Seyedeh Zahra Mirjalili

Evolutionary population dynamics and grey wolf optimizer

Evolutionary population dynamics (EPD) deal with the removal of poor individuals in nature. It has been proven that this operator is able to improve the median fitness of the whole population, a very effective and cheap method for improving the performance of meta-heuristics. This paper proposes the use of EPD in the grey wolf optimizer (GWO). In fact, EPD removes the poor search agents of GWO and repositions them around alpha, beta, or delta wolves to enhance exploitation. The GWO is also required to randomly reinitialize its worst search agents around the search space by EPD to promote exploration. The proposed GWO–EPD algorithm is benchmarked on six unimodal and seven multi-modal test functions. The results are compared to the original GWO algorithm for verification. It is demonstrated that the proposed operator is able to significantly improve the performance of the GWO algorithm in terms of exploration, local optima avoidance, exploitation, local search, and convergence rate.

Salp Swarm Algorithm: A bio-inspired optimizer for engineering design problems

A novel optimization algorithm called Salp Swarm Optimizer (SSA) is proposed.Multi-objective Salp Swarm Algorithm (MSSA) is proposed to solve multi-objective problems.Both algorithms are tested on several mathematical optimization functions.Two challenging engineering design problems are solved: airfoil design and marine propeller design.The qualitative and quantitative results prove the efficiency of SSA and MSSA. This work proposes two novel optimization algorithms called Salp Swarm Algorithm (SSA) and Multi-objective Salp Swarm Algorithm (MSSA) for solving optimization problems with single and multiple objectives. The main inspiration of SSA and MSSA is the swarming behaviour of salps when navigating and foraging in oceans. These two algorithms are tested on several mathematical optimization functions to observe and confirm their effective behaviours in finding the optimal solutions for optimization problems. The results on the mathematical functions show that the SSA algorithm is able to improve the initial random solutions effectively and converge towards the optimum. The results of MSSA show that this algorithm can approximate Pareto optimal solutions with high convergence and coverage. The paper also considers solving several challenging and computationally expensive engineering design problems (e.g. airfoil design and marine propeller design) using SSA and MSSA. The results of the real case studies demonstrate the merits of the algorithms proposed in solving real-world problems with difficult and unknown search spaces.

Designing evolutionary feedforward neural networks using social spider optimization algorithm

Training feedforward neural networks (FNNs) is considered as a challenging task due to the nonlinear nature of this problem and the presence of large number of local solutions. The literature shows that heuristic optimization algorithms are able to tackle these problems much better than the mathematical and deterministic methods. In this paper, we propose a new trainer using the recently proposed heuristic algorithm called social spider optimization (SSO) algorithm. The trained FNN by SSO (FNNSSO) is benchmarked on five standard classification data sets: XOR, balloon, Iris, breast cancer, and heart. The results are verified by the comparison with five other well-known heuristics. The results prove that the proposed FNNSSO is able to provide very promising results compared with other algorithms.

Full Optimizer for Designing Photonic Crystal Waveguides: IMoMIR Framework

This letter proposes a full optimizer framework (IMoMIR framework) for designing photonic crystal waveguide (PCW). In this framework, three main issues of PCWs operation [group index, bandwidth, and group velocity dispersion (GVD)] are considered and optimized simultaneously. In other words, multi-objective optimization is performed to maximize the group index and bandwidth, while minimizing the |GVD|. Since these three objectives are in conflict, the proposed framework finds the specific designs with considering all objectives, the so-called Pareto-optimal solutions. In addition, the band-mixing phenomenon is avoided during the optimization process. The merits of the proposed framework is demonstrated by designing an oval-shaped-hole PCW. The results show that the IMoMIR framework is able to determine a diverse range of PCWs that completely outperform the current PCW structure in the literature.

Grasshopper optimization algorithm for multi-objective optimization problems

This work proposes a new multi-objective algorithm inspired from the navigation of grass hopper swarms in nature. A mathematical model is first employed to model the interaction of individuals in the swam including attraction force, repulsion force, and comfort zone. A mechanism is then proposed to use the model in approximating the global optimum in a single-objective search space. Afterwards, an archive and target selection technique are integrated to the algorithm to estimate the Pareto optimal front for multi-objective problems. To benchmark the performance of the algorithm proposed, a set of diverse standard multi-objective test problems is utilized. The results are compared with the most well-regarded and recent algorithms in the literature of evolutionary multi-objective optimization using three performance indicators quantitatively and graphs qualitatively. The results show that the proposed algorithm is able to provide very competitive results in terms of accuracy of obtained Pareto optimal solutions and their distribution.

Optimization of problems with multiple objectives using the multi-verse optimization algorithm

This work proposes the multi-objective version of the recently proposed Multi-Verse Optimizer (MVO) called Multi-Objective Multi-Verse Optimizer (MOMVO). The same concepts of MVO are used for converging towards the best solutions in a multi-objective search space. For maintaining and improving the coverage of Pareto optimal solutions obtained, however, an archive with an updating mechanism is employed. To test the performance of MOMVO, 80 case studies are employed including 49 unconstrained multi-objective test functions, 10 constrained multi-objective test functions, and 21 engineering design multi-objective problems. The results are compared quantitatively and qualitatively with other algorithms using a variety of performance indicators, which show the merits of this new MOMVO algorithm in solving a wide range of problems with different characteristics.

Asymmetric Oval-Shaped-Hole Photonic Crystal Waveguide Design by Artificial Intelligence Optimizers

This paper proposes a new kind of photonic crystal waveguide (PCW) called asymmetric oval-shaped-hole PCW (AOPCW). In this PCW, 20 structural parameters are considered, which provides a very flexible PCW to design. The large number of variables makes the design process of this new PCW almost impossible by the current manual try and error methods. Therefore, the AOPCW is optimized automatically by the artificial intelligence optimization technique in three phases. First, the AOPCW is optimized to maximize normalized delay-bandwidth product. Second, it is optimized with respect to two objectives: averaged of group index (n̅g) and normalized bandwidth (Δω/ω0). Third, group velocity dispersion is also considered in addition to the two objectives in the second phase, so AOPCW is optimized with respect to three objectives. In all of the three phases, a band mixing avoidance mechanism is also considered and handled. The comparative study of the optimized designs proves that the proposed AOPCW is able to substantially outperform the current PCW structures in the literature. This paper also considers and discusses time-domain simulation issues of the PCW-based optical buffer.

Multi-objective versus single-objective optimization frameworks for designing photonic crystal filters

This paper proposes a novel framework for multi-objective optimization of photonic crystal (PhC) filters and compares it with a single-objective optimization approach. In this framework, an optimizer called the Multi-Objective Gray Wolf Optimizer has been utilized to automatically find the optimal designs. The proposed method is able to design any kind of PhC filter. As a case study, a new structure of super defect PhC filter for application in the wavelength-division multiplexer (WDM) is designed using the framework. The results show that the proposed framework is comprehensive and able to find a significantly wide range of optimal designs for general and specific application, such as WDM with respect to each defined WDM standard.