Journal: Automatika

Article Title: Model-Based Power Plant Master Control

doi: 10.7305/automatika.2014.12.434

Figure Lengend Snippet: Fig. 7. The Plant output performance for the case of in- creasing the value of hm by 100% for both the MPC con- troller (blue), and the existing PID controller (red). The variation in hm is a ramp starting at 500s and ending at 1000s.

Article Snippet: After successive iterations with some knowledge and experiences on power plants and MPC controller, the parameters of the Master MPC controller that give a good response and good robustness (as will be shown later) are as follows: Prediction horizon N = 30, Control horizon Nu = 12, Ts = 3 s, kint = 0.075, and weight matrices: Q = diag(40, 20, 9, 32), Pint = 0.01∗diag(4, 20, 3, 4), Qcon = 100∗I4∗4, Qinc = 0.125∗diag(40, 40, 0.01, 2) For the development of the optimization model and the simulation of the MPC controller, the MATLAB toolbox YALMIP [12] was used.

Techniques:

Journal: Automatika

Article Title: Model-Based Power Plant Master Control

doi: 10.7305/automatika.2014.12.434

Figure Lengend Snippet: Fig. 5. Plant response with the Master MPC controller (blue), and the existing PID controller (red) for load step from 27:29 MW at time = 200s.

Article Snippet: After successive iterations with some knowledge and experiences on power plants and MPC controller, the parameters of the Master MPC controller that give a good response and good robustness (as will be shown later) are as follows: Prediction horizon N = 30, Control horizon Nu = 12, Ts = 3 s, kint = 0.075, and weight matrices: Q = diag(40, 20, 9, 32), Pint = 0.01∗diag(4, 20, 3, 4), Qcon = 100∗I4∗4, Qinc = 0.125∗diag(40, 40, 0.01, 2) For the development of the optimization model and the simulation of the MPC controller, the MATLAB toolbox YALMIP [12] was used.

Techniques:

Journal: Automatika

Article Title: Model-Based Power Plant Master Control

doi: 10.7305/automatika.2014.12.434

Figure Lengend Snippet: Fig. 6. Plant output and Input variables for the on-site test performed at Idbäcken plant. Steps were carried out in set- point of Electric load, Steam Pressure, Steam temperature, O2 and Drum Level. Showing the performance with Mas- ter MPC controller (blue), and also with the Existing PID controller (red).

Article Snippet: After successive iterations with some knowledge and experiences on power plants and MPC controller, the parameters of the Master MPC controller that give a good response and good robustness (as will be shown later) are as follows: Prediction horizon N = 30, Control horizon Nu = 12, Ts = 3 s, kint = 0.075, and weight matrices: Q = diag(40, 20, 9, 32), Pint = 0.01∗diag(4, 20, 3, 4), Qcon = 100∗I4∗4, Qinc = 0.125∗diag(40, 40, 0.01, 2) For the development of the optimization model and the simulation of the MPC controller, the MATLAB toolbox YALMIP [12] was used.

Techniques: Serial Time-encoded Amplified Microscopy

Journal: Automatika

Article Title: Model-Based Power Plant Master Control

doi: 10.7305/automatika.2014.12.434

Figure Lengend Snippet: Fig. 9. The Plant output performance for the case of heat distribution ratio changed from 0.34 to 0.31, with the MPC controller (blue), and also with the existing PID controller (red). The heat distribution ratio disturbance for the su- perheater is shown in the bottom plot.

Article Snippet: After successive iterations with some knowledge and experiences on power plants and MPC controller, the parameters of the Master MPC controller that give a good response and good robustness (as will be shown later) are as follows: Prediction horizon N = 30, Control horizon Nu = 12, Ts = 3 s, kint = 0.075, and weight matrices: Q = diag(40, 20, 9, 32), Pint = 0.01∗diag(4, 20, 3, 4), Qcon = 100∗I4∗4, Qinc = 0.125∗diag(40, 40, 0.01, 2) For the development of the optimization model and the simulation of the MPC controller, the MATLAB toolbox YALMIP [12] was used.

Techniques:

Journal: Automatika

Article Title: Model-Based Power Plant Master Control

doi: 10.7305/automatika.2014.12.434

Figure Lengend Snippet: Fig. 8. The Plant output performance for the case of Fuel mass flow disturbance of value 15 kg/s, for both the MPC controller (blue), and the existing PID controller (red). The Fuel mass flow disturbance is shown in the lowest plot.

Article Snippet: After successive iterations with some knowledge and experiences on power plants and MPC controller, the parameters of the Master MPC controller that give a good response and good robustness (as will be shown later) are as follows: Prediction horizon N = 30, Control horizon Nu = 12, Ts = 3 s, kint = 0.075, and weight matrices: Q = diag(40, 20, 9, 32), Pint = 0.01∗diag(4, 20, 3, 4), Qcon = 100∗I4∗4, Qinc = 0.125∗diag(40, 40, 0.01, 2) For the development of the optimization model and the simulation of the MPC controller, the MATLAB toolbox YALMIP [12] was used.

Techniques:

Journal: Automatika

Article Title: Model-Based Power Plant Master Control

doi: 10.7305/automatika.2014.12.434

Figure Lengend Snippet: Fig. 10. Output variables for the case of the existence of Fuel flow disturbance, hm parameter variation and heat distribution ratio disturbances, with the MPC controller (blue), and with the existing PID controller (red).

Article Snippet: After successive iterations with some knowledge and experiences on power plants and MPC controller, the parameters of the Master MPC controller that give a good response and good robustness (as will be shown later) are as follows: Prediction horizon N = 30, Control horizon Nu = 12, Ts = 3 s, kint = 0.075, and weight matrices: Q = diag(40, 20, 9, 32), Pint = 0.01∗diag(4, 20, 3, 4), Qcon = 100∗I4∗4, Qinc = 0.125∗diag(40, 40, 0.01, 2) For the development of the optimization model and the simulation of the MPC controller, the MATLAB toolbox YALMIP [12] was used.

Techniques: