Ontology alignment is an essential and complex task to integrate heterogeneous ontology.The meta-heuristic algorithm has proven to be an effective method for ontology alignment.However,it only applies the inherent adv...Ontology alignment is an essential and complex task to integrate heterogeneous ontology.The meta-heuristic algorithm has proven to be an effective method for ontology alignment.However,it only applies the inherent advantages of metaheuristics algorithm and rarely considers the execution efficiency,especially the multi-objective ontology alignment model.The performance of such multi-objective optimization models mostly depends on the well-distributed and the fast-converged set of solutions in real-world applications.In this paper,two multi-objective grasshopper optimization algorithms(MOGOA)are proposed to enhance ontology alignment.One isε-dominance concept based GOA(EMO-GOA)and the other is fast Non-dominated Sorting based GOA(NS-MOGOA).The performance of the two methods to align the ontology is evaluated by using the benchmark dataset.The results demonstrate that the proposed EMO-GOA and NSMOGOA improve the quality of ontology alignment and reduce the running time compared with other well-known metaheuristic and the state-of-the-art ontology alignment methods.展开更多
实数编码的多目标进化算法常使用模拟二进制交叉(simulated binary crossover,称SBX)算子.通过对SBX以及进化策略中变异算子进行对比分析,并引入进化策略中的离散重组算子,提出了一种正态分布交叉(normal distribution crossover,称NDX...实数编码的多目标进化算法常使用模拟二进制交叉(simulated binary crossover,称SBX)算子.通过对SBX以及进化策略中变异算子进行对比分析,并引入进化策略中的离散重组算子,提出了一种正态分布交叉(normal distribution crossover,称NDX)算子.首先在一维搜索空间实例中对NDX与SBX算子进行比较和分析,然后将NDX算子应用于Deb等人提出的稳态多目标进化算法ε-MOEA(ε-dominance based multiobjective evolutionary algorithm)中.采用NDX算子的ε-MOEA(记为ε-MOEA/NDX)算法在多目标优化标准测试集ZDT和DTLZ的10个函数上进行了实验比较.实验结果和分析表明,采用NDX的ε-MOEA所求得的Pareto最优解集质量明显优于经典算法ε-MOEA/SBX和NSGA-Ⅱ.展开更多
针对多个风电机组接入配电网带来的不确定性问题,采用基于拉丁超立方采样的Monte Carlo概率潮流计算方法(correlation Latin hypercube sampling Monte Carlo simulation,CLMCS)以及场景缩减技术得到风机组输出功率的典型场景,将不确定...针对多个风电机组接入配电网带来的不确定性问题,采用基于拉丁超立方采样的Monte Carlo概率潮流计算方法(correlation Latin hypercube sampling Monte Carlo simulation,CLMCS)以及场景缩减技术得到风机组输出功率的典型场景,将不确定性问题转化为单场景确定性潮流问题。并建立以有功网损最小、电压偏差最小作为目标函数的配电网无功优化数学模型。采用e正交多目标差分进化算法(e-orthogonal differential evolution multi-objective algorithm,e-ODEMO)进行计算得到非劣解集,该算法基于一般差分演化算法,结合正交实验方法使初始个体均匀分布在决策变量空间,利用e占优技术对Archive群体进行更新,能得到均匀分布的非劣解集。应用IEEE 33节点以及PG&E 69节点配电网系统进行了测试,结果验证了所提方法和模型的可行性与有效性。展开更多
基金the Ministry of Education-China Mobile Joint Fund Project(MCM2020J01)。
文摘Ontology alignment is an essential and complex task to integrate heterogeneous ontology.The meta-heuristic algorithm has proven to be an effective method for ontology alignment.However,it only applies the inherent advantages of metaheuristics algorithm and rarely considers the execution efficiency,especially the multi-objective ontology alignment model.The performance of such multi-objective optimization models mostly depends on the well-distributed and the fast-converged set of solutions in real-world applications.In this paper,two multi-objective grasshopper optimization algorithms(MOGOA)are proposed to enhance ontology alignment.One isε-dominance concept based GOA(EMO-GOA)and the other is fast Non-dominated Sorting based GOA(NS-MOGOA).The performance of the two methods to align the ontology is evaluated by using the benchmark dataset.The results demonstrate that the proposed EMO-GOA and NSMOGOA improve the quality of ontology alignment and reduce the running time compared with other well-known metaheuristic and the state-of-the-art ontology alignment methods.
文摘实数编码的多目标进化算法常使用模拟二进制交叉(simulated binary crossover,称SBX)算子.通过对SBX以及进化策略中变异算子进行对比分析,并引入进化策略中的离散重组算子,提出了一种正态分布交叉(normal distribution crossover,称NDX)算子.首先在一维搜索空间实例中对NDX与SBX算子进行比较和分析,然后将NDX算子应用于Deb等人提出的稳态多目标进化算法ε-MOEA(ε-dominance based multiobjective evolutionary algorithm)中.采用NDX算子的ε-MOEA(记为ε-MOEA/NDX)算法在多目标优化标准测试集ZDT和DTLZ的10个函数上进行了实验比较.实验结果和分析表明,采用NDX的ε-MOEA所求得的Pareto最优解集质量明显优于经典算法ε-MOEA/SBX和NSGA-Ⅱ.
文摘针对多个风电机组接入配电网带来的不确定性问题,采用基于拉丁超立方采样的Monte Carlo概率潮流计算方法(correlation Latin hypercube sampling Monte Carlo simulation,CLMCS)以及场景缩减技术得到风机组输出功率的典型场景,将不确定性问题转化为单场景确定性潮流问题。并建立以有功网损最小、电压偏差最小作为目标函数的配电网无功优化数学模型。采用e正交多目标差分进化算法(e-orthogonal differential evolution multi-objective algorithm,e-ODEMO)进行计算得到非劣解集,该算法基于一般差分演化算法,结合正交实验方法使初始个体均匀分布在决策变量空间,利用e占优技术对Archive群体进行更新,能得到均匀分布的非劣解集。应用IEEE 33节点以及PG&E 69节点配电网系统进行了测试,结果验证了所提方法和模型的可行性与有效性。
