多种群协作差异演化算法及其应用

针对差异演化算法在求解高维、多峰问题时易陷入局部最优解发生"早熟"的缺陷,提出多种群协作差异演化算法。该算法将种群划分为多个子种群并设置一个精英群,精英群由子种群的最优个体产生,并与群体最优解进行竞争,子种群的最优个体只与精英群个体竞争而不与群体最优解竞争。构建了两种不同邻域结构的多种群协作差异演化算法,仿真结果及工程应用实例表明了算法的有效性。     

自适应差异演化算法及其应用

差异演化(Differential Evolution,DE)算法是一种基于群体差异的演化算法,具有良好的优化性能,但是对于高维复杂函数,DE算法易早熟收敛。为此,在对DE算法参数分析的基础上,提出自适应缩放因子及自适应交叉率两个概念,进而提出一种自适应差异演化(Adaptive Differential Evolution,ADE)算法。利用群体差异度对DE算法进行分期,一方面使缩放因子在前期较大,在进化的中期先变小,后增大,在进化的后期,缩放因子较小;另一方面使DE算法的交叉率在前期较小,中期在一定范围内随机取值,进化后期较大。仿真实验结果与工程应用实例表明,ADE算法在收敛速度和全局搜索能力方面得到了较好的平衡,不仅保证了ADE算法的收敛速度,而且具有较好的全局搜索能力。     

带自适应动态变异和二次变异的差分进化算法

为解决差分进化算法在解决多目标优化问题时的多样性与收敛性之间的平衡维持难题,首先提出了一种基于自适应动态变异和非支配解二次变异的改进差分进化算法。该算法的核心是将前N代进化的群体多样性值作为多样性判别准则,根据群体多样性变化情况自适应地选择对应的变异算子产生新个体;其次提出通过对所存档Pareto非支配解进行二次变异来增加新个体解群的优解质量和数量,以同时改进算法的多样性和收敛速度。仿真结果表明,与标准差分进化算法和改进的基于分类排序的Pareto遗传算法相比,所提算法在收敛性、分布性与分散度性能指标上都有较好的表现,多样性和收敛性之间的平衡维持能力则远优于另两种算法。     

混合变量非线性规划问题的差异演化算法研究

针对机械优化设计问题中的混合变量非线性规划问题,提出一种改进的差异演化算法.首先针对混合变量非线性规划问题的特点,采用离散变量与连续变量分别进化策略,以使差异演化算法能够适于解决这类问题.其次,依据基因突变原理,提出突变因子的概念,使群体中的一部分个体不经过交叉操作,而直接与父代竞争,提高了群体的多样性,增强了算法的全局搜索能力,并对差异演化算法的缩放因子进行线性递减计算,加快了算法的收敛速度.测试结果表明,所提出算法不仅能够解决混合变量非线性规划问题,而且有较快的收敛速度和较好的全局收敛能力,实用价值明显.     

Evolutionary sensitivity-based conceptual design and tolerance allocation for mechanical assemblies

This paper presents a new method based on evolutionary algorithms—Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II) and Differential Evolution (DE)—for the sensitivity-based conceptual design and the tolerance allocation for mechanical assemblies. The approach I of this paper moves the nominal values of non-critical dimensions to a less sensitive portion, and the approach II deals with the allocation of cost-based optimal tolerances. An improved optimization model that considers three objective functions (minimization of the deviation of critical dimensions, manufacturing cost, and the quality loss), eight constraints, and six variables is proposed. To show the effectiveness of the proposed methods, the stepped bar assembly is considered as a numerical example. The results obtained from NSGA-II and DE are compared and analyzed. The results show that the proposed methods are much effective, cost, and time saving than the ones considered in literature.     

简单差异演化算法及其在装配公差优化中的应用

差异演化算法是一种基于群体差异的演化算法,具有良好的优化性能,但是对于高维复杂函数,算法易早熟收敛.为此,在对算法参数以及关键算子分析的基础上,提出了自适应缩放因子及突变因子两个概念,进而提出了简单差异演化算法(A Simple Differential Evolution Algorithm)SDE.首先将缩放因子按照进化代数进行递减,一方面为了减少用户参与程度,另一方面为了平衡算法的收敛速度与全局搜索能力;其次在研究交叉算子的基础上,引入了灾变因子,使群体中的部分个体在进化过程中不进行交叉操作,而直接与父代个体进行竞争,简化了差异演化算法的步骤.仿真实验结果与工程应用实例表明,SDE算法在收敛速度和全局搜索能力方面得到了较好的平衡,不仅保证了算法的收敛速度,而且具有较好的全局搜索能力.     

Design and analysis of tilt integral derivative controller with filter for load frequency control of multi-area interconnected power systems

In this paper, a novel Tilt Integral Derivative controller with Filter (TIDF) is proposed for Load Frequency Control (LFC) of multi-area power systems. Initially, a two-area power system is considered and the parameters of the TIDF controller are optimized using Differential Evolution (DE) algorithm employing an Integral of Time multiplied Absolute Error (ITAE) criterion. The superiority of the proposed approach is demonstrated by comparing the results with some recently published heuristic approaches such as Firefly Algorithm (FA), Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) optimized PID controllers for the same interconnected power system. Investigations reveal that proposed TIDF controllers provide better dynamic response compared to PID controller in terms of minimum undershoots and settling times of frequency as well as tie-line power deviations following a disturbance. The proposed approach is also extended to two widely used three area test systems considering nonlinearities such as Generation Rate Constraint (GRC) and Governor Dead Band (GDB). To improve the performance of the system, a Thyristor Controlled Series Compensator (TCSC) is also considered and the performance of TIDF controller in presence of TCSC is investigated. It is observed that system performance improves with the inclusion of TCSC. Finally, sensitivity analysis is carried out to test the robustness of the proposed controller by varying the system parameters, operating condition and load pattern. It is observed that the proposed controllers are robust and perform satisfactorily with variations in operating condition, system parameters and load pattern.     

运用DE-RBFNN的直线伺服系统定位精度研究

提出一种基于差分进化算法(DE)的径向基函数神经网络(RBFNN)模型,用于预测直线伺服系统的定位误差.该模型用差分进化算法训练径向基函数(RBF)网络隐层中心位置、宽度和输出层连接权重.为了评价优化后RBF网络预测的精度,运用部分误差样本进行训练和仿真.构建了以数字信号处理器(DSP)为核心的直线电动机定位误差实验平台,根据误差校正值进行误差实时补偿实验.仿真和实验结果表明:经过DE算法训练的神经网络模型对工作台的误差具有良好的学习能力和泛化能力,与单纯RBF网络、基于遗传优化的RBF神经网络相比,该建模方法具有更高的定位精度.     

Predictive optimized adaptive PSS in a single machine infinite bus

Power System Stabilizer (PSS) devices are responsible for providing a damping torque component to generators for reducing fluctuations in the system caused by small perturbations. A Predictive Optimized Adaptive PSS (POA-PSS) to improve the oscillations in a Single Machine Infinite Bus (SMIB) power system is discussed in this paper. POA-PSS provides the optimal design parameters for the classic PSS using an optimization predictive algorithm, which adapts to changes in the inputs of the system. This approach is part of small signal stability analysis, which uses equations in an incremental form around an operating point. Simulation studies on the SMIB power system illustrate that the proposed POA-PSS approach has better performance than the classical PSS. In addition, the effort in the control action of the POA-PSS is much less than that of other approaches considered for comparison.     

Improving power system damping using a combination of optimal control theory and differential evolution algorithm

This paper presents a novel control strategy to improve the damping capability of sub-synchronous oscillations by tuning of Linear Quadratic Regulator (LQR) optimally in order to reduce the fluctuations in the power system. The proposed model includes the coordination of Power System Stabilizer (PSS) and Thyristor-Controlled Series Capacitors (TCSC) in combination with LQR controllers which is formulated as an optimal control problem. The problem is formulated as a linear regulator problem and then the Differential Evolution (DE) algorithm is utilized to optimize the proposed controlling parameters. Several cases have been studied to show the efficiency of the proposed strategy. Obtained results from a case study on a typical generator demonstrated that the proposed method has the best response and quickest function among conventional controller systems. Moreover, the coordination of the LQR with the other control systems, as an optimal linear regulator problem in order to damp system oscillations provides robust stability for optimizing system performance index. Besides, the flexibility and usability of the LQR controller guarantee the stability of the system to cope with the oscillations.     

复杂曲面测量数据的位姿配准方法

针对三维测量数据和自由曲面模型之间的位姿配准问题,研究了先粗后精的两步配准方法。在初始配准的基础上,融合最小二乘法和最小条件原则构造目标函数,应用微分进化算法对目标函数寻优,找出三维测量数据与理论曲面的最佳匹配矩阵以实现最优配准。实验结果表明,该方法与遗传算法相比具有运算速度快和精度高等特点,能较好的解决复杂曲面类零件测量数据的位姿配准问题,并且可用于逆向工程中曲面误差的分析及修正。     

永磁同步风力发电系统最大功率追踪新方法

风力发电系统的输出功率随着风速变化而变化,引起输出功率的波动。因此提出一种采用永磁同步电机作为发电机实现最大功率输出的永磁同步风力发电控制系统。建立了风力机模型、最大功率追踪算法模型,采用二分法和停止机制算法,使风机按最大功率点跟踪方式运行,通过二分法实现变步长,并加入停止机制降低转速振荡,防止在最大功率点的波动。介绍了传统爬山法和改进的爬山法,通过对比仿真研究,表明新方法可以有效提高最大功率追踪效率,并减少系统在最大功率点的波动。     

基于微分进化算法的双波长光纤F-P传感系统结构优化

针对光纤法布里-珀罗(F-P)传感器在实际应用中易偏离工作点,造成输出信号衰减,信噪比降低问题,提出了双波长稳定技术,建立了双波长稳定系统的数学优化模型.采用全局搜索能力强,收敛速度快,鲁棒性好,能有效解决复杂优化问题的微分进化算法(DE)进行了全局结构优化.结果表明,DE算法能在较短的运行时间内得到全局最优解,误差＜10-3,优化目标下降了89.46%,且运行时间有所减少;证明该算法正确可行,高效可靠,为光纤F-P传感器的结构优化设计提供了实验基础.     

Cat Swarm Optimization algorithm for optimal linear phase FIR filter design

In this paper a new meta-heuristic search method, called Cat Swarm Optimization (CSO) algorithm is applied to determine the best optimal impulse response coefficients of FIR low pass, high pass, band pass and band stop filters, trying to meet the respective ideal frequency response characteristics. CSO is generated by observing the behaviour of cats and composed of two sub-models. In CSO, one can decide how many cats are used in the iteration. Every cat has its′ own position composed of M dimensions, velocities for each dimension, a fitness value which represents the accommodation of the cat to the fitness function, and a flag to identify whether the cat is in seeking mode or tracing mode. The final solution would be the best position of one of the cats. CSO keeps the best solution until it reaches the end of the iteration. The results of the proposed CSO based approach have been compared to those of other well-known optimization methods such as Real Coded Genetic Algorithm (RGA), standard Particle Swarm Optimization (PSO) and Differential Evolution (DE). The CSO based results confirm the superiority of the proposed CSO for solving FIR filter design problems. The performances of the CSO based designed FIR filters have proven to be superior as compared to those obtained by RGA, conventional PSO and DE. The simulation results also demonstrate that the CSO is the best optimizer among other relevant techniques, not only in the convergence speed but also in the optimal performances of the designed filters.     

Multi-objective Optimization of a Parallel Ankle Rehabilitation Robot Using Modified Differential Evolution Algorithm

Dimensional synthesis is one of the most difficult issues in the field of parallel robots with actuation redundancy. To deal with the optimal design of a redundantly actuated parallel robot used for an...     

模糊制造系统中的不同尺寸工件单机批调度优化

将工件尺寸不同的单机批调度问题扩展到模糊制造系统中,建立了基于模糊批加工时间和模糊批间隔时间的制造跨度模型,提出了一种集成粒子群优化和差异演化的混合算法,将制造跨度最小化.为提高算法的收敛速度,设计了基于工件优先值向量的统一编码方式,并采用线性的缩放因子以确保足够的差异化信息;为解决差异演化算法早熟收敛的问题,将粒子群优化的全局搜索技术嵌入了差异演化算法;最后,在解码时利用批调度的启发式算法,将混合算法的个体加以优化分批.仿真实验结果验证了该混合算法的求解性能优于目前文献中的其他算法.     

工件圆度误差测量不确定度评定

为了实现工件圆度误差的不确定度评定,对基于三坐标测量机的工件圆度轮廓数据的采样策略、圆度评定方法及不确定度评定方法进行研究。首先,根据工件圆度轮廓特征进行实验测量,获取不同工件的多个样本。接着,基于最小二乘法和微分进化优化算法对样本的圆度误差进行了误差评定。然后,在分析比较误差大小的基础上,说明了采用的采样策略和微分进化评定算法。最后,基于圆度误差评定结果运用了测量不确定度表示指南(GUM)和蒙特卡洛方法(MCM)进行不确定度评定。实验结果表明:微分进化算法与最小二乘法相比均值差最大达到1.1μm, MCM方法比GUM方法得到的标准不确定度均值小0.02μm。合理的采样点数、微分进化算法及MCM不确定度评定方法可以得到更稳定可靠、精度高的评定结果。     

多输入多输出Hammerstein Wiener交流电弧炉电极系统模型

在交流电弧炉中对于电极系统的描述,目前大都采用针对单相电极的单输入单输出的Hammerstein-Wiener(H-W)模型,这种模型过于简化真实电极系统结构,导致模型的预测精度较低。针对该问题,提出一种基于多输入多输出H-W模型的电极系统建模方法,该模型的结构与实际电极系统结构一致,有利于模型预测精度的提高,另外在多输入多输出的静态非线性块不可逆的条件下,提出可分非线性最小二乘算法对H-W模型参数进行辨识。最后采用实际数据验证,在预测精度上,多输入多输出H-W电极系统模型优于传统的单输入单输出H-W电极系统模型。     

A road-adaptive control law for semi-active suspensions

This paper presents a road-adaptive control law for semi-active vehicle suspensions. In semi-active suspensions, damping coefficients are controlled so as to make the actual damper force as close to the desired damper force as possible at any time instance. The proposed control law consists of a road-adaptive sky-hook damping algorithm and a Road Detection Algorithm (RDA). This approach leads to the sprung mass and unsprung mass velocity feedback control law with time varying gains. The gains are tuned by the RDA. To evaluate the performance enhancement brought about by the proposed control law, the performance of a semi-active suspension with the proposed control law is compared to those of the sky-hook controlled semi-active suspension and a passive suspension. The controller has been implemented experimentally on a quarter car test rig and a semi-active damper with a 19 damping rates has been used to generate the desired semi-active force. The proposed control law provides adequate damping for the wheel hop frequency and improved performance compared to that of the sky-hook control law.     

Integrated optimal design of flexible mechanism and vibration control

A method of simultaneous optimal design of flexible mechanisms and control with multiple frequency constraints based on the independent modal space control theory is proposed. Firstly, the performance index for simultaneous mechanism and control design is discussed. Secondly, an integrated optimal design algorithm is developed on the basis of Kuhn–Tucker optimality criteria, in which three damping factors are introduced to guarantee the algorithm that possesses good stability and convergence. Finally, numerical simulations are carried out on a path generating mechanism. Simulation results show that the performance of the control input is improved significantly after integrated optimization.