Multi-objective Evolutionary Algorithm Method Research Articles

Numerical simulation and microchannels parameters optimization for thermal management of GaN HEMT devices

PurposeThis study aims to satisfy the thermal management of gallium nitride (GaN) high-electron mobility transistor (HEMT) devices, microchannel-cooling is designed and optimized in this work.Design/methodology/approachA numerical simulation is performed to analyze the thermal and flow characteristics of microchannels in combination with computational fluid dynamics (CFD) and multi-objective evolutionary algorithm (MOEA) is used to optimize the microchannels parameters. The design variables include width and number of microchannels, and the optimization objectives are to minimize total thermal resistance and pressure drop under constant volumetric flow rate.FindingsIn optimization process, a decrease in pressure drop contributes to increase of thermal resistance leading to high junction temperature and vice versa. And the Pareto-optimal front, which is a trade-off curve between optimization objectives, is obtained by MOEA method. Finally, K-means clustering algorithm is carried out on Pareto-optimal front, and three representative points are proposed to verify the accuracy of the model.Originality/valueEach design variable on the effect of two objectives and distribution of temperature is researched. The relationship between minimum thermal resistance and pressure drop is provided which can give some fundamental direction for microchannels design in GaN HEMT devices cooling.

Multi-Objective Optimization of Spatially Truss Structures Based on Node Movement

This paper discusses the solutions for topology optimization of spatially discrete structures. The optimization objects are the structural weight and the maximum displacement. The optimization variables include structural node coordinates, and the improved MOEA (Multi-objective Evolutionary Algorithm) method is used to optimize the structure. The innovation of this study is that it breaks through the shortage of constant node position in the optimization thought of traditionally discrete structure in the “Ground Structure Approach” and uses the coordinate of the node as the optimization variable for the optimization calculation. The result is not a single one but a set of optimal solutions through the evolution (i.e., Pareto optimal solutions); on this basis, the most suitable solution can be found according to the boundary conditions or other related requirements. Using the C# language to compile the calculation program, ANSYS finite element software is used to analyze the structure, and the Pareto front surface was automatically drawn to determine the optimal layout form of the discrete structure. The analysis results show that the improved MOEA method can provide an effective method to solve such optimization problems.

Optimal Design of Hydraulic Excavator Shovel Attachment Based on Multiobjective Evolutionary Algorithm

The optimal design of TriPower shovel attachment remains a challenging task although TriPower face shovel excavators work well in mines all over the world. To address this issue, we establish the multiobjective optimization model of TriPower shovel attachment, which sets three kinematic indicators and two mechanical indicators as the objective functions. An improved multiobjective evolutionary algorithm (MOEA) based on decomposition is proposed to solve the multiobjective optimization problem, which outperforms several representative MOEAs in the case study, and a lot of nondominated solutions are obtained. Then, the most satisfactory solution is selected based on the technique for order of preference by similarity to ideal solution method. Based on this solution, an excellent design of TriPower face shovel excavator is obtained, which can fully realize the three features of TriPower shovel attachment. These results demonstrate the feasibility and effectiveness of the MOEA and multicriteria decision-making method in the optimal design of hydraulic excavator working attachment.

Modifications of Fitness Assignment Parts of Some Multi-Objective Evolutionary Algorithms

Accepted: 03.12.2014 Evolutionary algorithms are very popular methods to solve multiobjective optimization problems. In literature, there are many multi-objective evolutionary algorithm methods. It is possible to improve these methods. This paper proposes a line of refinements and modifications on the fitness assignment part of well known SPEA method. The density information generated by k-NN (k-th nearest neighbor method) and the concept of domination power are used to modify SPEA in order to deliver a much refinement fitness values in terms of uniqueness and homogeneity. Two modified methods are proposed (named SPEAmod1 and SPEAmod2). The main goal is to extract more information from a population (and is to deliver to the decision maker) by using proposed variants than the original method. The original method and its modifications are tested on four well-known test functions in the literature. The proposed modifications are shown to be superior to the original method. Finally, fitness assignment capability of original SPEA method is significantly improved.

Capacitance Characteristic Optimization of Germanium MOSFETs with Aluminum Oxide by Using a Semiconductor-Device-Simulation-Based Multi-Objective Evolutionary Algorithm Method

This paper, for the first time, optimizes the characteristics of capacitance–voltage (C–V) of germanium (Ge) metal-oxide-semiconductor field effect transistors (MOSFETs) with aluminum oxide (Al2O3) by using a semiconductor-device-simulation-based multi-objective evolutionary algorithm (MOEA) technique. By solving a set of 2D semiconductor device transport equations, numerical simulation is intensively performed for the optimization of the C–V curve of Ge MOSFET devices. To optimize the capacitance of Ge MOSFETs with respect to the applied voltage, by minimizing the total errors of the C–V curve between the device simulation and a given specification (and experimentally measured data), the thicknesses of Al2O3 and GeO2, the work function of gate electrodes, the distribution range of channel doping, the dielectric constants of Al2O3 and GeO2, and the source/drain doping concentration are considered in the process of optimization. The semiconductor device simulation and the MOEA method are integrated and performed based on a unified optimization framework. According to the sharp variation characteristics of the C–V curve, except for using a residual sum of squares (RSS) (i.e., the sum of squares of residuals) as an objective function, physical key parts of the curve are also considered in the optimization problem. The engineering results of this study indicate that the semiconductor-device-simulation-based MOEA method shows great performance to optimize the parameters, which not only minimize the objective values but also match the curve shape.

A new probabilistic canopy dynamics model (SLCD) that is suitable for evergreen and deciduous forest ecosystems

There are strong uncertainties regarding LAI dynamics in forest ecosystems in response to climate change. While empirical growth & yield models (G&YMs) provide good estimations of tree growth at the stand level on a yearly to decennial scale, process-based models (PBMs) use LAI dynamics as a key variable for enabling the accurate prediction of tree growth over short time scales. Bridging the gap between PBMs and G&YMs could improve the prediction of forest growth and, therefore, carbon, water and nutrient fluxes by combining modeling approaches at the stand level.Our study aimed to estimate monthly changes of leaf area in response to climate variations from sparse measurements of foliage area and biomass. A leaf population probabilistic model (SLCD) was designed to simulate foliage renewal. The leaf population was distributed in monthly cohorts, and the total population size was limited depending on forest age and productivity. Foliage dynamics were driven by a foliation function and the probabilities ruling leaf aging or fall. Their formulation depends on the forest environment.The model was applied to three tree species growing under contrasting climates and soil types. In tropical Brazilian evergreen broadleaf eucalypt plantations, the phenology was described using 8 parameters. A multi-objective evolutionary algorithm method (MOEA) was used to fit the model parameters on litterfall and LAI data over an entire stand rotation. Field measurements from a second eucalypt stand were used to validate the model. Seasonal LAI changes were accurately rendered for both sites (R2=0.898 adjustment, R2=0.698 validation). Litterfall production was correctly simulated (R2=0.562, R2=0.4018 validation) and may be improved by using additional validation data in future work. In two French temperate deciduous forests (beech and oak), we adapted phenological sub-modules of the CASTANEA model to simulate canopy dynamics, and SLCD was validated using LAI measurements. The phenological patterns were simulated with good accuracy in the two cases studied. However, LAImax was not accurately simulated in the beech forest, and further improvement is required.Our probabilistic approach is expected to contribute to improving predictions of LAI dynamics. The model formalism is general and suitable to broadleaf forests for a large range of ecological conditions.

Quantitative Trait Loci Mapping Problem: An Extinction-Based Multi-Objective Evolutionary Algorithm Approach

The Quantitative Trait Loci (QTL) mapping problem aims to identify regions in the genome that are linked to phenotypic features of the developed organism that vary in degree. It is a principle step in determining targets for further genetic analysis and is key in decoding the role of specific genes that control quantitative traits within species. Applications include identifying genetic causes of disease, optimization of cross-breeding for desired traits and understanding trait diversity in populations. In this paper a new multi-objective evolutionary algorithm (MOEA) method is introduced and is shown to increase the accuracy of QTL mapping identification for both independent and epistatic loci interactions. The MOEA method optimizes over the space of possible partial least squares (PLS) regression QTL models and considers the conflicting objectives of model simplicity versus model accuracy. By optimizing for minimal model complexity, MOEA has the advantage of solving the over-fitting problem of conventional PLS models. The effectiveness of the method is confirmed by comparing the new method with Bayesian Interval Mapping approaches over a series of test cases where the optimal solutions are known. This approach can be applied to many problems that arise in analysis of genomic data sets where the number of features far exceeds the number of observations and where features can be highly correlated.