核动力装置蒸汽发生器水位的分层模糊自适应控制

针对压水堆核动力装置蒸汽发生器的水位控制提出一种分层模糊自适应控制方案，该方案中2个模糊控制器分层连接，每个模糊控制均采用典型模糊控制单元，使得模糊规则个数和可调参数个数大大减少，便于在线学习和实时控制，给出了分层模糊控制器的解析表达式及可调参数的在线学习方法，在快速加负荷和突然甩负荷的仿真实验中，该方案的控制效果明显优于已有的变参数PID控制，验证了该方案的有效性。     

一种基于遗传算法的模糊神经网络最优控制

通过对控制系统的过程模拟,提出一种模糊神经网络最优控制方案。离线化部分基于遗传算法,分三阶段实现模糊神经网络控制器结构和参数的优化。在线优化部分通过重构模糊神经网络控制器的去模糊化部分,进一步调整控制规则,实现在线去模糊优化。仿真结果表明该方案优于常模糊控制方案和基于专家经验的模糊神经网络控制方案。     

基于改进遗传算法的模糊控制器设计

针对模糊控制器的隶属度函数和模糊控制规则的选取及优化缺乏自学习能力与知识采集的手段,以及遗传算法具有自适应、启发式、概率性、迭代式全局收敛的特点,该文章将遗传算法与模糊控制相结合,给出了一种基于改进遗传算法的模糊控制器设计策略.改进算法引入了分裂算子来避免遗传算法在寻优过程中陷入局部最优解,同时对编码方式、选择算子、交叉算子以及变异算子做了相应的调整与改进.并将此改进算法用于优化模糊控制器的隶属度函数与模糊控制规则.仿真结果表明用该改进算法优化后的模糊控制器较用普通遗传算法优化后的模糊控制器具有更好的控制性能.     

基于遗传算法的机器人分层模糊控制

为解决模糊多变量控制中规则数随系统变量数呈指数增长的问题,针对机器人轨迹跟踪的特点,提出了一种分层模糊控制器的设计方法。该方法不仅减少了模糊规则数,而且使模糊控制逻辑变得清晰明了,其中控制参数采用遗传算法整定。实验结果证实:该方法控制效果好、系统跟踪速度快。     

采用遗传优化自适应模糊PID控制球杆系统

为解决球杆系统动态、静态性能不高的问题,提出了遗传算法优化自适应模糊PID控制器的控制方法.该模型在拉格朗日方程建立球杆系统数学模型的基础上,采用遗传算法优化模糊控制规则、隶属函数和自适应PID参数.在GBB1004系统中建立了遗传算法优化后的自适应模糊PID控制器以及控制模型,并对该控制器进行实验验证.实验结果证明了遗传算法优化后的模糊控制器有效地减小了系统的超调量,缩短了系统的调节时间,能够较好地控制球杆系统.     

一种基于遗传算法的模糊控制规则的优化方法

模糊控制规则的选择是模糊控制器设计的关键问题之一,文中在对现有应用遗传算法优化模糊控制规则的方法进行研究的基础上,以模糊控制规则的完整性和一致性为出发点,提出了一种用遗传算法来优化模糊控制规则的改进算法,具体给出了遗传算法设计中的各种函数和算子的确定,并将优化过的规则用于设计模糊控制器,进行仿真研究,取得了令人满意的效果。     

基于遗传模糊控制的多机器人避碰规划

设计了一种基于遗传算法优化的模糊逻辑控制的多机器人避碰规划方法,采用简化的三层行为结构:躲避机器人、躲避静态障碍物和趋向目标点,三个行为分别独立推理,将不同传感器信息作为输入,机器人动作作为输出,再通过优先级和加权的方法对三个行为输出进行综合.随后,针对模糊控制中构造全部的模糊规则比较复杂的问题,采用遗传算法对模糊规则的隶属度函数宽度和中心值进行优化,实现模糊控制器的离线自寻优,得到一组最优参数.从最终的仿真效果看,通过遗传算法优化提高了机器人的自导航性能.     

模糊控制规则优化方法研究

模糊控制规则的选择是模糊控制器设计的关键问题之一,在现有应用遗传算法优化模糊控制规则的方法进行研究的基础上,以模糊控制规则的完整性和一致性为出发点,提出了一种用遗传算法来优化模糊控制规则的改进算法,具体给出了遗传算法设计中的各种函数和算子的确定,并将优化过的规则用于设计模糊控制器,进行仿真研究,取得了令人满意的效果。     

氧乐果合成反应温度的模糊自学习控制方法

针对氧乐果合成反应温度控制具有参数时变、时滞后特性而不易控制的问题,提出一种基于遗传算法的模糊自学习控制方法。还提出了改进遗传算法的若干策略。改进的遗传算法仅根据在线获得的信息便能够实现控制器的多个加权因子的快速全局寻优,从而实现模糊控制规则的修正与完善。仿真结果表明,基于遗传算法设计的自学习模糊控制器能适应参数时变、时滞后系统的控制。     

基于模糊控制规则的建筑物结构主动控制的研究

针对地震中建筑物结构主动控制的问题,引入模糊控制规则。该方法采用分层结构,其中包括一个高层的控制器和多个为了降低层间位移的底层子控制器。模糊控制规则能够恰当地调节预先控制在每一时刻所估计的结构状态。改进的地震控制性能通过模糊调节过程将一个简单设计的静态增益转换为实时动态增益。本文在控制器的设计部分充分考虑作动器饱和的状态,并以三层剪切型建筑物结构模型来举例说明。最后模糊规则的运用和Matlab仿真证明这种控制策略的正确性和有效性。     

Hierarchical Fuzzy Control for C-Axis of CNC Turning Centers Using Genetic Algorithms

A combined PD and hierarchical fuzzy control is proposed for the low-speed control of the C-axis of CNC turning centers considering the effects of transmission flexibility and complex nonlinear friction. Learning of the hierarchical structure and parameters of the suggested control strategy is carried out by using the genetic algorithms. The proposed algorithm consists of two phases: the first one is to search the best hierarchy, and the second to tune the consequent center values of the constituent fuzzy logic systems into the hierarchy. For the least total control rule number, the hierarchical fuzzy controller is chosen to include only the simple two-input/one-output fuzzy systems, and both binary and decimal genes are used for the selection, crossover and mutation of the genetic algorithm. The proposed approach is validated by the computer simulation. Each generation consists of 30 individuals: ten reproduced from its parent generation, ten generated by crossover, and the other ten by mutation. In the simulations, the C-axis is assumed to be driven by a vector-controlled AC induction motor, and the dynamic friction model suggested by Canudas de Wit et al. in 1995 is used.     

Adaptive hierarchical tuning of fuzzy controllers

Fuzzy controller design includes both linear and non-linear dynamic analysis. The knowledge base parameters associated within the fuzzy rule base influence the non-linear control dynamics while the linear parameters associated within the fuzzy output signal influence the overall control dynamics. For distinct identification of tuning levels, an equivalent linear controller output and a normalized non-linear controller output are defined. A linear proportional-integral-derivative (PID) controller analogy is used for determining the linear tuning parameters. Non-linear tuning is derived from the locally defined control properties in the non-linear fuzzy output. The non-linearity in the fuzzy output is then represented in a graphical form for achieving the necessary non-linear tuning. Three different tuning strategies are evaluated. The first strategy uses a genetic algorithm to simultaneously tune both linear and non-linear parameters. In the second strategy the non-linear parameters are initially selected on the basis of some desired non-linear control characteristics and the linear tuning is then performed using a trial and error approach. In the third method the linear tuning is initially performed off-line using an existing linear PID law and an adaptive non-linear tuning is then performed online in a hierarchical fashion. The control performance of each design is compared against its corresponding linear PID system. The controllers based on the first two design methods show superior performance when they are implemented on the estimated process system. However, in the presence of process uncertainties and external disturbances these controllers fail to perform any better than linear controllers. In the hierarchical control architecture, the non-linear fuzzy control method adapts to process uncertainties and disturbances to produce superior performance.     

神经网络工况识别的混合动力电动汽车模糊控制策略

采用模糊控制可以改进混合动力电动汽车(HEV)的燃油经济性和排放性,但是对模糊控制器进行优化时通常只针对某一典型工况.不同的城市的行驶工况有一定差别,影响了模糊控制改善混合动力电动汽车性能的效果.研究中以广州和上海市主干道行驶工况为例,首先建立了一个模糊控制策略,并采用遗传算法,以汽车燃油经济性和排放性为优化目标,分别针对广州和上海主干道行驶工况对模糊控制器中隶属度函数进行优化.然后建立了一个基于模糊神经网络的行驶工况识别方法,通过识别广州和上海的主干道行驶工况,对控制策略中模糊控制器的隶属度参数进行相应调整,结果证明采用模糊神经网络识别行驶工况的HEV模糊控制策略可以进一步提高汽车的燃油经济性和排放性能.     

基于HGA的模糊神经控制器设计及其应用

将神经网络与模糊控制相结合,实现了模糊控制器的自学习和自适应。给出一种基于递阶遗传算法的模糊神经网络优化算法,通过对每个染色体采用递阶编码,可以同时优化模糊神经网络结构和权值参数。将这种模糊神经网络控制器应用于镍氢电池的充电控制中,证明了算法的有效性。     

A hybridization of genetic algorithms and fuzzy logic for the single-machine scheduling with flexible maintenance problem under human resource constraints

This work focuses on the problem of scheduling jobs on a single machine that requires flexible maintenance under human resource competence and availability constraints. To solve the problem we developed two fuzzy genetic algorithms that are based on respectively the sequential and total scheduling strategies. The one respecting the sequential strategy consists in two phases. In the first phase, the integrated production and maintenance schedules are generated. In the second one, the human resources are assigned to maintenance activities. The second algorithm respecting a total strategy consists in generating the integrated production and maintenance schedules that explicitly satisfy the human resource constraints. In regard to these two different strategies, we studied then two integrated fuzzy genetic algorithms that use the fuzzy logic framework to deal with the uncertain nature of both production and maintenance data. The proposed genetic algorithms have been implemented and applied to non-standard test problems which integrate production, maintenance and human resource data. The experimental results show that the consideration of human resource constraints and uncertainties allows to get more realistic and applicable solutions. Moreover, the comparison between the two proposed algorithms shows that the one based on the total strategy outperforms the one based on the sequential strategy regarding the objective functions’ optimization. However, this latter is better in terms of computational times.     

Hierarchical aggregation of multiple fuzzy controllers for global complex control problems

In this paper, we illustrate a proposed method for control that combines the outputs of several individual controllers to improve global control of complex nonlinear plants. In the first part of this paper, we illustrate the proposed method that consists of two levels, where in the top level a fuzzy system represents a superior control that is designed for adjusting the behavior of the individual fuzzy controllers at the lower level. To test the approach, we consider the problem of flight control because it requires several individual controllers. Also a comparison is performed, where the hierarchical control strategy is compared with a simple control approach using the t student test. In this paper, we show that the proposed method outperforms the conventional fuzzy control approach. In the optimal design of the proposed control architecture a genetic algorithm was also applied to tune the parameters of the fuzzy systems in an optimal fashion.     

基于改进模糊神经网络的PID参数自整定

针对传统PID整定控制效果差且单纯神经网络整定存在参数学习和调整困难等问题,提出了一种基于改进模糊神经网络的PID参数整定方法。在该方法中,PID控制器的控制参数采用基于Mamdani模型的模糊神经网络进行自适应整定,模糊神经网络参数采用混沌遗传算法离线粗调和BP算法在线细调的方式进行学习和调整,仿真结果表明该整定策略动态响应快、误差控制精度高且网络中各节点及参数物理意义明确。最后分别从模糊规则数的变化及适应度函数的选取两方面提出两种优化方案,仿真结果表明增加模糊规则数或采用不同的适应度函数都有利于进一步减小控制误差。     

基于MAS的分布式焦炉集气管压力解耦控制

针对焦炉集气管压力这类多变量非线性耦合系统,提出了一种基于multi-agent system(MAS)的焦炉集气管压力智能协调控制系统方案,给出了agent的分层组织结构和演化机制．在控制agent中,采用TS模糊神经回归网络对被控对象进行分布式建模,运用分布式智能协调解耦算法进行解耦控制,监督学习与强化学习相结合,采用遗传协进化算法对多个agent协调优化．通过agent模态变迁进行模式切换,以适应快速突变环境．工程应用表明,提出的控制策略有效地解决了集气管压力这类复杂对象的过程控制问题．