One of the key stages in
aircraft design process is the conceptual design phase. Conceptual design is
considered to be the most difficult phase of engineering design, with success
dependent to a great extent on the expertise of a designer. Automation of some
aspects of this phase would be of immense practical benefit [[1]]. During this phase, the
designer must devise an initial design that incorporates ‘‘working principles’’
or physical solutions to all the ‘‘essential’’ features of the problem, and
which has been evaluated to be acceptable and feasible [[2]]. This is the phase of the
design process ‘‘that makes the greatest demands on the designer, and where
there is the most scope for striking improvements and where the most important
decisions are taken’’ [[3]].
Conceptual design is a
fundamental and indispensable forerunner to further detailed design. Not only
is it well known that a design concept is the overwhelming factor influencing
the product life-cycle cost and level of innovation; but an excellent detailed
design based upon a poor and inappropriate concept can never compensate for the
inadequacy of that concept.
Conceptual design is an early phase in the design
process, which involves the generation of solution concepts to satisfy the
functional requirements of a design problem. There can be more than one
solution to a problem; this means that there is scope for producing improved
designs if one could explore a solution space larger than is possible at
present [[4]].
As shown in
Fig. 1, design uncertainty is the largest for entirely original designs and
decreases during design [[5]].
The project costs are minimal on a conceptual design
stage, but importance of accepted solutions is maximal. Therefore the
conceptual 
design
is the basic phase of design process (Fig.2).
For creation of competitive products in Aeronautics and Aerospace it is typically
necessary to synthesize from 50 up to 150 new engineering solutions (ES). [[6]]. In [[7]] 3 levels of optimization
are examined at creation of new ES. Performance characteristics of projected
systems at the third level of optimization can be improved on the average of
10-15% (Fig. 3). At level 1 and 2 characteristics are improved on the average
of 30-35 %, and sometimes more. The higher the level of optimization, the more
is the effect of optimization. In engineering practice there is usually no way
allowing the choice of an optimum engineering solution based on conditions of
the technical project at once.
Subject
of search in structural synthesis is achievement of some compromise levels for
lines of inconsistent criteria. The sequence of computing operations for
finding of an optimum algorithm for the design calculation is displayed by an
objective function. This function does not correspond to the basic requirements
of theoretical methods of optimization (including genetic algorithms).
The
objective function is discontinuous, exists in operator notation, it is not
based on analytic forms, is not differentiable, not unimodal,
not separable and also not additive. It is impossible to model analytically a
hyper surface of objective functions and to predict their change on an
increment of variables. Instead of step-by-step promotion in space of
attributes area-scanning searches with use of clusters seem to be more
efficient.
For the problems decision of structural synthesis
there are two groups of methods - morphological and heuristic. Heuristic
approaches yield unstable results and are subjective.
Morphological analysis (or General Morphological
Analysis) is a method developed by F. Zwicky for
exploring all the possible solutions to a multi-dimensional, non-quantified
problem complex [[8]]. Zwicky applied this
method to such diverse tasks as the classification of astrophysical objects and
the development of jet and rocket propulsion systems. More recently,
morphological analysis has been extended and applied by a number of researchers
in the U.S.A and Europe in the field of future studies, engineering system
analysis and strategy modeling [Odrin, 1998; Coyle et
al., 1994; Rhyne 1995; Ritchey 1997, 2003, 2006; Stenstroem & Ritchey 1999; Akimov,
2005]. The morphological approach serves as a standard when new systems are
being designed [[9],[10]].
The basic problem of morphological methods is the
so-called “damnation of dimension”, which means, that rapid generation of an
enormous set of variants is relatively simple, but selection of the best
variants is very difficult [[11]].
The disadvantages of morphological methods are as
follows:
·
Labor-intensive
selection from a set of variants and
·
Impossibility
of search and analysis of all variants.
To remove the aforementioned disadvantages a
methodology of the structural synthesis of engineering systems (SSES), based on
the system, morphological, and cluster approach has been developed.
Using the methodology, different engineering systems,
mostly in the area of Aeronautics, Astronautics, and Ecology, have been studied
[[12],[13],[14]] (Fig.4). Fields of generated variants for Reentry
vehicles [[15],[16]]
(Fig.5) is given below (s. synergistic effect in the Table 2).
|
|
|
|
Figure 4:
Areas of the SSES application |
Figure 5:
Solution field of the advanced reentry vehicles (2004) |
More and more works on electrical aircraft [[17],[18],[19]]
have been lately appearing while an integrated theory to study their design
process does not exist. To study issues concerning the design process of new
systems – electrical aircraft – it is expedient to use the advanced methodology
of SSES.
This project is focused
on performing a multidisciplinary research activity on the structural synthesis
and analysis of the next generation airplanes - electric aircraft.
The goals of the project
are as follows:
1.
Creation
of advanced structural synthesis methodology for conceptual design of
electrical aircraft.
2.
Software
development and implementation of visual interfaces to represent a strict
synthesis, evaluation and decision making process.
3.
Modeling, numerical simulation, and analysis for the synthesized ATS.
4.
Development of an electrical aircraft design
process theory
The multidisciplinary
aspect consists in applying methods of system and cluster analysis to ATS
modeling.