Over the past decade, several researchers have presented various optimisation algorithms for use in truss design. The no free lunch theorem implies that no optimisation algorithm fits all problems; therefore, the interest is not only in the accuracy and convergence rate of the algorithm but also the tuning effort and population size required for achieving the optimal result. The latter is particularly crucial for computationally intensive or high-dimensional problems. Contrast-based Fruit-fly Optimisation Algorithm (c-FOA) proposed by Kanarachos et al. in 2017 is based on the efficiency of fruit flies in food foraging by olfaction and visual contrast. The proposed Spontaneous Fruit Fly Optimisation (s-FOA) enhances c-FOA and addresses the difficulty in solving nonlinear optimisation algorithms by presenting standard parameters and lean population size for use on all optimisation problems. Six benchmark problems were studied to assess the performance of s-FOA. A comparison of the results obtained from documented literature and other investigated techniques demonstrates the competence and robustness of the algorithm in truss optimisation.
Deep Dive into Spontaneous Fruit Fly Optimisation for truss weight minimisation: Performance evaluation based on the no free lunch theorem.
Over the past decade, several researchers have presented various optimisation algorithms for use in truss design. The no free lunch theorem implies that no optimisation algorithm fits all problems; therefore, the interest is not only in the accuracy and convergence rate of the algorithm but also the tuning effort and population size required for achieving the optimal result. The latter is particularly crucial for computationally intensive or high-dimensional problems. Contrast-based Fruit-fly Optimisation Algorithm (c-FOA) proposed by Kanarachos et al. in 2017 is based on the efficiency of fruit flies in food foraging by olfaction and visual contrast. The proposed Spontaneous Fruit Fly Optimisation (s-FOA) enhances c-FOA and addresses the difficulty in solving nonlinear optimisation algorithms by presenting standard parameters and lean population size for use on all optimisation problems. Six benchmark problems were studied to assess the performance of s-FOA. A comparison of the result
Trusses have found significant applications in modern engineering. Such applications range from use in transmission towers, offshore wind turbine supports, offshore oil and gas platforms, to microstructural applications such as the lattice structures of additive manufacturing [1,2]. Truss optimisation aims to improve the performance of trusses while minimising the material resource [3]. The objective of the optimisation can be interpreted as a weight minimisation one, bounded by well-defined constraints. The constraints are the allowable stresses and displacements, as subject to high stress the truss members could fail through buckling or tension. There are many forms of optimisation, each with their unique design variables: this study focuses only on size optimisation. The design variable that is the most commonly investigated is the cross-sectional area of the truss member [4].
Optimisation algorithms are used in searching for the optimum solution to a problem. The application of optimisation algorithms to structures has proliferated in the last decade [5]. Many researchers have published on the applications of improved algorithms to truss weight minimisation problems. Kaveh and Mahdavi proposed a Multi-Objective Colliding Bodies Optimisation (MOCBO) algorithm for the optimisation of trusses bounded by an allowable stress limit [6]. A genetic programming methodology was used by Assimi et al. in the optimisation of the size and topology of trusses [7]. Another approach was taken by Cheng et al., proposing a Hybrid Harmony Search (HHS) algorithm for the design of truss structures with stress limits [8]. Tejani et al. made use of the Improved Passing Vehicle Search (IPVS), Improved Heat Transfer Search (IHTS), Improved Water Wave Optimization (IWWO) and Improved Heat Transfer Search (IHTS) to optimise the topology of truss structures with displacement, stress and kinematic stability constraints [9]. An adaptive Symbiotic Organism Search (SOS) was utilised by Tejani et al. in truss structural optimisation with frequency constraints [10]. A development of PSO was presented by Kaveh and Zolghadr: Democratic PSO (Particle Swarm Optimisation) algorithm for the optimisation of truss layout and size with frequency constraints [11]. Multi-Class Teaching-Learning-Based Optimisation algorithm (MC-TLBO) was utilised by Farshchin et al. for truss design with frequency constraints [12]. Rajan used a Genetic Algorithm (GA) to optimise the shape, size and topology of truss structures [13].
Through all these research studies, the efficiency of optimisation algorithms in solving structural design problems has been established. However, according to the “no free lunch theorem”, there exists no single algorithm to solve all optimisation problems. Hence the need to research lean algorithms [10].
In 2011, Pan proposed the FOA algorithm, a population-based technique that mimics the foraging activities of fruit flies [14]. Fruit-flies, compared to other species, possess a better sense of smell and vision which they use to find food efficiently. The algorithm has a framework which is simple, easy to understand, and is easily implementable in tackling various optimisation problems [15]. However, it is characterised by premature convergence (reduced accuracy) and is easily trapped in local optima.
FOA has been applied successfully to a variety of problems. In 2011, Pan applied the FOA algorithm to optimise the General Regression Neural Network and Multiple Regression utilised in modelling the financial distress problem of Taiwan’s enterprise [14]. Lu et al. in 2015 proposed an adaptive fruit fly optimisation algorithm based on velocity variable (VFOA). The algorithm utilised the particle velocity concept from PSO on FOA to improve its convergence speed and accuracy. The improved algorithm was used to solve 13 mathematical benchmark problems [16]. As another improvement, Kanarachos et al. modified the FOA algorithm by including a visual contrast phase. The modification was based on biological discoveries on the complexity of the fruit foraging activities of fruit flies, thus improving its exploration capabilities. The modified algorithm was applied for the first time to solve truss design problems with stress, displacement or frequency constraints [17]. The algorithm was also used to improve the shock performance of vehicle suspension systems to mitigate damages caused by potholes in the UK [18]. Since then it was applied successfully in a range of problems. Wu et al. solved 33 mathematical benchmark functions by modifying the FOA to improve its exploration capabilities. A normal cloud generator was introduced to generate new positions of the swarm based on parameters such as possible food position, search range and search stability. It was inspired by the fact that fruit flies are characterised by fuzziness and randomness as they fly towards the food source. A cloud model is a tool used to synthesise the randomness and fuzziness
…(Full text truncated)…
This content is AI-processed based on ArXiv data.
