Convergence Analysis of Particle Swarm Optimization via Illustration Styles

Particle swarm optimization (PSO) is the most important and popular algorithm to solving the engineering optimization problem due to its simple updating formulas and excellent searching capacity. This algorithm is one of evolutionary computations and is also a population-based algorithm. Traditionally, to demonstrate the convergence analysis of the PSO algorithm or its related variations, simulation results in a numerical presentation are often given. This way may be unclear or unsuitable for some particular cases. Hence, this paper will adopt the illustration styles instead of numeric simulation results to more clearly clarify the convergence behavior of the algorithm. In addition, it is well known that three parameters used in the algorithm, i.e., the inertia weight w, position constants c1 and c2, sufficiently dominate the whole searching performance. The influence of these parameter settings on the algorithm convergence will be considered and examined via a simple two-dimensional function optimization problem. All simulation results are displayed using a series of illustrations with respect to various iteration numbers. Finally, some simple rules on how to suitably assign these parameters are also suggested

Population Diversity Control of Genetic Algorithm Using a Novel Injection Method for Bankruptcy Prediction Problem

Exploration and exploitation are the two main concepts of success for searching algorithms. Controlling exploration and exploitation while executing the search algorithm will enhance the overall performance of the searching algorithm. Exploration and exploitation are usually controlled offline by proper settings of parameters that affect the population-based algorithm performance. In this paper, we proposed a dynamic controller for one of the most well-known search algorithms, which is the Genetic Algorithm (GA). Population Diversity Controller-GA (PDC-GA) is proposed as a novel feature-selection algorithm to reduce the search space while building a machine-learning classifier. The PDC-GA is proposed by combining GA with k-mean clustering to control population diversity through the exploration process. An injection method is proposed to redistribute the population once 90% of the solutions are located in one cluster. A real case study of a bankruptcy problem obtained from UCI Machine Learning Repository is used in this paper as a binary classification problem. The obtained results show the ability of the proposed approach to enhance the performance of the machine learning classifiers in the range of 1% to 4%.

Optimization of Generatively Encoded Multi-Material Lattice Structures for Desired Deformation Behavior

Natural systems achieve favorable mechanical properties through coupling significantly different elastic moduli within a single tissue. However, when it comes to man-made materials and structures, there are a lack of methods which enable production of artifacts inspired by these phenomena. In this study, a method for design automation based on alternate deposition of soft and stiff struts within a multi-material 3D lattice structure with desired deformation behavior is proposed. These structures, once external forces are applied, conform to the geometry given in advance. For that purpose, a population-based algorithm was proposed and integrated with a multi-material physics simulator. To reduce the amount of data processed during optimization, a generative encoding method based on discrete cosine transform (DCT) was proposed. This enabled a compressed topological description and promoted symmetry in material distribution. The simulation results showed different three-dimensional lattice structures designed with proposed algorithm to meet a set of desired deformation behaviors. The relation between residual deformation error, targeted deformation geometry, and material distribution is discussed.

A chaotic-based improved many-objective Jaya algorithm for many-objective optimization problems

The Jaya algorithm is a recently developed novel population-based algorithm. The proposed work presents the modifications in the existing many-objective Jaya (MaOJaya) algorithm by integrating the chaotic sequence to improve the performance to optimize many-objective benchmark optimization problems. The MaOJaya algorithm has exploitation more dominating, due to which it traps in local optima. The proposed work aims to reduce these limitations by modifying the solution update equation of the MaOJaya algorithm. The purpose of the modification is to balance the exploration and exploitation, improve the divergence and avoid premature convergence. The well-known chaotic sequence - a logistic map integrated into the solution update equation. This modification keeps the MaOJaya algorithm simple as well as, preserves its parameterless feature. The other component of the existing MaOJaya algorithm, such as non-dominated sorting, reference vector and tournament selection scheme of NSGA-II is preserved. The decomposition approach used in the proposed approach simplifies the complex many-objective optimization problems. The performance of the proposed chaotic based many-objective Jaya (C-MaOJaya) algorithm is tested on DTLZ benchmark functions for three to ten objectives. The IGD and Hypervolume performance metrics evaluate the performance of the proposed C-MaOJaya algorithm. The statistical tests are used to compare the performance of the proposed C-MaOJaya algorithm with the MaOJaya algorithm and other algorithms from the literature. The C-MaOJaya algorithm improved the balance between exploration and exploitation and avoids premature convergence significantly. The comparison shows that the proposed C-MaOJaya algorithm is a promising approach to solve many-objective optimization problems.

Energy-efficient scheduling of streaming applications in VFI-NoC-HMPSoC based edge devices

Energy-aware high-performance computing is becoming a challenging facet for streaming applications at edge devices in Internet-of-Things (IoT) due to the high computational complexity involved. Therefore, the number of processors has increased significantly on the multiprocessor system subsequently, Voltage Frequency Island (VFI) recently adopted for an effective energy management mechanism in the large scale multiprocessor chip designs. In this paper, energy-aware scheduling of real-time streaming applications on edge-devices is investigated. First, an innovative re-timing based technique is developed to transform the dependent workload into an independent task model to avail resources and the wasted slack in the processors with a possible minimal prologue. Moreover, unlike the existing population-based optimization algorithms, a novel population-based algorithm, ARSH-FATI is proposed that can dynamically switch between explorative and exploitative search modes at run-time for performance trade-off. Finally, a communication contention-aware Earliest Edge Consistent Deadline First (EECDF) scheduling algorithm is presented. Our static scheduler ARHS-FATI collectively performs task mapping and ordering. Consequently, its performance is superior to the existing state-of-the-art approach proposed for homogeneous VFI based MPSoCs.

An Efficient ACO+RVND for Solving the Resource-Constrained Deliveryman Problem

In this paper, Resource-Constrained Deliveryman Problem (RCDMP) is introduced. The RCDMP problem deals with finding a tour with minimum waiting time sum so that it consumes not more than the $R_{max}$ unites of the resources, where $R_{max}$ is some constant. Recently, an algorithm developed in a trajectory-based metaheuristic has been proposed. Since the search space of the problem is a combinatorial explosion, the trajectory-based sequential can only explore a subset of the search space, therefore, they easily fall into local optimal in some cases. To overcome the drawback of the current algorithms, we propose a population-based algorithm that combines an Ant Colony Algorithm (ACO), and Random Variable Neighborhood Descent (RVND). In the algorithm, ACO explores the promising solution areas while RVND exploits them with the hope of improving a solution. Extensive numerical experiments and comparisons with the state-of-the-art metaheuristic algorithms in the literature show that the proposed algorithm reaches better solutions in many cases.

State Estimation of Power Using the Whale Optimization Algorithm

In power systems, the process of attaining a better prediction from a set of variables from state variables is called state estimation (SE). These variables consist of magnitudes of bus voltage and the corresponding angles of all the buses. Because of the non-linearity and intricacy of ever-developing power systems, it has become important to apply upgraded techniques for the dissolution and supervision in practical environments. The discussed analysis evaluates the appositeness of a new metaheuristic technique called the whale optimization algorithm (WOA) which is a population-based algorithm, to reduce the measurement errors so as to gauge the optimal point of the power system when some susceptible values are inadequate. WOA displays admirable attainment in global optimization. It employs a bubble-net hunting approach and it mimics the social behaviour of humpback whales to get the best candidate solution. The approach is tested on IEEE-14, IEEE-30, and IEEE-57 bus test systems and the potency is validated by comparison with the biogeography based optimization algorithm (BBO).