| This Paper:Browse 20 Download 0 |
码上扫一扫！ |
| Application of feedforward and recurrent neural networks for model-based control systems |
| MarekKrok1,WojciechP.Hunek1,SzymonMielczarek1,FilipBuchwald1,AdamKolender1 |
|
|
| (Department of Control Science and Engineering, Opole University of Technology, Opole, Poland) |
|
| 摘要: |
| In this paper, a new study concerning the usage of artificial neural networks in the control application is given. It is shown,
that the data gathered during proper operation of a given control plant can be used in the learning process to fully embrace the
control pattern. Interestingly, the instances driven by neural networks have the ability to outperform the original analytically
driven scenarios. Three different control schemes, namely perfect, linear-quadratic, and generalized predictive controllers
were used in the theoretical study. In addition, the nonlinear recurrent neural network-based generalized predictive controller
with the radial basis function-originated predictor was obtained to exemplify the main results of the paper regarding the
real-world application. |
| 关键词: Predictive control · Linear-quadratic control · Inverse problems · Feedforward network · Recurrent neural network · Optimization |
| DOI：https://doi.org/10.1007/s11768-024-00234-6 |
|
| 基金项目: |
|
| Application of feedforward and recurrent neural networks for model-based control systems |
| Marek Krok1,Wojciech P. Hunek1,Szymon Mielczarek1,Filip Buchwald1,Adam Kolender1 |
| (Department of Control Science and Engineering, Opole University of Technology, Opole, Poland) |
| Abstract: |
| In this paper, a new study concerning the usage of artificial neural networks in the control application is given. It is shown,
that the data gathered during proper operation of a given control plant can be used in the learning process to fully embrace the
control pattern. Interestingly, the instances driven by neural networks have the ability to outperform the original analytically
driven scenarios. Three different control schemes, namely perfect, linear-quadratic, and generalized predictive controllers
were used in the theoretical study. In addition, the nonlinear recurrent neural network-based generalized predictive controller
with the radial basis function-originated predictor was obtained to exemplify the main results of the paper regarding the
real-world application. |
| Key words: Predictive control · Linear-quadratic control · Inverse problems · Feedforward network · Recurrent neural network · Optimization |
