资源列表
Fuzzy-Neural-Network-by-matlab
- 这是一个四个不同的S函数实现集合的递归模糊神经网络（RFNN）。该网络采用了4组可调参数,这使得它非常适合在线学习/操作，从而可应用到系统识别等方面。-This is a collection of four different S-function implementations of the recurrent fuzzy neural network (RFNN) described in detail in [1]. It is a four-layer, neuro-fuzzy net
single_node_multi_node
- 认知无线电中基于能量检测的协作频谱感知，考虑信噪比和检测概率的关系。分别 输出单结点和3结点的理论值与仿真值。-Cognitive radio based on energy detection cooperative spectrum sensing, consider the relationship between SNR and probability of detection. Respectively output nodes and three single-node theore
hao
- A算法解决15数码问题 应用在人工智能学科 用C#语言编程-digital hello world
GRNN-
- 基于广义回归神经网络的数据预测，使用交叉验证的GRNN神经网络预测程序，包含BP和GRNN效果比较程序。两网络用相同的数据进行训练。-Based on generalized regression neural network data prediction, using cross-validation GRNN neural network prediction program, including BP and GRNN effect comparison procedures. Two
Hybrid-code-GA
- 混合编码遗传算法，通过格雷编码，二进制编码，浮点数编码来实现混合编码遗传算算法-Hybrid coding genetic algorithm, through the Gray code, binary code, floating point code to implement hybrid coding algorithm genetic operators
MyBP-lxf
- 基于NWP数据和BP-人工神经网络的超短期风电功率预测-Based on NWP data and BP-ANN ultra short-term wind power prediction
cec05
- 智能优化算法标准测试函数，cec05单目标测试函数，包含技术文档-Problem Definitions and Evaluation Criteria for the CEC 2005 Special Session on Real-Parameter Optimization
BP-predict
- 针对中长期电力负荷预测样本量小、多因素影响的特点,利用灰色关联度筛选影响因素,建立基于BP 神经网络算法的负荷预测模型,通过多因素变量及历史负荷变量序列进行滚动预测,得到的预测值明显优于 单一预测方法,并通过马尔可夫过程对预测残差进行修正,使预测精度得到较大提高,研究实证表明,这种预 测方法具有进行推广应用的价值 -For long-term load forecasting small sample size, the characteristics of many facto
FFTnetwork
- 傅立叶神经网络，用于谐波检测和分析，精度高，可以检测间谐波-Fourier neural networks for harmonic detection and analysis, high precision, can detect inter-harmonics
SVM
- 用matlab实现支持向量机功能，可以进行四种核函数的选择，简单实用-support vector machine
program-two
- 静态K-S检验下的copula 分布估计算法边缘分布的研究- Marginal distribution in copula estimation of distribution algorithm based static K-S test
EM_Algorithm
- 介绍期望最大算法基本原理及聚类实现，可以很好的对多个高斯概率密度分布进行分类-Introduces the basic principle and expectation maximization clustering algorithm to achieve, can be good for multiple Gaussian probability density distribution of the classification