General optimisation - Scia Engineer Optimiser

Scia Engineer Optimiser is a cutting edge

software tool for the overall optimisation of
civil engineering structures. It represents a
combination of a widespread structural analysis
software (Scia Engineer) and a sophisticated
optimisation engine (EOT). The two programs
have been integrated together and offer a
versatile and complete optimisation solution for
DOOW\SHVRIFLYLOHQJLQHHULQJVWUXFWXUHV

 (27LVDQRSWLPLVDWLRQVROYHULQZKLFKWKHXVHU
GHQHVWKHJRDOIXQFWLRQIRUWKHRSWLPLVDWLRQ
determins relations between the parameters
and selects the suitable optimisation method.

If suitable or needed, it is possible to specify also

relations between individual parameters (e.g.
the relation between the width and height of a
cross-section).
'HQLWLRQRIWKHJRDOIXQFWLRQDQGVHOHFWLRQ
of the optimisation method
7KHJRDOIXQFWLRQGHQHVZKDWLVWREHRSWLPLVHG
It can be a price, weight, dimensions, position
of a support, location of a load. Furthermore,
it is necessary to select one of the available
optimisation methods. The selection of the
method may affect the time needed for the
solution of the sought-after result.
3. Optimisation cycle
The optimisation solver (EOT) generates the sets
of parameters used for the creation of a particular
variant of the model.
Scia Engineer receives these parameters, runs
the prescribed calculation, code-check and, if
UHTXLUHGDOVR$XWRGHVLJQ
In the next step, EOT gets back the results and
evaluates them to modify the parameters in order

to get closer to the desired solution.

$QGWKLVSURFHVVLVUHSHDWHGXQWLOWKHRSWLPXP
is found.
4. Evaluation of the optimum solution
$VDOUHDG\VWDWHGLWLVSRVVLEOHWKDWPRUHWKDQ
one optimum is found by the optimisation solver
(these optimums represent local extremes of the
goal function - it depends on the mathematical
representation of the goal function if it has one or
more local extremes). These individual optimums
can vary just a little in the value of the goal
IXQFWLRQEXWWKHUHPD\EHDVLJQLFDQWGLIIHUHQFH
in the values of the parameters for individual
optimums, which means that the structure takes
different shapes, forms, etc. In such a situation, it
is the user who must decide which of the variants
will be used in the end.

7KHVROYHUQGVWKHRSWLPDOVROXWLRQDFFRUGLQJWR
WKHXVHUVLQSXWWU\LQJWRQLVKWKHWDVNZLWKLQWKH
minimum possible number of steps.

2SWLPLVDWLRQZRUNRZ
The optimisation process can be clearly seen in
the picture. Once all the required input data are
entered, i.e. the model of the analysed structure
LVGHQHGWKHVHDUFKIRUWKHRSWLPXPVROXWLRQ
runs fully automatically and no interaction from
the user is required. For real-life problems
several optimum solutions can be found. In such
VLWXDWLRQVLWLVXSWRWKHXVHUWRPDNHWKHQDO
decision.
1. Creation of the model and its
parameterization
The model of the analysed structure is created
using standard Scia Engineer tools and functions.
The geometry, boundary conditions, loads, etc.
DUHGHQHG

5HTXLUHGPRGXOHHVD

esa.30

$QDO\VHU

 6FLD(QJLQHHULVDFRPSUHKHQVLYHVRIWZDUH
package for analysis, design and checks of
civil engineering structures. The integration of
Scia Engineer into the process of the overall
optimisation is enabled by its above-standard
IHDWXUHV
 3DUDPHWHUL]DWLRQRIWKHPRGHOGLUHFW
(numerical) values of individual properties of
entities in Scia Engineer can be replaced by
parameters. The parameters can be viewed
and edited directly in Scia Engineer or via
DQRSHQFRPPXQLFDWLRQLQWHUIDFH
 $XWRGHVLJQDXWRPDWLFVHDUFKIRUDQRSWLPDO
design for a particular structural entity - e.g.
DQRSWLPDOVL]HRIDVWHHOFURVVVHFWLRQ
optimal reinforcement in a concrete crosssection on the base of calculated internal
IRUFHV
 ;0/LQWHUIDFHIRUFRPPXQLFDWLRQZLWKRWKHU
DSSOLFDWLRQV

3DUDPHWHUVDUHDVVLJQHGWRWKHSURSHUWLHVWKDW
can vary during the optimisation. The parameter
indicates that a particular property becomes
YDULDEOHDQGWKHXVHUGHQHVLWVLQLWLDOYDOXHDQG
if required, also the limits.

General optimisation - Scia Engineer Optimiser

5. Final check
7KHQDOVHWRISDUDPHWHUVWKDWJLYHVWKH
RSWLPXPVROXWLRQLVWKHQXVHGWRFUHDWHWKHQDO
YDULDQWRIWKHVWUXFWXUH$OOUHTXLUHGW\SHVRI
calculation and checks that were not performed
during the optimisation can be executed now.

EOT Optimisation methods

$QDO\VHU

Several different methods have been

LPSOHPHQWHGLQWKH(27RSWLPLVDWLRQVROYHU
 Gradient method: Sequential quadratic
programming (SQP)
 *UDGLHQWPHWKRGVDUHNQRZQDVYHU\HIFLHQW
methods in case of continuous optimisation
problems. They are felicitous for example
when searching for the optimal positions of
nodes, supports, or geometry of cross-sections
etc. They cannot be used for optimisation
tasks working with discrete values, as a
VHOHFWLRQRIDUROOHGSUROHRUGHWHUPLQDWLRQRI
DQXPEHURIUHLQIRUFHPHQWEDUVHWF*UDGLHQW
methods could be very fast, on the other hand
convergence problems may occur in projects
with a large number of parameters and in tasks
ZLWKFRPSOLFDWHGVKDSHRIJUDGLHQWV
 6WRFKDVWLFPHWKRGV0RGLHGVLPXODWHG
annealing (MSA), Differential evolution (DE)
Simply said, stochastic methods search
for the result by means of trial-and-error
and evaluation of these trials. This group
contains methods that are also called genetic
algorithms. Stochastic methods are the most
stable, on the other hand, the necessary
calculation time is much higher with respect to
WKHJUDGLHQWPHWKRG
 Heuristic methods: Nelder-Mead (N-M)
Heuristic methods share the properties of both
gradient and stochastic methods. Their speed
is somewhere between stochastic and gradient
methods as well as the stability.

esa.30