CONTROL OF A FLUID CIRCUIT USING AN ESTIMATED SENSOR VALUE
TECHNICAL HELD
[0001] The present invention relates generally to the control of an electro-
hydraulic system, and in particular to an apparatus and method for maintaining
control and operation of an electro-hydraulic system or fluid circuit having a failed
pressure or position sensor.
BACKGROUND OF THE INVENTION
[0002] Electro-hydraulic systems or fluid circuits utilize various electrically-
actuated and hydraulically-actuated devices, alone or in combination, to provide open-
loop or closed loop feedback control. In a closed-loop system in particular, feedback
mechanisms or sensors can be used to monitor circuit output values. Each sensor can
generate a signal that is proportional to the measured output, and using a suitable
control logic device or controller the output can be compared to a particular input or
command signal to determine if any adjustments or control steps are required.
Sensors for use in an electro-hydraulic fluid circuit ordinarily include pressure
transducers, temperature sensors, position sensors, and the like.
[0003] In a conventional fluid circuit, the precise control of the operation of
the fluid circuit can be maintained by continuously processing the various measured
or sensed output values. Supply and tank pressures, as well as pressures operating on
particular ports or chambers of a control valve, cylinder, or fluid motor used within
the circuit, can be continuously fed to a control unit or controller. However, system
control can be lost or severely degraded in a conventional fluid circuit if any of the
required pressure or position sensors fails or ceases to function properly for whatever
reason. While certain code-based methods exist for detecting out-of-range sensor
operation, or for determining shorted or open circuits, such methods usually result in a
temporary shutdown of the process utilizing the fluid circuit, and therefore can be less
than optimal when continuous fluid circuit operation is required.
SUMMARY OF THE INVENTION
[0004] Accordingly, an electro-hydraulic system or fluid circuit includes a
sump or a tank configured for holding a supply of fluid, a hydraulic device having a

predetermined load configuration, and a puinp for drawing fluid from the tank and
delivering it under pressure to the hydraulic device. Sensors are adapted for
measuring a supply pressure, a tank pressure, and a position of a moveable spool
portion or other moveable portion of the hydraulic device, as well as one or more
additional valves, such as a fluid conditioning valve positioned in fluid parallel with
the hydraulic device. A controller has an algorithin suitable for estimating or
reconstructing an output value of a failed one of any of the plurality of sensors hi the
fluid circuit using the predetermined load configuration, thereby ensuring the
continued operation of the hydraulic device and the fluid circuit.
[0005] Using the method of the invention, which can be embodied by the
computer-executable algorithm mentioned above, at least some level of control can be
maintained over the fluid circuit despite the presence of the failed sensor. A quasi-
steady analysis of the fluid circuit can capture the fundamentals of the fluid circuit. In
a fluid circuit having a pump, a reservoir or tank, a plurality of check valves and/or
fluid conditioning valves, and a cylinder, fluid motor, or other device having a first
and a second work chamber or port, unknown variables Qa, Qb. and Qfcv are present,
wherein Qa describes the flow into and out of a first work chamber of the cylinder, Qb
is the flow into and out of a second work chamber of the cylinder, and Qfcv is the flow
through an orifice of a fluid conditioning valve positioned or connected in fluid
parallel with the cylinder and pump. In accordance with the invention, a fluid circuit
configured in this manner can be modeled via a predetermined set of non-linear
equations that differ depending on the failed state of the fluid circuit, i.e., a failure of a
sensor occurring when the fluid circuit, is active, that is, when fluid is flowing from
the work chamber a to the work chamber b, or from work port b to a, as described
below.
[0006] The method therefore allows for the estimating or reconstructing of an
otherwise lost or unavailable sensor signal using a calibrated, known, or
predetermined load configuration, e.g., in a two-port device such as a cylinder or fluid
motor, the relationship between the flow rates through the respective work chambers
or ports. A fluid circuit adapted for executing the method can include a controller
having an algorithm suitable for processing the output values from a plurality of
pressure and position sensors, calculating any required flow information using
calibrated volumetric and measured pressure and/or other required data in conjunction

with the pressure and position measurements, and estimating the missing sensor value
using a set of non-linear equations. The controller then automatically controls the
fluid circuit using the estimated value until such time as the sensor can be diagnosed,
repaired, or replaced.
[0007] More particularly, the method allows for the estimation or
reconstruction of an output value of any one sensor of a plurality of sensors in a fluid
circuit having a controller, a pump, a tank, a hydraulic device, and a fluid
conditioning valve. The conditioning valve is in fluid parallel with the hydraulic
device. The method includes sensing a set of output values from the plurality of
sensors, processing the output values using the controller to determine the presence of
a failed sensor, and using the controller to calculate an estimated output value of the
failed sensor using a predetermined load configuration of the hydraulic device. The
hydraulic device can be controlled using the estimated output value until the failed
sensor can be repaired or replaced, thereby ensuring continuous operation of the fluid
circuit.
[0098] The above features and advantages and other features and advantages
of the present invention are readily apparent from the following detailed description
of the best modes for carrying out the invention when taken in connection with the
accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] FIG. 1 is a schematic illustration of an exemplary fluid circuit in a first
sensory failure state having a controller in accordance with the invention;
[0010] FIG. 2 is a schematic illustration of the exemplary fluid circuit of FIG.
1 in a second sensory failure state: and
[0011] FIG. 3 is a flow chart describing a control method usable with the fluid
circuit of FIGS. 1-2.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0012] Referring to the drawings wherein like reference numbers correspond
to like or similar components throughout the several figures, and beginning with FIG.
1, a fluid circuit. 10 is shown in a first possible sensory failure state, as will be
described below. The fluid circuit 10 includes a pump (P) 12 and a low-pressure
reservoir, sump, or tank 14. The tank 14 holds or contains a supply of fluid 15, which

is drawn by the pump 12 and delivered under pressure (Ps) via a supply line 11 to a
hydraulic device 24. In the exemplary embodiment of FIG. 1, the hydraulic device 24
is configured as a dual-chamber cylinder 27 containing a spool or piston 26, with the
cylinder 27 having a first and a second work port, 31 and 33, respectively, in
communication with the work chambers a and b defined by and within the cylinder 27
and piston 26.
[0013] Control logic or an algorithm 100 for executing the method of the
invention can be programmed or recorded within a controller (C) 30 and implemented
to selectively control the various fluid control devices within the fluid circuit 10 as
needed to power a downstream fluid circuit (FC) 28, including items such as but not
limited to hydraulic machinery, valves, pistons, accumulators, etc. The FC 28 in turn
is in fluid communication with the tank 14 via a return line 13.
[0014] The controller 30, which can be directly wired to or in wireless
communication with the various components of the fluid circuit 10, receives a set of
pressure and position input signals (arrow 25) from sensors 18A-D and 19A-C, as
explained below. The fluid circuit 10 can be configured as a digital computer
generally including a CPU, and sufficient memory such as read only memory (ROM),
random access memory (RAM), electrically-programmable read only memory
(EPROM), etc. The controller 30 can include a high speed clock, analog to digital
(AID) and digital to analog (D/A) circuitry, and input/output circuitry and devices
(I/O), as well as appropriate signal conditioning and buffer circuitry. Any algorithms
resident in the controller 30 or accessible thereby, including the algorithm 100
described below with reference to FIG. 3, or any other required algorithms, can be
stored in ROM and automatically executed by the controller 30 to provide the
required circuit control functionality.
[001S] The fluid 15 is selectively admitted into the fluid circuit 10 via the
supply line 11 at the supply pressure (Ps). A fluid conditioning valve 16 is positioned
iu fluid parallel with the hydraulic device 24 between a pair of pressure sensors 18A
and 18B, e.g., pressure transducers or other suitable pressure sensing devices. The
sensor 18A is positioned and adapted for measuring the supply pressure (Pt), while
the sensor 18B is positioned and adapted for measuring the return line or tank
pressure (Pt). As needed, some or all of the fluid 15 flowing from the pump 32 can be

diverted from the hydraulic device 24 through, the conditioning valve 16 and back to
the tank 14.
[0016] The fluid circuit 10 includes position sensors 19A, 19B, and 19C
adapted for measuring the position of respective spools in the conditioning valve 16,
the valve 20, and the valve 22, respectively. Additional pressure sensors 18C, 18D
are positioned in fluid series with the hydraulic device 24. The sensor 18C is
positioned and adapted for measuring the fluid pressure (Pa) operating on work
chamber a or the first work port 31 of the hydraulic device 24, and is positioned
downstream of a first valve 20. The first valve 20 can be configured as any suitable
fluid control valve suitable for directing fluid 15 from the pump 12 in the direction of
arrow C, and into the first work port 31 of the hydraulic device 24 in order to move
the piston 26 in the direction of arrow C. A second valve 22 prevents a flow of fluid
15 into the work port 33. The sensor 18D is positioned and adapted for measuring the
fluid pressure (Pb) operating on work chamber b or the second work port 33 of the
hydraulic device 24.
[0017J Under normal operating conditions, the variables Ps, Pi Pa, and Pb are
known, being sensed or measured by the respective pressure sensors 18A-18D. The
position variables xa, xb, and xfcv are also known, being sensed by the position sensors
19A-C. The variables xa and xb describe the position of the piston 26 in work
chambers a and b, respectively, while xfcv describes the position of a spool portion of
the fluid conditioning valve 16. Three unknown variables include Qa, Qb, and Qfcv as
noted above, i.e., the flow into the first work port 31. the second work port 33, and the
conditioning valve 16, respectively. A unique solution is thus provided for these
values using the following three-function equation set:



is the discharge coefficient, p is the density of the fluid, and A is the orifice area as a
function of spool position.
[0018] However, in a sensory failure state in which one of the sensors 18A-D
or 19A-C fails, the set of equations above cannot be uniquely solved without resorting
to additional information. For example, if the pressure at work port 31 or Pa is
unavailable due to a failure of sensor 18C, the remaining known variables are Ps, Pb
We now have four unknown variables, i.e.,
before, as well as the unknown value of Pa.
[0019] In an observer-based model, state variables can be estimated by
comparing the model outputs to actual measurements. A signal can be easily
reconstructed only if the system itself is fully observable. However, observer-based
models are severely challenged in the face of unknown load conditions, such as the
velocity of a piston positioned within a fluid cylinder, a portion of a fluid motor, or
any moveable portion of a typical two-port fluid device.
[0020] For example, a fluid circuit can be modeled via the following equation:

wherein Pa refers to the change in fluid pressure at a first port or "work port a" of a 2-
port device,/ is the bulk modulus of the fluid used in the circuit, Vis the volume of
the cylinder, is the flow rate through work port a, Ps is the supply pressure, P„ is
the pressure at chamber a or work port 31, and xa is the spool position of a spool or
piston at chamber a or work port 31. Additionally, A is the cross-sectional area of the
cylinder, and is the rate of change in position of the cylinder, i.e., the velocity
thereof. The value is an unknown load condition hi such an exemplary cylinder.
[0021] Using the algorithm 100, the load configuration of the hydraulic device
24 can provide further constraints as determined using the unknown variables. For
example, Qa = -Qb for a cylinder/motor connection as shown in FIGS. 1 and 2, if the
work chambers on either side of the cylinder 27 are equally sized, or Qa = (Aa
where 4, is piston area in work chamber a and Ab is position area in work
chamber b, if the work chambers a and b are differently sized. Therefore, the

algorithm 100 can use non-linear equations to determine the unknown three variables
in a first sensory failure mode. Accordingly, any one of the sensor signals Pa Pb Pab
Pb, xa, and xb, can be estimated using the above equations.
[0022] Referring to FIG. 2, the fluid circuit 10 of FIG. 1 is shown in a second
failure sensory state, i.e., when fluid is being applied at work port 33 to move the
piston 26 in the direction of arrow D. As above, any one of the missing seasor signals
Ps, Pt, Pa, Pb, xa, and xb can be estimated or reconstructed using the known load
configuration for the hydraulic device 24.
[0023] Referring to FIG. 3 in conjunction with the fluid circuit 10 of FIGS. 1
and 2, the method of the invention can be executed via the algorithm 100. Beginning
at step 102, the controller 30 continuously or in accordance with a specified periodic
cycle time reads the output values from each of the sensors 18A - D and 19A - C. In
normal operation, the controller 30 processes these values using control logic, and
selectively actuates the hydraulic device 24 and, if used, any additional downstream
devices in the downstream fluid circuit 28 according to such control logic. The
algorithm 100 men proceeds to step 104.
[0024] At step 104, the controller 30 determines whether any of the sensors
1SA-D and 19A-C! has failed. If not, the algorithm 100 is finished, effectively
resuming with step 102 and repeating steps 102 and 104 until such a sensor failure is
determined to be present. If a sensor has failed, the algorithm 100 proceeds to step
106.
[0025] At step 106, the algorithm 100 estimates or reconstructs the value for
the failed sensor. This estimated value is represented in FIG. 3 as the value (e). For
example, if the sensor 18C has failed the output value Pa would be unavailable as a
result. Continuing with the example of sensor 18C, the unknown variables would be
However, given a known load configuration such as Qa = - Qb
for the cylinder or motor connection shown in FIGS. 2 and 3, the four unknowns
reduce to three: The algorithm 100 then uses the non-linear
[0026] Once the estimated value (e) has been determined or calculated at step
106, the algorithm 100 proceeds to step 108, wherein the controller 30 executes
control of the fluid circuit 10 of FIGS. 1 and 2 using the estimated value (e).

Continued control of die fluid circuit 10 can therefore be maintained. The algorithm
100 can then be finished, or can optionally proceed to step 110.
[0027] At step 110, an alarm can be activated, or another suitable control
action can be taken, to ensure that attention is draws to the presence of the failed
sensor. In this maimer, the sensor failure can be properly diagnosed, repaired, or
replaced as needed.
[0028] Accordingly, using the control algorithm 100 as set forth above as part
of the fluid circuit 10 of FIGS. 1 and 2, single sensor fault operation of the fluid
circuit 10 can be achieved. Given the load configuration, it is possible to reconstruct
most of a single failed sensor signal if service is running at the time of the sensor
failure. If service stops, i.e., if both work ports 31 and 33 of the hydraulic device 24
close, it can be difficult to accurately estimate the failed sensor signal.
[0029] While the best modes for carrying out the invention have been
described in detail, those familiar with the art. to which this invention relates will
recognize various alternative designs and embodiments for practicing the invention
within the scope of the appended claims. Likewise, while the invention has been
described with reference to a preferred embodiments), it will be understood by those
skilled in the art that various changes may be made and equivalents may be
substituted for elements thereof without departing from the scope of the invention, In
addition, many modifications may be made to adapt a particular situation or material
to the teachings of the invention without departing from the essential scope thereof.
Therefore, it is intended that, the invention not; be limited to the particular embodiment
disclosed as the best mode contemplated for carrying out this invention, but that the
invention will include all embodiments falling within the scope of the appended
claims.

we claims
1. A fluid circuit (10) comprisiag:
a tank (14) configured for holding fluid (15);
a hydraulic device (24) having a predetermined load configuration;
a pump (12) operable for drawing the fluid (15) from the tank (14) and
delivering the fluid (15) under pressure to the hydraulic device (24):
a plurality of sensors (18A-D, 19A-C) each adapted for measuring at
least one of a supply pressure (Ps) froui the pump (12), a tank pressure (Pt) at the tank
(14), and a position of a moveable portion (26) of the hydraulic device (24); and
a controller (30) having an algorithm (100) that is adapted for
estimating an output value of any one sensor of the plurality of sensors (18A-D. 19A-
C) using the predetermined load configuration when a predetermined failure occurs in
the one sensor, thereby ensuring continued operation of the hydraulic device (24).
2. The fluid circuit (10) of claim 1, wherein the hydraulic device
(24) is one of a cylinder-and-piston device and a fluid motor device.
3. The fluid circuit (10) of claim 1, further comprising a fluid
conditioning valve (16) in fluid parallel with the hydraulic device (24), wherein the
fluid conditioning valve (16) has a moveable portion, and wherein the plurality of
sensors (18A-D, 19A-C) includes a first position sensor (19C) for measuring a
position (xfcv) of the moveable portion of the fluid conditioning valve (16).
4. The fluid circuit (10) of claim 1, wherein the hydraulic device
(24) has a first and a second work port (31, 33), and wherein the predetermined failure
is a failure occurring when the fluid (15) is being delivered from the pump (12) to one
of the first work port (31) and the second work port (33).
5. The fluid circuit (10) of claim 1, wherein the algorithm (100) is
adapted for estimating the output value using a predetermined set of non-linear
equations.

6. A fluid control system adapted for use with a fluid circuit (10)
having a tank (14) configured for holding fluid (15), a hydraulic device (24) having a
piston (26) disposed in a cylinder (27) to define a first and a second work port (31, 33)
in conjunction therewith, a fluid conditioning valve (16) having a spool portion, and a
pump (12) operable for drawing the fluid (15) from the tank (14) and delivering the
fluid (15) under pressure to one of the first and the second work ports (31, 33), the
fluid control system comprising:
a set of pressure sensors (18A-D) each adapted for measuring one of a
supply pressxue (Ps) from the pump (12), a tank pressure (Pt) at the tank (14), a first
pressure (Pa) at the first work port (31), and a second pressure (Pb) at the second work
port (33);
a set of position sensors (19A-C) adapted for measuring a respective
position (xfcv) of the spool portion of the conditioning valve (16) and a position (xa,.xb)
of the piston (26); and
a controller (30) having an algorithm (100) that is adapted for
estimating an output value of any one sensor of the pressure and position sensors
(18A-D, 19A-C) using a predetermined load configuration of the hydraulic device
(24) in the event of a predetermined failure of the one sensor, thereby ensuring
continued operation of the hydraulic device (24).
7. The fluid control system of claim 6, wherein the predetermined
load configuration is modeled within the controller (30) as a calibrated equation
describing a ratio of the flow rates through the first and the second work ports (31,
33).
8. The fluid control system of claim 6, wherein the algorithm
(100) estimates the output value by calculating solutions to a set of three different
non-linear equations.
9. The fluid control system of claim 8, wherein each of the non-
linear equations is a function of a flow rate through one of the hydraulic device (24)
and the fluid conditioning valve (16).

10. The fluid control system of claim 9, wherein each of the non-
linear equations is a function of the tank pressure (Pt), the supply pressure (Ps), a
position (xa, Xb) of the piston (26), and a position (xfcv) of the spool portion of the
conditioning valve (16).
11. A method (100) for estimating or reconstructing an output,
value of any one sensor of a plurality of sensors (18A-D, 19A-C) in a fluid circuit
(10) having a controller (30), a pump (12), a tank (14), a hydraulic device (24), and a
fluid conditioning valve (16) in fluid parallel with the hydraulic device (24), the
method comprising:
sensing a set of output values (Ps, Pt, Pa, Pb, xa, xb xfcv) from the
plurality of sensors (18A-D, 19A-C);
processing the set of output values (Ps,, Pt, Pa, Pb, xa, xb, xfcv) using the
controller (30) to thereby determine the presence of a failed sensor among the
plurality of sensors (18A-D. 19A-C);
using the controller (30) to calculate an estimated output value of the
failed sensor in response to the determination of the failed sensor, wherein calculation
of the estimated value uses a predetermined load configuration of the hydraulic device
(24); and
automatically controlling an operation of the hydraulic device (24)
using the estimated output value until the failed sensor can be repaired or replaced,
thereby ensuring continuous operation of the fluid circuit (10).
12. The method (100) of claim 11, wherein processing the set of
output values (Ps, Pt, Pa, Pb. xa, xb, xfcv) includes comparing each output value in the
set of output values (Ps, Pt, Pa, Pb. xa, xb, xfcv) to a calibrated threshold to determine the
presence of the failed sensor.
13. The method (100) of claim 11, wherein calculating an
estimated output value of the failed sensor includes using the predetermined load
configuration to derive a set of non-linear equations having just three unknown
variables.

14. The method (100) of claim 13, wherein using the controller
(30) to calculate an estimated output value includes solving for one of the three
unknown variables to thereby determine the estimated output value.
15. The method (100) of claim 11, wherein the hydraulic device
(24) has a pair of work ports (31, 33), and wherein the predetermined load
configuration is a calibrated flow ratio of the pair of work ports (31, 33).

A fluid circuit (10) includes a tank (14) for holding
fluid (15), a hydraulic device (24) having a
predetermined load configuration, and a pump (12) for
delivering the fluid (15) under pressure to the
hydraulic device (24). Sensors (18A-D. 19A-C) measure
at least one of a supply pressure (Ps), a tank pressure
(Pt) , and a position (xa, xb) of a portion of the
hydraulic device (24). A controller (30) estimates or
reconstructs an output value of any one sensor using
the predetermined load configuration in the event of a
predetermined failure of that sensor, ensuring
continued operation of the hydraulic device (24). A
method (100) for estimating the output value includes
sensing output values using the sensors (18A-D, 19 A-
C) , processing the output values using the controller
(30) to determine the presence of a failed sensor, and
calculating an estimated output value of the failed
sensor using the predetermined load configuration.
Operation of the hydraulic device (24) is maintained
using the estimated output value until the failed
sensor can be repaired.