Multi-sensor fusion: an Evolutionary algorithm approach

In this paper, the evolutionary algorithms approach is applied to the parameterization of a mathematical model describing the Mössbauer spectra of nanogranular (or nanoparticle) magnetic systems. These systems exhibit physical properties very different from bulk specimens being of great interest for material science and its use as biosensors, magneto sensors, data storage, and magnetic fluids. The purpose of this work is to compare the performance between the Differential Evolution and the Evolutionary Strategies algorithms to optimize the model parameters which best fit the experimental Mössbauer spectra of nanoscale magnetic particles. Spectra of two samples (??iron foil and NiFe2O4 nanoparticles) were recorded, at room temperature, by a conventional Mössbauer spectrometer using a scintillation detector in transmission geometry with a 57Co/Rh source. Fits to Mössbauer spectra were done using spin hamiltonians to describe both the electronic and nuclear interactions; a model of superparamagnetic relaxation of two levels (spin ½) and stochastic theory; a lognormal particle size distribution function as well as a dependency of the magnetic transition temperature and the anisotropy constant on particle diameter. The evolutionary algorithms have been implemented using Python programming language. For comparison, the two algorithms obey the termination criterion of 6,000 evaluations of the objective function. The results presented show the efficiency of these algorithms in the optimization of the parameters and on the fits of the spectra.

A novel evolutionary algorithm (EA) approach is developed for the implementation of rule based controllers, which increases the diversity of individuals to reduce the opportunities of falling into local optima. The rule based controller is simply composed of a look up table whose dimensions are determined by the numbers of the subranges of error variables required for operation. Many experiments are successfully made on the evolution of the rule based controllers by the novel EA for three illustrative examples: DC servomotor control, ball and beam control, and cart pendulum control. The computer simulations show that the proposed EA is an effective approach to generating control rules which manipulate nonlinearly dynamical systems exceedingly well. It reveals that the EA's learning behavior is very like that of a human expert.

The core problem for networked systems for underwater surveillance revolves around the requirement to blend information from different sensors and platforms into a common operating picture. Data fusion of disparate data types can be problematic given the variety of potential surveillance systems in the network. Active and passive sonar in particular provide different data types with different accuracies. The acoustic environment is highly stochastic and errors are prevalent that can be promulgated throughout the system. An evolutionary algorithm approach for fusing active and passive sonar contact information is used to explore the issue of robustness when large contact errors prevail from multiple platforms. The algorithm may have the ability to recover from erroneous input at the expense of lower track accuracy. This may have a significant utility as an operator aid for construction of the global picture of fused active and passive data. A multitude of erroneous sources can still result in higher overall track errors and further research is required; however, the evolutionary approach may still provide significant benefit

WITHDRAWN: WITHDRAWN: An evolutionary algorithm approach for vehicle routing problems with backhauls

Tin monoxide, SnO, and its analog, lead monoxide, PbO, have the same tetragonal P4/nmm structure, shaped by nonbonding dispersion forces and lone pairs. The high-pressure phases of SnO and PbO have been explored in several experimental and theoretical studies, with conflicting results. In this study, the high-pressure structures of SnO and PbO are investigated using density functional theory calculations combined with an evolutionary algorithm to identify novel high-pressure phases. We propose that the monoclinic P21/m SnO and orthorhombic Pmmn PbO phases, which are metastable at 0 GPa, are a slight rearrangement of the tetragonal P4/nmm-layered structure. These orthorhombic (and their closely related monoclinic) phases become more favored than the tetragonal phase upon compression. In particular, the transition pressures to the orthorhombic γ-phase Pmn21 of SnO/PbO and the monoclinic phase P21/m of SnO are found to be consistent with experimental studies. Two new high-pressure SnO/PbO polymorphs are predicted: the orthorhombic Pbcm phase of SnO and the monoclinic C2/m of PbO. These phases are stabilized in our calculations when P > 65 GPa and P > 50 GPa, respectively. The weakening of the lone pair localization and elastic instability are the main drivers of pressure-induced phase transitions. Modulations of the SnO/PbO electronic structure due to structural transitions upon compression are also discussed.

This paper investigates the use of evolutionary and other meta-heuristic algorithms for routing problems where vehicle operating cost (VOC) and specifically, road roughness, has a significant impact. Three algorithms were implemented, namely a greedy heuristic, simulated annealing and CMA-ES. Simulated annealing delivered statistically significant results and was used to evaluate routes with and without VOC.

We attempt to find solutions to the fully fuzzified linear program where all the parameters and variables are fuzzy numbers. We first change the problem of maximizing a fuzzy number, the value of the objective function, into a multiobjective fuzzy linear programming problem. We prove that fuzzy flexible programming can be used to explore the whole undominated set to the multiobjective fuzzy linear program. An evolutionary algorithm is designed to solve the fuzzy flexible program. The program was applied to two applications with good solutions.

Evolutionary algorithm approaches to solving multiobjective optimization problems have received much attention in recent years. This is one of three books published in the last year or so on the subject, and Multiobjective Evolutionary Algorithms and Applications is distinguished by its claim to being written for a wide readership of engineers, researchers, senior undergraduate students and graduate students. The book is organized into 14 chapters, with the first and second chapters providing a brief introduction into multi-objective optimization and a historical summary of the basic similarities and differences between several significant multiobjective evolutionary algorithm (MOEA) approaches. Chapter 3 provides a more in-depth look at the important subject of distribution preservation. Chapters 4, 5 and 6 introduce the author's own MOEA features and approaches. In Chapter 4, the author's suggestions for implementing hard and soft goal prioritizations and problem constraints and an adaptive fitness sharing method are presented. Chapters 5 and 6 do the same for adaptive dynamic population sizes and distributed cooperative coevolution respectively. Chapter 7 introduces potential advantages from adaptive adjustment of the size of the search space.

Facility location under uncertain environments is an important and challenging problem. The problem deals with the optimal placement of facilities that serve a set of spatially distributed nodes. One way to deal with this problem is to model uncertainty by means of scenarios and to optimise some robustness criteria such as the average and maximum regrets over these scenarios. We propose to model the robust design as a bi-objective optimisation problem and to use a well-known multi-objective evolutionary algorithm, the NSGA-II, to solve it. We also propose to use the bi-objective optimisation framework to analyse the effects of variations in the number of facilities to install, and of nodes to be served, on the quality of the Pareto solutions. Computational experiments show that the proposal can be used to design robust solutions and to study the effects of changes in the system parameters on the quality of the generated solutions.

Constrained optimization is a computationally difficult task, particularly if the constraint functions are nonlinear and non-convex. As a generic classical approach, the penalty function approach is a popular methodology which degrades the objective function value by adding a penalty proportional to the constraint violation. However, the penalty function approach has been criticized for its sensitivity to the associated penalty parameters. Since its inception, evolutionary algorithms have been modified in various ways to solve constrained optimization problems. Of them, the recent use of a bi-objective evolutionary algorithm in which the minimization of the constraint violation is included as an additional objective has received significant attention. In this article, a combination of a bi-objective evolutionary approach with the classical penalty function methodology is proposed, in a manner complementary to each other. The evolutionary approach provides an appropriate estimate of the penalty parameter, while the solution of an unconstrained penalized function by a classical method induces a convergence property to the overall hybrid algorithm. The working of the procedure on a number of standard numerical test problems and an engineering design problem is demonstrated. In most cases, the proposed hybrid methodology is observed to take one or more orders of magnitude fewer function evaluations to find the constrained minimum solution accurately than some of the best reported existing methodologies.

PurposeThis paper aims to present an evolutionary algorithm (EA) to accelerate the convergence for the radiative transfer equation (RTE) numerical solution using high-order and high-resolution schemes by the relaxation coefficients optimization.Design methodology/approachThe objective function minimizes the residual value difference between iterations in each control volume until its difference is lower than the convergence criterion. The EA approach is evaluated in two configurations, a two-dimensional cavity with scattering media and absorbing media.FindingsExperimental results show the capacity to obtain the numerical solution for both cases on all interpolation schemes tested by the EA approach. The EA approach reduces CPU time for the RTE numerical solution using SUPERBEE, SWEBY and MUSCL schemes until 97% and 135% in scattering and absorbing media cases, respectively. The relaxation coefficients optimized every two numerical solution iterations achieve a significant reduction of the CPU time compared to the deferred correction procedure with fixed relaxation coefficients.Originality/valueThe proposed EA approach for the RTE numerical solution effectively reduces the CPU time compared to the DC procedure with fixed relaxation coefficients.

In this research, an evolutionary genetic algorithm based approach is proposed for designing independent machine cells in cellular manufacturing in the presence of operation based alternative process plans for parts. Manufacturing parameters, such as production volume, usage factor for process plans, machine flexibility and cell size and machine-operation compatibility in terms of a normalised rating factor are also considered during cell design. The problem is formulated as a multi-objective optimisation model and solved using a non-dominated sorting genetic algorithm (NSGA). A unique feature of the proposed method is that it takes into account the unequal efficiency of machines in performing different operations. Some new similarity and diversity measures among operations and machines are also proposed during the clustering of machines. The formation of machine cells has been treated as a minimisation of inter-cell traffic while maximising a defined efficiency function. An illustrated problem and comparisons with existing data are given.

In real-world applications, there exist a lot of multiobjective optimization problems whose Pareto-optimal solutions are sparse, that is, most variables of these solutions are 0. Generally, many sparse multiobjective optimization problems (SMOPs) contain a large number of variables, which pose grand challenges for evolutionary algorithms to find the optimal solutions efficiently. To address the curse of dimensionality, this article proposes an evolutionary algorithm for solving large-scale SMOPs, which aims to mine the sparse distribution of the Pareto-optimal solutions and, thus, considerably reduces the search space. More specifically, the proposed algorithm suggests an evolutionary pattern mining approach to detect the maximum and minimum candidate sets of the nonzero variables in the Pareto-optimal solutions, and uses them to limit the dimensions in generating offspring solutions. For further performance enhancement, a binary crossover operator and a binary mutation operator are designed to ensure the sparsity of solutions. According to the results on eight benchmark problems and four real-world problems, the proposed algorithm is superior over existing evolutionary algorithms in solving large-scale SMOPs.

With the rapid development of information technology, multi-sensor information fusion research has been made great progress, currently there are Bayesian test, the voting method, fuzzy sets, neural networks and other methods algorithm being used in multi-sensor information fusion commonly. However, there are many shortcomings which are difficult to overcome in the process of fusion by these methods, so now the international community has already proposed to apply the D-S evidence theory to information fusion. This method does not require the priori probability, and it's very useful when it is used in information fusion, but the information with the conflict can not be effectively fused by using D-S evidence theory. Larger conflict in the evidence, the fusion result is more easily prone to false consensus. Therefore, this paper presents an improved information fusion method based on original D-S evidence theory, and this method can deal with the evidences with conflict effectively. In this paper, the improvement is in the denominator of the D-S evidence theory. When the evidence is purely of conflict, the denominator is zero, the formula does not make sense. This improvement thinking is coupled with sum between the orthogonal sum of all of evidence and product, which is between the conflict factor and the inverse value of the standard deviation of the support degree, and then it subtract maximum difference of the basic probability of the evidence. If the conflict of the evidence is greater, this advantage of this method is more obvious. If there is no conflict in the evidence, the fusion result is consistent with item of original D-S evidence theory. Finally the simulation experiment gives a set of data with more conflict. The comparative data from the experiments show that the improved method of information fusion is essential to improved in resolving conflict issues, and it is effective.

K-Means is a well-established clustering algorithm widely used in data analysis and various real-world applications. However, its requirement for a predefined number of clusters limits its effectiveness in automatic clustering tasks. To address this, metaheuristic optimisation algorithms have been integrated into K-Means, leading to the development of Evolutionary K-Means clustering approaches. These methods often rely on internal validity indices as fitness functions to automatically determine both the optimal number of clusters and the clustering configuration. However, the effectiveness of internal validity indices is often data-dependent, as most are tailored to specific data characteristics. Consequently, the choice of validity index can significantly influence clustering outcomes. This study evaluates the performance of fifteen internal validity indices within the Enhanced Firefly Algorithm-K-Means (FA-K-Means) framework, an evolutionary approach that integrates Firefly metaheuristics with the classical K-Means algorithm. The performance of each index is assessed across a diverse collection of real-life and synthetic datasets with varying structures. The results reveal that the Calinski-Harabasz (CH) and Silhouette indices consistently outperform others, offering more reliable clustering performance. These findings provide practical guidance for selecting appropriate fitness functions in Evolutionary K-Means algorithms for automatic clustering tasks.