融合指标分组的高维混合多目标进化优化

高维混合多目标优化问题因包含多个不同类型指标,目前尚缺乏有效求解该问题的进化优化方法。提出一种基于目标分组的高维混合多目标并行进化优化方法。采用深度学习神经网络预测种群隐式性能指标;基于指标相关性,将高维混合多目标优化问题分解为若干子优化问题;采用多种群并行进化算法,求解分解后的每一子优化问题,并基于各子种群的非被占优解构建外部保存集;采用聚合函数对外部保存集个体进一步优化,得到Pareto最优解集。在室内布局优化问题中验证所提方法,实验结果表明,所提方法的Pareto最优解在收敛性、分布性以及延展性等方面均优于对比方法。     

基于自适应学习的多目标粒子群优化算法

将进化算法应用于某些多目标优化问题时,采用增加种群规模和进化代数的方法往往耗费大量的目标函数计算开销,且达不到提高种群进化效率的目的,为此提出了一种基于自适应学习最优搜索方向的多目标粒子群优化算法。采用自适应惯性权值平衡算法的全局和局部搜索能力,采用聚类排挤方法保持Pareto非支配解集的分布均匀性,使用最近邻学习方法为每个粒子在Pareto非支配解集中寻找一个最优飞行目标来提高其收敛速度并保持粒子群搜索方向的多样性。实验结果表明,提出的算法可在显著地降低函数评估成本的前提下实现快速的搜索,并使粒子群均匀地逼近Pareto最优面。     

基于Pareto解集分段预测策略的动态多目标进化算法

针对现有的动态多目标优化算法种群收敛速度慢、多样性难以保持等问题,提出了一种基于Pareto解集分段预测策略的动态多目标进化算法BPDMOP。当检测到环境变化时,对前一时刻进化得到的Pareto最优解根据任一子目标函数进行排序,并按照该子目标的大小均分为3段,分别计算出每一段Pareto解集中心点的移动方向;对每一段Pareto子集进行系统抽样得到Pareto前沿面的特征点,利用线性模型分段预测下一代种群;根据优化问题的难易程度,自适应地在预测的种群周围产生随机个体来增加种群的多样性。通过对3类标准测试函数的实验表明了该算法能够有效求解动态多目标优化问题。     

多目标差分进化在热连轧负荷分配中的应用

提出一种基于差分进化算法的多目标进化算法, 该算法个体的选择是通过非支配排序和拥挤度距离进行综合考虑. 保证了算法收敛到Pareto最优解集的同时, 提高了最优解个体分布的多样性. 通过与非支配排序遗传算法Ⅱ(NSGA Ⅱ)算法进行仿真对比, 结果显示基于拥挤度排序的多目标差分进化算法在收敛性和Pareto最优解集分布的多样性上均优于NSGA Ⅱ算法. 最后将其引入到热连轧负荷分配优化计算中, 给出了目标函数的表达方式, 对多目标进化算法在热连轧负荷分配计算中的应用进行了研究.     

基于差分进化粒子群算法的多目标无功优化

针对电力系统有功网损最小、电压水平最好和电压稳定裕度最大的多目标无功优化问题,提出一种基于差分进化的改进多目标粒子群优化算法。该算法通过对Pareto最优解集的差分进化来增加Pareto最优解的多样性,通过拥挤距离来控制精英集中非支配解的分布,以提高对种群空间的均匀采集;采用擂台赛法则构造多目标Pareto最优解集,较大程度的提高了算法的运行效率;自适应惯性权重和加速度因子的动态变化可增强算法的全局搜索能力。将该算法在IEEE14、IEEE30节点标准测试系统上进行了无功优化仿真,结果表明,基于差分进化的改进多目标粒子群优化算法能够在保持Pareto最优解的多样性的同时具有较好的收敛性能,为多目标无功优化提供了一种新的方法。     

基于人工免疫系统的多目标函数优化

为克服传统遗传算法退化和早熟等缺点,同时降低优化算法的复杂度,提出基于人工免疫系统（Artificial Immune System, AIS）实现无约束多目标函数的优化。使用随机权重法和自适应权重法计算种群个体的适应值,使Pareto最优解均匀分布的同时,加快算法的收敛;通过引入人工免疫系统的三个基本算子：克隆、超变异和消亡,保持种群的多样性;在进化种群外设立Pareto 解集,保存历代的近似最优解。使用了两个典型的多目标检测函数验证了该算法的有效性。优化结果表明,基于AIS的多目标优化算法可使进化种群迅速收敛到Pareto前沿,并能均匀分布,是实现多目标函数优化的有效方法。     

求解多目标作业车间调度问题的混合变异杂草优化算法

针对多目标作业车间调度问题,提出一种混合变异杂草优化算法。该算法采用基于各子目标熵值权重的欧氏贴近度作为适应度值计算方法,引导种群向Pareto前端进化。在进化过程中,运用快速非支配排序策略构建Pareto档案,并利用进化种群中最优个体实时更新Pareto最优解集,提升算法的优化性能;同时通过引入变异算子增加种群多样性,避免算法陷入局部最优。最后,基于Benchmark算例的仿真实验,验证了该算法求解多目标作业车间调度问题的有效性。     

基于弱关联的自适应高维多目标进化算法

对现有的分解方法进行改进,提出一种基于弱关联的自适应高维多目标进化算法(WAEA).首先,提出一种基于夹角子空间的关联策略,使得一个解能与多个参考向量相关联;其次,提出弱关联概念,并基于此概念设计双模态标量函数,使算法能够更好地处理复杂PF问题,此外,算法通过检测参考向量子空间内解的数量,自适应调整惩罚参数大小,使其能有效处理各类多目标问题;最后,将WAEA算法与8种代表性的高维多目标算法进行比较,实验结果表明WAEA算法在处理复杂Pareto前沿的高维多目标问题时能更好地平衡Pareto最优解的收敛性与多样性.     

基于群智能的模糊多目标软件可靠性冗余分配

针对软件可靠性冗余分配问题,建立了一种模糊多目标分配模型,并提出了基于分布估计的细菌觅食优化算法求解该模型。将软件可靠性和成本作为模糊目标函数,通过三角形隶属函数对模糊多目标进行处理,用高斯分布对细菌觅食算法进行优化,并将该优化算法用来求解多目标软件可靠性冗余分配问题,设置不同的隶属函数参数可以得到不同的Pareto最优解,实验数据验证了该群智能算法对解决多目标软件可靠性分配的有效性和正确性,Pareto最优解可为在可靠性和成本之间决策提供依据。     

记忆增强的动态多目标分解进化算法

现实世界中的一些多目标优化问题经常受动态环境影响而不断发生变化,要求优化算法不断地及时跟踪时变的Pareto 最优解集.提出了一种记忆增强的动态多目标分解进化算法.将动态多目标优化问题分解为若干个动态单目标优化子问题并同时优化这些子问题,以便快速逼近Pareto 最优解集.给出了一个改进的环境变化检测算子,以便更好地检测环境变化.设计了一种基于子问题的串式记忆方法,利用过去类似环境下搜索到的最优解来有效地响应新的环境变化.在8 个标准的测试问题上,将新算法与其他3 种记忆增强的动态进化多目标优化算法进行了实验比较.结果表明,新算法比其他3 种算法具有更快的运行速度、更强的记忆能力与鲁棒性能,并且新算法所获得的解集还具有更好的收敛性与分布性.     

基于GA-NN的复杂工艺生产过程多目标优化研究

在复杂非线性多目标优化问题求解中,非线性模型结构很难事先给定,需要检验的参数也非常繁多,应用传统的建模方法和优化模型已难以解决更为复杂的现实问题。人工神经网络技术为解决复杂非线性系统建模问题提供了一条新的途径。将神经网络响应面作为目标函数或者约束条件,加上其他常规约束条件进行系统模型的建立,再应用遗传算法进行优化,从而实现设计分析与设计优化的分离。以某化工企业的生产过程优化问题为例,利用BP神经网络建立了工艺参数与性能目标之间的模型,然后利用遗传算法搜索最优工艺参数,获取了用于指导生产的样本点数据。研究结果表明,该方法能够获得高精度的多目标优化模型,从而使优化效率大为提高。     

A surrogate-assisted evolutionary algorithm based on the genetic diversity objective

In this work, a novel surrogate-assisted memetic algorithm is proposed which is based on the preservation of genetic diversity within the population. The aim of the algorithm is to solve multi-objective optimization problems featuring computationally expensive fitness functions in an efficient manner. The main novelty is the use of an evolutionary algorithm as global searcher that treats the genetic diversity as an objective during the evolution and uses it, together with a non-dominated sorting approach, to assign the ranks. This algorithm, coupled with a gradient-based algorithm as local searcher and a back-propagation neural network as global surrogate model, demonstrates to provide a reliable and effective balance between exploration and exploitation. A detailed performance analysis has been conducted on five commonly used multi-objective problems, each one involving distinct features that can make the convergence difficult toward the Pareto-optimal front. In most cases, the proposed algorithm outperformed the other state-of-the-art evolutionary algorithms considered in the comparison, assuring higher repeatability on the final non-dominated set, deeper convergence level and higher convergence rate. It also demonstrates a clear ability to widely cover the Pareto-optimal front with larger percentage of non-dominated solutions if compared to the total number of function evaluations.     

A reliable multi-objective control strategy for batch processes based on bootstrap aggregated neural network models

This paper presents a reliable multi-objective optimal control method for batch processes based on bootstrap aggregated neural networks. In order to overcome the difficulty in developing detailed mechanistic models, bootstrap aggregated neural networks are used to model batch processes. Apart from being able to offer enhanced model prediction accuracy, bootstrap aggregated neural networks can also provide prediction confidence bounds indicating the reliability of the corresponding model predictions. In addition to the process operation objectives, the reliability of model prediction is incorporated in multi-objective optimisation in order to improve the reliability of the obtained optimal control policy. The standard error of the individual neural network predictions is taken as the indication of model prediction reliability. The additional objective of enhancing model prediction reliability forces the calculated optimal control policies to be within the regions where the model predictions are reliable. By such a means, the resulting control policies are reliable. The proposed method is demonstrated on a simulated fed-batch reactor and a simulated batch polymerisation process. It is shown that by incorporating model prediction reliability in the optimisation criteria, reliable control policy is obtained.     

Modeling,analysis and multi-objective optimization of twist extrusion process using predictive models and meta-heuristic approaches,based on finite element results

Recently, twist extrusion has found extensive applications as a novel method of severe plastic deformation for grain refining of materials. In this paper, two prominent predictive models, response surface method and artificial neural network (ANN) are employed together with results of finite element simulation to model twist extrusion process. Twist angle, friction factor and ram speed are selected as input variables and imposed effective plastic strain, strain homogeneity and maximum punch force are considered as output parameters. Comparison between results shows that ANN outperforms response surface method in modeling twist extrusion process. In addition, statistical analysis of response surface shows that twist extrusion and friction factor have the most and ram speed has the least effect on output parameters at room temperature. Also, optimization of twist extrusion process was carried out by a combination of neural network model and multi-objective meta-heuristic optimization algorithms. For this reason, three prominent multi-objective algorithms, non-dominated sorting genetic algorithm, strength pareto evolutionary algorithm and multi-objective particle swarm optimization (MOPSO) were utilized. Results showed that MOPSO algorithm has relative superiority over other algorithms to find the optimal points.     

A polar-based guided multi-objective evolutionary algorithm to search for optimal solutions interested by decision-makers in a logistics network design problem

In practical multi-objective optimization problems, respective decision-makers might be interested in some optimal solutions that have objective values closer to their specified values. Guided multi-objective evolutionary algorithms (guided MOEAs) have been significantly used to guide their evolutionary search direction toward these optimal solutions using by decision makers. However, most guided MOEAs need to be iteratively and interactively evaluated and then guided by decision-makers through re-formulating or re-weighting objectives, and it might negatively affect the algorithms performance. In this paper, a novel guided MOEA that uses a dynamic polar-based region around a particular point in objective space is proposed. Based on the region, new selection operations are designed such that the algorithm can guide the evolutionary search toward optimal solutions that are close to the particular point in objective space without the iterative and interactive efforts. The proposed guided MOEA is tested on the multi-criteria decision-making problem of flexible logistics network design with different desired points. Experimental results show that the proposed guided MOEA outperforms two most effective guided and non-guided MOEAs, R-NSGA-II and NSGA-II.     

基于自适应多目标进化CNN的图像分割方法

卷积神经网络已经成为强大的分割模型,但通常为手动设计,这需要大量时间并且可能导致庞大而复杂的网络.人们对自动设计能够准确分割特定领域图像的高效网络架构越来越感兴趣,然而大部分方法或者没有考虑构建更加灵活的网络架构,或者没有考虑多个目标优化模型.鉴于此,提出一种称为AdaMo-ECNAS的自适应多目标进化卷积神经架构搜索算法,用于特定领域的图像分割,在进化过程中考虑多个性能指标并通过优化模型的多目标适应特定的数据集. AdaMo-ECNAS可以构建灵活多变的预测分割模型,其网络架构和超参数通过基于多目标进化的算法找到,算法基于自适应PBI实现3个目标进化问题,即提升预测分割的F₁-score、最大限度减少计算成本以及最大限度挖掘额外训练潜能.将AdaMo-ECNAS在两个真实数据集上进行评估,结果表明所提出方法与其他先进算法相比具有较高的竞争性,甚至是超越的.     

智能集成控制在大功率电弧炉系统中的应用研究

针对冶金行业大功率电弧炉的控制，提出了一种基于模糊控制、神经网络和多目标 优化决策相结合的智能集成控制方案．首先采用变结构模糊神经网络控制来设计温度外环控制器，给三相电极电流平衡内环提供电流指令信号，然后在内环控制中综合各种优化目标，构造优化目标函数，运用多目标模糊优化决策来实现整个系统的平衡．现场数据表明运用该控制方案的系统目前已在广东韶关冶炼厂成功运行．     

面向深度卷积网络的多目标神经演化算法

传统的深度卷积神经网络设计方法依赖于人工设计以及反复试错,只能采用形式单一的网络结构,导致其参数过分冗余,乘法次数巨大.为了自动化地设计出结构灵活多变,网络规模及计算量较小的深度卷积神经网络,本文提出了一种面向深度卷积网络的多目标神经演化算法.该算法将深度神经网络表达成有向图,使用神经演化和多目标优化算法实现了深度、计算量和识别率下的多目标同时优化,同时还引入了线性规划用于将基因编码翻译为卷积层的配置参数,使得演化算法可以自动调整各个网络层的具体配置.演化得到的模型其最深路径上含有36个卷积层,CIFAR-100上Top5精度为86.1%,Top1精度为60.2%,与识别率相近的网络相比,具有结构新颖,乘法次数低等特点.综上,本文提出的方法能够自动生成一系列各具特色的深度神经网络,可根据在深度、计算量和识别率3个指标上的不同应用需求选择适合的深度神经网络,为深度神经网络部署于资源受限的无线传感器网络上提供了一种快速、经济、自动化的设计方法.     

Multi-objective community detection in complex networks

Community detection in social network analysis is usually considered as a single objective optimization problem, in which different heuristics or approximate algorithms are employed to optimize a objective function that capture the notion of community. Due to the inadequacy of those single-objective solutions, this paper first formulates a multi-objective framework for community detection and proposes a multi-objective evolutionary algorithm for finding efficient solutions under the framework. After analyzing and comparing a variety of objective functions that have been used or can potentially be used for community detection, this paper exploits the concept of correlation between objective which charcterizes the relationship between any two objective functions. Through extensive experiments on both artifical and real networks, this paper demonstrates that a combination of two negatively correlated objectives under the multi-objective framework usually leads to remarkably better performance compared with either of the orignal single objectives, including even many popular algorithms..     

求解随机机会约束规划的混合智能算法及应用

为更有效地求解随机机会约束规划问题,提出一种基于克隆选择算法(CSA)、随机模拟技术及神经网络的混合智能算法。采用随机模拟技术产生随机变量样本矩阵训练反向传播(BP)网络以逼近不确定函数,之后在CSA中利用神经网络检验个体的可行性、计算适应度,从而得到优化问题的最优解。为保证算法搜索的快速性和有效性,CSA采用双克隆和双变异策略。仿真结果表明,与已有算法相比,混合智能算法在500代时已取得比较满意的结果,且其精度在单目标优化问题中提高了2.2%,在多目标优化问题中提高了65%;将该算法应用于求解水库优化调度的难题上,结果也表明所建立的模型及算法的可行性和有效性。