实现了预测控制的三个环节:多步预测、滚动优化、反馈校正。
This method implements the three laches of GPC: multistep prediction, recursive optimization, feedback emendation .
本文通过修正CPN模型,提出了一种模糊多步预测控制算法。
Through modifying the CPN model, a kind of fuzzy multistep predictive control algorithm is proposed in this paper.
多步预测的误差和滞后现象显示AR模型难以预测突发性的喘振现象。
It shows that an ar model is difficult to predict a sudden surge, because errors and delay are too large to be ignored in multi-step predication.
与已有神经网络预测模型相比,具有更高的一步预测和多步预测精度。
Comparison with the ANN model shows that both one step forecasting accuracy and multistep forecasting accuracy by the online model are higher than that by ANN model.
针对工业过程的特点和控制要求,提出一种基于多步预测的神经网络自适应控制算法。
Adaptively controlling algorithm based on neural network of multi-step prediction was proposed for the industrial processes.
设计了确定性网络控制系统的输出反馈控制器和具有时延补偿功能的多步预测控制器;
The output feedback controller and multiple-step predicting controller for compensating time delay are designed.
试图用BP神经网络建立轴承寿命预测模型,并在该模型上进行多特征参数和多步预测方法的研究。
The bearing life forecast model based on BP network is researched. The multi step and multi feature forecasts can be realized concurrently.
采用神经网络逆控制的思想设计小波神经网络控制器,引入多步预测性能指标函数对控制器权值进行在线训练。
WNN controller with the idea of neural networks inverse control is applied. Using multi-step predictive index function to train the weights of controller.
仿真实验表明,基于多步预测的PID型神经网络控制系统能有效抑制随机干扰,具有较强的适应性和鲁棒性。
Simulation res ults prove that this new multi-step prediction based on PID-like neural network control system can effectively attenuate random noise interference and is more robust and adaptive.
针对交通流量混沌时间序列多步预测的问题,提出了一种基于混沌机理的小波神经网络(WNN)快速学习算法。
Aiming at the issue about multi-step prediction of the traffic flow chaotic time series, a fast learning algorithm of wavelet neural network (WNN) based on chaotic mechanism is proposed.
针对自适应阈值的检测方法,为提高检测系统的计算速度和鲁棒性,提出了基于模型多步预测方法的自适应阈值设计。
A new adaptive threshold technique was proposed by using multi-step ahead prediction based on model for improving speed and robustness of fault detection system.
针对单个BP神经网络作为预测模型时,递推多步预测误差积累大的缺点,本文提出多BP神经网络并行预测控制算法。
As the recursive error is accumulated largely when the single BP network is regarded as predictive model, we propose the paratactic predictive control algorithm based on multi-BP networks.
本文针对一类非线性动态系统,提出了一种新的基于后向回归网络的自适应多步预测方法,并对基于神经网络的自适应预测机理进行了分析。
This paper develops a novel backpropagation networks based adaptive multistep prediction technique for a class of nonlinear dynamical systems, and the prediction mechanism is analyzed.
提出一种新的基于神经网络多步时序预测的非线性系统故障诊断方法。
A new approach to fault diagnosis of nonlinear systems, which USES multistep prediction of time series based on neural network, is presented in this paper.
针对单弧度弯曲大变形无裂纹轴校直问题,根据广义预测控制理论,建立了多步校直行程预测算法。
However to the large deformation shaft without crack, general predictive theory has been applied to built up the multi-steps straightening stroke predictive algorithm.
提出了一种基于多步递推预测的广义预测自适应控制算法。
In this paper, a generalized predictive adaptive control algorithm based on multi-step regressive predication is presented.
本文导出了预测反卷积法(PDC)的多步迭代形式,以改进参数估计。
The algorithm uses a predictive model that makes whitening treatment of the obsevations and a predictive deconvolution (PDC) method to estimate the MA parameters.
本文导出了预测反卷积法(PDC)的多步迭代形式,以改进参数估计。
The algorithm uses a predictive model that makes whitening treatment of the obsevations and a predictive deconvolution (PDC) method to estimate the MA parameters.
应用推荐