U.S. patent application number 15/625591 was filed with the patent office on 2018-12-20 for system and method for measuring and diagnosing initial offsets of an analog input sensor.
This patent application is currently assigned to GM GLOBAL TECHNOLOGY OPERATIONS LLC. The applicant listed for this patent is GM GLOBAL TECHNOLOGY OPERATIONS LLC. Invention is credited to Siddharth Ballal, Brian A. Welchko, Wesley G. Zanardelli.
Application Number | 20180367077 15/625591 |
Document ID | / |
Family ID | 64457685 |
Filed Date | 2018-12-20 |
United States Patent
Application |
20180367077 |
Kind Code |
A1 |
Welchko; Brian A. ; et
al. |
December 20, 2018 |
SYSTEM AND METHOD FOR MEASURING AND DIAGNOSING INITIAL OFFSETS OF
AN ANALOG INPUT SENSOR
Abstract
An electric power system includes a polyphase electric machine,
battery pack, power inverter module, analog input sensor, and
diagnostic controller executing a method. The sensor measures an
electrical parameter that differs from a true value of the
parameter by an initial offset value. The controller collects
sample sets of the parameter, compares the initial offset of each
sample to an outlier threshold in a first diagnostic loop, and
transmits a bit flag indicative of an outlier sample to a slower
second diagnostic loop when the initial offset of a sample exceeds
the outlier threshold. The second control loop calculates a rolling
average of the initial offsets of the sample sets, discards the set
containing the outlier sample in response to the bit flag, and
executes a control action when the average exceeds a threshold that
is lower than the outlier threshold.
Inventors: |
Welchko; Brian A.; (Oakland,
MI) ; Ballal; Siddharth; (Shelby Township, MI)
; Zanardelli; Wesley G.; (Rochester, MI) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
GM GLOBAL TECHNOLOGY OPERATIONS LLC |
Detroit |
MI |
US |
|
|
Assignee: |
GM GLOBAL TECHNOLOGY OPERATIONS
LLC
Detroit
MI
|
Family ID: |
64457685 |
Appl. No.: |
15/625591 |
Filed: |
June 16, 2017 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
B60L 2240/429 20130101;
B60L 2250/10 20130101; H02P 29/0241 20160201; H02P 23/0077
20130101; B60L 3/003 20130101; B60L 2240/427 20130101; B60L 3/0046
20130101; B60L 2240/527 20130101; B60L 2240/547 20130101; B60L
2240/529 20130101; H02P 23/14 20130101; B60L 2250/16 20130101; B60L
2240/526 20130101; Y02T 10/64 20130101; B60L 3/12 20130101 |
International
Class: |
H02P 23/00 20060101
H02P023/00; H02P 23/14 20060101 H02P023/14; B60L 11/18 20060101
B60L011/18 |
Claims
1. An electric power system comprising: a polyphase electric
machine having a plurality of phase windings; a battery pack
connected to a direct current (DC) voltage bus; a power inverter
module (PIM) connected to the battery pack via the DC voltage bus,
and to the electric machine via the phase windings; an analog input
sensor configured to measure an electrical parameter of the
electric power system, wherein the measured electrical parameter
differs from a true value of the electrical parameter by an initial
offset value; and a diagnostic controller in communication with the
analog input sensor, and configured to: collect sample sets of the
electrical parameter; compare the initial offset of each sample in
the collected sample sets to a calibrated outlier threshold using a
first diagnostic loop; and transmit a bit flag indicative of an
outlier sample from the first diagnostic loop to a slower second
diagnostic loop when the initial offset of one or more of the
samples exceeds the outlier threshold, the second control loop
further being configured to: calculate a rolling average of the
initial offsets of the collected sample set; discard the sample set
containing the outlier sample in response to the bit flag; and
execute a control action with respect to the system when the
calculated rolling average exceeds a rolling average threshold that
is lower than the outlier threshold.
2. The electric power system of claim 1, wherein the analog input
sensor includes a plurality of phase current or phase voltage
sensors each electrically connected to a corresponding one of the
phase windings.
3. The electric power system of claim 2, wherein the outlier
threshold is at least three times greater than the rolling average
threshold.
4. The electric power system of claim 1, wherein the analog input
sensor includes a voltage bus sensor connected to the DC voltage
bus.
5. The electric power system of claim 1, wherein the first
diagnostic loop has a cycle time of that is 1 and 2 orders of
magnitude faster than a cycle time of the second diagnostic
loop.
6. The electric power system of claim 1, wherein the controller
executes the control by setting or transmitting a diagnostic code
to memory of the controller or to a remote device.
7. The electric power system of claim 1, wherein the electric power
system is used as part of a motor vehicle having road wheels and a
transmission, and wherein the electric machine has an output shaft
connected to the road wheels via the transmission.
8. A method for measuring and diagnosing an initial offset of an
analog input sensor in an electric power system, wherein the
measured electrical parameter differs from a true value of the
electrical parameter by the initial offset, the method comprising:
collecting sample sets of the electrical parameter via the analog
input sensor; calculating and comparing the initial offset of each
sample in the collected sample sets to a calibrated outlier
threshold using a first diagnostic loop of a controller;
transmitting a bit flag indicative of an outlier sample from the
first diagnostic loop to a slower second diagnostic loop of the
controller when the initial offset of one or more of the samples
exceeds the outlier threshold; calculating a rolling average of the
initial offsets of the collected sample sets via the second
diagnostic control loop; discarding the sample set containing the
outlier sample in response to the bit flag; and executing a control
action with respect to the system via the controller when the
calculated rolling average exceeds a rolling average threshold that
is lower than the outlier threshold, including recording a
diagnostic code in memory of the controller or transmitting the
diagnostic code to a remote device.
9. The method of claim 8, wherein the analog input sensor includes
a plurality of phase sensors each connected to a corresponding one
of the phase windings and the parameter is a corresponding phase
current or phase voltage of the electric machine.
10. The method of claim 8, wherein the outlier threshold is at
least three times greater than the rolling average threshold.
11. The method of claim 8, wherein the analog input sensor includes
a voltage bus sensor connected to the DC voltage bus.
12. The method of claim 8, wherein the wherein the first diagnostic
loop has a cycle time of that is 1 or 2 orders of magnitude faster
than a cycle time of the second diagnostic loop.
13. The method of claim 8, further comprising using the electric
machine to rotate an output shaft connected via a transmission to a
set of road wheels of a vehicle.
14. A vehicle comprising: a transmission; and an electric power
system comprising: a polyphase electric machine having a plurality
of phase windings, and further having an output shaft connected to
the transmission; a battery pack connected to a direct current (DC)
voltage bus; a power inverter module (PIM) connected to the battery
pack via the DC voltage bus, and to the electric machine via the
phase windings; a plurality of analog input sensors each configured
to measure a corresponding phase current or phase voltage of the
electric machine, wherein the measured phase current differs from a
true value of the phase current or phase voltage by an initial
offset; and a diagnostic controller in communication with the
analog input sensors and configured, in response to an ignition-on
or key-on event of the vehicle, to collect sample sets of the phase
currents or phase voltages and compare the initial offset of each
sample in the collected sample sets to a calibrated outlier
threshold using a first diagnostic loop, transmit a bit flag
indicative of an outlier sample from the first diagnostic loop to a
second diagnostic loop that has a slower cycle speed than the first
diagnostic loop when the initial offset of one or more of the
samples exceeds the outlier threshold, the second control loop
further being configured to calculate a rolling average of the
initial offsets of the collected sample sets, discard the sample
set containing the outlier sample in response to the bit flag, and
execute a control action with respect to the system when the
calculated rolling average exceeds a rolling average threshold.
15. The vehicle of claim 14, wherein the outlier threshold is at
least three times larger than the rolling average threshold.
16. The vehicle of claim 14, wherein the controller executes the
control by setting or transmitting a diagnostic code to memory of
the controller or to a remote device.
Description
INTRODUCTION
[0001] Electric power systems are used to provide a reliable source
of torque in a variety of beneficial applications. An electric
power system typically includes a polyphase electric machine that
is connected to a battery pack via a power inverter. When the
individual phase windings of the electric machine are energized
using pulse-width modulation or other types of high-speed
semiconductor switching control of the power inverter, an output
shaft of the electric machine delivers a defined torque. The
rotating output shaft alternatively powers a coupled load or
generates electricity depending on the operating mode and
configuration of the electric machine.
[0002] Precise operational control over the functions of the
electric machine and other connected power electronic components of
the electric power system is predicated on the accurate real-time
determination of certain electrical parameters. To this end, analog
input sensors may be used within the electric power system to
directly measure and report such parameters. Analog sensors have a
performance quality known as an initial offset that describes a
bias-based variance of the sensor's voltage output reading from an
expected reading. For instance, when using a current sensor at 0
amps, the sensor may output a voltage corresponding to .+-.15 amps,
in which case the initial offset of the sensor is 15 amps. For
improved accuracy, therefore, the initial offset of a given sensor
is determined and subtracted from the sensor's reported values
before use of the measured value in a control system.
SUMMARY
[0003] An electric power system of the type described herein has
one or more analog input sensors, such as but not limited to phase
current sensors or voltage bus sensors. Also described is an
associated method for measuring and diagnosing the initial offset
of such sensors. The present approach is intended to improve the
robustness of existing sensor diagnostic methodologies,
particularly in response to transient voltage dips or sags that may
occur on an auxiliary (low-voltage) bus upon system start-up.
[0004] In an example embodiment of the disclosed method, separate
but interrelated diagnostic loops of a diagnostic controller
function together to collect and average sets of electrical data
samples, with the diagnostic loop performing the data collection
function at a faster cycle speed/loop speed than the diagnostic
loop that is performing the core diagnostic function. For
illustrative clarity, the diagnostic control loops are described
herein as "fast" and "slow", with the fast loop possibly operating
in the kilohertz (kHz) range in some embodiments and the slow loop
functioning at about 100 Hz, or with the fast loop being about 1 or
2 orders of magnitude faster than the slow loop.
[0005] The fast loop is configured to set a bit flag when the
initial offset of a given electrical data sample exceeds a
calibrated offset threshold, with the calibrated offset threshold
set well in excess of a historic average offset. As used herein,
the term "well in excess of" means at least about 3-4 times the
historic rolling average of the initial offset, with such a sample
referred to herein as an "outlier sample". When the fast loop bit
flag is set, the slow loop may respond by automatically discarding
the set of data samples to which the outlier sample belongs. The
fast loop may thereafter attempt to collect another set of
electrical data samples to replace the discarded set.
[0006] At the same time, the slow loop performs an X/Y fast-pass
process, i.e., with a control action performed with respect to the
system when an average of a predetermined threshold (X) of the
collected data samples fails within a predetermined number (Y) of
consecutive sample sets. In this manner, the present approach
provides a specific improvement to the overall operational accuracy
of analog input sensor diagnostic computers in electric power
systems of the type described herein.
[0007] An electric power system is also disclosed that is
controlled via the method noted above. The system may include a
polyphase electric machine having a plurality of phase windings and
a rotatable output shaft, a battery pack connected to a direct
current (DC) voltage bus, a power inverter module (PIM), an analog
input sensor, and a diagnostic controller. The PIM is connected to
the battery pack via the DC voltage bus, and to the electric
machine via the phase windings. The analog input sensor measures an
electrical parameter of the electric power system, such as a phase
current or voltage or a DC bus voltage.
[0008] The diagnostic controller in this embodiment is configured
to diagnose the initial offset value upon start-up of the system.
The controller does so by collecting sample sets of the electrical
parameter and comparing an initial offset of each sample in the
collected sample sets to a calibrated outlier threshold using a
first diagnostic loop, and then transmitting a bit flag indicative
of an outlier sample from the first diagnostic loop to a slower
second diagnostic loop when the initial offset of one or more of
the samples exceeds the calibrated outlier threshold. The second
control loop is also configured to calculate a rolling average of
the initial offsets of the collected sample sets, discard the
sample set containing the outlier sample in response to the bit
flag, and execute a control action with respect to the system when
the calculated rolling average exceeds an average threshold that is
lower than the outlier threshold.
[0009] The diagnostic controller executes the control action in
certain embodiments by setting or transmitting a diagnostic code to
memory of the controller or to a remote device, such as via a
vehicle telematics unit.
[0010] The electric power system may be used as part of a motor
vehicle having road wheels and a transmission, with the electric
machine having an output shaft connected to the road wheels via the
transmission.
[0011] In another example embodiment, a vehicle includes a
transmission and the electric power system, in this instance with
the sensors embodied as a plurality of analog current sensors each
configured to measure a corresponding phase current or voltage of
the electric machine. The controller is configured, in response to
an ignition-on or key-on event of the vehicle, to perform the
above-described method, and to execute a control action with
respect to the system when the calculated rolling average exceeds a
rolling average threshold, e.g., of between 25 and 35 percent of
the calibrated outlier threshold.
[0012] The above-described and other features and advantages of the
present disclosure are readily apparent from the following detailed
description of the best modes for carrying out the disclosure when
taken in connection with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0013] FIG. 1 is a schematic illustration of an example motor
vehicle having an electric power system, an analog input sensor,
and a diagnostic controller configured to execute a method for
diagnosing initial offset values of the sensor.
[0014] FIGS. 2A and 2B are logic flow diagrams describing an
example embodiment of the present method with respect to respective
slow and fast control loops of the controller.
[0015] The present disclosure is susceptible to various
modifications and alternative forms, and some representative
embodiments have been shown by way of example in the drawings and
will be described in detail herein. Novel aspects of this
disclosure are not limited to the particular forms illustrated in
the drawings. Rather, the disclosure is intended to cover
modifications, equivalents, combinations, or alternatives falling
within the spirit and scope of the disclosure as defined by the
appended claims.
DETAILED DESCRIPTION
[0016] Referring to the drawings, wherein like reference numbers
refer to like components throughout the several views, FIG. 1
depicts an example electric power system 20 having one or more
analog input sensors 32. The sensors 32 may be variously embodied
as individual phase current or phase voltage sensors, as direct
current (DC) voltage bus sensors 32A, or other types of analog
input sensors having a predefined output voltage range, e.g.,
.+-.0-5 V.sub.DC, with each value in the range corresponding to a
particular measured input value. For example, when using the
sensors 32, an actual/true value of .+-.200 A-600 A may be measured
a high-current embodiment or .+-.60-300 V.sub.DC or more in a
high-voltage embodiment. The measured parameter differs from the
true value by an initial offset value, as noted above and
understood in the art.
[0017] The electric power system 20 further includes a diagnostic
controller (C) 50 that is programmed and otherwise configured to
perform instructions of a method 100 for measuring and diagnosing
the initial offset values of the analog input sensors 32 or 32A,
with the method 100 described in detail below with references to
respective subroutines 100A and 100B of FIGS. 2A and 2B. For
illustrative consistency, an example application is described for
the method 100 in which the electric power system 20 of FIG. 1 is
used as part of a motor vehicle 10, such as a battery electric
vehicle or a hybrid electric vehicle, with such a vehicle 10 having
a vehicle body 12 and road wheels 14 in rolling contact with a road
surface 16. However, the electric power system 12 is not limited to
mobile applications in general or automotive applications in
particular. Possible applications include stationary power plants,
appliances, robots, and other such systems constructed using the
analog input sensors 32 or 32A and other basic electrical
components as shown in FIG. 1.
[0018] The electric power system 20 includes one or more electric
machines (M.sub.E) 22 energized when a polyphase/alternating
current output voltage (V.sub.AC) is applied by a power inverter
module (PIM) 26 to individual phase windings 31 of the electric
machine 22. The PIM 26 is electrically connected to a DC voltage
bus 33 providing a DC voltage (V.sub.DC) from a high-voltage
battery pack 24 (B.sub.HV). A DC-DC voltage converter 28 may be
connected to the DC voltage bus 33 and controlled so as to reduce
the voltage level from the battery pack 24 to lower voltage
auxiliary levels on an auxiliary voltage bus (V.sub.AUX), e.g.,
12-15 V.sub.DC. A lead-acid or other suitable auxiliary battery
(B.sub.AUX) 30 may be connected to the DC-DC converter 28 and used
to power connected auxiliary devices (not shown) such as radios,
lights, and auxiliary motors.
[0019] With respect to the electric machine 22, this device
includes an output shaft 27 that rotates when the electric machine
22 is energized, which occurs in response to a controlled internal
semiconductor switching operation of the PIM 26. Rotation of the
output shaft 27 provides motor output torque (arrow T.sub.M) to a
connected load, e.g., the road wheels 14 in the example motor
vehicle 10. In such an embodiment, a transmission (T) 23 may be
disposed between the electric machine 22 and the road wheels 14,
with the motor output torque (arrow T.sub.M) ultimately transferred
through one or more gear sets or a continuously variable pulley
configuration to provide a transmission output torque (arrow
T.sub.O). Similarly, an internal combustion engine (E) 25 may be
cranked and started in some embodiments using the motor output
torque (arrow T.sub.M), e.g., in a belted alternator starter
configuration, with the engine 25 coupled with and delivering
engine torque (arrow T.sub.E) to the transmission 23 via a suitable
drive connection 36, for example via an input clutch (CI) in the
form of a friction clutch or a hydrodynamic torque converter.
[0020] The diagnostic controller 50 of FIG. 1, which is in
communication with the electric power system 20 over a controller
area network (CAN) bus or other suitable communications channels,
includes a processor (P) and memory (M). The memory (M) may include
tangible, non-transitory memory, e.g., read only memory, whether
optical, magnetic, flash, or otherwise. The controller 50 also
includes sufficient amounts of random access memory,
electrically-erasable programmable read only memory, etc., as well
as a high-speed clock, analog-to-digital and digital-to-analog
circuitry, and input/output circuitry and devices, as well as
appropriate signal conditioning and buffer circuitry.
[0021] In particular, the diagnostic controller 50 is programmed or
otherwise configured to execute instructions embodying the method
100, an example of which is depicted in FIGS. 2A and 2B as slow
loop and fast loop subroutines 100A and 100B, respectively. To this
end, the controller 50 is in communication with the analog input
sensors 32, such as the three separate phase sensors 32
electrically connected to a corresponding one of the phase windings
of the electric machine 22 as shown. Thus, the controller 50 is
configured to receive the measured phase currents I.sub.A, I.sub.B,
and I.sub.C from the three sensors 32 in a three-phase current
sensing embodiment, or three corresponding measured phase voltages
(not shown) in a possible voltage sensor configuration.
[0022] Alternatively, two such sensors 32 may be used to measure
two of the three possible phase currents or voltages in a 3-phase
embodiment of the electric machine 22, with the third phase current
or voltage calculated in logic of the controller 50 using the two
measured values. However, the illustrated three-sensor embodiment
may be used to provide increased fault tolerance. Also illustrated
is an optional embodiment in which the sensor 32A is disposed on
the DC voltage bus and used to measure the DC bus voltage
(V.sub.DC), with the method 100 being readily modified for such use
as noted below with particular reference to FIGS. 2A and 2B.
[0023] In the execution of the method 100, the controller 50
receives input signals (arrow 11) from the electric power system 20
in addition to the measured sensor values from the analog input
sensors 32 or 32A. For instance, the controller 50 receives input
signals (arrow 11) such as ambient temperature, key/ignition switch
on/off position, running state of the electric machine 22 or PIM
26, or other information pertaining to determining entry conditions
for execution of the method 100. The controller 50 may also be in
communication with a remote device 35 such as an indicator lamp,
telematics unit, or display screen, and configured to selectively
output a diagnostic code (arrow D) to the remote device 35 in
response to certain diagnostic results.
[0024] Operation of the controller 50 in the ongoing diagnosis of
the analog input sensors 32 or 32A will now be described with
reference to FIGS. 2A and 2B, which respectively describe the
above-noted slow and fast diagnostic loops in the form of
subroutines 100A and 100B. While the actual cycle time of each loop
may vary with the intended application and number of samples per
sample set, for illustrative purposes the slow loop may have a
cycle time on the order of about 100 Hz, with the cycle time of the
fast loop being 1 or 2 orders of magnitude greater than that of the
slow loop.
Diagnostic Slow Loop
[0025] Beginning with initialization (*) of the controller 50 of
FIG. 2A and continuing with step S102 of the subroutine 100A, the
controller 50 determines whether certain diagnostic entry
conditions are satisfied. As the method 100 captures data
concurrently with certain operating conditions in which the output
voltage of the auxiliary battery 30 may temporarily sag or dip, the
entry conditions may include a detected key-on or ignition-on
state, as reported to the controller 50 of FIG. 1 via the input
signals (arrow 11). Also as part of the entry conditions, the
controller 50 may determine whether a rotational output speed of
the electric machine 22 is zero and switching control of the PIM 26
has not commenced. The subroutine 100A proceeds to step S104 if
these entry conditions are not satisfied, and to step S106 in the
alternative when the entry conditions are satisfied.
[0026] Step S104 includes setting a corresponding bit flag to 0 or
"FALSE", resetting the counters used in the various steps below,
and returning to the initialization step (*).
[0027] Step S106 includes setting a corresponding bit flag to 1 or
"TRUE" and proceeding to step S108. Also as part of step S106, the
controller 50 may instruct the fast loop to commence collection of
electrical data from the analog input sensors 32, e.g., via a
handshake communication protocol.
[0028] At step S108, the controller 50 determines whether a
sufficient sample size of initial offset values of the sensors 32
has been collected, such as by comparing the sample size to a
predetermined threshold. In certain applications, a sample set of
about 40 to 125 samples may be deemed sufficient for the purpose of
average calculation. Subroutine 100A repeats step 108 until the
threshold sample size has been collected and then proceeds to step
S110.
[0029] Step S110 includes setting the corresponding bit flag to
0/FALSE and commencing averaging of the collected samples. The
controller 50 may add the values of the sampled offsets and then
divide this number by the total number of samples in the sample
set. The controller 50 then resets the corresponding bit flag to
1/TRUE and increments a sample counter to indicate a successful
averaging function. The subroutine 100A then proceeds to step
S112.
[0030] At step S112, the controller 50 next determines whether the
calculated average offset from step S110 exceeds a calibrated
average threshold over a predetermined number of sample sets, for
instance by performing an X/Y fast-pass process. At step S112, the
controller 50 also determines, via receipt of a bit flag from the
diagnostic fast loop described below with reference to FIG. 2B,
whether the fast loop has detected, for the present sample set, one
or more discrete samples exceeding a calibrated outlier threshold.
The subroutine 100A proceeds to step S114 if either condition is
true, and to step S116 in the alternative.
[0031] Step S114 includes incrementing a failure counter and
proceeding to step S116. When step S112 described above determines
that the diagnostic fast loop has detected a sample exceeding an
outlier threshold, step S114 may also include resetting the bit
flag and discarding the sample set having such a sample and
attempting to collect a clean data set lacking such an outlying
sample. Doing so may allow the output of the sensor 32 to recover
from a transient voltage spike or dip.
[0032] Step S116 includes determining, via the controller 50,
whether more than a calibrated number of allowable failures have
been detected. If so, the subroutine 100A proceeds to step S118.
Otherwise, the subroutine 100A proceeds to step S120.
[0033] At step S118, the controller 50 generates a diagnostic code
indicative of a failing average (high average over a calibrated
number of sample sets). The diagnostic code may be recorded in
memory (M) of the controller 50, or such a code may be communicated
to the remote device 35 of FIG. 1 as the diagnostic code (arrow D).
As the particular problem of a failing battery 30 of FIG. 1 may
manifest itself upon start up, the diagnostic code (arrow D) may
report a need for servicing or replacement of the battery 30. The
subroutine 100A thereafter proceeds to step S124.
[0034] Step S120 includes determining whether a passing threshold
has been achieved. Similar to step S118, step S120 may include
comparing the number of passing sample sets to a calibrated passing
threshold, and repeating step S102 if such a threshold is not
achieved. Step S122 is executed in the alternative when the passing
threshold is achieved.
[0035] Step S122 is analogous to step S118, however with the
controller 50 in this instance generating a diagnostic code
indicative of a passing average (acceptable average over a
calibrated number of sample sets). The passing diagnostic code may
be recorded in memory (M) of the controller 50, or such a code may
be communicated to the remote device 35 of FIG. 1 as the diagnostic
code (arrow D). The subroutine 100A thereafter proceeds to step
S124.
[0036] Step S124 entails resetting the above-noted counters to
zero, and then exiting the slow loop (**). The subroutine 100A may
commence anew at the next key-on or ignition-on event satisfying
the entry conditions of step S102.
Diagnostic Fast Loop
[0037] FIG. 2B depicts an example embodiment of a subroutine 100B
suitable for performing data collection and discrete threshold
comparison as part of the method 100. Upon initialization (***) of
the controller 50, the subroutine 100B determines at step S101 if
the average flag noted above with reference to step S106 is TRUE,
which once again means that entry conditions for executing method
100 have been satisfied. If so, the subroutine 100B proceeds to
step S103. Otherwise, the fast loop exists (****) and awaits
satisfaction of the entry conditions.
[0038] Step S103 includes collecting the sensor data from the
analog input sensors 32. As part of step S103, the controller 50
may also add the present measurements to a sum of
previously-collected measurements for a given sample set, as well
as increment a sample counter for the present sample set. Such
counter values may be used by the slow loop of subroutine 100A to
evaluate when a sufficient sample set has been collected as well as
to ascertain the average offset value for the data set. The
subroutine 100B then proceeds to step S105.
[0039] At step S105 the controller 50 determines, for the present
data sample, whether the sample exceeds a calibrated outlier
threshold. As stated above, in an illustrative embodiment in which
the sensors 32 are connected to a respective one of the phase
windings of the electric machine 22 of FIG. 1 and configured to
measure and report a corresponding phase value, e.g., a phase
current I.sub.A, I.sub.B, or I.sub.C, an example outlier threshold
may be an absolute current magnitude that is set sufficiently
higher, e.g., 3-4 times higher, than the threshold used for the
rolling average of collected offset values. The outlier threshold
should be set high enough that inclusion of the sample value in the
sample set could skew the average for the data set, and thus the
actual threshold may be expected to vary with the intended
application. The subroutine 100B proceeds to step S107 when the
measured sample exceeds the outlier threshold, with the subroutine
100B exiting (****) when all samples in the sample set are below
such a threshold.
[0040] Step S107 includes setting a bit flag that is indicative of
the outlier threshold having been exceeded in step S105. The bit
flag is communicated to the slow loop and subroutine 100A of FIG.
2A for use in step S112. The subroutine 100B then exits (****) and
starts anew with the next iteration or ignition/key-on event
satisfying the requisite entry conditions.
[0041] The method 100, an example of which is described above with
reference to subroutines 100A and 100B of FIGS. 2A and 2B, may
therefore be used to improve existing diagnostic approaches. That
is, a check is added in the diagnostic fast loop of FIG. 2B that
communicates to the slow loop of FIG. 2A that one of the sample
data points falls outside of a measurement norm, i.e., exceeds the
above-described outlier threshold. One or more such elevated values
may skew the measured offset while still passing under pure
average-based diagnostic methods. The present approach, while
taking slightly more time to collect a replacement data set to
replace the discarded set containing outliers, is intended to
improve the robustness of the system 12 by allowing for recovery of
the diagnostic method 100 from transient voltage dips on the
auxiliary voltage bus (V.sub.AUX) of FIG. 1.
[0042] While aspects of the present disclosure have been described
in detail with reference to the illustrated embodiments, those
skilled in the art will recognize that modifications may be made
without departing from the scope of the present disclosure. The
present disclosure is not limited to the precise construction and
compositions disclosed herein, as modifications or variations
apparent from the foregoing descriptions are within the scope of
the disclosure as defined in the appended claims. Moreover, the
present concepts may expressly include combinations and
sub-combinations of the preceding elements and features.
* * * * *