Browsing by Author "Riley, Mike J. W."

Now showing 1 - 3 of 3
  • Thumbnail Image
    ItemOpen Access
    Evaluating cascade correlation neural networks for surrogate modelling needs and enhancing the Nimrod/O toolkit for multi-objective optimisation
    (Cranfield University, 2011-03) Riley, Mike J. W.; Jenkins, Karl W.
    Engineering design often requires the optimisation of multiple objectives, and becomes significantly more difficult and time consuming when the response surfaces are multimodal, rather than unimodal. A surrogate model, also known as a metamodel, can be used to replace expensive computer simulations, accelerating single and multi-objective optimisation and the exploration of new design concepts. The main research focus of this work is to investigate the use of a neural network surrogate model to improve optimisation of multimodal surfaces. Several significant contributions derive from evaluating the Cascade Correlation neural network as the basis of a surrogate model. The contributions to the neural network community ultimately outnumber those to the optimisation community. The effects of training this surrogate on multimodal test functions are explored. The Cascade Correlation neural network is shown to map poorly such response surfaces. A hypothesis for this weakness is formulated and tested. A new subdivision technique is created that addresses this problem; however, this new technique requires excessively large datasets upon which to train. The primary conclusion of this work is that Cascade Correlation neural networks form an unreliable basis for a surrogate model, despite successes reported in the literature. A further contribution of this work is the enhancement of an open source optimisation toolkit, achieved by the first integration of a truly multi-objective optimisation algorithm.
  • Thumbnail Image
    ItemOpen Access
    Improving the performance of cascade correlation neural networks on multimodal functions
    (Newswood Limited, 2010-12-31) Riley, Mike J. W.; Thompson, Christopher P.; Jenkins, Karl W.
    Intrinsic qualities of the cascade correlation algorithm make it a popular choice for many researchers wishing to utilize neural networks. Problems arise when the outputs required are highly multimodal over the input domain. The mean squared error of the approximation increases significantly as the number of modes increases. By applying ensembling and early stopping, we show that this error can be reduced by a factor of three. We also present a new technique based on subdivision that we call patchworking. When used in combination with early stopping and ensembling the mean improvement in error is over 10 in some cases.
  • Thumbnail Image
    ItemOpen Access
    Multi-objective engineering shape optimization using differential evolution interfaced to the Nimrod/O tool
    (IOP Publishing, 2010-07-05) Riley, Mike J. W.; Peachey, Tom; Abramson, D.; Jenkins, Karl W.
    This paper presents an enhancement of the Nimrod/O optimization tool by interfacing DEMO, an external multiobjective optimization algorithm. DEMO is a variant of differential evolution – an algorithm that has attained much popularity in the research community, and this work represents the first time that true multiobjective optimizations have been performed with Nimrod/O. A modification to the DEMO code enables multiple objectives to be evaluated concurrently. With Nimrod/O’s support for parallelism, this can reduce the wall-clock time significantly for compute intensive objective function evaluations. We describe the usage and implementation of the interface and present two optimizations. The first is twoobjective mathematical function in which the Pareto front is successfully found after only 30 generations. The second test case is the three-objective shape optimization of a rib-reinforced wall bracket using the Finite Element software, Code_Aster. The interfacing of the already successful packages of Nimrod/O and DEMO yields a solution that we believe can benefit a wide community, both industrial and academic.