Include Parameters or Data

Typically, you have data or values to pass to the solver. Place those values in the input section (where you see x0) and run the section by choosing Section > Run Section or pressing Control+Enter.

Set the initial point x0 and scale a for the optimization.

x0 = [2;1];
a = 100;

Place the x0 value and any other problem data into the workspace by running this section before proceeding.

Optimize Live Editor Task

This task has objective and nonlinear constraint functions included. To change these functions, edit the function listings below the task.

To change the constraints, select appropriate constraint types and enter values in the input boxes. You might need to enter values in the section containing x0 above, and run the section to put values in the workspace.

Run the task by clicking the striped bar to the left, or by choosing Run or Section > Run Section, or by pressing Control+Enter.

Local minimum found that satisfies the constraints.

Optimization completed because the objective function is non-decreasing in 
feasible directions, to within the value of the optimality tolerance,
and constraints are satisfied to within the value of the constraint tolerance.

<stopping criteria details>

Results

Optimize saves the solution to the workspace variable solution, and saves the objective function value at the solution to the workspace variable objectiveValue. You can see and modify these variable names at the top of the Optimize task.

View these variables.

solution

solution = 2×1

    1.1413
    1.3029

objectiveValue

objectiveValue = 0.0200

View the nonlinear constraint function values at the solution.

[ccons,ceqcons] = constraintFcn(solution)

ccons = 1×2

   -2.0000   -0.0000

ceqcons =

     []

Helper Functions — Local Functions

The following code creates the objective function. Modify this code for your problem.

function f = objectiveFcn(x,a)
f = a*(x(2) - x(1)^2)^2 + (1 - x(1))^2;
end

The following code creates the constraint function. Modify this code for your problem.

function [c,ceq] = constraintFcn(x)
c(1) = x(1)^2 + x(2)^2 - 5;
c(2) = 3 - x(1)^2 - x(2)^2;
ceq = [];  % No equality constraints
end