神经网络模糊预测优化控制在VAV系统中的应用

将神经网络、模糊控制与非线性预测优化控制结合起来,提出了神经网络模糊预测优化控制方法,采用前馈神经网络作为预测模型,利用贝叶斯正则化方法对模型进行了辨识,以自调整模糊控制器作为优化控制器,通过多步预测方式,系统的优化性能指标综合考虑温度偏差最小和能耗最小这两方面因素,应用该方法对制冷工况变风量空调系统的送风温度和回风温度（室内温度）进行了仿真控制研究。控制结果表明了该方法的有效性,控制效果良好,并且可以达到节省能耗的目的。

Abstract：

Artificial neural network,fuzzy control and nonlinear optimal predictive control were combined.The algorithm of neural network nonlinear fuzzy predictive optimal control was proposed.Feed-forward neural network was adopted as the predictive model of the cooling VAV system.The model was identified by the method of Bayesian regularization.The self-adjusting fuzzy controller was adopted as optimal controller.The algorithm was applied in the cooling VAV system with multi-step predictive method.Indoor temperature and supply air temperature was controlled aimed at minimum temperature deviation and minimum energy consumption by this scheme in Matlab.Simulation results illustrate the effectiveness of this technique,and in the meantime illustrate that this technique can save energy consumption.     

AN INTELLIGENT CONTROL SYSTEM BASED ONRECURRENT NEURAL FUZZY NETWORK AND ITS APPLICATION TO CSTR

In this paper, an intelligent control system based on recurrent neural fuzzy network is presented for complex, uncertain and nonlinear processes, in which a recurrent neural fuzzy network is used as controller (RNFNC) to control a process adaptively and a recurrent neural network based on recursive predictive error algorithm (RNNM) is utilized to estimate the gradient information ρy/ρu for optimizing the parameters of controller.Compared with many neural fuzzy control systems, it uses recurrent neural network to realize the fuzzy controller. Moreover, recursive predictive error algorithm (RPE) is im-plemented to construct RNNM on line. Lastly, in order to evaluate the performance of the proposed control system, the presented control system is applied to continuously stirred tank reactor (CSTR). Simulation comparisons, based on control effect and output error,with general fuzzy controller and feed-forward neural fuzzy network controller (FNFNC),are conducted. In addition, the rates of convergence of RNNM respectively using RPE algorithm and gradient learning algorithm are also compared. The results show that the proposed control system is better for controlling uncertain and nonlinear processes.     

Self-organizing fuzzy clustering neural network and application to electronic countermeasures effectiveness evaluation

A self-organizing fuzzy clustering neural network by combining the self-organizing Kohonen clustering network with the fuzzy theory is proposed. This network model is designed for the effectiveness evaluation of electronic countermeasures, which not only exerts the advantages of the fuzzy theory, but also has a good ability in machine learning and data analysis. The subjective value of sample versus class is computed by the fuzzy computing theory, and the classified results obtained by self-organizing learning of Kohonen neural network are represented on output layer. Meanwhile, the fuzzy competition learning algorithm keeps the similar information between samples and overcomes the disadvantages of neural network which has fewer samples. The simulation result indicates that the proposed algorithm is feasible and effective.     

Novel backstepping design for blended aero and reaction-jet missile autopilot

The advanced missile uses blended control of nero-fin and reaction-jet to improve missile maneuverability. The blended control design, which is multi-inputs and multi-outputs （MIMO）, severe nonlinear, and model uncertain, is much more complex than conventional nero-fin control. A novel nonlinear backstepping control approach is proposed to design the blended autopilot. Missile model is reformed to a new one by state reconstruction technique so that it is easy to be handled by the backstepping method. Then a Lyapunov function is chosen to avoid oscillation caused in normal backstepping way when control parameters are mismatched. In distribution of both inputs, optimal energy logic is proposed. In addition, a fuzzy cerebellar model articulation controller （FCMAC） neural network is used to guarantee controller robustness to uncertainties. Finally, simulation results demonstrate the efficiency and advantages of the proposed method.     

Fuzzy neural network image filter based on GA

A new nonlinear image filter using fuzzy neural network based on genetic algorithm is proposed. The learning of network parameters is performed by genetic algorithm with the efficient binary encoding scheme. In the following, fuzzy reasoning embedded in the network aims at restoring noisy pixels without degrading the quality of fine details. It is shown by experiments that the filter is very effective in removing impulse noise and significantly outperforms conventional filters.     

Autotuning algorithm of particle swarm PID parameter based on D-Tent chaotic model

An improved particle swarm algorithm based on the D-Tent chaotic model is put forward aiming at the standard particle swarm algorithm. The convergence rate of the late of proposed algorithm is improved by revising the inertia weight of global optimal particles and the introduction of D-Tent chaotic sequence. Through the test of typical function and the autotuning test of proportionalintegral-derivative （PID） parameter, finally a simulation is made to the servo control system of a permanent magnet synchronous motor （PMSM） under double-loop control of rotating speed and current by utilizing the chaotic particle swarm algorithm. Studies show that the proposed algorithm can reduce the iterative times and improve the convergence rate under the condition that the global optimal solution can be got.     

Application of quantum neural networks in localization of acoustic emission

Due to defects of time-difference of arrival localization,which influences by speed differences of various model waveforms and waveform distortion in transmitting process,a neural network technique is introduced to calculate localization of the acoustic emission source.However,in back propagation(BP) neural network,the BP algorithm is a stochastic gradient algorithm virtually,the network may get into local minimum and the result of network training is dissatisfactory.It is a kind of genetic algorithms with the form of quantum chromosomes,the random observation which simulates the quantum collapse can bring diverse individuals,and the evolutionary operators characterized by a quantum mechanism are introduced to speed up convergence and avoid prematurity.Simulation results show that the modeling of neural network based on quantum genetic algorithm has fast convergent and higher localization accuracy,so it has a good application prospect and is worth researching further more.     

Adaptive nonuniformity correction for IRFPA sensors based on neural network framework

For infrared focal plane array sensors, imagery is degraded during signal acquisition, particularly nonuniformity. In this paper, an adaptive nonuniformity correction technique is proposed which simultaneously estimates detector-level and readoutchannel-level correction parameters using neural network approaches. Firstly, an improved neural network framework is designed to compute the desired output. Secondly, an adaptive learning rate rule is used in the gain and offset parameter estimation process. Experimental results show the proposed algorithm can achieve a faster convergence speed and better stability, remove nonuniformity and track parameters drift effectively, and present a good adaptability to scene changes and nonuniformity conditions.     

Constrained predictive control based on T-S fuzzy model for nonlinear systems

A constrained generalized predictive control （GPC） algorithm based on the T-S fuzzy model is presented for the nonlinear system. First, a Takagi-Sugeno （T-S） fuzzy model based on the fuzzy cluster algorithm and the orthogonalleast square method is constructed to approach the nonlinear system. Since its consequence is linear, it can divide the nonlinear system into a number of linear or nearly linear subsystems. For this T-S fuzzy model, a GPC algorithm with input constraints is presented. This strategy takes into account all the constraints of the control signal and its increment, and does not require the calculation of the Diophantine equations. So it needs only a small computer memory and the computational speed is high. The simulation results show a good performance for the nonlinear systems.     

Learning algorithm and application of quantum BP neural networks based on universal quantum gates

A quantum BP neural networks model with learning algorithm is proposed. First, based on the universality of single qubit rotation gate and two-qubit controlled-NOT gate, a quantum neuron model is constructed, which is composed of input, phase rotation, aggregation, reversal rotation and output. In this model, the input is described by qubits, and the output is given by the probability of the state in which （1） is observed. The phase rotation and the reversal rotation are performed by the universal quantum gates. Secondly, the quantum BP neural networks model is constructed, in which the output layer and the hide layer are quantum neurons. With the application of the gradient descent algorithm, a learning algorithm of the model is proposed, and the continuity of the model is proved. It is shown that this model and algorithm are superior to the conventional BP networks in three aspects： convergence speed, convergence rate and robustness, by two application examples of pattern recognition and function approximation.     

基于灰色神经网络组合模型的故障预测

针对装备故障预测存在有效样本少、模型预测精度低等问题，集成灰色理论和神经网络方法，提出基于灰色神经网络的故障预测组合模型。基于新信息优先原理和重构背景值方法优化灰色GM（1，1）模型的初始值与背景值，利用Levenberg-Marquardt算法改进反向传播神经网络模型；采用组合预测思想，将多方法融合改进灰色模型和神经网络模型，分别构建基于权重分配、基于误差修正和基于结构优化的3种灰色神经网络组合模型。以某雷达发射机的故障预测为例，验证上述方法在故障预测中的有效性。结果表明，灰色神经网络组合模型的预测精度优于单一预测模型，可用于装备的故障预测和预测性维修。     

基于先验知识和神经网络的非线性建模与预测控制

神经网络模型是模拟非线性系统的有力工具，它的缺陷是难以利用已有的先验知识。利用通用学习网络的建模方法，提出了一种利用先验知识和神经网络建立非线性系统模型的方法，具有简化神经网络结构、减小计算量的优点。基于这种模型利用改进的遗传算法进行优化计算，从而实现了基于先验知识和神经网络的非线性建模和预测控制。对一个悬吊系统的仿真实验说明了该算法的有效性。     

一类基于神经网络的非线性模型预测控制

在研究非线性对象输入/输出数据的基础上，将对象输出的Taylor级数展开式取线性项作为预测模型，提出一种非线性系统模型预测控制算法，为了保证预测模型的准确性，以神经网络做辩识器估计系统建模误差，对非线性对象进行单频预测控制，理论上已证明三层BP网能任意逼近L2上的非线性函数，本文通过仿真研究也表明了当神经网络逼近系统建模误差时，所提出的预测控制算法对复杂非线性对象能达到良好的控制效果。     

基于遗传预估策略的智能主动队列管理

针对TCP传输过程中的典型时滞特性,提出了一种智能主动队列管理算法.该算法以自学习预估机制模型为核心来克服大时滞特征对网络稳定性能的影响,拥塞控制系统以两条信息通道分别实现模型补偿和预测控制功能.模型补偿通道采用了Smith预估嚣实现对网络时滞特征的动态补偿,并进一步设计迭代进化算法实现对Smith预估模型未建模特征的估计过程.预测控制通道采用基于神经网络的PID智能丢弃算法,通过神经网络的学习预测功能自适应调整预测控制通道的控制行为.通过仿真研究表明了提出的控制方法显著提高了拥塞控制机制的稳定性能和自适应性能.     

一种规则简化的模糊神经网络控制器

构造了一种实时模糊神经网络控制器，为解决模糊规则组合爆炸问题提供一个新方案。控制器基于T-S模糊模型，由前后件分离的网络结构实现。前件参数通过移动小论域法创建，每个变量仅在工作小论域上生成两个模糊子集，有效减少模糊规则，增强实时性；后件参数通过有ki,kp,kd修正因子的BP改进算法在线更改，控制意义明确，确保系统动态性能。仿真结果证实该控制器实时性好，控制性能优，鲁棒性强。     

基于结构自适应模糊神经网络的前列腺癌诊断方法

前列腺癌是近年来严重危害男性健康的疾病.利用模糊神经网络方法可以实现前列腺癌诊断,并将诊断模型表示为模糊规则集合.针对模糊神经网络所提取规则解释性差的问题,提出结构自适应模糊神经网络方法,通过改进损失函数,在训练中控制相似隶属度函数的合并,实现模糊神经网络模型结构自适应调整,减少模糊规则数量,在保证诊断准确性情况下,提取出容易理解的可解释性规则.同时该方法在模型的训练过程中引入粒子群优化(PSO)算法进行结构和参数学习,有效减少计算量,提高训练效率.最后,使用临床医学科学数据中心提供的前列腺疾病检查数据进行数值实验,验证了所提出方法在前列腺癌诊断和可解释性规则提取中的有效性.     

基于改进D-S的汽轮机组集成故障诊断研究

在分析与比较多个D-S合成规则的基础上,结合汽轮机组故障的特点,提出了一种基于改进D-S证据理论的集成故障诊断方法。该方法利用改进的D-S理论来表示和处理不确定的、模糊的信息,利用灰色理论和GRNN（广义回归神经网络）网络来处理证据理论中的基本概率分配问题,充分发挥灰色理论和GRNN的优点,提高故障诊断率。仿真结果表明,所提曲的集成故障诊断方法能有效地诊断汽轮机纽的故障,决策合理,可信度高,且能避免误诊现象,具有庭好的应用前景。