Nonlinear Control Design Blockset    

Nonlinear Control Design Blockset Global Variable ncdStruct

This section discusses the fields in the global variable ncdStruct that is created in your workspace when using the Nonlinear Control Design Blockset. Clearing these variables or changing them in unexpected ways causes the Nonlinear Control Design Blockset to generate errors from which you may not be able to recover.

These variables are associated with the Nonlinear Control Design Blockset optimization problem.

ncdStruct.TvarStr
Single row string containing the names of the tunable variables. This string is changed via the Tunable Variables edit field on the Tunable Variables dialog box.
ncdStruct.TvlbStr
Single row string which, when evaluated, defines a lower bound on the tunable variables. This string is changed via the Lower bounds edit field on the Tunable Variables dialog box.
ncdStruct.TvubStr
Single row string which, when evaluated, defines an upper bound on the tunable variables. This string is changed via the Upper bounds edit field on the Tunable Variables dialog box
ncdStruct.Tdelta
Scalar defining the discretization interval. This scalar is changed via the Discretization interval edit field on the Tunable Variables dialog box.
ncdStruct.CostFlag
Boolean equal to 1 if optimization should continue even after achieving the constraints. This Boolean is changed via the Stop optimization as soon as the constraints are achieved check box on the Tunable Variables dialog box.
ncdStruct.GradFlag
Boolean equal to 1, if the optimization routine is specified to use an accurate method for gradients. This Boolean is changed via the Compute gradients with better accuracy (slower) check box on the Tunable Variables dialog box.
ncdStruct.OptmOptns
Vector sets various optimization options as defined in the Optimization Toolbox documentation. Some elements of this vector are changed via the Display check box and Terminate for x and Terminate for g edit fields on the Tunable Parameters dialog box.

Variables associated with plant uncertainty.

ncdStruct.UvarStr
Single row string containing the names of the uncertain variables. This string is changed via the Uncertain Variables edit field on the Uncertain Variables dialog box.
ncdStruct.UvlbStr
Single row string which, when evaluated, defines a lower bound on the uncertain variables. This string is changed via the Lower bound edit field on the Uncertain Variables dialog box.
ncdStruct.UvubStr
Single row string which, when evaluated, defines a lower bound on the uncertain variables. This string is changed via the Upper bound edit field on the Uncertain Variables dialog box.
ncdStruct.NumMC
Scalar number of Monte Carlo plants to be generated.
ncdStruct.PlntON
1-by-4 vector of Booleans relating whether nominal plant, lower bound uncertain plant, upper bound uncertain plant, and Monte Carlo plants are to be included in the optimization.

Variables associated with the interface.

ncdStruct.RefSgnl
String matrix used for plotting reference signals for the various outputs.
ncdStruct.RngLmts
A 3-by-2N matrix, where N is the number NCD blocks, containing the x- and y-axis limits for constraint figures. The first row specifies the port number, with two columns for each NCD block. The second and third rows contain [xmin xmax] and [ymin ymax] axis limits, respectively, for the port specified in the corresponding column in the first row.
ncdStruct.SavdFlg
Boolean equal to 1 if no changes have been made to the constraint bounds, optimization variables, or uncertain variables since data has last been saved.
ncdStruct.SysName
Boolean equal to 1 if no changes have been made to the constraint bounds, optimization variables, or uncertain variables since data has last been saved.

Additional global variables associated with program flow (and communication between constr and the Nonlinear Control Design Blockset).

OPT_STOP
Boolean equal to 1 if the Stop button has been pushed.
OPT_STEP
Boolean equal to 1 if the optimization is presently evaluating the cost function of a major iteration. Used as a flag for plotting during optimization.


  Representation of Time-Domain Constraints LQR Design for Inverted Pendulum