共查询到20条相似文献,搜索用时 218 毫秒
1.
2.
基于GEP优化的RBF神经网络算法 总被引:1,自引:0,他引:1
RBF神经网络作为一种采用局部调节来执行函数映射的人工神经网络,在逼近能力、分类能力和学习速度等方面都有良好的表现,但由于RBF网络的隐节点的个数和隐节点的中心难以确定,从而影响了整个网络的精度,极大地制约了该网络的广泛应用.为此本文提出基于GEP优化的RBF神经网络算法,对其中心向量及连接权值进行优化.实验表明,本文所提算法比RBF算法的预测误差平均减少了48.96% . 相似文献
3.
基于Sugeno型神经模糊系统的交通流状态预测算法 总被引:1,自引:1,他引:0
从交通流状态的模糊特性出发,设计基于Sugeno型神经模糊系统的交通流状态预测算法.选择交通流状态的影响指标作为模糊推理系统的输入、交通流状态作为输出;据经验对输入、输出划分模糊子集,给出相应的隶属度函数并制定模糊规则;建立具有5层结构的神经模糊推理系统,利用神经网络优化调整模糊推理系统的隶属度函数和模糊规则.仿真实验表明,神经网络可直接优化模糊推理系统的隶属度函数,通过对连接权值的训练间接优化模糊规则,故Sugeno型神经模糊系统相比常规模糊系统具有更好的交通流状态预测性能. 相似文献
4.
5.
针对和声搜索算法参数影响其优化BP神经网络的性能问题,提出了一种可有效提高BP神经网络收敛速度和准确度的基于BtW参数动态变化的改进和声算法,同时用于BP网络优化。算法根据和声搜索参数的特点,采用以BtW为自变量的非线性函数变换方法,对微调概率PAR和微调幅度BW进行动态调整,利用改进的和声搜索算法对BP神经网络的连接权和偏置值进行优化。实验结果表明,该算法有效改善了和声搜索算法在BP神经网络优化中的性能,提高了BP网络的训练速度和预测的准确度。 相似文献
6.
针对神经网络结构设计问题,提出一种基于神经网络复杂度的修剪算法.其实质是在训练过程中,利用网络连接权矩阵的协方差矩阵计算网络的信息熵,获得网络的复杂度;在保证网络信息处理能力的前提下,删除对网络复杂度影响最小的隐节点.该算法不要求训练网络到代价函数的极小点,适合在线修剪网络结构,并且避免了结构调整前的网络权值预处理.通过对典型函数逼近的实验结果表明,该算法在保证网络逼近精度的同时,可有效地简化网络结构. 相似文献
7.
文章介绍了一种基于进化式模糊神经网络时间预测系统,它是一种快速自适应的局部学习模型;进化式模糊神经网络是一个特殊类型的神经网络,它能通过进化其结构和参数来容纳新的数据.文章重点介绍了网络结构、学习方法及创建、修剪、聚合规则节点的算法;实验结果表明:模糊隶属函数的个数,规则的修剪和聚合等训练参数,与网络的行为和预测结果有很重要的关系. 相似文献
8.
曹邦兴 《计算机工程与应用》2010,46(2):224-226
提出了一种基于蚁群算法的径向基函数神经网络,用它来进行地下水位预测,既具有神经网络广泛映射能力,又具有蚁群算法全局寻优、分布式计算等特点。实验表明,蚁群算法与径向基函数神经网络相融合能达到良好的预测效果。 相似文献
9.
神经网络和改进粒子群算法在地震预测中的应用 总被引:1,自引:1,他引:0
提出了一种基于神经网络与改进粒子群算法的地震预测方法,该方法采用前向神经网络作为地震震级的预测模型,引入改进的粒子群算法对前向网络的连接权值进行修正。为了设计在全局搜索和局部搜索之间取得最佳平衡的惯性权重,基于粒子动态变异思想对粒子群优化算法进行改进,提出了一种动态变异粒子群优化算法,并将其应用于地震震级预测神经网络模型优化。在仿真实验中,将所提出的方法与另外两个采用不同算法的前向网络预测方法进行了比较。结果表明所提出的优化算法收敛速度最快,所得模型的预测误差最小,泛化能力最强,对地震的中期预测有很好的参考作用。 相似文献
10.
自适应变系数粒子群—径向基神经网络模型在负荷预测中的应用 总被引:1,自引:0,他引:1
为了提高短期电力负荷预测精度,提出了一种自适应变系数粒子群-径向基函数神经网络混合优化算法(AVCPSO-RBF).实现了径向基神经网络参数优化.建立了基于该优化算法的短期负荷预测模型,利用贵州电网历史数据进行短期负荷预测.仿真表明,该方法的收敛速度和预测精度优于传统径向基神经网络方法和粒子群-RBF神经网络方法及基于混沌理论的神经网络模型,该优化算法克服了径向基神经网络和传统的粒子群优化方法的缺点,改善了径向基神经网络的泛化能力,提高了贵州电网短期负荷预测的精度,各日预测负荷的平均百分比误差可控制在1.7%以内.该算法可有效用于电力系统的短期负荷预测. 相似文献
11.
The classification ability of a single-layer perceptron could be improved by considering some enhanced features. In particular, this form of neural networks is called a functional-link net. In the output neuron's activation function, such as the sigmoid function, an inner product of a connection weight vector with an input vector is computed. However, since the input features are not independent of each other for the enhanced pattern, an assumption of the additivity is not reasonable. This paper employs a non-additive technique, namely the fuzzy integral, to aggregate performance values for an input pattern by interpreting each of the connection weights as a fuzzy measure of the corresponding feature. A learning algorithm with the genetic algorithm is then designed to automatically find connection weights. The sample data for bankruptcy analysis obtained from Moody's Industrial Manuals is considered to examine the classification ability of the proposed method. The results demonstrate that the proposed method performs well in comparison with traditional functional-link net and multivariate techniques. 相似文献
12.
A fuzzy set based solution method for multiobjective optimal design problem of mechanical and structural systems using functional-link net 总被引:1,自引:0,他引:1
Hong-Zhong Huang Ping Wang Ming J. Zuo Weidong Wu Chunsheng Liu 《Neural computing & applications》2006,15(3-4):239-244
The principle of solving multiobjective optimization problems with fuzzy sets theory is studied. Membership function is the key to introduce the fuzzy sets theory to multiobjective optimization. However, it is difficult to determine membership functions in engineering applications. On the basis of rapid quadratic optimization in the learning of weights, simplification in hardware as well as in computational procedures of functional-link net, discrete membership functions are used as sample training data. When the network converges, the continuous membership functions implemented with the network. Membership functions based on functional-link net have been used in multiobjective optimization. An example is given to illustrate the method. 相似文献
13.
14.
15.
Petri net language is a powerful tool for describing dynamic behaviors of physical systems. However, it is not easy to obtain the language expression for a given Petri net especially a structure-complex net. In this paper, we first analyze the behaviors of S-nets, which are structure-simple. With the decomposition method based on a given index function on the place set, a given structure-complex Petri net can be decomposed into a set of structure-simple S-nets. With the language relationships between the original system and the decomposed subnets, an algorithm to obtain the language expression of a given structure-complex net system is presented, which benefits the analysis of physical systems based on the Petri net language. 相似文献
16.
Wei Lu Gabriel Pui Cheong Fung Xiaoyong Du Xiaofang Zhou Lijiang Chen Ke Deng 《World Wide Web》2011,14(2):157-186
We study the problem of efficiently extracting K entities, in a temporal database, which are most similar to a given search query. This problem is well studied in relational
databases, where each entity is represented as a single record and there exist a variety of methods to define the similarity
between a record and the search query. However, in temporal databases, each entity is represented as a sequence of historical
records. How to properly define the similarity of each entity in the temporal database still remains an open problem. The
main challenging is that, when a user issues a search query for an entity, he or she is prone to mix up information of the
same entity at different time points. As a result, methods, which are used in relational databases based on record granularity,
cannot work any further. Instead, we regard each entity as a set of “virtual records”, where attribute values of a “virtual
record” can be from different records of the same entity. In this paper, we propose a novel evaluation model, based on which the similarity between each “virtual record” and the query can be effectively quantified, and the maximum
similarity of its “virtual records” is taken as the similarity of an entity. For each entity, as the number of its “virtual
records” is exponentially large, calculating the similarity of the entity is challenging. As a result, we further propose
a Dominating Tree Algorithm (DTA), which is based on the bounding-pruning-refining strategy, to efficiently extract K entities with greatest similarities. We conduct extensive experiments on both real and synthetic datasets. The encouraging
results show that our model for defining the similarity between each entity and the search query is effective, and the proposed
DTA can perform at least two orders of magnitude improvement on the performance comparing with the naive approach. 相似文献
17.
S Fiori 《Network (Bristol, England)》1999,10(2):171-186
The aim of this paper is to present a study of polynomial functional-link neural units that learn through an information-theoretic-based criterion. First the structure of the neuron is presented and the unsupervised learning theory is explained and discussed, with particular attention being paid to its probability density function and cumulative distribution function approximation capability. Then a neural network formed by such neurons (the polynomial functional-link artificial neural network, or PFANN) is shown to be able to separate out linearly mixed eterokurtic source signals, i.e. signals endowed with either positive or negative kurtoses. In order to compare the performance of the proposed blind separation technique with those exhibited by existing methods, the mixture of densities (MOD) approach of Xu et al, which is closely related to PFANN, is briefly recalled; then comparative numerical simulations performed on both synthetic and real-world signals and a complexity evaluation are illustrated. These results show that the PFANN approach gives similar performance with a noticeable reduction in computational effort. 相似文献
18.
Interest matching is an important data-filtering mechanism for a large-scale distributed virtual environment. Many of the existing algorithms perform interest matching at discrete timesteps. Thus, they may suffer the missing-event problem: failing to report the events between two consecutive timesteps. Some algorithms solve this problem, by setting short timesteps, but they have a low computing efficiency. Additionally, these algorithms cannot capture all events, and some spurious events may also be reported. In this paper, we present an accurate interest matching algorithm called the predictive interest matching algorithm, which is able to capture the missing events between discrete timesteps. The PIM algorithm exploits the polynomial functions to model the movements of virtual entities, and predict the time intervals of region overlaps associated with the entities accurately. Based on the prediction of the space–time intersection of regions, our algorithm can capture all missing events and does not report the spurious events at the same time. To improve the runtime performance, a technique called region pruning is proposed and used in our algorithm. In experiments, we compare the new algorithm with the frequent interest matching algorithm and the space–time interest matching algorithm on the HLA/RTI distributed infrastructure. The results prove that although an additional matching effort is required in the new algorithm, it outperforms the baselines in terms of event-capturing ability, redundant matching avoidance, runtime efficiency and scalability. 相似文献
19.
20.
本文介绍了一种新型的人工神经网络的改进,并把它运用于非线性系统的辨识中。这种新型的网络就是带有内部动态元的FLNN(Functional-Link Neural Network),其中内部动态元分别由带有局部激活反馈和局部输出反馈的自回归滑动平均滤波器构成。其具体的动态网络参数寻优由遗传算法来决定。仿真结果表明,把这种改善了的FLNN与原有的外部带动态元的FLNN分别应用于系统辨识中,前者具有更好的泛化能力和鲁棒性。 相似文献