Papers by Keyword: Genetic Algorithm (GA)

Paper TitlePage

Authors: Zi Сheng Li, Li Xu, Bao Shan Yuan
Abstract: The purpose in this paper is the design of the control to switching power supply for small perturbations. By the theoretical analysis and calculation, with the output filter inductor current and filter capacitor voltage switching power supply as two state variables, the conclusion is that control of the output filter inductor current sampling do well in the anti-jamming. The simulation is made for verification. And comparing the results, the current control mode shows a very strong anti-interference ability.
Authors: Rajesh Kanna, P. Malliga, K. Sarukesi
Abstract: This paper presents a combination of Genetic Algorithm (GA) and Tuning algorithm, for optimizing Three Dimensional (3D) arbitrary sized heterogeneous box packing into a container, by considering practical constraints facing in the logistics industries. Objective of this research is to pack four different shapes of boxes of varying sizes into a container of standard dimension, without violating various practical constraints. Inorder to obtain a real time feasible packing pattern, Genetic Algorithm is developed to maximize the container volume utilization and inturn profit [9]. It significantly improves the search efficiency with less computational time and loads most of the heterogeneous boxes into the container by considering its optimal position and orientation. Tuning algorithm is used to decode the genetic output into user understandable sequential packing pattern and to fill the left-out empty space inside the container. In general, GA in conjunction with the tuning algorithm is substantially better than those obtained by applying heuristics to the bin packing directly.
Authors: Ovidiu Buiga
Abstract: In conventional methods, designing a gear drive such as a 2K-h single-row planetary gearbox requires a very large number of calculations based on recommended gear standards, trial and error methods, etc. This time consuming process may often finish up with inadequate design outcomes. Therefore, in this paper a Genetic Algorithm (GA) methaeuristics is considered in order to resolve this complex design problem. The GA was used to find the optimal values of 14 genes (i.e. design variables) that define the planetary gearbox. The optimal design of the power transmission was evaluated considering the mass minimization criterion. The results of the optimised planetary gearbox suggest a reduction of the mass with 15.16% as compared with the situation when the traditional design was used.
Authors: Si Yuan Zhao, Wang Tao, Ge Xin, Yun Liu
Abstract: A novel bearing fault diagnosis method based on Lie group was proposed, and genetic algorithm (GA) was introduced to optimize feature amount. This method was applied to inner ring fault, outer ring fault and rolling element fault of rolling bearing. Firstly, the rolling bearing vibration signal was decomposed as intrinsic model functions (IMF) by using the empirical mode decomposition (EMD) method. The energy of every IMF and the singular value of the IMF matrix were chosen as features. The Shannon and Renyi entropy of the energy and singular value distribution were also extracted. Secondly genetic algorithm was used to reduce feature redundancy, with lowest classifier error rate and least feature amount as finess function. At last, a comparison was made between this method and least square support vector machine (LSSVM).The results showed that Lie group clkassifier was more sensitivce to feature. This method could use less feature amount to diagnose fault.
Authors: Xi Chen, Jian Wu, Yang Zhao, Hong Tao Bai
Abstract: In the design of CAN network system, as CAN bus topology will affect network performance and cost, it is important to optimize the network topology. This paper analyzes the CAN busload based on CAN protocol, calculates the upper limit of transmission message frames in a single CAN bus. As the amount of information on the bus is increasing, a single CAN bus cannot meet the communication requirements, we put forward dividing the network into multiple homogeneous segments via multi-objective optimization method, developing a genetic algorithm strategy and solving the problem in a MATLAB platform. Finally utilize the method to design a pure electric vehicle network topology.
Authors: Qing Feng Xia
Abstract: Extreme Learning Machine-Radial Basis Function (ELM-RBF) not only inherit RBF’s merit of not suffering from local minima, but also ELM’s merit of fast learning speed, Nevertheless, it is still a research hot area of how to improve the generalization ability of ELM-RBF network. Genetic Algorithms (GA) to solve optimization problem has its unique advantage. Considered on these, the paper adopted GA to optimize ELM-RBF neural network hidden layer neurons center and biases value. Experiments data results indicated that our proposed combined algorithm has better generalization performance than classical ELM-RBF, it achieved the basic anticipated task of design.
Authors: Yu Hua Zhu, Dian Zheng Zhuang, Ping Li, Wei Yan Tong
Abstract: It will be face some problems about the complicated reaction mechanism, environment uncertainty, serious nonlinear in nitric acid process .a method of creating steady-state model of nitric acid process using neural network. and used genetic algorithm to optimize parameter on neural network model. The result can provide reference for analyzing and optimizing the parameters of nitric acid process.
Authors: Song Chai, Yu Bai Li, Chang Wu, Jian Wang
Abstract: Real-time task schedule problem in Chip-Multiprocessor (CMP) receives wide attention in recent years. It is partly because the increasing demand for CMP solutions call for better schedule algorithm to exploit the full potential of hardware, and partly because of the complexity of schedule problem, which itself is an NP-hard problem. To address this task schedule problem, various of heuristics have been studied, among which, Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Simulated Annealing (SA) are the most popular ones. In this paper, we implement these 3 schedule heuristics, and compare their performance under the context of real-time tasks scheduling on CMP. According to the results of our intensive simulations, PSO has the best fitness optimization of these 3 algorithms, and SA is the most efficient algorithm.
Authors: An Ping Pan
Abstract: Evacuation is a process in which threatened people are displaced from dangerous places to safer places in order to reduce the health and life vulnerability of affected people. Constructing typhoon emergency shelters is one of the ways to improve the areas anti-disaster capacity. There are many questions in emigration resettlement work. Such as: the layout of the resettlement places unreasonable, the fixed resettlement is insufficient, temporary resettlement site lacks the necessary living facilities and so on. Planning and construction of emergency shelters is a systematic project, which needs scientific and rational design. This paper proposed a mathematical model applied to the location selection, a solution of genetic algorithm was suggested. Then bring forward a decision-making model for the location-choosing of urgency response system and carried out the design procedure of genetic algorithm to finalize the optimized combination plan.
Authors: Kun Lei Lian, Chao Yong Zhang, Liang Gao, Shao Tan Xu, Yi Sun
Abstract: Process planning is an essential component of computer aided process planning (CAPP), which involves operations selection from design features and operations sequencing of these selected operations. It makes process planning a complex combinatorial optimization problem to conduct of these two steps simultaneously. In this paper, we propose a cooperative simulated annealing (CoSA) approach for the process planning problem to minimize total manufacturing cost. The proposed CoSA algorithm employed a novel optimization strategy different from all the existing approaches in the literature. Simulated annealing was utilized to optimize the four components of a process plan individually and sequentially. The approach is tested on two parts from the literature and compared with other approaches. The computation results validate the effectiveness of the proposed algorithm.
Showing 1 to 10 of 1809 Paper Titles