Rule Generation Based on Dominance Matrices and Functions

Rough set theory has proved to be a useful tool for rule induction. But, the theory based on indiscemibility relation or similarity relation cannot induce rules from decision tables with criteria. Greco et al have proposed a new rough set approach based on dominance relation to handle the problems. In this paper, the concept of dominance matrix is put forward and the dominance function is constructed to compute the minimal decision rules that are more general and applicable than the ones induced by the classical rough set theory. In addition,the methodology of simplification is presented to eliminate the redundancy in the rule set.     

A Method for Reducing Noise Radiated from Structures with Vibration Absorbers by Using an Accelerated Neural Network

A method for reducing noise radiated from structures by vibration absorbers is presented. Since usual design method for the absorbers is invalid for noise reduction, the peaks of noise power in the frequency domain as cost functions are applied. Hence, the equations for obtaining optimal parameters of the absorbers become nonlinear expressions. To have the parameters, an accelerated neural network procedure has been presented.Numerical calculations have been carried out for a plate-type cantilever beam with a large width, and experimental tests have been also performed for the same beam. It is clarified that the present method is valid for reducing noise radiated from structures. As for the usual design method for the absorbers, model analysis has been given, so the number of absorbers should be the same as that of the considered modes. While the nonlinear problem can be dealt with by the present method, there is no restriction on the number of absorbers or the model number.     

Reinforcement-Based Fuzzy Neural Network ontrol with Automatic Rule Generation

A reinforcemen-based fuzzy neural network control with automatic rule generation RBFNNC) is pro-posed. A set of optimized fuzzy control rules can be automatically generated through reinforcement learning based onthe state variables of object system. RBFNNC was applied to a cart-pole balancing system and simulation resultshows significant improvements on the rule generation.     

模糊逼近神经网络摄动系统的若干进展

本文对本世纪80年代中期兴起并紧密结合现代科学技术进步的一门新兴学科--模糊神经网络进行了综述,分析了所取得的主要成果及其特点,并指出了今后模糊神经网络研究中有待解决的许多问题.针对这些问题,介绍了笔者的工作--模糊逼近神经网络摄动系统,对开展模糊神经网络的研究将具有启迪作用和现实意义.     

基于模糊神经网络PID的无刷直流电机控制系统研究

针对传统PID控制在无刷直流电机控制系统中达不到良好的控制效果的问题,在无刷直流电机的数学模型基础之上,设计一种模糊神经网络自适应PID控制器,该控制器利用模糊控制非线性控制作用和BP神经网络的学习能力及适应能力相结合对PID参数进行在线实时调整。对基于模糊神经网络自适应PID控制器的无刷直流电机的双闭环控制系统进行仿真实验,实验结果表明,可以提高控制系统的响应速度,减小超调量,对负载及电机参数的变化都有较强的鲁棒性。     

Multi-agent reinforcement learning using modular neural network Q-learning algorithms

1 Introduction Reinforcement learning is a machine learningmethod for agents to acquire the optimal policyautonomously from the environment of their behaviors.When an action is executed, the agent receives areinforcement signal by interacting with theenvironment. This technology has recently been used inmany fields, such as robot control [1], artificialintelligence [2], especially multi-agent system [3,4].Generally, when the state space of the environment issmall enough and all states can be e…     

Simultaneous perturbation stochastic approximation algorithm combined with neural network and fuzzy simulation

In order to solve three kinds of fuzzy programm model, fuzzy chance-constrained programming mode ng models, i.e. fuzzy expected value and fuzzy dependent-chance programming model, a simultaneous perturbation stochastic approximation algorithm is proposed by integrating neural network with fuzzy simulation. At first, fuzzy simulation is used to generate a set of input-output data. Then a neural network is trained according to the set. Finally, the trained neural network is embedded in simultaneous perturbation stochastic approximation algorithm. Simultaneous perturbation stochastic approximation algorithm is used to search the optimal solution. Two numerical examples are presented to illustrate the effectiveness of the proposed algorithm.     

基于禁忌搜索的模糊规则自动获取

本文采用集中性和多样性策略对禁忌搜索进行改进,提出了一种基于模糊神经网络的混合禁忌搜索优化算法(FNN-based Hybrid Tabu Search Algorithm,FNN-HTS),用于同时优化模糊神经网络的结构和参数以提取出一组尽量精练的模糊规则。在FNN-HTS中,禁忌搜索用于同时优化网络结构和隶属函数参数,结合最小二乘法快速求解规则后件的线性参数。非线性函数逼近的实验结果表明所提出的方法能获得一组更精练的规则和更小的误差。     

基于模糊观测数据的RBF神经网络回归模型

提出了一种基于模糊观测数据的RBF神经网络（FORBFNN）,用于解决一类输出不可精确测量但可用模糊隶属度来表征的非线性系统建模问题．神经网络模型中各隐层神经单元的权重系数采用一种新的模糊EM算法辨识获得;隐层神经单元的数量及径向基函数的中心和宽度基于一种数据驱动的方法自适应确定,即首先初始生成一个隐层单元,然后根据一定的规则逐步加入新的单元,该过程不断迭代直到模型满足预设要求．该方法同时考虑了模型的复杂度及预测精度．数值模拟实验结果表明该建模方法是有效的,且建立的模型具有较高的预测精度．     

模糊综合评判的神经网络方法在损害分析中的应用

本文针对反倾销中损害案件的调查,在满足法律条文规定的基础上,采用模糊综合评判的神经网络方法,优化损害和损害程度的综合评判模型中企业经营指标的评判指标权重,进而采用优化后的权重进行损害和损害程度的综合评判.使评判过程和结果更加具有科学性,也使反倾销损害调查的裁决更加具有公共性和透明度.此方法对政府有关部门裁决倾销和反倾销案件有一定的辅助决策.     

A rapid fuzzy rule extraction method for fuzzy controller

Based on division of the three-dimensional space from data samples, the method proposed in this paper can rapidly extract fuzzy rules by using the fuzzy information of the samples. The principle of this approach is proved theoretically. Due to its simplicity this method can be used to extract fuzzy rules in real-time for an adaptive control system. Simulation results showed that this approach is effective and practical. Project (69775013) supoorted by Natural Science Foundation of China.     

基于模糊神经网络的智能能量管理策略研究——以混合动力电动汽车模型为例

针对当前在标准行驶工况下开发的混合动力电动汽车优化控制策略不能根据变化的行驶工况动态调整控制策略的问题,提出一种基于模糊神经网络的混合动力电动汽车动态能量管理策略：先利用模糊神经网络进行工况识别,然后根据识别的工况类型动态调整自身控制参数.仿真实验显示,该策略可以有效提高混合动力汽车的燃油消耗,并降低污染物的排放量.