一种模拟退火和粒子群混合优化算法

针对粒子群优化算法(PSO)容易陷入局部极值点、进化后期收敛慢和优化精度较差等缺点.把模拟退火技术(SA)引入到PSO箅法中,提出了一种混合优化算法.混合优化算法在各温度下依次进行PSO和SA搜索,是一种两层的串行结构.由于PSO提供了并行搜索结构,所以,混合优化算法使SA转化成并行SA算法.SA的概率突跳性保证了种群的多样性,从而防止PSO算法陷入局部极小.混合优化算法保持了PSO算法简单容易实现的特点,改善了算法的全局优化能力,提高了算法的收敛速度和计算精度.仿真结果表明,混合优化算法的优化性能优于基本PSO算法.     

三维适形放疗计划系统中的交互式测量方法

为了在制定三维适形放疗计划时合理地确定射线角度,并根据给定射线方向准确计算射线剂量,需要对三维模型进行精确的交互式测量。结合放疗计划系统的需求和医学图像特点,分析三维交互式测量过程,应用平行投影理论实现二维坐标到三维坐标的转换。提出以一个测量端点为起始点的三维任意叠加旋转算法,在重离子放疗计划系统中实现三维精确测量。实验结果证明了该方法的可行性和有效性。     

多体动力学优化的并行SQP方法

并行计算的发展大大提高计算机的计算效率,降低计算时间.针对多体动力学的优化问题,分析了求解灵敏度的三种方法的并行性,建立了有限差分法与直接微分法的并行算法.同时采用并行Armijo线性搜索,构成了完整的并行序列二次规划(SQP)算法.将上述算法应用到曲柄滑块的优化中,并与串行SQP算法进行了比较,证实了并行SQP算法可以大大降低计算时间.上述研究为多体动力学优化提供了一种并行求解思路.     

蚁群算法的并行性研究

并行计算作为现代计算机的一种重要的计算方法,在很大程度上优化了蚁群算法的计算过程.蚁群算法本身隐含着一定的并行性,从本质上来说,蚁群算法是以并行式的协同优化计算方式为特征,利用并行计算求出最优解.本文重点讨论蚁群算法的并行实现,并通过一个仿真实验验证并行优化蚁群算法在解决一个具有时变动态、连续、多输入、非线性系统的最优控制问题上的最优解决方法,得出蚁群算法在加速比上更具有优势.     

基于分层并行灾变粒子群算法的交通控制

提出了分层并行策略结合灾变模型的混合粒子群算法--分层并行灾变粒子群算法(HPCPSO),它能提高算法的收敛性和稳定性.同时通过对交叉口交通情况的研究,把车辆延误、车辆停车数和能源消耗都纳入性能指标值PI,建立了区域交通协调控制优化模型.在此模型的基础上,应用分层并行灾变粒子群算法实现了交通信号优化控制及验证算法.仿真结果表明,分层并行灾变粒子群算法相对于基本粒子群算法提高了寻找全局最优解的能力,能够有效实现交通信号优化控制.     

多核并行编程技术在中文分词程序优化中的应用

为了充分利用多核处理器的硬件资源和计算能力,提出了多核并行编程技术在中文分词程序中的优化方案.根据中文分词最大正向匹配算法的特点,由传统的串行程序,改为并行程序.利用多核并行编程模式的思想,设计了一个混合并行编程模式,通过Intel的性能分析工具,找出了该算法的热点和瓶颈,对其进行优化.实验结果表明,优化过后的执行时间较原来串行程序的执行时间缩短了50%～60%,同时提高了程序的加速性能,取得了良好的效果.     

并行定向扰动的混合粒子群优化算法

针对全向变异易使粒子失去已有的有利搜索信息的问题, 提出了一种并行定向变异的混合粒子群优化算法。该算法以当前群体最优位置为基准, 用变异信息矩阵和混沌位置变异矩阵对群体进行并行定向扰动, 有效利用了现有的有利搜索信息。该算法将并行定向变异与序列二次规划法融为一体, 实现了全局搜索和局部寻优的统一。仿真实验和比较分析结果表明并行定向变异混合粒子群优化算法具有良好的、稳定的优化效果。     

基于多核系统的混合神经网络及应用

借助混沌免疫遗传优化算法对于BP神经网络进行训练,建立基于混沌免疫遗传算法的混合神经网络模型.针对混沌免疫遗传神经网络计算工作量大,训练速度慢的缺点,利用Matlab的Parallel Computing Toolbox对于所建立的混沌免疫遗传神经网络模型进行并行化算法设计实现,并对渤海海区年极值冰厚数据进行预测,对比分析了串行和并行算法的计算效率和加速比,表明基于多核系统的并行化设计算法可以提高加速比和计算效率.     

分层并行遗传算法和遗传复合形算法及其应用

基于复合形算法、遗传算法、分层和并行思想,设计了一种求解复杂多目标、多约束和多变量工程优化问题的分层并行遗传或复合形算法,编制了界面友好和计算可靠性高的VC++软件。对于一类复杂三多工程综合优化问题,进行了遗传算法、复合形算法、分层并行遗传算法和分层并行遗传复合形算法的大量计算,结果表明：分层并行遗传算法计算效率最高;为解决复杂的三多工程综合优化问题提供了有效的可行方法。     

基于逆向分层的网格工作流调度算法

有向无环图DAG(Directed Acrylic Graph)描述的工作流时间费用优化问题是计算网格下一个基本的且难以求解的问题.通过分析DAG图中活动的并行和同步完成特征,采取由后向前方法将活动逆向分层(BottomLevel,BL),将工作流截止期转化为层截止时间,提出截止期约束的逆向分层费用优化算法DBL(Deadline BottomLevel).算法中同层活动的开始时间不同于DTL(Deadline Top Level)算法中设置相同的策略,而是分别由其前驱活动确定,时间浮差被平均分配到各分层,以尽量增大活动的费用优化区间.通过大量模拟实验将DBL和MCP(mini mumCritical Path)、DTL两算法比较,结果表明DTL将MCP的平均费用降低15.62%,而DBL将MCP的平均费用降低24.74%.最后讨论了截止期和分组参数对算法性能的影响.     

基于正则化的混合准则放疗规划模型的改进

针对调强放射治疗中基于等效均匀剂量线性目标函数的物理生物混合准则模型的不足,利用正则化理论构造了最大函数的平滑和凸正则化函数,改进了基于等效均匀剂量线性目标函数的混合准则放疗规划模型,解决了原模型限制优化算法寻优能力和难以确定梯度算法步长的问题。实验证明,该方法可以在保证相似靶区剂量覆盖特性的前提下,更好地保护危及器官,提高了放疗计划质量。     

异构多目标差分-动态窗口算法 及其在移动机器人中的应用

为了实现在多移动机器人和多窄通道的复杂动态环境中机器人的节能运动规划,提出异构多目标差分-动态窗口法(heterogeneous multi-objective differential evolution-dynamic window algorithm,HMODE-DWA).首先,建立行驶时间、执行器作用力和平滑度的3目标优化模型,设计具有碰撞约束的异构多目标差分进化算法来获得3个目标函数的最优解,进而在已知的静态环境中获得帕累托前沿,利用平均隶属度函数获得起点与终点间最优的全局路径;其次,定义基于环境缓冲区域的模糊动态窗口法使机器人完成动态复杂环境中避障,利用所提出的HMODE-DWA算法动态避障的同时实现节能规划.仿真和实验结果表明,所提出的混合路径规划控制策略能够有效降低移动机器人动态避障过程中的能耗.     

Optimization strategies for non-linear material parameters identification in metal forming problems

The quality of Finite Element Analysis (FEM) results relies on the input data, such as the material constitutive models. In order to achieve the best material parameters for the material constitutive models assumed a priori to represent the material, parameter identification inverse problems are considered. These inverse problems attempt to lead to the most accurate results with respect to physical experiments, i.e. minimizing the difference between experimental and numerical results.In this work three constitutive models were considered, namely, a non-linear elastic–plastic hardening model, a hyperelastic model -more specifically the Ogden model- and an elasto-viscoplastic model with isotropic and kinematic work-hardening.For the determination of the best suited material parameter set, two different optimization algorithms were used: (i) the Levenberg–Marquardt algorithm, which is gradient-based and (ii) a real search-space evolutionary algorithm (EA).The robustness and efficiency of classical single-stage optimization methods can be improved with new optimization strategies. Strategies such as cascade, parallel and hybrid approaches are analysed in detail. In hybrid strategies, cascade and parallel approaches are integrated. These strategies were implemented and analysed for the material parameters determination of the above referred material constitutive models.It was observed that the developed strategies lead to better values of the objective function when compared with the single-stage optimizers.     

预制构件并行生产线资源配置与生产调度集成优化

并行生产线和特定工序生产资源共享模式可以显著改善客户满意度并节约成本.针对预制构件并行生产线资源配置与生产调度集成优化问题,基于分解策略和交替迭代优化思想,提出一种交替式混合果蝇-禁忌搜索算法(AHFOA＿TS)以最小化拖期惩罚费用.首先,通过快速启发式方法产生一较好初始解;然后,固定资源配置方案,为提高算法局部搜索能力,通过集成多种局部搜索方式,设计一种离散果蝇优化算法优化订单指派及调度方案;最后,固定订单指派及调度方案,为减少无效搜索次数,设计一种基于双层变异算子和精英劣解交叉策略的混合禁忌搜索算法以优化资源配置方案,如此两个阶段交替运行直至满足终止条件.此外,设计4种基于交替搜索框架的智能优化算法用于比较.计算结果表明, AHFOA＿TS算法能够更有效求解预制构件生产线资源配置和生产调度集成优化问题.     

基于一类混合策略的模型参数估计和控制器参数整定研究

融合遗传算法的并行搜索结构和模拟退火的可控性概率突跳行性,构造出一类高效的混优化策略,该策略适用于多种类型模型的参数估计和控制器参数整定。对典型类型问题的仿真结果验证了混合策略的有效有和初值鲁棒性,且其优化性能明显好于单一的遗传算法和传统方法。     

Embedded Parallelization Approach for Optimization in Aerodynamic Design

A new efficient parallelization strategy for optimization of aerodynamic shapes is proposed. The optimization method employs a full Navier-Stokes solver for accurate estimation of the objective function. As such it requires huge computational resources which makes efficient parallelization crucial for successful promotion of the method to an engineering environment. The algorithm is based on a multilevel embedded parallelization approach, which includes (1) parallelization of the multiblock full Navier-Stokes solver with parallel CFD evaluation of objective function, (2) parallelization of optimization process with parallel optimal search on multiple search domains and, finally, (3) parallel grid generation. Applications (implemented on a 144-processors distributed memory cluster) include various transonic profile optimizations in the presence of nonlinear constraints. The results demonstrate that the approach combines high accuracy of optimization with high parallel efficiency. The proposed multilevel parallelization which efficiently makes use of computational power supplied by multiprocessor systems, leads to a significant computational time-saving and allows application of the method to practical aerodynamic design in the aircraft industry.     

Simultaneous estimations of temperature-dependent thermal conductivity and heat capacity using a time efficient novel strategy based on MEGA–NN

The aim of the presented novel strategy is to find the best values of input parameters, while the objective functions are not explicitly known in terms of input parameters and their values only can be calculated by a time-consuming simulation. In this paper, a hybrid modified elitist genetic algorithm–neural network (MEGA–NN) strategy is proposed for such optimization problems. The good approximation performance of neural network (NN) and the effective and robust evolutionary searching ability of modified elitist genetic algorithm (MEGA) are applied in hybrid sense, where NNs are employed in predicting the objective value, and MEGA is adopted in searching optimal designs based on the predicted fitness values. The proposed strategy (MEGA–NN) is used to estimate the temperature-dependent thermal conductivity and heat capacity using inverse heat transfer method. In order to demonstrate the accuracy and time efficiency of the proposed strategy, the results are compared to those of pre-selected parameters and MEGA. Finally, the results show that proposed MEGA–NN could save a great deal of time depending on the case.     

Process planning optimization for parallel drilling of blind holes using a two phase genetic algorithm

Determining the optimal process parameters and machining sequence is essential in machining process planning since they significantly affect the cost, productivity, and quality of machining operations. Process planning optimization has been widely investigated in single-tool machining operations. However, for the research reported in process planning optimization of machining operations using multiple tools simultaneously, the literature is scarce. In this paper, a novel two phase genetic algorithm (GA) is proposed to optimize, in terms of minimum completion time, the process parameters and machining sequence for two-tool parallel drilling operations with multiple blind holes distributed in a pair of parallel faces and in multiple pairs of parallel faces. In the first phase, a GA is used to determine the process parameters (i.e., drill feed and spindle speed) and machining time for each hole subject to feed, spindle speed, thrust force, torque, power, and tool life constraints. The minimum machining time is the optimization criterion. In the second phase, the GA is used to determine the machining sequence subject to hole position constraints (i.e., the distribution of the hole locations on each face is fixed). The minimum operation completion time is the optimization criterion in this phase. Simulation results are presented to demonstrate the effectiveness of the proposed algorithm in solving the process planning optimization problem for parallel drilling of blind holes on multiple parallel faces. In order to evaluate the performance of proposed algorithm, the simulation results are compared to a methodology that utilizes the exhaustive method in the first phase and a sorting algorithm.     

A Greedy Ant Colony System for Defensive Resource Assignment Problems

The weapon-target assignment (WTA) problem is crucial for strategic planning in military decision-making operations. It defines the best way to assign defensive resources against threats in combat scenarios. This is a NP-complete problem where no exact solution is available to deal with all possible scenarios. A critical issue in modeling the WTA problem is the time performance of the developed algorithms, subject only recently contemplated in related publications. This paper presents a hybrid approach which combines an ant colony optimization with a greedy algorithm, called the Greedy Ant Colony System (GACS), in which a multi colony parallel strategy was also implemented to improve the results. Aiming at large scale air combat scenarios, simulations controlling the algorithm time performance were executed achieving good quality results.     

An effective hybrid optimization strategy for job-shop scheduling problems

Simulated annealing is a naturally serial algorithm, but its behavior can be controlled by the cooling schedule. Genetic algorithm exhibits implicit parallelism and can retain useful redundant information about what is learned from previous searches by its representation in individuals in the population, but GA may lose solutions and substructures due to the disruptive effects of genetic operators and is not easy to regulate GA's convergence. By reasonably combining these two global probabilistic search algorithms, we develop a general, parallel and easily implemented hybrid optimization framework, and apply it to job-shop scheduling problems. Based on effective encoding scheme and some specific optimization operators, some benchmark job-shop scheduling problems are well solved by the hybrid optimization strategy, and the results are competitive with the best literature results. Besides the effectiveness and robustness of the hybrid strategy, the combination of different search mechanisms and structures can relax the parameter-dependence of GA and SA.Scope and purposeJob-shop scheduling problem (JSP) is one of the most well-known machine scheduling problems and one of the strongly NP-hard combinatorial optimization problems. Developing effective search methods is always an important and valuable work. The scope and purpose of this paper is to present a parallel and easily implemented hybrid optimization framework, which reasonably combines genetic algorithm with simulated annealing. Based on effective encoding scheme and some specific optimization operators, the job-shop scheduling problems are well solved by the hybrid optimization strategy.