ADAPTIVE MULTI-OBJECTIVE OPTIMIZATION BASED ON NONDOMINATED SOLUTIONS

An adaptive hybrid model (AHM) based on nondominated solutions is presented in this study for multi-objective optimization problems (MOPs). In this model, three search phases are devised according to the number of nondominated solutions in the current population: 1) emphasizing the dominated solutions when the population contains very few nondominated solutions; 2) maintaining the balance between nondominated and dominated solutions when nondominated ones become more; 3) when the population consists of adequate nondominated solutions, dominated ones could be ignored and the isolated nondominated ones are allocated more computational budget by their crowding distance values for heuristic search. To exploit local information efficiently, a local incremental search algorithm, LISA, is proposed and merged into the model. This model maintains the adaptive mechanism between the optimization process by the online discovered nondominated solutions. The proposed model is validated using five ZDT and five DTLZ problems. Compared with three other state-of-the-art multi-objective algorithms, namely NSGA-II, SPEA2, and PESA-II, AHM achieves comparable results in terms of convergence and diversity metrics. Finally, the sensitivity of introduced parameters and scalability to the number of objectives are investigated.     

一种基于JADE改进的差分演化算法

差分演化算法有局部搜索能力不足、容易跌入局部最优等缺点,其搜索性能主要依赖于对杂交概率和缩放因子的设置。为了改善上述缺陷,对带归档的自适应差分演化算法JADE进行深入的研究与分析,提出了改进的自适应差分演化算法ZJADE。该算法采用斜帐篷混沌映射函数初始化种群,在每次迭代中为每个个体分别产生满足正态分布、柯西分布的杂交概率和满足正态分布的缩放因子,并且记录成功变异个体的杂交概率和缩放因子,引入统计杂交概率,采用两种策略自适应地更新杂交概率。在13个经典测试函数上将ZJADE算法与多种经典自适应差分演化算法进行对比,实验结果表明,ZJADE算法在解的精度与收敛速度上更优,具有更好的搜索性能。     

自适应差分进化算法优化BP神经网络的时间序列预测

针对BP神经网络学习算法随机初始化连接权值和阈值易使模型陷入局部极小点的缺点,设计了一种自适应差分进化算法优化BP神经网络的混合算法。该混合算法中,差分进化算法采用自适应变异和交叉因子优化BP神经网络的初始权值和阈值,再用预寻优得到的初始权值和阈值训练BP神经网络得到最优的权值和阈值。首先对改进的自适应差分进化算法运用测试函数进行性能测试,然后用一个经典时间序列问题对提出的混合算法进行了检验,并与一般的神经网络、ARIMA预测模型及其它混合预测模型进行了对比,实验结果表明,本文提出的混合算法有效并且明显提高了预测精度。     

具有Pbest引导机制的适应性多策略差分进化算法

为进一步提升差分进化算法的收敛性能, 提出一种具有Pbest引导机制的适应性多策略差分进化算法(AMSDE)。AMSDE设计交叉概率控制参数库、变异尺度参数库及差分变异策略库, 其中, 交叉概率采用Logistic混沌序列来模拟, 尺度参数采用线性变化机制产生, 差分变异策略库采用6个常用的差分变异策略组成, 并给出算法运行的框架。最后, 基于25个标准测试函数的仿真结果表明, AMSDE的收敛性能优于现存的其它差分进化算法, 具有较好的收敛精度及收敛速度。     

AN ANALYSIS OF MULTIOBJECTIVE EVOLUTIONARY ALGORITHMS FOR OPTIMIZATION PROBLEMS WITH TIME CONSTRAINTS

Many multiobjective optimization problems in the engineering field are required to be solved within more or less severe time restrictions. Because the optimization criteria, the parameters, and/or constraints might change with time, the optimization solutions must be recalculated when a change takes place. The time required by the optimization procedure to arrive at the new solutions should be bounded accordingly with the rate of change observed in these dynamic problems. This way, the faster the optimization algorithm is to obtain solutions, the wider is the set of dynamic problems to which that algorithm can be applied. Here, we analyze the performance of the nondominated sorting algorithm (NSGA-II), strength Pareto evolutionary algorithm (SPEA2), and single front genetic algorithms (SFGA, and SFGA2) on two different multiobjective optimization problems, with two and three objectives, respectively. For these two studied problems, the single front genetic algorithms have obtained adequate quality in the solutions in very little time. Moreover, for the second and more complex problem approached, SFGA2 and NSGA-II obtain the best hypervolume in the found set of nondominated solutions, but SFGA2 employs much less time than NSGA-II. These results may suggest that single front genetic algorithms, especially SFGA2, could be appropiate to deal with optimization problems with high rates of change, and thus stronger time constraints.     

基于拥挤距离的多目标粒子群算法

该算法通过引用NSGA-II中的拥挤距离,确定外部档案中非支配解的拥挤度,依据竞标赛选择方法选出每个粒子的全局最优位置,引导每个粒子向处于较稀松区域的非支配解搜索,提高了解的多样性。动态变异算子的引入,减缓了算法的收敛速度,增大了解的搜索区域,避免了算法早熟收敛或陷入局部最优。实验结果表明,算法CDMOPSO比NSGA-II具有更好的收敛性和维持种群多样性的能力。     

A clustering based forecasting algorithm for multivariable fuzzy time series using linear combinations of independent variables

There are two popular types of forecasting algorithms for fuzzy time series (FTS). One is based on intervals of universal sets of independent variables and the other is based on fuzzy clustering algorithms. Clustering based FTS algorithms are preferred since role and optimal length of intervals are not clearly understood. Therefore data of each variable are individually clustered which requires higher computational time. Fuzzy Logical Relationships (FLRs) are used in existing FTS algorithms to relate input and output data. High number of clusters and FLRs are required to establish precise input/output relations which incur high computational time. This article presents a forecasting algorithm based on fuzzy clustering (CFTS) which clusters vectors of input data instead of clustering data of each variable separately and uses linear combinations of the input variables instead of the FLRs. The cluster centers handle fuzziness and ambiguity of the data and the linear parts allow the algorithm to learn more from the available information. It is shown that CFTS outperforms existing FTS algorithms with considerably lower testing error and running time.     

An enhanced self-adaptive differential evolution based solution methodology for multiobjective optimal power flow

This paper presents an Enhanced Self-adaptive Differential Evolution with Mixed Crossover (ESDE-MC) algorithm to solve the multiobjective optimal power flow problems with conflicting objectives that reflect the minimization of total production cost, emission pollution, L-index, and active power loss. In this algorithm, a combination of eigenvector and binomial crossovers has been used to move the current population towards better search positions to provide good quality solutions. Besides, an adaptive dynamic parameter adjusting strategy is adopted to obtain the appropriate parameter settings in differential evolution algorithm during the evolution process. Further, an external archive is used to preserve all the nondominated solutions evaluated in each iteration and a fuzzy decision-making technique is applied to extract the best compromise solution from all the nondominated solutions in the archive set. Finally, in order to investigate the usefulness of the proposed algorithm, IEEE 30-bus, IEEE 57-bus and Algerian 59-bus systems with different single and multiobjective OPF problems have been solved and the simulation results are evaluated and compared with the other algorithms recently reported in the literature. The results indicate that the proposed algorithm is competent, effective and quite suitable for solving single/multi objective optimal power flow problems.     

A hybrid ANFIS model based on empirical mode decomposition for stock time series forecasting

Time series forecasting is an important and widely popular topic in the research of system modeling, and stock index forecasting is an important issue in time series forecasting. Accurate stock price forecasting is a challenging task in predicting financial time series. Time series methods have been applied successfully to forecasting models in many domains, including the stock market. Unfortunately, there are 3 major drawbacks of using time series methods for the stock market: (1) some models can not be applied to datasets that do not follow statistical assumptions; (2) most time series models that use stock data with a significant amount of noise involutedly (caused by changes in market conditions and environments) have worse forecasting performance; and (3) the rules that are mined from artificial neural networks (ANNs) are not easily understandable.To address these problems and improve the forecasting performance of time series models, this paper proposes a hybrid time series adaptive network-based fuzzy inference system (ANFIS) model that is centered around empirical mode decomposition (EMD) to forecast stock prices in the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) and Hang Seng Stock Index (HSI). To measure its forecasting performance, the proposed model is compared with Chen's model, Yu's model, the autoregressive (AR) model, the ANFIS model, and the support vector regression (SVR) model. The results show that our model is superior to the other models, based on root mean squared error (RMSE) values.     

基于区间相似度的模糊时间序列预测算法

针对现有模糊时间序列预测算法无法适应预测中新关系出现的问题,提出了一种基于区间相似度的模糊时间序列预测（ISFTS）算法。首先,在模糊理论的基础上,采用基于均值的方法二次划分论域的区间,在论域区间上定义相应模糊集将历史数据模糊化;然后建立三阶模糊逻辑关系并引入逻辑关系相似度的计算公式,计算未来数据变化趋势值得到预测的模糊值;最后对预测模糊值去模糊化得到预测的确定值。由于ISFTS算法是预测数据变化趋势,克服了目前预测算法的逻辑关系的缺陷。仿真实验结果表明,与同类的预测算法相比,ISFTS算法预测误差更小,在误差相对比(MAPE)、绝对误差均值(MAE)和均方根误差(RMSE)三项指标上均优于同类的对比算法,因此ISFTS算法在时间序列预测中尤其是大数据量情况下的预测具有更强的适应性。     

Reducing the run-time complexity of multiobjective EAs: The NSGA-II and other algorithms

The last decade has seen a surge of research activity on multiobjective optimization using evolutionary computation and a number of well performing algorithms have been published. The majority of these algorithms use fitness assignment based on Pareto-domination: Nondominated sorting, dominance counting, or identification of the nondominated solutions. The success of these algorithms indicates that this type of fitness is suitable for multiobjective problems, but so far the use of Pareto-based fitness has lead to program run times in O(GMN/sup 2/), where G is the number of generations, M is the number of objectives, and N is the population size. The N/sup 2/ factor should be reduced if possible, since it leads to long processing times for large population sizes. This paper presents a new and efficient algorithm for nondominated sorting, which can speed up the processing time of some multiobjective evolutionary algorithms (MOEAs) substantially. The new algorithm is incorporated into the nondominated sorting genetic algorithm II (NSGA-II) and reduces the overall run-time complexity of this algorithm to O(GN log/sup M-1/N), much faster than the O(GMN/sup 2/) complexity published by Deb et al. (2002). Experiments demonstrate that the improved version of the algorithm is indeed much faster than the previous one. The paper also points out that multiobjective EAs using fitness based on dominance counting and identification of nondominated solutions can be improved significantly in terms of running time by using efficient algorithms known from computer science instead of inefficient O(MN/sup 2/) algorithms.     

基于改进的NSGA-II多目标优化方法研究*

针对传统多目标优化算法在其领域存在的多个子目标不能同时取优的问题,提出了一种基于改进的非支配排序遗传算法（Non-dominated Sorting Genetic Algorithm-II, NSGA-II）多目标优化方法,以多目标优化遗传算法为基础,多输入多输出的反向传播（Back-Propagation, BP）神经网络为适应度函数评价体系,保证算法快速收敛并搜索到全局最优解集,该算法在建模前对实验数据进行主成分分析,降低了运算时间和算法难度,通过在遗传进化过程中引进正态分布交叉算子（Normal Distribution Crossover, NDX）和改进的自适应调整变异算子,实现了多个目标同时取优,保证Pareto最优解集快速、准确地获取。仿真实验使用UCI数据集,通过与其他常用的多目标优化算法对比,验证了改进的NSGA-II算法精确度更高、收敛速度更快、稳定性更强。     

采用差分进化算法优化模糊Petri网参数

如何有效地确定模糊Petri网（FPN）的各项参数、摆脱自学习能力差的缺点,一直是悬而未决的问题。针对此问题,将差分进化算法首次引入到FPN参数优化中,根据FPN的实际特征,提出了一种改进的差分进化算法。算法采用混沌策略产生初始种群,融合自适应变异因子及早熟惩罚策略提高种群多样性,同时保证很强的收敛性与全局性。仿真实验表明,将改进的差分进化算法与传统算法相比较,收敛到理想参数值的速度提高了5倍。     

Multiobjective differential evolution using homeostasis based mutation for application in software cost estimation

The problem in software cost estimation revolves around accuracy. To improve the accuracy, heuristic/meta-heuristic algorithms have been known to yield better results when it is applied in the domain of software cost estimation. For the sake of accuracy in results, we are still modifying these algorithms. Here we have proposed a new meta-heuristic algorithm based on Differential Evolution (DE) by Homeostasis mutation operator. Software development requires high prediction and low Root Mean Squared Error (RMSE) and mean magnitude relative error(MMRE). The problem in software cost estimation relates to accurate prediction and minimization of RMSE and MMRE, which are used to solve multiobjective optimization. Many versions of DE were proposed, however multi-objective versions where the concept of Pareto optimality is used, are most popular. Pareto-Based Differential Evolution (PBDE) is one of them. Although the performance of this algorithm is very good, its convergence rate can be further improved by minimizing the time complexity of nondominated sorting, and by improving the diversity of solutions. This has been implemented by using efficient nondominated algorithm whose time complexity is better than the previous one and a new mutation scheme is implemented in DE which can provide more diversity among solutions. The proposed variant multiplies the Homeostasis value with one more vector, named the Homeostasis mutation vector, in the existing mutation vector to provide more bandwidth for selecting effective mutant solutions. The proposed approach provides more promising solutions to guide the evolution and helps DE escape the situation of stagnation. The performance of the proposed algorithm is evaluated on twelve benchmark test functions (bi-objective and tri-objective) on the Pareto-optimal front. The performance of the proposed algorithm is compared with other state-of-the-art algorithms on five multi-objective evolutionary algorithms (MOEAs). The result verifies that our proposed Homeostasis mutation strategy performs better than other state-of-the-art algorithms. Finally, application of MODE-HBM is applied to solve in terms of Pareto front, representing the trade-off between development RMSE, MMRE, and prediction for COCOMO model.     

A novel hybrid multi-objective immune algorithm with adaptive differential evolution

In this paper, we propose a novel hybrid multi-objective immune algorithm with adaptive differential evolution, named ADE-MOIA, in which the introduction of differential evolution (DE) into multi-objective immune algorithm (MOIA) combines their respective advantages and thus enhances the robustness to solve various kinds of MOPs. In ADE-MOIA, in order to effectively cooperate DE with MOIA, we present a novel adaptive DE operator, which includes a suitable parent selection strategy and a novel adaptive parameter control approach. When performing DE operation, two parents are respectively picked from the current evolved and dominated population in order to provide a correct evolutionary direction. Moreover, based on the evolutionary progress and the success rate of offspring, the crossover rate and scaling factor in DE operator are adaptively varied for each individual. The proposed adaptive DE operator is able to improve both of the convergence speed and population diversity, which are validated by the experimental studies. When comparing ADE-MOIA with several nature-inspired heuristic algorithms, such as NSGA-II, SPEA2, AbYSS, MOEA/D-DE, MIMO and D²MOPSO, simulations show that ADE-MOIA performs better on most of 21 well-known benchmark problems.     

一种求解旅行商问题的进化多目标优化方法

为了克服传统小生境(Niching)策略中的参数设置难题,提出一种求解旅行商问题的进化多目标优化方法:建立以路径长度和平均离群距离为目标的双目标优化模型,利用改进非支配排序遗传算法(NSGAII)进行求解.为了在全局探索能力与局部开发能力之间保持平衡,算法中采用一种使路径长度相同的可行解互不占优的评价策略,并通过一种新的离散差分进化算子和简化的2-Opt策略生成候选解.与已有算法的数值试验结果比较表明,求解旅行商问题(TSP)的改进非支配排序遗传算法(NSGAII-TSP)能够更好地保持种群多样性,从而克服局部最优解的吸引并具有更鲁棒的全局探索能力.通过借助特殊的个体评价策略,所提出的算法可以更好地进行全局优化,甚至同时得到多个全局最优解.     

Multi-objective differential evolution with performance-metric-based self-adaptive mutation operator for chemical and biochemical dynamic optimization problems

Each mutation operator of differential evolution (DE) algorithm is generally suitable for certain specific types of multi-objective optimization problems (MOPs) or particular stages of the evolution. To automatically select an appropriate mutation operator for solving MOPs in different phases of the evolution, a multi-objective differential evolution with performance-metric-based self-adaptive mutation operator (MODE-PMSMO) is proposed in this study. In MODE-PMSMO, a modified inverted generational distance (IGD) is utilized to evaluate the performance of each mutation operator and guide the evolution of mutation operators. The proposed MODE-PMSMO is then compared with seven multi-objective evolutionary algorithms (MOEAs) on five bi-objective and five tri-objective optimization problems. Generally, MODE-PMSMO exhibits the best average performance among all compared algorithms on ten MOPs. Additionally, MODE-PMSMO is employed to solve four typical multi-objective dynamic optimization problems in chemical and biochemical processes. Experimental results indicate that MODE-PMSMO is suitable for solving these actual problems and can provide a set of nondominated solutions for references of decision makers.     

基于分解和差分进化的多目标粒子群优化算法

为了提高多目标优化算法解集的分布性和收敛性,提出一种基于分解和差分进化的多目标粒子群优化算法(dMOPSO-DE).该算法通过提出方向角产生一组均匀的方向向量,确保粒子分布的均匀性;引入隐式精英保持策略和差分进化修正机制选择全局最优粒子,避免种群陷入局部最优Pareto前沿;采用粒子重置策略保证群体的多样性.与非支配排序(NSGA-II)算法、多目标粒子群优化(MOPSO)算法、分解多目标粒子群优化(dMOPSO)算法和分解多目标进化-差分进化(MOEA/D-DE)算法进行比较,实验结果表明,所提出算法在求解多目标优化问题时具有良好的收敛性和多样性.     

一种基于进化方向的新的差分进化算法

变异策略对差分进化算法（DE）算法的成功与否起到至关重要的作用.然而,方向信息在DE变异策略的设计当中并没有被充分地挖掘,且对于如何平衡进化速度和种群多样性这两者之间的矛盾也没有得到很好的解决方案.研究了个体在进化选择操作前后产生的差量信息在变异操作上的导向作用,提出了一种新的基于进化方向的变异策略“DE/current-to-pbest/1/Gvector”.同时,为了测试我们这种新的方向信息能否提高算法的优化能力,我们在自适应差分进化算法（JADE）的基础上提出了一种新的算法DVDE.对CEC2005常用的12个测试函数做了仿真实验,实验结果证明DVDE的算法性能平均优于其他5个目前来说性能最好的DE算法（JADE,SaDE,CoDE,jDE,EPSDE）,特别是对于单峰函数,效果更为明显.实验结果也说明进化方向的加入对于提高算法的收敛速度以及保护种群的多样性避免算法过早陷入局部最优起到了较好的作用.     

基于正交差分演化无迹卡尔曼滤波的短时交通流量预测算法

针对复杂交通路段下的短时交通流量模型的参数估计问题,建立了基于宏观交通流量预测的状态空间模型,提出了基于正交自适应差分演化的无迹卡尔曼滤波(UKF)算法,解决交通流量预测动态模型的参数优化问题.对差分演化算法(DE)的初始化过程,使用基于正交设计和量化技术的交叉算子最大限度地提高种群的多样性,平衡差分演化算法的开采性和勘探性,更高效地搜索无迹卡尔曼滤波的模型参数.并针对UKF、DE的不同情况,分别采用不同的自适应策略提高调节算法性能.实验结果表明,相对于单独使用随机分布的方式初始化,或者根据经验设置模型参数的方法,使用正交设计方法的初始化策略、变异算子以及参数自适应控制策略的差分演化算法能够有效地节省计算资源,提升预测性能和精度,具有更高的鲁棒性.