Modified Particle Swarm Optimization with Chaotic Initialization Scheme for Unconstrained Optimization Problems

Different variants of particle swarm optimization (PSO) algorithms were introduced in recent years with various improvements to tackle different types of optimization problems more robustly. However, the conventional initialization scheme tends to generate an initial population with relatively inferior solution due to the random guess mechanism. In this paper, a PSO variant known as modified PSO with chaotic initialization scheme is introduced to solve unconstrained global optimization problems more effectively, by generating a more promising initial population. Experimental studies are conducted to assess and compare the optimization performance of the proposed algorithm with four existing well-establised PSO variants using seven test functions. The proposed algorithm is observed to outperform its competitors in solving the selected test problems.

Sequential Monte Carlo methods for multi-sensor tracking with applications to radar systems

The aim of this thesis is to explore specific sequential Monte Carlo (SMC) methods and their application to the unique demands of radar and bearing only tracking systems. Asynchronous radar networks are of special interest and a novel algorithm, the multiple imputation particle filter (MIPF), is formulated to perform data fusion and estimation using asynchronous observations. Convergence analysis is carried out to show that the algorithm will converge to the optimal filter. Simulations are performed to demonstrate the effectiveness of this filter. Next, the problem of multi-sensor bearing only tracking is tackled. A particle based tracking algorithm is derived and a new filter initialization scheme is introduced for the specific task of multi-sensor bearing only tracking. Simulated data is used to study the efficiency and performance of the initialization scheme.

Sequential Monte Carlo methods for multi-sensor tracking with applications to radar systems

The aim of this thesis is to explore specific sequential Monte Carlo (SMC) methods and their application to the unique demands of radar and bearing only tracking systems. Asynchronous radar networks are of special interest and a novel algorithm, the multiple imputation particle filter (MIPF), is formulated to perform data fusion and estimation using asynchronous observations. Convergence analysis is carried out to show that the algorithm will converge to the optimal filter. Simulations are performed to demonstrate the effectiveness of this filter. Next, the problem of multi-sensor bearing only tracking is tackled. A particle based tracking algorithm is derived and a new filter initialization scheme is introduced for the specific task of multi-sensor bearing only tracking. Simulated data is used to study the efficiency and performance of the initialization scheme.

A Review on Initialization Methods for Nonnegative Matrix Factorization: Towards Omics Data Experiments

Nonnegative Matrix Factorization (NMF) has acquired a relevant role in the panorama of knowledge extraction, thanks to the peculiarity that non-negativity applies to both bases and weights, which allows meaningful interpretations and is consistent with the natural human part-based learning process. Nevertheless, most NMF algorithms are iterative, so initialization methods affect convergence behaviour, the quality of the final solution, and NMF performance in terms of the residual of the cost function. Studies on the impact of NMF initialization techniques have been conducted for text or image datasets, but very few considerations can be found in the literature when biological datasets are studied, even though NMFs have largely demonstrated their usefulness in better understanding biological mechanisms with omic datasets. This paper aims to present the state-of-the-art on NMF initialization schemes along with some initial considerations on the impact of initialization methods when microarrays (a simple instance of omic data) are evaluated with NMF mechanisms. Using a series of measures to qualitatively examine the biological information extracted by a given NMF scheme, it preliminary appears that some information (e.g., represented by genes) can be extracted regardless of the initialization scheme used.

Modified Migrating Birds Optimization for Solving the Low-carbon Scheduling Problem

Production scheduling problems have historically emphasized cycle time without involving the environmental factors. In the study, a low-carbon scheduling problem in a flexible job shop is considered to minimize the energy consumption, which mainly consists of two parts: the useful part and the wasted part. First, a mathematical model is built based on the features of the workshop. Second, a modified migrating bird's optimization (MMBO) is developed to obtain the optimal solution. In the MMBO, a population initialization scheme is designed to enhance the solution quality and convergence speed. Five types of neighborhood structures are introduced to create neighborhood solutions. Furthermore, a local search method and a reset mechanism are developed to improve the computational results. Finally, experimental results validate that the MMBO is effective and feasible.