分析了动态递归神经网络系统辨识的参数学习算法。
The parameter learning algorithm of dynamic recurrent neural network based on system identification is analyzed. D.
分析了动态递归神经网络系统辨识的参数学习算法。
The parameter learning algorithm of dynamic recurrent neural network based on system identification is analyzed.
对所提出的动态递归神经网络进行了分析,以及如何利用它们来进行系统辨识。
The dynamic recurrent neural network is analyzed, and how to use it for system identification is also analyzed.
与改进BP算法相结合,各取所长,形成集成化动态递归神经网络建模辨识算法。
The identification algorithm integrating the forward evolutionary algorithm and improved BP algorithm for the dynamic recursive neural network model is formed.
辨识结果表明,动态递归网络模型优于传统辨识模型,适于非线性、不确定结构的辨识。
Results of identification show that the Elman's recurrent model is superior to the traditional model. It is adaptive to the identification of the non linear and uncertain structure.
接着,结合其存在的问题,对动态递归神经网络、R BF神经网络和自适应逆控制进行了算法研究。
Then, aiming at the existing problem, the algorithm of dynamic recurrent neural network, RBF neural network and adaptive inverse control is studied in the paper.
针对仿射非线性系统,提出了一种新型的基于动态递归模糊神经网络(DRFNN)的间接自适应控制器。
A novel indirect adaptive controller based on dynamic recurrent fuzzy neural network (DRFNN) is proposed for affine nonlinear system.
提出一种基于动态递归神经网络的自适应pid控制方案,该控制系统由神经网络辨识器和神经网络控制器组成。
This paper presents an adaptive PID control scheme based on dynamic recurrent neural network. The control system is consisted of the neural network identifier and the neural network controller.
提出一种基于动态递归神经网络的自适应pid控制方案,该控制系统由神经网络辨识器和神经网络控制器组成。
An adaptive PID control scheme based on dynamic recurrent neural network is presented. The control system is consisted of the neural network identifier and the neural network controller.
提出了动态递归数字滤波器的鲁棒稳定性检验定理,将动态递归数字滤波器划分为两类,单调系统簇与非单调系统簇。
The robust stability test theorems of dynamic recursive digital filters have been proposed in the paper. The digital filters can be classified into two classes, monotone and non-monotone.
应用该模型对线性结构和非线性结构在变阻尼控制和外荷载激励下结构的响应进行了数值仿真,表明所提的动态递归神经网络可以达到较高的预测精度。
Simulations on linear and nonlinear structures demonstrate that RDRNN is very effective on predicting the response of a structure subject to semi-active control and external excitation.
蝴蝶和马利筋,犹如立足笔尖的铅笔,依靠共同进化的递归动态而立得笔直。
They are held together, poised upright like a pencil standing on its point, by the recursive dynamics of coevolution.
在可以使用递归,但因为递归重复解决相同的子问题造成效率低下的时候,则可以采用动态编程。
Dynamic programming is used when recursion could be used but would be inefficient because it would repeatedly solve the same subproblems.
大多数问题与递归结构以及如何处理动态输入和输出数据有关。
The majority of the issues relate to recursive structures and how to cope with dynamic input and output data.
由于其反馈特征,使得递归神经网络模型能获取系统的动态响应特性。
With the feedback behavior, the recurrent neural network can catch up with the dynamic response of the system.
提出一个新颖的递归算法,用于实现动态的网格负载平衡。
A novel recursion algorithm is proposed to implement the dynamic grid load balancing.
经过多次修改,目前也有管理结构化数据、动态数据配置、递归(过程调用自身)等等功能。
It has been revised several times and now includes capabilities for handling structured data, dynamic data allocation, recursions (procedures that call themselves), and other features.
由于其反馈特征,使得递归神经网络模型能获取系统的动态响应。
With the feedback behavior, the recursive neural network can catch up with the dynamic response of the system.
为了考察该系统的动态性能,采用递归BP神经网络对该系统进行辨识。
To test the dynamic property, this pneumatic fatigue test system was identified by a recursive BP neural network.
利用对角递归神经网络在线自适应调整PID控制器的参数,从而使系统的静态和动态性能指标较为理想。
DRNN is used to adjust the parameters of PID control on-line, accordingly it can make static and dynamic performance index comparatively ideal.
针对静态网络无法处理暂态问题,对具有递归环节的动态模糊神经网络进行了研究。
Since a static fuzzy neural network cannot deal with the temporal problem, a dynamic fuzzy neural network (DFNN) with recurrent units is proposed.
复杂意谓它就是那样,非线性意谓递归和更高数学法则,而动态意谓非常数及非周期。
Complex implies just that, nonlinear implies recursion and higher mathematical algorithms, and dynamic implies nonconstant and nonperiodic.
本文介绍了动态对角递归网络,并针对BP算法收敛慢的缺点,将递推预报误差学习算法应用到神经网络权值和域值的训练。
To overcome the slow convergence of the BP algorithm, recursive prediction error algorithm is proposed, which can train both the weight and the bias.
提出了动态八叉树结构和相应的自相似裂变递归算法,实现了虚拟环境下的可加工工件动态建模。
A dynamic 8-tree structure and a relative self-similar recursion algorithm are presented to model a dynamic workpiece object, which is machinable.
提出了动态八叉树结构和相应的自相似裂变递归算法,实现了虚拟环境下的可加工工件动态建模。
A dynamic 8-tree structure and a relative self-similar recursion algorithm are presented to model a dynamic workpiece object, which is machinable.
应用推荐