PID Control for Multivariable System Based on Ameliorative RBF Neural Networks

Aiming at systems which are of characteristics of multi-input and multi-output, nonlinearity and time-variation in the industrial control fields, this paper presents a intelligent PID control method based on ameliorative RBF neural networks, which constructs RBF neural networks identifier on-line and identifies a controlled object on-line by means of adopting the nearest neighbor-clustering algorithm, and adjusts parameters of PID controller on-line and realizes decoupiing control of multivariable, nonlinear and time-variation system. The simulation result indicates that the controller can get parameters which are optimal under some control law, it makes the decoupled system, compared to the PID control method based on the conventional RBF neural networks, has perfect dynamic and static performances, possesses the advantages of high precision, quick response speed and is of great adaptability and robustness.     

A Prediction Model Using Parallel Kernel RBF Neural Networks

Radial Basis Function Neural Networks(RBF NN) are frequently used for regression prediction.But kernel matrix computation for high dimensional data source demands heavy computing power.To shorten computing time,the paper designs a parallel algorithm to compute the kernel function matrix of RBF NN and applies it to the prediction of converter re-vanadium modeling.The paper then implements the algorithm on a cluster of computing workstations using MPI.Finally,experiment is done with the practical data to study the speedups and accuracy of the algorithm.     

基于RBF神经网络的蔬菜价格预报研究

准确预测农产品市场价格对于农户生产决策与政府调控等具有重要意义。针对蔬菜市场价格预报的复杂性,利用RBF神经网络的特性,应用2003-2007年的香菇市场价格数据建立蔬菜价格预报模型,并对RBF神经网络模型的参数选择进行分析。最后应用模型对2008-2009年的香菇市场价格数据进行预报,通过与BP神经网络模型预报结果进行比较,表明RBF神经网络模型具有更高的预报准确度。     

Application of RBF Algorithm Based on Generalized Genetic Optimization in Prediction Model for Molten Iron Desulfuration

Desulfuration process is a very sophisticate reaction which is not only diverse but also non-line. A RBF algorithm based on generalized genetic optimization is proposed after studying the standard genetic and RBF algorithm. The authors also introduce its application in prediction Model for molten Iron Desulfuration. The algorithm perfectly resolve the problem of random selection of RBF cluster center number. Furthermore, it also reduces the time which GA uses. Comparison between the simulation results of RBF and RBF algorithm Based on GGA optimization further proves the efficiency and precision of its application in Prediction Model for Molten Iron Desulfuration. Finally the result of the test shows that after adopting the algorithm, the end-point hitting ratio can reach eighty-five percent. This indicate the algorithm has the engineering practicability.     

基于 SPA-RBF神经网络的小麦蛋白质含量无损检测

传统半微量凯氏法测量小麦蛋白质含量繁琐费时,应用近红外光谱分析技术结合SPA-RBF神经网络对小麦蛋白质含量进行快速、无损检测.采用SPXY算法划分校正集和预测集样本,运用连续投影算法(SPA)对一阶微分和SNV预处理后的光谱数据提取敏感波点作为RBF神经网络的输入,建立小麦蛋白质含量的SPA-RBF神经网络校正模型.模型的预测均方根误差和预测相关系数可达到0.26576和0.975,预测效果较好,基本上可以完成粮食储备和食品加工行业对小麦及其制品品质的划分以及育种上的前期世代筛选.研究表明:近红外光谱技术结合SPA-RBF神经网络可实现对小麦蛋白质含量的检测,满足现代农业发展对小麦无损、实时、大量检测的需要.     

Self-adaptation Optimize BP Neural Network Design Based on the Genetic Algorithms

BP(Back Propagation) Neural networks is in the presence of the local optimization in the Neural networks training.The algorithm have slow convergence and the local convergence problem which impact the neural networks work performance.In order to cover these shortcomings and solves the size's hugeness and the low efficiency of the net problem in the traditional NN designing,the action principles of BP-Neural network's structure are analyzed,and a new method is formed which is confirmed from the Enhance genetic algorithms(EGA).The method can identify network configuration and network training methods.By adopting the number coding,self-adaptable multi-point variations operation,this method can effectively reduce the network size and the network convergence time,increase the network training speed.Tomatoes disease diagnosis examples illustrate the feasibility of this approach.     

Application of Incremental Genetic RBF NN in Converter Vanadium Recover

Converter vanadium recover is a very sophisticate reaction which is diverse and non- line. From the point of view of statistics and reaction mechanism, it is difficult to build up end- point control static model. Aim at this problem, the paper puts forward a model identify method based on incremental genetic RBF neural network to build up such a model, which can perfectly resolve the problem of random selection of RBF cluster center number and sample data clustering. Furthermore, in order to ensure structure of neural network to fit with continuous incremental data set, the paper presents a method of incremental data dealing, which is applied to amend the parameters of neural network. Then the request of continuous production is satisfied. Finally the result of test shows that after adopting the algorithm, the error of result is less than before and end- point hitting ratio satisfies to ninety percent. These indicate the algorithm has the engineering practicability.     

Features Recognition of Parts Drawing Based on BP Neural Network of Genetic Algorithms

This paper gives a new approach to solve the problem of features recognition of parts drawing form by means of multi - layers BP neural network. The concept and arithmetic of expanding property matrix of parts drawing are brought forward. A feature recognition and extraction system of 2D revolving parts drawing is developed by means of secondary development of AutoCAD R2000. These works have high theoretical and practical value.     

基于云遗传RBF神经网络的储粮温度预测研究

采用基于云理论的遗传算法对RBF神经网络算法进行优化,将优化后的算法引入到储粮温度参数分析中,对粮堆内部温度环境的变化情况进行预测。实验结果表明:优化后的算法对粮堆内温度预测具有较好的效果,预测的拟合程度很高,进而证明优化后的温度预测模型的有效性和可行性。     

Analysis on Application in Profile Optimizing Based on Genetic Algorithms

In order to achieving the Automation of profile optimization and offering more reasonable schemes,the author proposes a kind of computer method based on hereditary algorithm,which is using mathematic methods to simplify the profile lines as a list of points and using cut-fill gross as aim's function and using restriced conditions to control the profile lines,the authors optimize the list of points of changing slopes,the results of the method are more accurate than the artificial design and can provide many kinds of schemes within short time.In the actual projects we can properly increase the condition restraints to design more high-quality lines.If GIS platform is added to it,the profile can be optimized more effectively and more intuitionisticly.     

A Kind of Fuzzy System Modelling and Control Based on Neural Networks

Combining artificial neural networks(ANN) with fuzzy system theory,a kind of modelling & control method of fuzzy system based on ANN is presented.The simulation researches have verified that the proposed approach can be applied effectively to a number of control systems which are defficult to build strict mathematical model.     

The Inherent Fault Tolerance of Multi-Layer Neural Networks

The inherent fault tolerance of multilayer neural networks (MLNs) is due to thedistributed properties of information processing and information storage in an MLN , and the inher-ent fault-masking capability of every neuron. Also,the concept of inherent fault tolerance of MLNsis formalized, Finally, perceptrons and three-layer nets are shown inherently fault tolerant underreasonable statistical conditions,respectively.     

遗传算法优化人工神经网络模型在日光温室湿度预报中的应用

提出一种基于遗传算法优化BP神经网络的方法预测日光温室湿度环境因子。实测日光温室内影响空气湿度的环境因子组成数据样本作为神经网络的输入,采用基于实数编码的遗传算法替代随机设定神经网络的初始权阈值,然后通过改进的BP算法在由遗传算法确定的搜索空间中对网络进行精确训练。模型预报值和实测值基于1：1线的决定系数R2和预测平均相对误差MSE分别为0.9857和3．1％。结果表明,遗传算法优化BP神经网络预报模型收敛速度快、预测精度高。可为日光温室的湿度环境调控制提供理论依据和决策支持。     

基于NLPCA-RBF神经网络的番茄蒸散量预测

蒸散量(ET)是水文循环中的重要组成部分。精确的ET预测在水资源管理和灌溉系统设计等方面的研究是十分必要的。利用非线性主成分分析法(NLPCA)和径向基(RBF)神经网络组成的模型(NLPCA-RBF)对番茄蒸散量进行估算。在既保证ET影响因素信息完整,又可消除影响因素之间相关性的前提下,利用NLPCA将影响ET的7个气象因素简化为3个综合成分,并以此为网络训练的输入数据,根据实测的蒸散量作为网络输出建立了RBF神经网络,并且经非训练样本点数据检验。结果表明,与传统RBF网络模型较,NLPCA-RBF网络预测模型能够更好的反应影响因子与蒸散量之间的关系,取得更为精确的结果。     

Fault Section Diagnosis Algorithm of Distribution Networks Based On Topology Identification

An algorithm of fault section diagnosis based on topology identification for distribution networks is presented. By decomposing the topologic matrix which describing the distribution network into two parts,one part only contain the complex coupling factor ,and the other ignore the complex coupling factor. Using this method, the fault zones in distribution network can be identified and isolated efficiently, and the vertexes of the zones can also be identified automatically. This approach adapts to the changefully network structures.     

基于智能算法的储粮通风过程中温度场预测

粮食的储备对于一个国家来说十分重要,是关乎民生的重要战略资源。储粮对于温度要求十分严格,机械通风是保持粮仓环境温度的重要方法。针对传统机械通风不能准确识别通风时机而导致的风机损耗和粮食损害问题,选用RBF神经网络建立温度预测模型,利用粒子群算法（Particle Swarm Optimization,PSO）及列文伯格-马夸尔特算法（Levenberg-Marquard,LM）优化网络的参数。仿真结果可以看出,用PSO算法和LM算法优化后的神经网络模型预测精度更加准确,与用PSO优化后的RBF神经网络进行比较后可知,PSO-LM-RBF算法预测误差小,模型更加稳定。     

Design and Implementation of Fingerprint Identification Module Based on DSP Chip

An automatic fingerprint identification module based on the fixed DSP chip TMS320VC5410 and the fingerprint sensor FPS200 is presented. The program of fingerprint identification algorithm mixes the assemble language and C, which implements a high effective and low cost embedded system. The software and hardware design and implementation of this system are introduced later. In experiments, the module is fast speed and accurate, and takes a very good performance.     

一种基于距离和角度的无线传感器网络GPSR协议

传感器节点能量是影响无线传感器网络的生存周期的关键因素,基本的GPSR协议所使用的贪婪算法忽略了角度因素对下一跳选择的影响,在转发失败遇到空洞问题时只能进行周边边界转发算法,以损耗节点能量为代价。为了节约节点能量,延长网络生存周期,从改进GPSR协议出发,综合考虑距离和角度这两个因素解决路由过程中的空洞问题。仿真实验表明,改进后的协议是网络节点消耗大大减少,网络生存时间为300 s,长于基本GPSR协议下的200 s。该协议缩短了路径长度,节约节点能量,减少能量的消耗,延长了网络生存时间。     

基于RBF的安徽省资源环境压力动态预警

基于1996—2011年安徽省资源环境的相关数据,从资源压力、环境压力2方面构建安徽省资源环境压力预警指标体系。运用熵权法测算资源环境压力指数,进而运用RBF模型对1996—2011年安徽省资源环境压力演变趋势进行模拟预测,并结合预警指数和警度标准对1996—2011年安徽省资源环境压力预警状况进行分析,旨在为安徽省社会经济与资源环境和谐发展提供参考,使社会经济健康、快速、可持续的发展。结果表明：RBF模型有较高的模拟仿真精度,能够较好地拟合1996—2011年安徽省资源压力、环境压力和资源环境压力指数的发展趋势。安徽省资源压力预警指数总体呈现波动上升趋势,警度由“中警”上升到“巨警”。环境压力指数在1996—2011年一直呈现上升趋势,其预警指数在2004年以前基本处于“中警”格局,但2004年以来,这一指数明显增大,警度由“中警”逐步转变为“重警”和“巨警”状态。可得出结论,影响安徽省资源环境压力的主要因素包括人均用电量、人均用水量、自然灾害受灾面积比重、能源消费标煤总量、工业废气排放量、固体废弃物排放量等,是今后调控的重点。     

Investigations of Distribution System Planning of Loop Configurations Using a Genetic Algorithm

The distribution system planning is performed with a loop configuration constraint from scratch.With Evolutionary Algorithm,a sample distribution network is designed with a loop configuration,in which the system is operated as radial configuration.It firstly introduces the mathematical model for distribution network planning,then the application of Evolutionary Algorithm for the planning of distribution network.By using the theory mentioned above, Visual C++ is applied to develop the distribution network optimization software.Finally with the optimization process of 17 nodes system being showed,the presented algorithm's utility and validity is verified.