Arithmetic or functional expression in terms of optimization variables


An OptimizationExpression is an arithmetic or functional expression in terms of optimization variables. Use an OptimizationExpression as an objective function, or as a part of an inequality or equality in a constraint or equation.


Create an optimization expression by performing operations on OptimizationVariable objects. Use standard MATLAB® arithmetic including taking powers, indexing, and concatenation of optimization variables to create expressions. See Supported Operations on Optimization Variables and Expressions and Examples.

You can also create an optimization expression from a MATLAB function applied to optimization variables by using fcn2optimexpr. For examples, see Create Expression from Nonlinear Function and Problem-Based Nonlinear Optimization.

Create an empty optimization expression by using optimexpr. Typically, you then fill the expression in a loop. For examples, see Create Optimization Expression by Looping and the optimexpr function reference page.

After you create an expression, use it as either an objective function, or as part of a constraint or equation. For examples, see the solve function reference page.


expand all

Index names, specified as a cell array of strings or character vectors. For information on using index names, see Named Index for Optimization Variables.

Data Types: cell

This property is read-only.

Optimization variables in the object, specified as a structure of OptimizationVariable objects.

Data Types: struct

Object Functions

evaluateEvaluate optimization expression
showDisplay information about optimization object
writeSave optimization object description


collapse all

Create optimization expressions by arithmetic operations on optimization variables.

x = optimvar('x',3,2);
expr = sum(sum(x))
expr = 
  Linear OptimizationExpression

    x(1, 1) + x(2, 1) + x(3, 1) + x(1, 2) + x(2, 2) + x(3, 2)

f = [2,10,4];
w = f*x;
(1, 1)

  2*x(1, 1) + 10*x(2, 1) + 4*x(3, 1)

(1, 2)

  2*x(1, 2) + 10*x(2, 2) + 4*x(3, 2)

Create an optimization expression by transposing an optimization variable.

x = optimvar('x',3,2);
y = x'
y = 
  2x3 Linear OptimizationExpression array with properties:

    IndexNames: {{}  {}}
     Variables: [1x1 struct] containing 1 OptimizationVariable

  See expression formulation with show.

Simply indexing into an optimization array does not create an expression, but instead creates an optimization variable that references the original variable. To see this, create a variable w that is the first and third row of x. Note that w is an optimization variable, not an optimization expression.

w = x([1,3],:)
w = 
  2x2 OptimizationVariable array with properties:

  Read-only array-wide properties:
          Name: 'x'
          Type: 'continuous'
    IndexNames: {{}  {}}

  Elementwise properties:
    LowerBound: [2x2 double]
    UpperBound: [2x2 double]

  Reference to a subset of OptimizationVariable with Name 'x'.

  See variables with show.
  See bounds with showbounds.

Create an optimization expression by concatenating optimization variables.

y = optimvar('y',4,3);
z = optimvar('z',4,7);
f = [y,z]
f = 
  4x10 Linear OptimizationExpression array with properties:

    IndexNames: {{}  {}}
     Variables: [1x1 struct] containing 2 OptimizationVariables

  See expression formulation with show.

Use optimexpr to create an empty expression, then fill the expression in a loop.

y = optimvar('y',6,4);
expr = optimexpr(3,2);
for i = 1:3
    for j = 1:2
        expr(i,j) = y(2*i,j) - y(i,2*j);
(1, 1)

  y(2, 1) - y(1, 2)

(2, 1)

  y(4, 1) - y(2, 2)

(3, 1)

  y(6, 1) - y(3, 2)

(1, 2)

  y(2, 2) - y(1, 4)

(2, 2)

  y(4, 2) - y(2, 4)

(3, 2)

  y(6, 2) - y(3, 4)

Create an optimization expression corresponding to the objective function


x = optimvar('x');
f = @(x)x^2/10 + exp(-exp(-x));
fun = fcn2optimexpr(f,x)
fun = 
  Nonlinear OptimizationExpression



    anonymousFunction1 = @(x)x^2/10+exp(-exp(-x));

Find the point that minimizes fun starting from the point x0 = 0.

x0 = struct('x',0);
prob = optimproblem('Objective',fun);
[sol,fval] = solve(prob,x0)
Solving problem using fminunc.

Local minimum found.

Optimization completed because the size of the gradient is less than
the value of the optimality tolerance.
sol = struct with fields:
    x: -0.9595

fval = 0.1656

Create an optimization expression in two variables.

x = optimvar('x',3,2);
y = optimvar('y',1,2);
expr = sum(x,1) - 2*y;

Evaluate the expression at a point.

xmat = [3,-1;
sol.x = xmat;
sol.y = [4,-3];
val = evaluate(expr,sol)
val = 1×2

    -3    12

More About

expand all

Introduced in R2017b