ID 原文 译文
10614 其次,通过构建各响应变量在信号因子不同水平下的联合位置,尺度与偏度的响应曲面模型,进而建立基于多元偏正态分布的期望损失函数; Secondly, this paper builds the response variables in the signal factor under different levels of joint position, scale and partial degrees of response surface model, which based on multivariate skewed normal distribution of expected loss function;
10615 然后,利用混合遗传算法对所构建的综合期望损失函数进行全局优化求解; Then, by using hybrid genetic algorithm to construct comprehensive global optimization to solve the expected loss function;
10616 最后,通过对具体的工业实例进行分析研究,结果表明本文所提出的方法能够有效地解决具有偏度特征的动态多响应稳健参数设计问题。 Finally, through the analysis of specific instances of industrial research, the results show that the presented method can effectively solve the dynamic characteristic of partial degrees response more robust parameter design problems.
10617 通过采用粒子群算法(particle swarm optimization,PSO),求解出不等质量4星库仑卫星编队在平面和立体情况下的最优静态构型。 By adopting the particle swarm algorithm (particle swarm optimization, PSO), to solve the unequal quality 4 star coulomb satellite formation under the condition of plane and solid is the optimal static configurations.
10618 当库仑卫星编队中各卫星的质量、带电量以及位置均为变量时,库仑卫星编队的静态构型模型具有很强的非线性特点。 ‭When the coulomb satellite formation of each satellite in quality, with power and position are variable, the coulomb static configuration of satellite formation model has strong non-linear characteristics.
10619 此时可以将非线性模型的求解问题转化为约束条件下的参数优化问题,通过PSO求得各种编队构型下的最优静态构型。 ‭At this point to solve the problem of nonlinear model can be transformed to parameter optimization problem under the constraint condition, optimum under various formation configuration by PSO static configurations.
10620 同时,所提方法可以应用于求解PSO复杂多星编队情况下的最优静态构型,并为卫星队形重构问题等提供理论基础。 ‭At the same time, the proposed method can be applied to solve the PSO star formation conditions of the optimal more complex than static configuration, and provide theoretical basis for satellite formation reconfiguration problem, etc. ‭
10621 针对复杂网络中的社区检测问题,提出了一种基于节点影响力的离散粒子群社区检测方法。 For community detection problem in complex networks, this paper proposes a discrete particle swarm community detection method based on node influence.
10622 该方法以模块度密度作为目标函数,利用离散粒子群算法对其进行优化,在优化过程中提出了节点影响力的概念,其充分利用了网络中节点的相互关系检测网络中的社区结构。 Module density as objective function, the method using the discrete particle swarm algorithm to optimize, and puts forward the concept of node influence in the process of optimization, the full use of the network node correlation detection in a network of community structure.
10623 同时,在此基础上提出了基于节点影响力的粒子群初始化方法和粒子状态更新方法。 At the same time, based on the proposed particle swarm initialization method based on node power method and particle state update.