一种高效的匹配场反演混合优化算法

将下山单纯形算法引入差异进化算法,提高了差异进化算法对目标函数梯度信息的利用,改善了差异进化算法的优化效率,由于下山单纯形算法与差异进化算法都是并行算法,混合算法同时具备了并行高效的特点.     

流体饱和孔隙介质参数反演的模拟退火算法

本文研究了模拟退火算法在流体饱和孔隙介质参数反演中的应用。通过计算响应数据与实测响应数据的拟合将参数反问题归结为最优化问题。由于流体饱和孔隙介质运动方程的复杂性，动力响应与材料参数之间呈复杂的非线性关系，优化目标函数是非凸多峰函数。传统的梯度类优化方法一方面受局部极值的困扰难以搜索到全局最优解； 另一方面确定搜索方向须进行复杂的参数敏度分析。为克服这些困难，本文应用模拟退火算法进行了多参数反演数值模拟，模拟结果表明了模拟退火算法的可行性和稳健性。     

浅海环境参数宽带匹配场反演研究

宽带匹配场反演是获取海洋环境参数信息的有效方法 ,选择合适的反演目标函数对环境参数的估计结果十分重要。文章研究了浅海环境参数宽带匹配场反演的各种目标函数建立方法 ,同时结合亚洲海实验 (ASIAEX2 0 0 1)数据采用遗传全局寻优算法对该海域环境参数进行了反演 ,并对反演结果尤其是沉积层参数的合理性进行了分析。     

多级分销网络设计及库存控制整合优化模型

从供应链集成的角度出发,基于多目标规划,根据排队论探讨了随机性需求下多级分销网络设计与库存控制的整合优化问题,提出了多级分销网络设计和库存控制整合优化的多目标规划模型.针对遗传算法收敛速度慢、易陷入局部最优等缺点,采用了基于并列选择法的遗传-模拟退火算法混合优化策略.实验证明,模拟退火算法提高了遗传算法的全局搜索能力,改善了遗传算法的求解性能.     

基于改进遗传算法的水轮发电机振动荷载参数识别

根据水轮发电机现场振动测试实验数据,采用改进的遗传算法研究了水轮发电机运行过程中振动荷载反演问题。与传统的参数反演方法相比,遗传算法并不是基于对目标函数梯度方向搜索,而是在解的整个区域随机搜索.将遗传算法与模拟退火算法相结合,提高了种群在进化过程中个体多样性,可以有效地防止简单遗传算法早熟问题。同时,将遗传算法与梯度优化方法相结合,使得混合型遗传算法有效地解决了梯度算法局部极小问题和简单遗传算法的收敛速度慢问题。工程实际应用表明,采用本文所建立改进遗传算法所反演的水轮发电机振动荷载参数,预报其它振动观测点的位移具有较高的预报精度。     

改进型遗传算法在环境噪声反演海底问题中的应用

利用浅海环境噪声进行海底特性反演一直是学者们研究的热门课题之一,当考虑两层海底时,由于待反演参数较多,基于传统遗传算法的反演策略往往使反演过程陷入局部最小值,从而无法找到全局最优解。针对两层海底的情况,计算得到了浅海环境噪声场空间相干性的表达式,并对传统遗传算法进行了改进,通过引进自适应交叉和变异算子以及能够保持种群多样性的小生境技术来提高反演算法的性能。通过数值仿真,验证了改进后的算法较传统算法能更好地收敛到真实值;并针对一次海试数据进行了海底特性反演,获得了较好的结果。     

一种用阈值提取子空间的多步匹配场反演方法

在分析已有的匹配场反演方法的基础上,构造了一种用阈值提取子空间的多步匹配场反演方法。它根据一定反演环境下参数的不同敏感性将参数划分为子集（子空间）,并依次在各敏感子空间内反演。反演时用一定的阈值将目标函数优于阈值的参数区域提取出,最后在提取出的已相对缩减的区域和最后一个子空间（通常是不敏感参数子空间）内联合反演全部参数,求得最优值。这样既可减少反演参数空间又能可靠地保证精确度,避免了已有的子空间方法反演结果受非反演参数失配影响的问题。仿真研究结果表明,本算法比已有的两类算法性能上有明显提高。     

海底介质特性反演方法研究

海底介质特性反演一直是水声研究里很活跃的领域,本文应用两种不同的反演方法,差分进化方法(DE)和自适应下山模拟退火算法(ASSA)分别对海底介质特性进行反演,并比较了这两种算法在解决海底介质特性反演问题的优缺点,特别是通过对噪声数据和试验数据的反演发现,差分进化方法虽然是一种效率高,鲁棒性好,收敛速度快的反演算法,但是在抗噪声能力方面显然不如自适应下山模拟退火算法.     

一种用于匹配场反演的遗传算法

遗传算法在接近全局最优解时,存在搜索速度变慢、过早收敛、个体的多样性减少很快、甚至陷入局部最优解等问题。通过在遗传算法中引入模拟退火因子、混沌因子和多样性测度因子,在很大程度上克服了原有遗传算法的早熟、局部搜索能力差的缺点。同时,又能发挥原有遗传算法的强大的全局搜索能力,保证了改进后的混合遗传算法能较好地收敛于其全局最优值。     

复杂非线性函数最优化问题的一种实用智能算法

对浮点编码遗传算法加以改进,并与DFP变尺度算法相结合,经加速循环,构建新型混合加速遗传算法(以下简称NHAGA);协同求解具有变量边界约束的非凸、高度非线性的复杂函数最优化问题。算例测试表明,该法兼顾了改进浮点编码遗传算法全局搜索能力和DFP算法快速局部搜索能力的优点,成功搜索全局最优点的概率较高,是一种求解非凸、高度非线性全局优化问题的有效智能算法。     

Performance comparison of several optimization algorithms in matched field inversion

Optimization efficiencies and mechanisms of simulated annealing, genetic algorithm, differential evolution and downhill simplex differential evolution are compared and analyzed. Simulated annealing and genetic algorithm use a directed random process to search the parameter space for an optimal solution. They include the ability to avoid local minima, but as no gradient information is used, searches may be relatively inefficient. Differential evolution uses information from a distance and azimuth between individuals of a population to search the parameter space, the initial search is effective, but the search speed decreases quickly because differential information between the individuals of population vanishes. Local downhill simplex and global differential evolution methods are developed separately, and combined to produce a hybrid downhill simplex differential evolution algorithm. The hybrid algorithm is sensitive to gradients of the object function and search of the parameter space is effective. These algorithms are applied to the matched field inversion with synthetic data. Optimal values of the parameters, the final values of object function and inversion time is presented and compared.     

混合遗传算法在结构动力反演中的应用研究

研究了输入荷载未知条件下的结构参数识别及荷载反演问题,该问题最终归结为一个非线性的优化问题求解,根据目标函数、约束条件的具体特性,采用BFGS算法作为局部搜索算子,构造了基于浮点编码的混合遗传算法。针对系统输入未知的激励特性,采用分解反演的计算策略,从而提高了动力反演中混合遗传算法的稳健性和收敛速度。数值算例表明,这种方法具有很好的参数识别精度及荷载反演效果,对测试噪声有较强的适应能力。     

遗传禁忌搜索算法在混流装配线排序中的应用

针对混流装配线排序问题，提出了一种混合遗传禁忌搜索算法，在每一代遗传演化之后，按一定比例随机选择部分解进行禁总搜索，以提高算法的全局搜索能力和收敛性。通过一个混流装配线排序实验，分别利用遗传算法和遗传禁忌搜索算法进行求解，结果表明遗传禁忌搜索算法具有更好的全局搜索能力和收敛性能。     

Multidisciplinary design and optimization of an air launched satellite launch vehicle using a hybrid heuristic search algorithm

A multidisciplinary design and optimization strategy for a multistage air launched satellite launch vehicle comprising of a solid propulsion system to low earth orbit with the implementation of a hybrid heuristic search algorithm is proposed in this article. The proposed approach integrated the search properties of a genetic algorithm and simulated annealing, thus achieving an optimal solution while satisfying the design objectives and performance constraints. The genetic algorithm identified the feasible region of solutions and simulated annealing exploited the identified feasible region in search of optimality. The proposed methodology coupled with design space reduction allows the designer to explore promising regions of optimality. Modules for mass properties, propulsion characteristics, aerodynamics, and flight dynamics are integrated to produce a high-fidelity model of the vehicle. The objective of this article is to develop a design strategy that more efficiently and effectively facilitates multidisciplinary design analysis and optimization for an air launched satellite launch vehicle.     

Hybrid simplex search and particle swarm optimization for the global optimization of multimodal functions

This article proposes the hybrid Nelder–Mead (NM)–Particle Swarm Optimization (PSO) algorithm based on the NM simplex search method and PSO for the optimization of multimodal functions. The hybrid NM–PSO algorithm is very easy to implement, in practice, since it does not require gradient computation. This hybrid procedure performed the exploration with PSO and the exploitation with the NM simplex search method. In a suite of 17 multi-optima test functions taken from the literature, the computational results via various experimental studies showed that the hybrid NM–PSO approach is superior to the two original search techniques (i.e. NM and PSO) in terms of solution quality and convergence rate. In addition, the presented algorithm is also compared with eight other published methods, such as hybrid genetic algorithm (GA), continuous GA, simulated annealing (SA), and tabu search (TS) by means of a smaller set of test functions. On the whole, the new algorithm is demonstrated to be extremely effective and efficient at locating best-practice optimal solutions for multimodal functions.     

A hybrid differential evolution algorithm for a two-stage flow shop on batch processing machines with arbitrary release times and blocking

Batch processing machines (BPMs) have important applications in a variety of industrial systems. This paper considers the problem of scheduling two BPMs in a flow shop with arbitrary release times and blocking such that the makespan is minimised. The problem is formulated as a mixed integer programming model. Subsequently, a hybrid discrete differential evolution (HDDE) algorithm is proposed. In the algorithm, individuals in the population are first represented as discrete job sequences, and mutation and crossover operators are applied based on the representation. Second, after using the first-fit rule to form batches, a novel least idle/blocking time heuristic is developed to schedule the batches in the flow shop. Furthermore, an effective local search technique is embedded in the algorithm to enhance the exploitation ability. The performance of the proposed algorithm is evaluated by comparing its results to a commercial solver (CPLEX), a genetic algorithm and a simulated annealing algorithm. Computational experiments demonstrate the superiority of the HDDE algorithm in terms of solution quality, robustness and run time.     

基于改进遗传算法的海底地层参数反演计算

为了提高对海底地层参数变量的反演计算能力,设计了一种基于双种群协同进化策略的改进遗传算法。针对标准遗传算法局部搜索能力差且易于出现早熟现象的缺点,在标准遗传算法基础上引入双种群同时进行优化搜索,两个种群分别给予不同的控制参数,实现协同进化,最终给出一个综合的最优解。通过两个算例对遗传算法的寻优能力进行测试,实验结果表明,提出的改进算法不仅提高了搜索性能,并且对遗传控制参数的依靠度大大降低,特别是对大型复合参数反演问题的求解计算更为有效。     

Reducing the computational cost of local search in the hybrid evolutionary algorithm with application to electronic imaging

This article focuses on the efficiency problems associated with the use of local search in the hybrid evolutionary algorithm. A two-phase cyclic local search is proposed that alternates the random search and the downhill simplex method (DSM), and helps prevent the algorithm from converging to a sub-optimal solution in multidimensional optimization. The algorithm utilizes a novel micro-model of image local response, in order to reduce the number of fitness evaluations during the local DSM search, with the application to the global optimization problem arising in electronic imaging. The problem is stated as the search for the feasible transformation parameters that minimize the difference between two images. Image local response is defined as the variation of the fitness function that occurs because of a small variation of the parameters, and is computed over a small pixel area. The computed response coefficients specify a contraction transformation applied to the vector of the regular DSM coefficients that control the movement and the shape of the simplex. The transformation adjusts the length of the vector, making the step size of the simplex adaptive to the local properties of the fitness landscape. The computational experiments with two-dimensional grayscale images provide the experimental support and justification of the analytical model of image local response and its utilization for the reduction of the computational cost of local search, without the loss of the quality of the final solution.