Aircraft
Heat Exchange Test Conditions Simulator
by
Sundaram Raghuraman Engineering Team Leade
Roger Hartzell Testing Laboratory Manager
T.W. Sutterfield Technical Services Manager (SWSW)
V I Engineering, Inc. and Stewart Warner South Wind Corp.
Products Used:
Lookout
FieldPoint Ethernet Network Module
FieldPoint Thermocouple Input Module
FieldPoint Analog Output Module
FieldPoint Analog Input Module
FieldPoint SPST Relay Module
The Challenge: Simulate test conditions on an aircraft
heat exchanger. Flow, temperature and pressure had to
be accurately cycled according to a user specified test
recipe.
The Solution: Employing Lookout and FieldPoint modules,
V I Engineering, Inc. developed a test system which simultaneously
manipulates eight valves, utilizing several PID loops
to modulate the temperatures, pressures and flow rates
according to the test recipe.
Introduction
V I Engineering, Inc., a National Instruments Select Integrator
and Alliance member was contracted to develop a control
system for an Aircraft Heat Exchanger Test System. As
part of the design validation of a heat exchanger, it
is necessary to subject a test unit to conditions experienced
during a flight. This is established by valves, electric
heater, furnace, and compressors to adjust the temperatures,
pressures and flow rates of both hot and cold airflows.
Using Lookout, a test system was developed that provided
PID control of the eight valves to regulate hot air flow
(also called buffer air) and cold air flow. The user can
define a “recipe” that describes a complete
cycle by specifying the temperatures, pressures, and flow-rates
at various times. The system then executes the cycle repeatedly,
and logs data to the Citadel database as well as in a
Microsoft Excel file. Simultaneously, the Main GUI, as
seen in figure 1, is updated with the latest input and
output values. The trend charts, seen in Figure 2, communicate
directly with the Citadel database and run in parallel
with the test system.
Approach and Solution
Lookout’s object-based approach enabled creating
a solution by integrating available PID and Fieldpoint
objects. The combination of these tools and VIE’s
and SWSW’s expertise allowed for design of a tailor-made
solution that permitted ease of operation, flexibility
and stability necessary to execute 35,000 cycles of a
user defined test recipe.
Heat Exchanger Test System
The purpose of the system is to perform tests to help
validate design of prototype aircraft heat exchangers.
A test consists of executing a recipe 35,000 times. A
test that performs PID control using six variables, with
no variable completely independent of others, would tax
any system for a day. However a system had to be developed
that could perform precise temperature, pressure and flow
control for several days as well as log data to a database
and provide feedback to the user during execution. Lookout
proved to be up to the task.
As shown in Figure 1, Lookout provides the ability to
design an intuitive user interface for a real-world system.
Security features available in Lookout were implemented.
This allowed access control such that certain actions
(pressing buttons, changing set-points etc) can be performed
only by authorized users.
A very useful feature of Lookout was the ability to create/edit
objects on the fly without stopping test execution. This
proved to be a crucial factor during software testing
because this obviated the need to shut down the system
to make changes. Considering that it takes about 30 minutes
to bring the system up to temperature and pressure after
each start-up, shutting down the system to make software
changes would have made the development and testing process
excruciatingly slow.
Problems arose during the system testing because of interference
between various control valves. Although, it was initially
assumed that each valve would only control one process
variable (for eg. V07 would control only Buffer Air Pressure),
in reality, the process variables were inter-related according
to the laws of physics of gas flows. Another problem was
caused by the different time constants for each process
variable. For example, pressures and flow-rates change
much more rapidly than temperatures. This resulted in
instability problems during PID control.
This necessitated a more complex control scheme, such
that the role of each control valve was also made to vary
during the test cycle. For example, valve V04 would control
Cooling Air Flow in steps 4-6 and 9-11, and then Cooling
Air Temperature in steps 1-3 and 7-8. Another aspect of
the new control scheme was staging and direct-control
(non-PID control) of the valves during set-point changes,
and then switching to PID control of the valves after
the error between set-points and process variables were
minimized. In other words, flows, temperatures and pressures
were no longer the only variables, the valves themselves
have also become a factor in this control scheme.
Considerable access was provided to user to help tune
the PID loops, and to have complete control over the test
execution as shown in Figure 3. The objects were linked
to the PID parameters of the selected valve using the
DataTable object. The Control Mode is used to set a valve
to PID or Direct control. There is also a feature for
importing direct-control settings of valves from a file,
as well as ramping of set-points (PID) or manual outputs
(direct). The Set point Mode allows the user to control
set points of the cycle while under PID control. Under
normal test conditions, the system will be in cycle control,
where the system will follow the recipe defined by the
user. In set-point mode, user can change set-points interactively,
thus providing the ability to over-ride the set-points
defined in the recipe at any point within the test cycle.
Conclusion
Although the system presented many challenges and a few
surprises, successful development and testing of the control
software was possible with Lookout. Amazingly, the software
developed for the system will easily fit on a standard
1.44 MB diskette!
|