基于PSO和LSSVM的生化过程建模研究

针对最小二乘支持向量机(LSSVM)在生化过程建模中的重要建模参数值选择问题.提出利用具有较强的全局搜索能力的粒子群(PSO)优化算法.对最小二乘支持向量机建模过程中的重要参数进行优化调整,每一个粒子的位置向量对应一组最小二乘支持向量机建模的参数.利用参数优化调整后得到的具有较优拟合预测效果的模型对谷氨酸发酵过程进行预测,仿真结果表明,该方法能使模型取得较好的预测效果.     

一种改进的混沌优化算法

为了克服遗传算法的早熟现象以及混沌优化的搜索时间过长的缺点,将遗传算法、混沌优化和变尺度方法相结合,提出了一种改进的混沌优化算法.该算法利用混沌的随机性、遍历性和规律性来避免陷入局部极小值,从而也克服了遗传算法中的早熟现象,同时引入了变尺度方法提高该算法的搜索速度.本文还给出了算法的收敛性分析.对典型测试函数的仿真结果表明此算法优于变尺度混沌优化和遗传算法.     

基于捕食搜索策略遗传算法的SVM参数优化方法

基于支持向量机(SVM)模型的泛化能力和拟合精度与其相关参数的选取有关,提出将捕食搜索策略的遗传算法(PSGA)运用到SVM的参数选取中。该算法以最小化输出量的拟合误差为目标,以SVM的3个参数作为决策变量。通过对谷氨酸发酵过程建模的实验表明,该方法可以提高谷氨酸浓度的训练精度及预测精度,是一种优化SVM参数的有效方法。     

基于混沌“微变异”自适应遗传算法

遗传算法可以较好地解决复杂的组合优化问题,但也存在两方面不足:一是搜索效率比其他优化算法低;二是容易过早收敛,陷入局部最优.对此,提出一种混沌“微变异”遗传算法.利用混沌优化算法具有随机性和遍历性的特点,解决遗传算法容易陷入局部最优解的早熟问题,使得新算法同时具有较强的局部搜索能力和完成全局寻找最优解的能力.同时,对遗传算法的选择算子增加了混沌扰动,对交叉算子和变异算子进行自适应调整,对适应度函数进行改进,使遗传算法整体性能得到提高.最后,通过经典函数验证表明,混沌“微变异”遗传算法比一般的混沌遗传算法和经典遗传算法的进化速度更快,搜索精度更高.     

一种改进变尺度混沌优化的模糊量子遗传算法

针对量子遗传算法存在的易陷入局部极小等问题,提出一种模糊量子遗传算法。该算法采用一种变尺度混沌优化方法,只需设 2个循环,内循环进行混沌搜索,外循环负责缩小区间,通过改进它的收敛策略,可以避免混沌优化在区间内的盲目重复搜索。利用改进的变尺度混沌优化方法,对量子遗传操作产生的种群进行混沌搜索寻优,同时模糊控制更新,加快种群的进化。仿真结果表明,该方法的寻优效果优于量子遗传算法及遗传算法。     

发酵过程混合建模与优化控制新方法

针对当前微生物发酵过程存在因为生物传感器不具备足够的准确性和灵敏性,实验时的菌液和产物浓度等生化指标难以实时监测和控制等缺点,提出了采用量子粒子群优化算法(QPSO)优化最小二乘支持向量机(LSSVM)参数的QPSO-LSSVM混合建模新方法,并用于多粘菌素的发酵过程建模;同时,基于此模型,采用QPSO算法对pH值与溶解氧浓度Do控制轨线进行优化研究;首先,利用LSSVM进行发酵过程的建模,然后采用QPSO对LSSVM建模过程中的重要参数进行优化调整,形成QPSO-LSSVM混合建模与优化控制方法;仿真结果表明,该方法得到的模型能取得更好的预测效果,优化后的pH值与Do浓度控制轨线能够提高最终的产物浓度;该方法用于发酵过程的建模和重要参数的优化控制是可行的、有效的。     

基于最小二乘支持向量机和混沌优化的非线性预测控制

针对非线性多入多出（MIMO）系统,提出一种基于最小二乘支持向量机（LSSVM）和混沌优化的预测 控制策略．预测模型是预测控制的三要素之一．本文给出了基于混沌优化的Chaos-LSSVM 算法,在可行域内反复搜 索,从而得到最优的LSSVM 算法参数,以及最优的LSSVM 模型．在线优化是另一个要素．提出了基于变尺度混沌 优化的MSC-MPC（变尺度混沌－模型预测控制）算法,可根据控制误差的大小,决定是否缩小搜索范围,从而迅速 收敛到最优解．该算法计算简单,容易实现,避免了同类方法复杂的求导、求逆运算．仿真结果显示：Chaos-LSSVM 算法和MSC-MPC 算法分别具有良好的建模、控制性能．     

基于APSO—LSSVM的软测量建模研究

针对最小二乘支持向量机在生化过程建模中的重要建模参数值选择问题,提出利用具有较强的全局搜索能力的自适应粒子群（APSO）优化算法,对最小二乘支持向量机建模过程中的重要参数进行优化调整,每一个粒子的位置向量对应一组最小二乘支持向量机建模的参数。利用参数优化调整后得到的具有较优拟和预测效果的模型对谷氨酸发酵过程进行预测,仿真结果表明该方法能使模型取得较好的预测效果。     

超混沌遗传算法在多约束QoS组播路由中的应用研究

针对多约束QoS组播路由的优化问题,提出了一种超混沌遗传混沌算法.该算法利用遗传算法中的改进的适应度函数,通过结合超混沌映射优越性的搜索能力,对遗传算法选出的个体进行混沌优化,以改善遗传算法过早陷入早熟的情况.通过仿真实验表明,该算法有效地改进了搜索效率,且收敛速度更快更稳定,是一种解决多约束QoS路由问题可行和有效的方法.     

基于QPSO-LSSVM的L-缬氨酸发酵建模

最小二乘支持向量机(least square support vector machine,LSSVM)是1种新的机器学习算法,它采用结构风险最小化准则,能有效提高模型的泛化能力,且具有运算速度快、抗噪能力强等特点.本文针对最小二乘支持向量机发酵建模中,选择重要模型参数值的问题,提出利用全局搜索能力强的量子粒子群优化算法,优化LSSVM建模过程的重要参数,并将该混合建模方法应用于L-缬氨酸发酵,建立L-缬氨酸产物浓度、菌体浓度和底物浓度等重要过程变量的预测模型,在线预估这些不能在线测量的生化状态变量.实验表明,混合算法所建立起的L-缬氨酸发酵模型在模拟菌体生长、底物消耗及发酵产酸过程的变化等方面都比BP神经网络建模方法具有较小的拟合误差和较好的推广性能,可以为L-缬氨酸发酵生产过程提供动态模拟,具有重要的实用价值.     

Combining Bayesian optimization and Lipschitz optimization

Machine Learning - Bayesian optimization and Lipschitz optimization have developed alternative techniques for optimizing black-box functions. They each exploit a different form of prior about the...     

Multi-verse optimization algorithm- and salp swarm optimization algorithm-based optimization of multilevel inverters

Neural Computing and Applications - Renewable energy sources are installed into both distribution and transmission grids more and more with the introduction of smart grid concept. Hence, efficient...     

Elephant herding optimization using dynamic topology and biogeography-based optimization based on learning for numerical optimization

With the increasing complexity of optimization problems in the real world, more and more intelligent algorithms are used to solve these problems. Elephant herding optimization (EHO), a recently proposed metaheuristic algorithm, is based on the nomadic habits of elephants on the grassland. The herd is divided into multiple clans, each individual drawing closer to the patriarchs (clan updating operator), and the adult males are separated during puberty (separating operator). Biogeography-based optimization (BBO) is inspired by the principles of biogeography, and finally achieves an equilibrium state by species migration and drifting between geographical regions. To solve the numerical optimization problems, this paper proposes an improved elephant herding optimization using dynamic topology and biogeography-based optimization based on learning, named biogeography-based learning elephant herding optimization (BLEHO). In BLEHO, we change the topological structure of the population by dynamically changing the number of clans of the elephants. For the updating of each individual, we use the update of the operator based on biogeography-based learning or the operator based on EHO. In the separating phase, we set the separation probability according to the number of clans, and adopt a new separation operator to carry out the separation operation. Finally, through elitism strategy, a certain number of individuals are preserved directly to the next generation without being processed, thus ensuring a better evolutionary process for the population. To verify the performance of BLEHO, we used the benchmarks provided by IEEE CEC 2014 for the test. The experimental results were compared with some classical algorithms (ABC, ACO, BBO, DE, EHO, GA, and PSO) and the most advanced algorithms (BBKH, BHCS, CCS, HHO, PPSO, SCA, and VNBA) and analyzed by Friedman rank test. Finally, we also applied BLEHO to the simple traveling salesman problem (TSP). The results show that BLEHO has better performance than other methods.

     

Large scale structural optimization: Computational methods and optimization algorithms

Summary The objective of this paper is to investigate the efficiency of various optimization methods based on mathematical programming and evolutionary algorithms for solving structural optimization problems under static and seismic loading conditions. Particular emphasis is given on modified versions of the basic evolutionary algorithms aiming at improving the performance of the optimization procedure. Modified versions of both genetic algorithms and evolution strategies combined with mathematical programming methods to form hybrid methodologies are also tested and compared and proved particularly promising. Furthermore, the structural analysis phase is replaced by a neural network prediction for the computation of the necessary data required by the evolutionary algorithms. Advanced domain decomposition techniques particularly tailored for parallel solution of large-scale sensitivity analysis problems are also implemented. The efficiency of a rigorous approach for treating seismic loading is investigated and compared with a simplified dynamic analysis adopted by seismic codes in the framework of finding the optimum design of structures with minimum weight. In this context a number of accelerograms are produced from the elastic design response spectrum of the region. These accelerograms constitute the multiple loading conditions under which the structures are optimally designed. The numerical tests presented demonstrate the computational advantages of the discussed methods, which become more pronounced in large-scale optimization problems.     

混合性能指标优化问题的进化优化方法及应用

针对混合性能指标优化问题的普遍性及其处理过程中的特点,提出一种混合性能指标优化问题的进化优化方法,首先给出混合性能指标优化问题的定义;然后．确定不同类型和标度的性能指标的转换策略、混合性能指标的个体适应度的赋值方法、以及混合性能指标优化问题的进化优化流程;最后．通过服装设计这一典型的混合性能指标优化问题的仿真验证了算法的有效性．     

Backtracking biogeography-based optimization for numerical optimization and mechanical design problems

As a novel Evolutionary Algorithm (EA), Biogeography-Based Optimization (BBO), inspired by the science of biogeography, draws much attention due to its significant performance in both numerical simulations and practical applications. In BBO, the features in poor solutions have a large probability to be replaced by the features in good solutions. The replacement operator is termed migration. However, the replacement causes a loss of the features in poor solutions, breaks the diversity of population and may lead to a local optimal solution. To overcome this, we design a novel migration operator to propose Backtracking BBO (BBBO). In BBBO, besides the regular population, an external population is employed to record historical individuals. The size of external population is the same as the size of regular population. The external population and regular population are used together to generate the next population. After that, the individuals in external population are randomly selected to be updated by the individuals in current population. In this way, the external population in BBBO can be considered as a memory to take part in the evolutionary process. The memory takes into account both current and historical data to generate next population, which enhances algorithm’s ability in exploring searching space. In numerical simulation, 14 classical benchmarks are employed to test BBBO’s performance and several classical nature inspired algorithms are use in comparison. The results show that the strategy in BBBO is feasible and very effective to enhance algorithm’s performance. In addition, we apply BBBO to mechanical design problems which involve constraints in optimization. The comparison results also exhibit that BBBO is very competitive in solving practical optimization problems.     

The particle swarm optimization algorithm in size and shape optimization

Shape and size optimization problems instructural design are addressed using the particle swarm optimization algorithm (PSOA). In our implementation of the PSOA, the social behaviour of birds is mimicked. Individual birds exchange information about their position, velocity and fitness, and the behaviour of the flock is then influenced to increase the probability of migration to regions of high fitness. New operators in the PSOA, namely the elite velocity and the elite particle, are introduced. Standard size and shape design problems selected from literature are used to evaluate the performance of the PSOA. The performance of the PSOA is compared with that of three gradient based methods, as well as the genetic algorithm (GA). In attaining the approximate region of the optimum, our implementation suggests that the PSOA is superior to the GA, and comparable to gradient based algorithms. Received December 18, 2000     

Multiobjective optimization using population-based extremal optimization

In recent years, a general-purpose local-search heuristic method called Extremal Optimization (EO) has been successfully applied in some NP-hard combinatorial optimization problems. In this paper, we present a novel Pareto-based algorithm, which can be regarded as an extension of EO, to solve multiobjective optimization problems. The proposed method, called Multiobjective Population-based Extremal Optimization (MOPEO), is validated by using five benchmark functions and metrics taken from the standard literature on multiobjective evolutionary optimization. The experimental results demonstrate that MOPEO is competitive with the state-of-the-art multiobjective evolutionary algorithms. Thus MOPEO can be considered as a viable alternative to solve multiobjective optimization problems.     

LCD optimization and modeling

Abstract— The application of the LCD modeling and optimization system, LCD DESIGN, for the design and development of new advanced LCD configurations was demonstrated. The software includes a powerful optimization module that allows for spectral and angular averaging, thus enabling the production of LCDs with wide viewing angles, achromatic (black/white) switching, and fast response time. We describe the basic principles of the software development and present several examples of LCD optimization in various electro‐optical modes. A brief review of our results of LCD optimization and modeling using LCD DESIGN software is also given.