针对传统支持向量数据描述(support vector data description,SVDD)训练中存在的训练速度慢、存储核矩阵需要的空间开销大、计算量大、算法效率低等问题,提出一种基于改进序贯最小优化(SMO)算法的SVDD快速训练方法。该算法针对原有SMO...针对传统支持向量数据描述(support vector data description,SVDD)训练中存在的训练速度慢、存储核矩阵需要的空间开销大、计算量大、算法效率低等问题,提出一种基于改进序贯最小优化(SMO)算法的SVDD快速训练方法。该算法针对原有SMO算法仅能处理单类样本的缺陷,提出一种可以处理负样本的改进方法,给出详细的计算推导过程,并针对KKT判定条件、工作集选择等关键问题进行改进。试验证明:与传统的SVDD训练算法相比,基于改进SMO算法的SVDD快速训练方法训练时间短,计算量小,分类准确度高,空间开销小,更适合于大规模数据的快速训练,具有较高的工程应用价值。展开更多
Homing trajectory planning is a core task of autonomous homing of parafoil system.This work analyzes and establishes a simplified kinematic mathematical model,and regards the homing trajectory planning problem as a ki...Homing trajectory planning is a core task of autonomous homing of parafoil system.This work analyzes and establishes a simplified kinematic mathematical model,and regards the homing trajectory planning problem as a kind of multi-objective optimization problem.Being different from traditional ways of transforming the multi-objective optimization into a single objective optimization by weighting factors,this work applies an improved non-dominated sorting genetic algorithm Ⅱ(NSGA Ⅱ) to solve it directly by means of optimizing multi-objective functions simultaneously.In the improved NSGA Ⅱ,the chaos initialization and a crowding distance based population trimming method were introduced to overcome the prematurity of population,the penalty function was used in handling constraints,and the optimal solution was selected according to the method of fuzzy set theory.Simulation results of three different schemes designed according to various practical engineering requirements show that the improved NSGA Ⅱ can effectively obtain the Pareto optimal solution set under different weighting with outstanding convergence and stability,and provide a new train of thoughts to design homing trajectory of parafoil system.展开更多
In order to reduce both the weight of vehicles and the damage of occupants in a crash event simultaneously, it is necessary to perform a multi-objective optimal design of the automotive energy absorbing components. Mo...In order to reduce both the weight of vehicles and the damage of occupants in a crash event simultaneously, it is necessary to perform a multi-objective optimal design of the automotive energy absorbing components. Modified non-dominated sorting genetic algorithm II(NSGA II) was used for multi-objective optimization of automotive S-rail considering absorbed energy(E), peak crushing force(Fmax) and mass of the structure(W) as three conflicting objective functions. In the multi-objective optimization problem(MOP), E and Fmax are defined by polynomial models extracted using the software GEvo M based on train and test data obtained from numerical simulation of quasi-static crushing of the S-rail using ABAQUS. Finally, the nearest to ideal point(NIP)method and technique for ordering preferences by similarity to ideal solution(TOPSIS) method are used to find the some trade-off optimum design points from all non-dominated optimum design points represented by the Pareto fronts. Results represent that the optimum design point obtained from TOPSIS method exhibits better trade-off in comparison with that of optimum design point obtained from NIP method.展开更多
A new hybrid optimization algorithm was presented by integrating the gravitational search algorithm （GSA） with the sequential quadratic programming （SQP）, namely GSA-SQP, for solving global optimization problems a...A new hybrid optimization algorithm was presented by integrating the gravitational search algorithm （GSA） with the sequential quadratic programming （SQP）, namely GSA-SQP, for solving global optimization problems and minimization of factor of safety in slope stability analysis. The new algorithm combines the global exploration ability of the GSA to converge rapidly to a near optimum solution. In addition, it uses the accurate local exploitation ability of the SQP to accelerate the search process and find an accurate solution. A set of five well-known benchmark optimization problems was used to validate the performance of the GSA-SQP as a global optimization algorithm and facilitate comparison with the classical GSA. In addition, the effectiveness of the proposed method for slope stability analysis was investigated using three ease studies of slope stability problems from the literature. The factor of safety of earth slopes was evaluated using the Morgenstern-Price method. The numerical experiments demonstrate that the hybrid algorithm converges faster to a significantly more accurate final solution for a variety of benchmark test functions and slope stability problems.展开更多
基金国家自然科学基金(the National Natural Science Foundation of China under Grant No.10471045 No.60433020)+8 种基金广东省自然科学基金(the Natural Science Foundation of Guangdong Province of China under Grant No.970472 No.000463 No.04020079)广东科技公关计划(the Key Technologies R&D Program of Guangdong Province China under Grant No.2005B10101010)霍英东基金( No.91005)教育部人文社科基金(No.2005-241)广州市天河区科技攻关项目(No.051G041)华南理工大学自然科学基金(No.B13-E5050190)
文摘针对传统支持向量数据描述(support vector data description,SVDD)训练中存在的训练速度慢、存储核矩阵需要的空间开销大、计算量大、算法效率低等问题,提出一种基于改进序贯最小优化(SMO)算法的SVDD快速训练方法。该算法针对原有SMO算法仅能处理单类样本的缺陷,提出一种可以处理负样本的改进方法,给出详细的计算推导过程,并针对KKT判定条件、工作集选择等关键问题进行改进。试验证明:与传统的SVDD训练算法相比,基于改进SMO算法的SVDD快速训练方法训练时间短,计算量小,分类准确度高,空间开销小,更适合于大规模数据的快速训练,具有较高的工程应用价值。
基金Project(61273138)supported by the National Natural Science Foundation of ChinaProject(14JCZDJC39300)supported by the Key Fund of Tianjin,China
文摘Homing trajectory planning is a core task of autonomous homing of parafoil system.This work analyzes and establishes a simplified kinematic mathematical model,and regards the homing trajectory planning problem as a kind of multi-objective optimization problem.Being different from traditional ways of transforming the multi-objective optimization into a single objective optimization by weighting factors,this work applies an improved non-dominated sorting genetic algorithm Ⅱ(NSGA Ⅱ) to solve it directly by means of optimizing multi-objective functions simultaneously.In the improved NSGA Ⅱ,the chaos initialization and a crowding distance based population trimming method were introduced to overcome the prematurity of population,the penalty function was used in handling constraints,and the optimal solution was selected according to the method of fuzzy set theory.Simulation results of three different schemes designed according to various practical engineering requirements show that the improved NSGA Ⅱ can effectively obtain the Pareto optimal solution set under different weighting with outstanding convergence and stability,and provide a new train of thoughts to design homing trajectory of parafoil system.
文摘In order to reduce both the weight of vehicles and the damage of occupants in a crash event simultaneously, it is necessary to perform a multi-objective optimal design of the automotive energy absorbing components. Modified non-dominated sorting genetic algorithm II(NSGA II) was used for multi-objective optimization of automotive S-rail considering absorbed energy(E), peak crushing force(Fmax) and mass of the structure(W) as three conflicting objective functions. In the multi-objective optimization problem(MOP), E and Fmax are defined by polynomial models extracted using the software GEvo M based on train and test data obtained from numerical simulation of quasi-static crushing of the S-rail using ABAQUS. Finally, the nearest to ideal point(NIP)method and technique for ordering preferences by similarity to ideal solution(TOPSIS) method are used to find the some trade-off optimum design points from all non-dominated optimum design points represented by the Pareto fronts. Results represent that the optimum design point obtained from TOPSIS method exhibits better trade-off in comparison with that of optimum design point obtained from NIP method.
文摘A new hybrid optimization algorithm was presented by integrating the gravitational search algorithm （GSA） with the sequential quadratic programming （SQP）, namely GSA-SQP, for solving global optimization problems and minimization of factor of safety in slope stability analysis. The new algorithm combines the global exploration ability of the GSA to converge rapidly to a near optimum solution. In addition, it uses the accurate local exploitation ability of the SQP to accelerate the search process and find an accurate solution. A set of five well-known benchmark optimization problems was used to validate the performance of the GSA-SQP as a global optimization algorithm and facilitate comparison with the classical GSA. In addition, the effectiveness of the proposed method for slope stability analysis was investigated using three ease studies of slope stability problems from the literature. The factor of safety of earth slopes was evaluated using the Morgenstern-Price method. The numerical experiments demonstrate that the hybrid algorithm converges faster to a significantly more accurate final solution for a variety of benchmark test functions and slope stability problems.
