vegy__yhe

使用BP神经网络预测农产品的需求，论文在知网上应该就一篇。(Using neural network to predict the demand of agricultural products)

用MATLAB实现单因素/两因素重复测量方差分析，自带工具包中只有非重复测量的方差分析函数ANOVA。anova_rm实现单因素；rm_anova2实现两因素( function stats = rm_anova2(Y,S,F1,F2,FACTNAMES) Two-factor, within-subject repeated measures ANOVA.For designs with two within-subject factors. function [p, table] = anova_rm(X, displayopt)Single factor, repeated measures ANOVA.)

最新的关于贝叶斯聚焦算法IEEE文章。该算法通过利用信号DOA概率密度函数实现加权的阵列波前差值法，能实现宽带相干信号的DOA估计，并能减小阵列聚焦误差，提高宽带阵列频域波束形成鲁棒性。(In this paper, we present and study a Bayesian focusing transformation (BFT) for coherent wideband array processing, which takes into account the uncertainty of the direction of arrivals (DOAs). The Bayesian focusing method minimizes the mean-square error of the transformation over the probability densityfunctions(pdfs)oftheDOAs,thusachievingimprovedfocusing accuracy over the entire bandwidth)

对该模型参数辨识采用递推最小二乘法，即 RLS（ recurisive least square）， 它是一种能够对模型参数进行在线实时估计的辨识方法。(The recursive least square method is used to identify the parameters of the model, that is, rls (recurisive least square), which is an online real-time estimation method for the model parameters)

椭圆偏微分方程求解函数，调用该函数可以解任意椭圆偏微分方程(Elliptic partial differential equations to solve the function, call this function can solve any elliptic partial differential equations)

说明：  模拟退火算法，地球物理反演，非线性反演算法(Simulated annealing algorithm)

compute Bhattacharyya distance between two Gaussian classes

说明：  计算时程信号数据的快速傅里叶变换，进行频谱分析(The fast Fourier transform of time history signal data is calculated and the spectrum analysis is carried out)

：针对三容水箱是较为典型的非线性、时延对象，具有很强的代表性这一特点， 通过机理建模方法建立了三容水箱的一般数学模型。分析了线性和非线性阻力板的流量特性， 并通过实验测定了阻力板流量系数。最后给出了三容水箱的线性化模型。由仿真实验可知，非线性模型的阶跃响应和实验测得的响应曲线相吻合，(: For three-tank water tank is a more typical non-linear, time-delay object, has a strong representation of this characteristic, through the mechanism of modeling methods allow the establishment of three general mathematical model of water tanks. Analysis of linear and nonlinear flow resistance characteristics of plate, and through experimental determination of the resistance plate flow coefficient. Finally, the three-linear model of water tanks. By simulation experiments, we can see that the nonlinear model of step response and the experimentally measured response curve coincide)

灰色预测模型，对未来数据进行预测，同时还有误差检测(GM(1,1),Grey prediction model, which can forecast future data and detect errors at the same time)