Narma ( nonlinear autoregressive moving average ) neutral network current controller of pmsm is proposed and it in company with neutral network speed regulator composes the pmsm vector control system 提出永磁同步电动机的narma神经网络电流控制器，并以神经网络速度控制器和narma电流控制器构成电动机矢量控制系统。
So we apply momentum threshold autoregressive model ( mtar ) in this paper to analyze bubble - driven run - ups in stock prices followed by a crash in a cointegration framework with asymmetric adjustment 因此，本文引入mtar模型，通过检验协整残差的非对称调整假设，对我国股票市场发展的不同阶段是否存在泡沫现象进行对比分析。
Although the chronicle patterns inherent in the electric loads could be approached by the autoregressive model from the historical records , factors other than the chronicle patterns could still have significant impacts on the accuracy of the load forecast 因此本论文以提升预测之准确性为目标，提出一个可考虑非时间因素影响之负载预测架构。
A new method for short time traffic flow ( sttf ) prediction is presented in this thesis based on wavelet , fuzzy neural network and autoregressive model . first , we review the history and status on sttf prediction 首先对交通工程的研究现状进行了综述，由此阐明了短时交通流预测在城市集成交通控制系统和作为未来交通模式的智能交通系统( its )中的重要地位。
R / s models are established using the pressure fluctuation signals of gas - solid fluidized bed . analytic results show that the autoregressive orders of the pressure fluctuation signals are different in various flow regimes of gas - solid fluidized bed 利用压力波动信号建立了ar模型，对气固流化床的流型进行了分析研究，发现在不同流型下， ar模型阶数有明显不同。
Synthesizing the two identification methods of weighted least square and resricted memory , the mutivariable system recursive estimate algorithems of unknown parameter of autoregressive models in the presence of controlled inputare are given 摘要将加权最小二乘法和限定记忆两种参数估计方法相综合，给出了多变量系统带控制输入的自回归模型未知参数的递推估计算法。
Immunity mix algorithm based on the continuous differential function is proposed in this paper , and its astringency is proved . from here we get the accurate estimator method of the autoregressive moving average model coefficient 本文首先提出了通用的基于连续可导函数的免疫混合算法，并证明了其收敛性，由此我们得到了自回归滑动均值模型系数的精确估计方法。
Experimental results show that wls is better than classical autoregressive ( ar ) models for 0 . 5 < h < 1 . it also can be seen that the disadvantage of wls is most salient for the lower of estimate accuracy of hurst exponent 本文对wls法进行了仿真研究，研究结果表明wls法对h 0 . 5的1 f噪声的滤除作用明显优于传统的ar模型，但同时发现wls法对hurst指数的估计误差较大。
Two - stage algorithms of parameter estimation for the autoregressive moving average ( arma ) models are presented , which are called two - stage recursive least squares algorithm ( 2 - rls ) and recursive least squares - pseudoinverse algorithm ( rls - pi ) 本文提出了自回归滑动平均( arma )模型的两段参数估计算法：两段递推最小二乘算法( 2 - rls )和递推最小二乘-伪逆算法( rls - pi ) 。
In the basis of that , the autoregressive model , which is very mature in the theory of linear time series , was used for establishing the future cash dividend predicted model . parameters were estimated by computer simulation 在这个基础上，本文利用线性时间序列理论上十分成熟的自回归模型( ar )建立了未来现金股利的预报模型，并通过计算机模拟技术对参数进行了估算。
