Home Engineering Small Unmanned Fixed-Wing Aircraft Design. A Practical Approach
In order to demonstrate that the overall system meets the requirements, it needs to be tested. Waiting until the overall system has been built is a very high risk way of carrying out compliance testing, and hence systems engineers tend to follow the “V” model of systems engineer- ing illustrated in Figure 8.7.
The “V” model ensures that engineers (a) know what the requirements flow down for their subsystem or part is and (b) think about how they can demonstrate compliance (i.e., test their contribution). This should result in a lower risk overall system. When the overall systems test takes place, it should involve a set of already tested subsystems.
Figure 8.6 Example military system requirements flowdown . Defence Acquisition University.
Figure 8.7 Systems engineering “V” model.
Cost and Weight Management
For an aerospace team-design activity, it is extremely important to manage holistic parameters such as weight and cost very closely. It is often useful to allocate management and monitoring of weight and cost to a single individual.
Weight and cost are aggregating characteristics and can easily exceed aircraft-level targets unless a very disciplined approach is taken. They have to be managed continuously throughout the design process by using different techniques at each design stage depending on the level of detail available. During the concept design stage, for example, parametric or empirical methods are necessary. This involves the use of actual historical data for products that are sufficiently similar to the concept being worked on. During the preliminary design phase, the product definition becomes more refined and allows some of the parts and subsystems to be based on “bottom-up” calculations. This might involve the use of supplier cost and weight figures for “bought-out” parts and subsystems as well as estimates of structural parts based on emerging geometry and material selection.
Finally, in the detail design phase a full product definition starts to emerge and accurate “bottom-up” estimates can be developed. Even at this stage it is hard to make calculations which are 100% accurate, particularly for weight estimates. For example, Figure 8.8 shows a plot of the predicted weight of the SPOTTER UAV calculated during the detail design phase against the actual measured weight. This shows that most of the time engineers underpredict weight mainly because of missing detail. Hence it is vital to use a mix of “bottom-up” estimates which are validated against relevant “top-down” historical actual data.
Figure 8.9 shows the same data plotted in a pie chart form. For anyone developing a similar class of UAV to SPOTTER, the actual data is very useful to help set weight targets during the concept design phase of a new design.
Figure 8.10 shows an extract from the SPOTTER weight breakdown table. This illustrates a useful approach whereby “traffic-light” color coding is used to indicate the fidelity of the estimate. Green indicates that the figure used is very accurate/reliable (ideally the actual mass of a part/subsystem validated using calibrated weighing scales). Amber is an estimate that is reasonably accurate (ideally within 10%). Finally, red is used to signify rough estimates which
Figure 8.8 Weight prediction of SPOTTER UAV.
Figure 8.9 Pie chart plots of SPOTTER weight.
Figure 8.10 Example weight and cost breakdown.
are a source of significant uncertainty (and therefore risk). An important ambition during the preliminary design stage should be to eliminate all significant mass red entries. Similarly, during the detail design phase all the significant amber entries should be eliminated.
|< Prev||CONTENTS||Next >|