A neural control strategy for nonlinear processes with time-variant time-delay is proposed in this paper. In this strategy, a dynamic neural network based nonlinear Smith predictor is constructed to compensate for the...
详细信息
A neural control strategy for nonlinear processes with time-variant time-delay is proposed in this paper. In this strategy, a dynamic neural network based nonlinear Smith predictor is constructed to compensate for the effect of time-delay of a class of nonlinear processes. An on-line optimizing controller is illustrated based on the neural Smith predictor. It is known that the performance of the Smith predictor may be deteriorated if the time-delay of the process changes with time. In order to improve the performance of the Smith predictor, a time-delay adaptation mechanism is introduced into the control structure to track the variation of the time-delay. The simulation, comparing with the classical Smith predictive control, on a continuous-stirred-tank-reactor (CSTR), where the time-delay of the manipulating flow changes with time, is used for the test.
The estimation for the nonlinear dynamic system with time-varying input time-delay is an important issue for system identification. In order to estimate the dynamics of the process, a dynamic neural network with exter...
详细信息
The estimation for the nonlinear dynamic system with time-varying input time-delay is an important issue for system identification. In order to estimate the dynamics of the process, a dynamic neural network with external recurrent structure is applied to the modelling procedure. In the case where time-delay is time varying, a useful way is to develop on-line time-delay estimation mechanisms to track the input time-delay variation. In this paper, two schemes respectively called direct as well as indirect time-delay estimators are proposed. Finally, two numerical examples are illustrated for the test of the proposed methods.
A control strategy for discrete time systems preceded with dominant hysteresis is presented in this paper. In this strategy, a recurrent neural network with hysteron (RNNH) is applied for the construction of the nonli...
详细信息
A control strategy for discrete time systems preceded with dominant hysteresis is presented in this paper. In this strategy, a recurrent neural network with hysteron (RNNH) is applied for the construction of the nonlinear compensator to remove the effect of the hysteresis. Then the incremental pole placement control approach is used for the control of the systems with the neural compensator. Finally, numerical simulation examples are illustrated to show the performance of the proposed strategy.
A dynamic neural network with a hidden layer that consists of wavelets for nonlinear dynamic system identification is presented. In order to model the dynamics of nonlinear dynamic systems, the external auto-regressiv...
详细信息
A dynamic neural network with a hidden layer that consists of wavelets for nonlinear dynamic system identification is presented. In order to model the dynamics of nonlinear dynamic systems, the external auto-regressive connection is introduced into the wavelet based neural network. For fast training of the wavelet neural network, a PID backpropagation algorithm is proposed. The procedure of using a wavelet neural network for modeling is described in detail in the article. Finally, an example for identification of a continuous-stirred-tank-reactor is given.
暂无评论