Experimental Flight Test Management. Optimization procedures and flight test techniques

3 
INTRODUCTION 
Approaching to flight test is paramount to keep clear in mind that accurate test 
management is the cornerstone between failure and success. Flight testing remains 
an essential element of sound air vehicle development. The current emphasis on 
expanding the use of M&S has been promulgated with the intention that it can 
help to reduce flight test time and cost, enhance test safety, and increase testing 
efficiency.  
The “predict-test-validate” (a.k.a. ”model-test-model”) paradigm is held forth 
as the most efficient combination of these development tools. In this paradigm the 
initial modelling and simulation guides the planning and conduct of flight testing, 
with incremental test results then used to enhance the accuracy and/or fidelity of 
the simulation before the process is repeated. The cycle would be repeated many 
times during the course of the test program, especially in an effort to avoid the 
"fly-fix-fly" paradigm that commonly proves inefficient and trying to avoid future 
operational shortfalls. 
Although much of the technical leadership in the NATO aerospace industry 
and Italian Defence Department insist that M&S is not intended to replace flight 
testing, there remains concern among flight test practitioners that the result will be 
an overreliance on simulation. This has a potential for neglecting invaluable 
empirical test data verifying system performance. In addition, detrimental and 
potentially hazardous system characteristics may not be uncovered, and overall 
assessment of vehicle worthiness vis-a-vis its mission will suffer. Appreciation for 
a sound balancing of flight testing with simulation must be promulgated. In 
addition, a methodology appears to be needed to help insure this sound balance. 
The term M&S is taken to include,[1]: 
 Digital models and computer simulations using those models; 
 Mathematical analytical tools such as Computational Fluid Dynamics; 
 Simulated flight testing such as in wind tunnels and engine altitude test 
chambers; 
 Hardware-in-the-loop simulations; 
 Pilot-in-the-loop simulations, with and without hardware-in-the-loop; 
 In-flight simulation;
Introduction  
 
4 
 
 
 Other large-scale ground tests. 
Each of these initially employ simplified system representations that become 
more complex as the systems engineering process defines the system during the 
course of development and as test data becomes available to improve model and 
simulation fidelity and accuracy. Present initiatives are expanding the application 
of verification and validation of M&S resources to ensure that they function as 
intended and suitably represent real-world behaviour. Flight testing itself can be 
considered a simulation if the test article is an experimental system or early 
prototype, if some internal or external system functions are contrived, and if test 
conditions do not truly match actual in-service scenarios (such as simulated 
combat). OT&E flight test relies heavily on constructive simulation and PITL 
tactical simulations. All this has become more popular as simulation capabilities 
have increased and flight test budgets and schedules have decreased.  
However, the flight environment, with systems interacting and with a pilot 
(perhaps) in control, is not a simulation. Flight test remains the most dynamic and 
credible medium for collecting actual system performance data. 
Test management holistic concept is much more, taking into account also the 
relevant phase of actual test preparation, test matrix identification (totality of test 
points to be performed), coordination, FTTs set-up, generation of new validation 
techniques and reporting, of course. 
The purpose of this thesis work is to show how an accurate test management 
based on alternative geometry acquisition processes, test matrix generation 
algorithms, M&S, new FTTs and validation procedures can be used effectively and 
efficiently to support flight testing. In particular, in order to reduce the scope of 
the subject activity the focus is kept on a specific branch of the test field known as 
Store Integration and Safe Separation; the approach could be expanded to other 
branches of flight test, but customization would be required. 
The question becomes how much flight testing can really be replaced by 
simulation before jeopardizing the safety of flight and increasing the cost of 
simulation prohibitively to make it worthwhile. Simulation is not a panacea for all 
test problems, but a valuable tool that must be used cautiously and wisely in the 
course of a test program, the key word is “balance” and its maidservants are 
“optimization, synchronization and coordination”.
5 
CHAPTER 1 
AIRCRAFT STORE INTEGRATION AND 
CERTIFICATION PROCESS 
Approaching the optimization process for the whole flight test area of expertise 
could result pretty cumbersome and arduous, therefore a choice is necessary. This 
study is focused on the store integration, evaluating experimental relevant steps 
ride along the certification process starting from the task mandate up to the 
production of outcome necessary to the reporting phase, Figure 1. 
 
 
Figure 1: FT contribution to Store Integration Certification Process. 
 
New FTTs 
Identification 
 
Data Gathering 
and Model 
Validation 
 
Numerical 
Prediction - FSI 
Geometry 
Acquisition 
(Reverse 
Engineering/CAD 
generation) 
 
Test Matrix 
Identification and 
Test Management 
 
Store Integration 
Certification 
Process
Aircraft store integration and certification process  
 
6 
 
 
Historically, the management of the aforementioned process was based on the 
experience of few very skilful people, leading the activity, depriving the process 
itself from being standardized and depersonalized. Moreover, some steps of the 
process were not evaluated for many reasons: time deficiency, lack of necessary 
know-how, high costs, risk mitigation under-evaluation. 
The purpose of the new “predict-test-evaluate” approach is to force the 
decision makers to take in due account all relevant steps concerning the 
experimental evaluation of the store integration certification process and to 
standardize, to the limit of depersonalization, the execution phase. The “predict-
test-evaluate” represents a circular approach, Figure 2, producing the effect of a 
self-test process, leading the test team to the challenge-and-response test. 
 
 
 
Figure 2: Predict-Test-Validate model. 
 
Optimization of the entire process, cost reduction, time efficiency and risk 
mitigation effectiveness pass through the standardization and optimization of the 
single steps listed in Figure 1. 
1.1 Test matrix identification and test management 
Sizing and identifying the test matrix is the most time consuming and 
uncertain activity faced by every test team when approaching a new flight 
test task. The unkowns are countless and the proposed methods promising 
a solution in the academic world are much more. Different project 
managers in the same organization could decide to pursue different
Aircraft store integration and certification process  
 
7 
 
 
solution tracks, all valid, but requiring different means, way and outcomes 
for the same problem. Trying to avoiding this uncertainty related to the 
store integration process test points identification, in this study are 
examined the two most recognized method proposed by the academic and 
industrial world and new ways to think and perform are proposed in order 
to meet the requirement of a tailorable, efficient and effective tool for 
planning test activity. Theory of games and Theory of fields are the pillars 
of the new proposed algorithm.  
 
1.2 Geometry acquisition techniques 
Nowadays, the global market is extremely competitive, therefore 
product enterprises are constantly seeking new ways to shorten lead times 
for new product developments matching customers’ expectations. Product 
enterprise has invested in CADCAM, rapid prototyping, and a range of 
new technologies that provide business benefits. 
Engineering is the process of designing, manufacturing, assembling, and 
maintaining products and systems. Two major types of engineering could 
be identified: forward engineering and reverse engineering, Figure 3.  
 
 
 Figure 3: Re-engineering/Model.
Aircraft store integration and certification process  
 
8 
 
 
Forward engineering is the traditional process of moving from high-
level abstractions and logical designs to the physical implementation of a 
system. 
The process of duplicating an existing part, subassembly, or product, 
without referencing to drawings, documentation, or a computer model is 
known as reverse engineering. RE is also defined as the process of obtaining 
a geometric CAD model from 3-D points acquired by scanning/digitizing 
existing parts/products. 
In aeronautics environment it is very important to determine the aim of 
the CAD generation activity, in fact the required quality of a product, the 
CAD drawing, is dramatically relevant to determine the success of a 
computational fluid dynamic analysis or not. Where a precise fine-tuning, 
asymptotically a zero-error correction, is required, as for the case of a 
propeller or wing profile, the RE could not be the best fitting method, of 
course every consideration is related also to the available tools. In fact, the 
operator judgement and intervention required in order to reproduce the 
lifting surface starting from the laser scanning acquired cloud of points 
brings in itself some relevant error; filling the uncertainty means to 
misrepresent the truth, Figure 4. 
However, a reconstruction of the profile via forward engineering, for 
example interpolating a certain number of the cross section area airfoils, 
could place the analysis on realm of realism instead of on the plane of pure 
simulation academic training analysis. 
 
 
 
Figure 4: Propeller discrete laser scanned cloud of points.