Nikhil Padhye

Empirical comparison of MOPSO methods - Guide selection and diversity preservation -

In this paper, we review several proposals for guide selection in Multi-Objective Particle Swarm Optimization (MOPSO) and compare them with each other in terms of convergence, diversity and computational times. The new proposals made for guide selection, both personal best (‘pbest’) and global best (‘gbest’), are found to be extremely effective and perform well compared to the already existing methods. The combination of selection methods for choosing ‘gbest’ and ‘pbest’ is also studied and it turns out that there exist certain combinations which yield an overall superior performance outperforming the others on the tested benchmark problems. Furthermore, two new proposals namely velocity trigger (as a substitute for “turbulence operator”) and a new scheme of boundary handling is made.

Improving differential evolution through a unified approach

Only a few attempts in past have been made in adopting a unified outlook towards different paradigms in evolutionary computation (EC). The underlying motivation of these studies was aimed at gaining better understanding of evolutionary methods, both at the level of theory as well as application, in order to design efficient evolutionary algorithms for solving wide-range of complex problems. However, the past descriptions have either been too general or sometimes abstract in issuing a clear direction for improving an evolutionary paradigm for a task-specific. This paper recollects the ‘Unified Theory of Evolutionary Computation’ from past and investigates four steps—Initialization, Selection, Generation and Replacement, which are sufficient to describe traditional forms of Evolutionary Optimization Systems such as Genetic Algorithms, Evolutionary Strategies, Evolutionary Programming, Particle Swarm Optimization and differential evolution (DE). Then, a relatively new evolutionary paradigm, DE, is chosen and studied for its performance on a set of unimodal problems. Discovering DEs inability as an efficient solver, DE is reviewed under ‘Unified Framework’ and functional requirements of each step are evaluated. Targeted towards enhancing the DE’s performance, several modifications are proposed through borrowing of operations from a benchmark solver G3-PCX. Success of this exercise is demonstrated in a step-by-step fashion via simulation results. The Unified Approach is highly helpful in understanding the role and re-modeling of DE steps in order to efficiently solve unimodal problems. In an avalanching-age of new methods in EC, this study outlines a direction for advancing EC methods by undertaking a collective outlook and an approach of concept-sharing.

Multi-objective optimisation and multi-criteria decision making in SLS using evolutionary approaches

Purpose – The goal of this study is to carry out multi‐objective optimization by considering minimization of surface roughness (Ra) and build time (T) in selective laser sintering (SLS) process, which are functions of “build orientation”. Evolutionary algorithms are applied for this purpose. The performance comparison of the optimizers is done based on statistical measures. In order to find truly optimal solutions, local search is proposed. An important task of decision making, i.e. the selection of one solution in the presence of multiple trade‐off solutions, is also addressed. Analysis of optimal solutions is done to gain insight into the problem behavior.Design/methodology/approach – The minimization of Ra and T is done using two popular optimizers – multi‐objective genetic algorithm (non‐dominated sorting genetic algorithm (NSGA‐II)) and multi‐objective particle swarm optimizers (MOPSO). Standard measures from evolutionary computation – “hypervolume measure” and “attainment surface approximator” have ...

Boundary Handling Approaches in Particle Swarm Optimization

In recent years, Particle Swarm Optimization (PSO) methods have gained popularity in solving single objective and other optimization tasks. In particular, solving constrained optimization problems using swarm methods has been attempted in past but arguably stays as one of the challenging issues. A commonly encountered situation is one in which constraints manifest themselves in form of variable bounds. In such scenarios the issue of constraint-handling is somewhat simplified.This paper attempts to review popular bound handling methods, in context to PSO, and proposes new methods which are found to be robust and consistent in terms of performance over several simulation scenarios. The effectiveness of bound handling methods is shown PSO; however, the methods are general and can be combined with any other optimization procedure.

Enhancing performance of particle swarm optimization through an algorithmic link with genetic algorithms

Evolutionary Algorithms (EAs) are emerging as competitive and reliable techniques for several optimization tasks. Juxtapositioning their higher-level and implicit correspondence; it is provocative to query if one optimization algorithm can benefit from another by studying underlying similarities and dissimilarities. This paper establishes a clear and fundamental algorithmic linking between particle swarm optimization (PSO) algorithm and genetic algorithms (GAs). Specifically, we select the task of solving unimodal optimization problems, and demonstrate that key algorithmic features of an effective Generalized Generation Gap based Genetic Algorithm can be introduced into the PSO by leveraging this algorithmic linking while significantly enhance the PSO’s performance. However, the goal of this paper is not to solve unimodal problems, neither is to demonstrate that the modified PSO algorithm resembles a GA, but to highlight the concept of algorithmic linking in an attempt towards designing efficient optimization algorithms. We intend to emphasize that the evolutionary and other optimization researchers should direct more efforts in establishing equivalence between different genetic, evolutionary and other nature-inspired or non-traditional algorithms. In addition to achieving performance gains, such an exercise shall deepen the understanding and scope of various operators from different paradigms in Evolutionary Computation (EC) and other optimization methods.

Multi-objective Optimisation and Multi-criteria Decision Making for FDM Using Evolutionary Approaches

In this chapter, we methodologically describe a multi-objective problem solving approach, concurrently minimising two conflicting goals—average surface roughness—Ra and build time—T, for object manufacturing in Fused Deposition Method (FDM) process by usage of evolutionary algorithms. Popularly used multi-objective genetic algorithm (NSGA-II) and recently proposed multi-objective particle swarm optimisation (MOPSO) algorithms are employed for the optimisation purposes. Statistically significant performance measures are employed to compare the two algorithms and approximate the Pareto-optimal fronts. To refine the solutions obtained by the evolutionary optimisers, an effective mutation-driven hill-climbing local search is proposed. Three new proposals and several suggestions pertaining to the issue of decision making in the presence of multiple optimal solutions are made. The overall procedure is integrated into an engine called MORPE—multi-objective rapid prototyping engine. Sample objects are considered and several case studies are performed to demonstrate the working of MORPE. Finally, a careful investigation of the optimal build orientations for several considered objects is done or selected basis and a trend is discovered, which can be considered highly useful for various practical rapid prototyping (RP) applications.

Interplanetary trajectory optimization with swing-bys using evolutionary multi-objective optimization

Interplanetary trajectory optimization studies mostly considered a single objective of minimizing the travel time between two planets or the launch velocity of spacecraft at the departure planet. In this paper, we have considered a simultaneous minimization study of both launch velocity and time of travel between two specified planets with and without the use of gravitational advantage (swingby) of some intermediate planets. Using careful consideration of a Newton-Raphson based root finding procedure of developing a trajectory based on a given set of decision variables (departure date, swing-by planets, altitude of spacecraft at the first swing-by planet, etc.), a number of derived parameters such as time of flight between arrival and destination planet, date of arrival, and launch velocity are computed. A popularly used evolutionary multi-objective optimization algorithm (NSGA-II) is then employed to find a set of trade-off solutions. The accuracy of the developed software (we called GOSpel) is first demonstrated by matching the trajectories with known missions and then the efficiency of the software is shown by solving a number complex, real-world like missions.

Evolutionary approaches for real world applications in 21st century

Evolutionary computation is notably one of the fast growing fields of research and application and is becoming pervasive in several streams of science and engineering. This paper reviews its origin, investigates the reason for its growth and widespread applicability. The reasons for existence and advantages of other paradigms are also discussed. The goal of this paper is to underline root-causes necessary for successful deployment of evolutionary methods in diverse applications and challenges that need to be addressed in any such endeavour.

Enhancing the performance of the T-peel test for thin and flexible adhered laminates

Symmetrically bonded thin and flexible T-peel specimens, when tested on vertical travel machines, can be subject to significant gravitational loading, with the associated asymmetry and mixed-mode failure during peeling. This can cause erroneously high experimental peel forces to be recorded which leads to uncertainty in estimating interfacial fracture toughness and failure mode. To overcome these issues, a mechanical test fixture has been designed, for use with vertical test machines, that supports the unpeeled portion of the test specimen and suppresses parasitic loads due to gravity from affecting the peel test. The mechanism, driven by the test machine cross-head, moves at one-half of the velocity of the cross-head such that the unpeeled portion always lies in the plane of the instantaneous center of motion. Several specimens such as bonded polymeric films, laminates, and commercial tapes were tested with and without the fixture, and the importance of the proposed T-peel procedure has been demonstrated.

A New Phenomenon: Sub-T g , Solid-State, Plasticity-Induced Bonding in Polymers

Polymer self-adhesion due to the interdiffusion of macromolecules has been an active area of research for several decades. Here, we report a new phenomenon of sub-Tg, solid-state, plasticity-induced bonding; where amorphous polymeric films were bonded together in a period of time on the order of a second in the solid-state at ambient temperatures, up to 60 K below their glass transition temperature (Tg), by subjecting them to active plastic deformation. Despite the glassy regime, the bulk plastic deformation triggered the requisite molecular mobility of the polymer chains, causing interpenetration across the interfaces held in contact. Quantitative levels of adhesion and the morphologies of the fractured interfaces validated the sub-Tg, plasticity-induced, molecular mobilization causing bonding. No-bonding outcomes (i) during the uniaxial compressive straining of films (a near-hydrostatic setting which strongly limits plastic flow) and (ii) between an ‘elastic’ and a ‘plastic’ film further established the explicit role of plastic deformation in this newly reported sub-Tg solid-state bonding.