Nonlinear systems identification and control using dynamic multi-time scales neural networks

2013 Fourth International Conference on Intelligent Control and Information Processing (ICICIP)

2013

Self Cite

This paper deals with the adaptive nonlinear identification and trajectory tracking via dynamic multilayer neural network with different time-scales. By means of a Lyapunov-like analysis we determine stability conditions for the identification. Based on the identification results, we design a sliding mode controller for the nonlinear system to track the trajectory of a reference model. The main contributions of the paper are: First, we extend our prior results of single-layer dynamic neural networks with multi-time scales to the multilayer case. Second, the e-modification in standard use in adaptive control is introduced in the on-line update laws to guarantee bounded weights, bounded identification and tracking errors. Simulation results are presented confirming the validity of the above approach.

“…The RMS values of all state variables demonstrate that the proposed NN algorithm has better performance than [18].…”

Section: Applicationmentioning

confidence: 88%

Section: Applicationmentioning

confidence: 97%

Section: Applicationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Indirect adaptive control of nonlinear system via dynamic multilayer neural networks with multi-time scales

Liu

2013 Fourth International Conference on Intelligent Control and Information Processing (ICICIP)

2013

Self Cite

“…Second, the adaptive control and identification for singularly perturbed dynamic systems via multiple time scales neural networks had not been taken into consideration. In our previous work [17], the structure of the neural identifier depends on the output signals from the actual system. This may risk the stability of the neural network because it is related to that of the real system.…”

mentioning

confidence: 99%

On-line nonlinear systems identification via dynamic neural networks with multi-time scales

Han

49th IEEE Conference on Decision and Control (CDC)

Ren³

2010

Self Cite

In this paper, an new on-line identification algorithm with dead-zone function is proposed for nonlinear systems identification via dynamic neural networks with different time-scales including the aspects of fast and slow phenomenon. The main contribution of the paper is that the Lyapunov function and singularly perturbed techniques are used to develop the on-line update laws for both dynamic neural networks weights and the linear part parameters of the neural network model. On example is also given to demonstrate the effectiveness of the proposed identification algorithm.

Indirect adaptive control of nonlinear system via dynamic multilayer neural networks with multi‐time scales

Zheng

Adaptive Control & Signal

et al. 2014

Self Cite

This paper deals with adaptive nonlinear identification and trajectory tracking problem via dynamic multilayer neural network with different time scales. By means of a Lyapunov-like analysis, we determine stability conditions for the on-line identification. Then, a sliding mode controller is designed for trajectory tracking with consideration of the modeling error and disturbance. The main contributions of the paper lie in the following aspects. First, we extend our prior identification results of single-layer dynamic neural networks with multi-time scales to those of multilayer case. Second, the e-modification in standard use in adaptive control is introduced in the on-line update laws to guarantee bounded weights and bounded identification errors. Third, the potential singularity problem in controller design is solved by using new update laws for the NN weights so that the control signal is guaranteed bounded. The stability of proposed controller is proved by using Lyapunov function. Simulation results demonstrate the effectiveness of the proposed algorithm. that a gradient descent algorithm for weight adjustment was stable and robust to any bounded uncertainties, including the optimal network approximation error [6]. All these researches are focused on the stability property analysis, such as the conditions that the system should satisfy to achieve global asymptotic stability, exponential stability, input-to-state stability, and short-term memory or long-term memory behavior of the neural networks. The research on identification and control for nonlinear systems by using dynamic multi-time scales neural networks is rare.Adaptive neural networks control is classified into two kinds of structure: indirect and direct adaptive control. In direct adaptive control, the parameters of the controller are changed directly without determining the characteristics of the process and its disturbance first [7,8,[14][15][16][17][18]. In indirect adaptive control, the parameters of controller are designed on the basis of the process model and possibly the disturbance model, which are determined first [9,10,[19][20][21][22][23][24][25]. In this paper, we focus on developing the indirect adaptive neural network controller for the nonlinear systems. So, the nonlinear system identification process turns out to be one of the central parts in constructing successful tracking controllers.System identification using neural networks has been investigated by many researchers. For instance, in [26], a multilayer feed forward neural network is utilized to identify the moving-average model of an induction motor. In [27], a robust identification method is proposed for nonlinear systems with unknown driving noise. Abdollahi et al. proposed a stable identification approach for multivariable nonlinear system based on nonlinear-in-parameters neural network in [28]. Neural networks are also utilized for modeling and identification of nonlinear dynamics for freeway traffic in [29]. Nevertheless, most researches performed before are based on feed ...