1887
banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Identification and control of plasma vertical position using neural network in Damavand tokamak
Rent:
Rent this article for
USD
10.1063/1.4791925
/content/aip/journal/rsi/84/2/10.1063/1.4791925
http://aip.metastore.ingenta.com/content/aip/journal/rsi/84/2/10.1063/1.4791925

Figures

Image of FIG. 1.
FIG. 1.

Block diagram of closed loop system to control plasma vertical position in Damavand tokamak.

Image of FIG. 2.
FIG. 2.

Tracking of plasma vertical position with PD controller (k p = 1, k d = 5).

Image of FIG. 3.
FIG. 3.

A sample of shots for model training.

Image of FIG. 4.
FIG. 4.

Execution of preprocessor program for filtering, offset rejection, time interval selection, and increasing of the sampling time.

Image of FIG. 5.
FIG. 5.

(a) Block diagram of the closed loop system and the general structure of simulator model. (b) Simulator model structure used for identification.

Image of FIG. 6.
FIG. 6.

Error of test data with respect to the number of neurons for different sample times.

Image of FIG. 7.
FIG. 7.

%RMSE with respect to the number of epochs for different sampling time.

Image of FIG. 8.
FIG. 8.

Comparison of output of the simulator model with experimental output for shot #5.

Image of FIG. 9.
FIG. 9.

Comparison of output of simulator model with experimental output in test for shot #1.

Image of FIG. 10.
FIG. 10.

Execution of the simulator model using real control input signal of the system.

Image of FIG. 11.
FIG. 11.

Comparison of Z p and I cz with the simulator model outputs and in the closed loop system.

Image of FIG. 12.
FIG. 12.

Comparison of Z p and I cz with the simulation model outputs of and in the closed loop system.

Image of FIG. 13.
FIG. 13.

Block diagram of complete structure of the closed loop simulator using neural controller.

Image of FIG. 14.
FIG. 14.

Block diagram of neural identifier G c (s).

Image of FIG. 15.
FIG. 15.

Input PRBS signal, u output of G c (s) and output of neural controller are shown.

Image of FIG. 16.
FIG. 16.

Comparison of experimental data of shot #10 with results of closed loop control of model with offline NNC.

Image of FIG. 17.
FIG. 17.

Comparison of the performance of the closed loop control of the model with NNC controller tuned online with experimental data of PD controller for shots #10, 22, and 26.

Image of FIG. 18.
FIG. 18.

The effect of command pulse frequency of driver on its operating frequency with NNC to improve the quality of the plasma vertical position control.

Tables

Generic image for table
Table I.

Performance of the models for data of shot 5.

Generic image for table
Table II.

Results of RMSE for shots of Figure 17 using NNC and PD controllers.

Generic image for table
Table III.

Comparison between the RMSE values of existed controller in shots #22 and 26 using NNC and PD controllers.

Loading

Article metrics loading...

/content/aip/journal/rsi/84/2/10.1063/1.4791925
2013-02-21
2014-04-19
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Identification and control of plasma vertical position using neural network in Damavand tokamak
http://aip.metastore.ingenta.com/content/aip/journal/rsi/84/2/10.1063/1.4791925
10.1063/1.4791925
SEARCH_EXPAND_ITEM