins and normalizing the one or more
attributes. In one non-limiting embodiment, for said recording and normalizing, the at
least one processor 206 may be used. Further, it may be noted by a skilled person that
for normalizing the one or more attributes, the at least one processor 206 may calculate
a sensitivity factor based on the attribute and feature and utilize the calculated
sensitivity factor in combination with a standard value for the feature.
[059] At step 612, the method 600 may include comparing one or more normalized attributes
with corresponding one or more defined reference attributes to identify each of the one
or more activated bins as healthy or unhealthy. In one non-limiting embodiment, for
said comparing, the at least one processor 206 may be used that may employ a suitable
comparison logic. Further, in one aspect, for defining the one or more defined reference
attributes for each bin of plurality of pre-defined bins (i.e., the defined reference
attribute for each bin is unique), the method 600 may include may setting one or more
nominal reference attribute values in accordance with the specification of the
component being evaluated. The method 600 may further include modulating the set
one or more nominal reference attribute values based on the one or more normalized
attributes and a predefined calibration factor for a predefined engine operation duration
threshold based on the equation (2) described in preceding paragraphs.
[060] At step 614, the method 600 may include calculating, from a total number of activated
bins in each priority group, a number of unhealthy bins in each priority group and
comparing the calculated number of unhealthy bins against a predefined unhealthy bin
weightage. In one non-limiting embodiment, for said calculating and comparing, the at
least one processor 206 may be used.
19
[061] At step 616, the method 600 may include identifying a health status of each priority
group based on the comparison. In one non-limiting embodiment, for said identifying
the at least one processor 206 may be used. Further, in one aspect, the at least one
processor 206 may identify a priority group as healthy if the number of unhealthy bins
in said priority group is less than the predefined unhealthy bin weightage for said
priority group, otherwise said priority group is identified as unhealthy.
[062] At step 618, the method 600 may include providing final performance prognostic
judgement for the component being evaluated based on the identification. In one nonlimiting
embodiment, for providing the final performance prognostic judgement, the at
least one processor 206 may be used in conjunction with the I/O interface 204. To
provide the final performance prognostic judgement, the at least one processor 206 may
provide a remedial action suggestion associated with the higher priority group. For
instance, in accordance with the example described in table-4, the final performance
prognostic judgement for the component being evaluated is “Need to repair” as
associated with the priority group – 2. Therefore, based on the health status of bins in
each priority group and a degree of priority of each priority group, the at least one
processor 206 may provide a final performance evaluation.
[063] The illustrated steps are set out to explain the exemplary embodiments shown, and it
should be anticipated that ongoing technological development will change the manner
in which particular functions are performed. These examples are presented herein for
purposes of illustration, and not limitation. Further, the boundaries of the functional
building blocks have been arbitrarily defined herein for the convenience of the
description. Alternative boundaries can be defined so long as the specified functions
and relationships thereof are appropriately performed.
[064] Alternatives (including equivalents, extensions, variations, deviations, etc., of those
described herein) will be apparent to persons skilled in the relevant art(s) based on the
teachings contained herein. Such alternatives fall within the scope and spirit of the
disclosed embodiments.
[065] Furthermore, one or more computer-readable storage media may be utilized in
implementing embodiments consistent with the present disclosure. A computer-
20
readable storage medium refers to any type of physical memory on which information
or data readable by a processor may be stored. Thus, a computer-readable storage
medium may store instructions for execution by one or more processors, including
instructions for causing the processor(s) to perform steps or stages consistent with the
embodiments described herein. The term “computer- readable medium” should be
understood to include tangible items and exclude carrier waves and transient signals,
i.e., are non-transitory. Examples include random access memory (RAM), read-only
memory (ROM), volatile memory, non-volatile memory, hard drives, CD ROMs,
DVDs, flash drives, disks, and any other known physical storage media.
[066] Suitable processors include, by way of example, a general-purpose processor, a special
purpose processor, a conventional processor, a digital signal processor (DSP), a graphic
processing unit (GPU), a plurality of microprocessors, one or more microprocessors in
association with a DSP core, a controller, a microcontroller, Application Specific
Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) circuits, any
other type of integrated circuit (IC), and/or a state machine.
21
REFERENCE NUMERALS
Reference
Numeral
Component
102a, 102b.
102c
Components of engine and power train
104 Telematics Device
106 System
108 User-authorized mobile widget
110 OEM service center
202 System
204 I/O Interface
206 At least one processor
208 Memory
210 Telematics Device
600 Method , Claims:We Claim:
1. A method for adaptive prognostics of performance of one or more components associated
with powertrain in a vehicle, the method comprising:
receiving (602), in real-time during each driving cycle, telematics data associated with
each of the one or more components (102a, 102b … 102n) wherein the telematics data
comprises one or more parameters associated with a component being evaluated, attributes
corresponding to the component being evaluated and feature(s) data associated with the
attributes;
populating (604), at the end of each driving cycle, a matrix consisting of plurality of
predefined bins with data points received for each component separately, indicating an
component operation time in each bin during the driving cycle, wherein each of the plurality
of predefined bins are aligned into one or more priority groups;
comparing (606) data density value of each populated bin with a corresponding precalibrated
data density value;
activating (608) one or more populated bins when the data density value in a populated
bin is found greater than the pre-calibrated data density value;
recording (610) one or more attributes corresponding to the one or more activated bins
and normalizing the one or more attributes;
comparing (612) one or more normalized attributes with corresponding one or more
defined reference attributes to identify each of the one or more activated bins as healthy or
unhealthy;
calculating (614), from a total number of activated bins in each priority group, a number
of unhealthy bins in each priority group and comparing the calculated number of unhealthy
bins against a predefined unhealthy bin weightage of each priority group;
identifying (616) a health status for each priority group based on the comparison; and
providing (618) final performance prognostic judgement for the component being
evaluated based on the identification.
2. The method as claimed in claim 1, wherein the plurality of predefined bins are aligned into
one or more priority groups based on an operational behavior or stability of the component
being evaluated in each bin of the plurality of predefined bins,
wherein when two or more priority groups are identified as unhealthy, the final
prognostic performance judgement is based on a higher priority group among the two or more
priority groups, and
wherein providing the final performance prognostic judgement comprises providing a
remedial action suggestion to either one of a user-authorized mobile widget or OEM service
center.
3. The method as claimed in claim 1, wherein the plurality of predefined bins are created by
dividing a component operating zone for a given driving cycle using the one or more
23
parameters, associated with the component, into a plurality of equally sized grids, wherein each
grid represents a bin, and wherein a number of the plurality of predefined bins varies with
respect to a type of the component being evaluated.
4. The method as claimed in claim 1, wherein normalizing the one or more attribute values
further comprises:
normalizing each of the one or more attribute values based on a sensitivity factor
corresponding to each of the one or more attribute values and a standard value associated with
a feature associated with the one or more attribute values.
5. The method as claimed in claim 1, further comprises defining the one or more defined
reference attribute values by:
setting one or more nominal reference attribute values based on a specification or predefined
initial operation of the component being evaluated; and
modulating the set one or more nominal reference attribute values based on the one or
more normalized attribute values and a predefined calibration factor,
wherein the one or more defined reference attributes are unique for each of the plurality
of predefined bins.
6. A system for adaptive prognostics of performance of one or more components associated
with powertrain in a vehicle, the system comprising:
an I/O interface (204);
a memory (208); and
at least one processor (206) operatively coupled to the I/O interface (204) and the
memory (208), wherein the at least one processor (206) is configured to:
receive, in real-time during each driving cycle, from a telematics device (104, 210),
telematics data associated with each of the one or more components (102a, 102b … 102n)
associated with the powertrain, wherein the telematics data comprises one or more
parameters associated with a component being evaluated, attributes corresponding to the
component being evaluated and feature(s) data associated with the attributes;
populate, at the end of each driving cycle, a matrix consisting of plurality of
predefined bins with data points received for each component separately, indicating a
component operation time in each bin during the driving cycle, wherein each of the
plurality of predefined bins are aligned into one or more priority groups;
compare data density value of each populated bin with a corresponding precalibrated
data density value;
activate one or more populated bins when the data density value in a populated bin
is found greater than the pre-calibrated data density value;
record one or more attributes corresponding to the one or more activated bins and
normalize the one or more attributes;
compare one or more normalized attributes with corresponding one or more
defined reference attributes to identify each of the one or more activated bins as healthy
or unhealthy;
24
calculate, from a total number of activated bins in each priority group, a number of
unhealthy bins, in each priority group and compare the calculated number of unhealthy
bins against a predefined unhealthy bin weightage of each priority group;
identify a health status for each priority group based on the comparison; and
provide a final performance prognostic judgement for the component being
evaluated based on the identification.
7. The system as claimed in claim 6, wherein the at least one processor (206) is configured to
align the plurality of predefined bins into one or more priority groups based on an operational
behavior or stability of the component being evaluated in each bin of the plurality of predefined
bins,
wherein when two or more priority groups are identified as unhealthy, the final
prognostic performance judgement is based on a higher priority group among the two or more
priority groups, and
wherein to provide the final performance prognostic judgement, the at least one
processor (206) is further configured to:
provide a remedial action suggestion to either one of a user-authorized mobile
widget or OEM service center.
8. The system as claimed in claim 6, wherein to create the plurality of predefined bins, the at
least one processor (206) is configured to divide a component operating zone for a given
driving cycle using the one or more parameters into a plurality of equally sized grids, wherein
each grid represents a bin, and wherein a number of the plurality of predefined bins varies with
respect to a type of the component being evaluated.
9. The system as claimed in claim 6, wherein to normalize the one or more attribute values, the
at least one processor (206) is further configured to:
normalize each of the one or more attribute values based on a sensitivity factor
corresponding to each of the one or more attribute values and a standard value associated with
a feature associated with the one or more attribute values.
10. The system as claimed in claim 6, wherein the at least one processor (206) is further
configured to define the one or more defined reference attribute values, wherein to define the
one or more defined reference attributes, the at least one processor (206) is further configured
to:
set one or more nominal reference attribute values based on a specification or predefined
initial operation of the component being evaluated; and
modulate the set one or more nominal reference attribute values based on the one or
more normalized attribute values and a predefined calibration factor,
wherein the one or more defined reference attributes are unique for each of the plurality
of predefined bins.