Logical Conditions in Optimization

Main.LogicalConditions History

Hide minor edits - Show changes to output

April 30, 2021, at 09:32 PM by 10.35.117.248 -
Deleted lines 3-4:

!!! Conditional Statements in Optimization
Changed line 61 from:
m.Obj((y+3)**2)
to:
m.Minimize((y+3)**2)
Changed line 86 from:
m.Obj((y+3)**2)
to:
m.Minimize((y+3)**2)
June 21, 2020, at 04:43 AM by 136.36.211.159 -
Deleted lines 318-336:

----

(:html:)
 <div id="disqus_thread"></div>
    <script type="text/javascript">
        /* * * CONFIGURATION VARIABLES: EDIT BEFORE PASTING INTO YOUR WEBPAGE * * */
        var disqus_shortname = 'apmonitor'; // required: replace example with your forum shortname

        /* * * DON'T EDIT BELOW THIS LINE * * */
        (function() {
            var dsq = document.createElement('script'); dsq.type = 'text/javascript'; dsq.async = true;
            dsq.src = 'https://' + disqus_shortname + '.disqus.com/embed.js';
            (document.getElementsByTagName('head')[0] || document.getElementsByTagName('body')[0]).appendChild(dsq);
        })();
    </script>
    <noscript>Please enable JavaScript to view the <a href="https://disqus.com/?ref_noscript">comments powered by Disqus.</a></noscript>
    <a href="https://disqus.com" class="dsq-brlink">comments powered by <span class="logo-disqus">Disqus</span></a>
(:htmlend:)
March 31, 2019, at 05:34 AM by 45.56.3.173 -
Added lines 227-251:
plt.ylabel('y')
plt.show()
(:sourceend:)
(:divend:)

(:toggle hide gekko6 button show='Show Python GEKKO if3 example':)
(:div id=gekko6:)

%width=550px%Attach:if3_gekko.png

(:source lang=python:)
import numpy as np
import matplotlib.pyplot as plt
from gekko import GEKKO
m = GEKKO(remote=False)
p = m.Param(value=np.linspace(0,10,41))
# conditional statement
y = m.if3(p-4,p**2,-0.2*(p-4)+7)
m.options.IMODE = 2
m.solve()
lbl = r'$y=\mathrm{if3}(p-4,p^2,-0.2(p-4)+7)$'
plt.plot(p,y,'bo',label=lbl)
plt.text(1,5,r'$p^2$')
plt.text(5,10,r'$-0.2 (p-4)+7$')
plt.legend(loc=4)
March 31, 2019, at 01:19 AM by 12.29.221.219 -
Added lines 202-230:

(:toggle hide gekko5 button show='Show Python GEKKO min2 and max3 examples':)
(:div id=gekko5:)

%width=550px%Attach:min2_max3_gekko.png

(:source lang=python:)
import numpy as np
import matplotlib.pyplot as plt
from gekko import GEKKO
m = GEKKO(remote=False)
p = m.Param(value=np.linspace(10,20,21))
x = m.Var()
m.Equation(x==p)
# with MPCCs
y2 = m.min2(p,15)
# with integer variables
y3 = m.max3(p,16)
m.options.IMODE = 2
m.solve()
plt.plot(p,x,'b-',label='x')
plt.plot(p,y2,'g:',label='MPCC')
plt.plot(p,y3,'r--',label='Integer Switch')
plt.legend()
plt.xlabel('x')
plt.ylabel('y')
plt.show()
(:sourceend:)
(:divend:)
March 31, 2019, at 01:15 AM by 63.110.156.42 -
Changed lines 196-197 from:
* Max Function
* Min Function
to:
* Max Function (see Python GEKKO max2 and max3 functions)
* Min Function (see Python GEKKO min2 and min3 functions)
Changed line 199 from:
* Piece-wise Linear Functions
to:
* Piece-wise Linear Functions (see Python GEKKO pwl function)
February 14, 2019, at 12:30 PM by 10.37.111.106 -
Changed line 108 from:
# calculate x to minimize objective
to:
# parameter
Changed lines 127-135 from:
(:divend:)

* [[https://apmonitor.com/online/view_pass.php?f=abs4.apm|Example 4: Absolute Value at the Discontinuity]]

(:toggle hide gekko4 button show='Show Example 4 with Python GEKKO Binary Variable':)
(:div id=gekko4:)

'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] binary variable in optimization (Success)'''

to:
Added line 129:
# with abs3 object
Changed lines 133-143 from:
# calculate x to minimize objective
x =
m.Var(1.0)
# define new binary variable
intb
= m.Var(0,lb=0,ub=1,integer=True)
# define y
y = m.Var()
# define equations
m.Equation((1-intb)*x <= 0)
m.Equation(intb * (-x) <= 0)
# output
m.Equation(y==(1-intb)*(-x) + intb*
x)
to:
# parameter
x = m.Param(-0.5)
# calculate y=abs(x) with abs3
y
= m.abs3(x)
Deleted line 137:
m.options.SOLVER=1
Added lines 141-171:
print('y: ' + str(y.value))
(:sourceend:)
(:divend:)

* [[https://apmonitor.com/online/view_pass.php?f=abs4.apm|Example 4: Absolute Value at the Discontinuity]]

(:toggle hide gekko4 button show='Show Example 4 with Python GEKKO Binary Variable':)
(:div id=gekko4:)

'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] binary variable in optimization (Success)'''

(:source lang=python:)
from gekko import GEKKO
# define new GEKKO model
m = GEKKO()
# calculate x to minimize objective
x = m.Var(1.0)
# define new binary variable
intb = m.Var(0,lb=0,ub=1,integer=True)
# define y
y = m.Var()
# define equations
m.Equation((1-intb)*x <= 0)
m.Equation(intb * (-x) <= 0)
# output
m.Equation(y==(1-intb)*(-x) + intb*x)
# solve with APOPT (MINLP solver)
m.options.SOLVER=1
m.solve()
# print solution
print('x: ' + str(x.value))
Added lines 173-188:
print('y: ' + str(y.value))
(:sourceend:)

(:source lang=python:)
# with abs3 object
from gekko import GEKKO
# define new GEKKO model
m = GEKKO()
# variable
x = m.Var(-0.5)
# calculate y=abs(x) with abs3
y = m.abs3(x)
# solve with APOPT (MINLP solver)
m.solve()
# print solution
print('x: ' + str(x.value))
February 12, 2019, at 03:58 PM by 10.37.233.114 -
Deleted lines 8-9:
Attach:download.jpg [[Attach:mpcc_sine.zip|Sine Wave Clipping]]
Added lines 231-232:

Attach:download.jpg [[Attach:mpcc_sine.zip|Sine Wave Clipping]]
February 12, 2019, at 03:57 PM by 10.37.233.114 -
Changed line 136 from:
'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] binary variable in simulation (Success)'''
to:
'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] binary variable in optimization (Success)'''
February 12, 2019, at 03:56 PM by 10.37.233.114 -
Changed lines 49-51 from:
(:toggle hide gekko2 button show='Show Example 2 with Python GEKKO ABS (Failure)':)
(:div id=gekko2:)
to:
(:toggle hide gekko2a button show='Show Example 2 with Python GEKKO ABS (Failure)':)
(:div id=gekko2a:)
Changed lines 74-76 from:
(:toggle hide gekko2 button show='Show Example 2 with Python GEKKO ABS2 (Success)':)
(:div id=gekko2:)
to:
(:toggle hide gekko2b button show='Show Example 2 with Python GEKKO ABS2 (Success)':)
(:div id=gekko2b:)
Changed line 143 from:
x = m.Param(-0.5)
to:
x = m.Var(1.0)
February 12, 2019, at 03:54 PM by 10.37.233.114 -
Changed lines 23-24 from:
'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] ABS function (abs2)'''
to:
(:toggle hide gekko1 button show='Show Example 1 with Python GEKKO':)
(:div id=gekko1:)
Changed lines 30-32 from:
# use abs2 to define a new variable
y = m.abs2(x)
# use abs2 in an equation
to:
# use abs to define a new variable
y = m.abs(x)
# use abs in an equation
Changed lines 34-35 from:
m.Equation(z==m.abs2(x)+1)
to:
m.Equation(z==m.abs(x)+1)
Changed lines 43-44 from:
to:
(:divend:)
Added lines 49-51:
(:toggle hide gekko2 button show='Show Example 2 with Python GEKKO ABS (Failure)':)
(:div id=gekko2:)

Changed lines 70-71 from:
to:
(:divend:)
Added lines 74-76:
(:toggle hide gekko2 button show='Show Example 2 with Python GEKKO ABS2 (Success)':)
(:div id=gekko2:)

Changed lines 95-96 from:
to:
(:divend:)
Added lines 100-130:

(:toggle hide gekko3 button show='Show Example 3 with Python GEKKO Binary Variable':)
(:div id=gekko3:)

'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] binary variable in simulation (Success)'''

(:source lang=python:)
from gekko import GEKKO
# define new GEKKO model
m = GEKKO()
# calculate x to minimize objective
x = m.Param(-0.5)
# define new binary variable
intb = m.Var(0,lb=0,ub=1,integer=True)
# define y
y = m.Var()
# define equations
m.Equation((1-intb)*x <= 0)
m.Equation(intb * (-x) <= 0)
# output
m.Equation(y==(1-intb)*(-x) + intb*x)
# solve with APOPT (MINLP solver)
m.options.SOLVER=1
m.solve()
# print solution
print('x: ' + str(x.value))
print('intb: ' + str(intb.value))
print('y: ' + str(y.value))
(:sourceend:)
(:divend:)

Added lines 132-161:

(:toggle hide gekko4 button show='Show Example 4 with Python GEKKO Binary Variable':)
(:div id=gekko4:)

'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] binary variable in simulation (Success)'''

(:source lang=python:)
from gekko import GEKKO
# define new GEKKO model
m = GEKKO()
# calculate x to minimize objective
x = m.Param(-0.5)
# define new binary variable
intb = m.Var(0,lb=0,ub=1,integer=True)
# define y
y = m.Var()
# define equations
m.Equation((1-intb)*x <= 0)
m.Equation(intb * (-x) <= 0)
# output
m.Equation(y==(1-intb)*(-x) + intb*x)
# solve with APOPT (MINLP solver)
m.options.SOLVER=1
m.solve()
# print solution
print('x: ' + str(x.value))
print('intb: ' + str(intb.value))
print('y: ' + str(y.value))
(:sourceend:)
(:divend:)
February 12, 2019, at 03:40 PM by 10.37.233.114 -
Changed lines 23-24 from:
'''GEKKO ABS function (abs2)'''
to:
'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] ABS function (abs2)'''
Changed lines 48-88 from:
In this case, the IPOPT solver reaches the default maximum number of iterations (100) and reports a failure to converge. Now if the problem is reformulated with an additional variable ''intb'', the optimizer can find a solution to the same problem that failed with the abs(x) operator. The value of ''intb'' is a binary variable that can be either ''0'' or ''1''. The value of ''intb'' is zero when ''x<0'' and is one when ''x>=0''. The following two examples show that this reformulation allows the solver to quickly find a solution either away from or at the discontinuity.
to:
'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] ABS function in optimization (Failure)'''

(:source lang=python:)
from gekko import GEKKO
# define new GEKKO model
m = GEKKO()
# calculate x to minimize objective
x = m.Var(1)
# use abs to define y
y = m.abs(x)
# define objective to minimize
m.Obj((y+3)**2)
# solve
m.solve()
# print solution
print('x: ' + str(x.value))
print('y: ' + str(y.value))
(:sourceend:)

In this case, the IPOPT solver reaches the default maximum number of iterations (100) and reports a failure to converge. However, if another form with continuous first and second derivatives ('''abs2''' in GEKKO) is used, a solution is found.

'''[[https://apmonitor.com/wiki/index.php/Main/GekkoPythonOptimization|Python GEKKO]] ABS2 function in optimization (Success)'''

(:source lang=python:)
from gekko import GEKKO
# define new GEKKO model
m = GEKKO()
# calculate x to minimize objective
x = m.Var(1)
# use abs2 to define y
y = m.abs2(x)
# define objective to minimize
m.Obj((y+3)**2)
# solve
m.solve()
# print solution
print('x: ' + str(x.value))
print('y: ' + str(y.value))
(:sourceend:)

Also, if the problem is reformulated with an additional binary variable ''intb'', the optimizer can find a solution to the same problem that failed with the abs(x) operator. The value of ''intb'' is a binary variable that can be either ''0'' or ''1''. The value of ''intb'' is zero when ''x<0'' and is one when ''x>=0''. The following two examples show that this reformulation allows the solver to quickly find a solution either away from or at the discontinuity.
February 12, 2019, at 03:29 PM by 10.37.233.114 -
Added lines 22-42:

'''GEKKO ABS function (abs2)'''

(:source lang=python:)
from gekko import GEKKO

m = GEKKO()
x = m.Param(-1)
# use abs2 to define a new variable
y = m.abs2(x)
# use abs2 in an equation
z = m.Var()
m.Equation(z==m.abs2(x)+1)

# solve
m.solve()

print('x: ' + str(x.value))
print('y: ' + str(y.value))
print('z: ' + str(z.value))
(:sourceend:)
April 13, 2017, at 01:01 PM by 45.56.3.173 -
Changed line 9 from:
Attach:download.png [[Attach:mpcc_sine.zip|Sine Wave Clipping]]
to:
Attach:download.jpg [[Attach:mpcc_sine.zip|Sine Wave Clipping]]
April 13, 2017, at 01:01 PM by 45.56.3.173 -
Changed lines 7-15 from:
Conditional statements can cause problems for gradient-based optimization algorithms because they are often posed in a form that gives discontinuous functions, first derivatives, or second derivatives. For example, the absolute value operator is a continuous function but gives a discontinuous first derivative at the origin.
to:
Conditional statements can cause problems for gradient-based optimization algorithms because they are often posed in a form that gives discontinuous functions, first derivatives, or second derivatives.

Attach:download.png [[Attach:mpcc_sine.zip|Sine Wave Clipping]]

(:html:)
<iframe width="560" height="315" src="https://www.youtube.com/embed/NPUNixLGlh0" frameborder="0" allowfullscreen></iframe>
(:htmlend:)

For example, the absolute value operator is a continuous function but gives a discontinuous first derivative at the origin.
April 11, 2017, at 10:32 AM by 10.5.113.121 -
Changed lines 83-93 from:
!!!! MPECs: Mathematical Programs with Equilibrium Constraints

Mathematical Programs with Equilibrium Constraints (MPECs) are formulations that can also be used to model certain classes of discrete events.  MPECs can be more efficient than solving mixed integer formulations of the optimization problems because it avoids the combinatorial difficulties of searching for optimal discrete variables.

* [[https://apmonitor.com/online/view_pass.php?f=abs_mpec.apm|ABS: Absolute Value Operator (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=max_mpec.apm|MAX: Maximum Operator (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=min_mpec.apm|MIN: Minimum Operator (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl0.apm|Piecewise Linear Function without Object Use (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl1.apm|Piecewise Linear Function with PWL Object (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl2.apm|Piecewise Linear Function with LOOKUP and PWL Objects (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=sign_mpec.apm|SIGN: Signum Operator (MPEC Form)]]
to:
!!!! MPCCs: Mathematical Programs with Complementarity Constraints

Mathematical Programs with Complementarity Constraints (MPCCs) are formulations that can also be used to model certain classes of discrete events.  MPCCs can be more efficient than solving mixed integer formulations of the optimization problems because it avoids the combinatorial difficulties of searching for optimal discrete variables.

* [[https://apmonitor.com/online/view_pass.php?f=abs_mpec.apm|ABS: Absolute Value Operator (MPCC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=max_mpec.apm|MAX: Maximum Operator (MPCC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=min_mpec.apm|MIN: Minimum Operator (MPCC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl0.apm|Piecewise Linear Function without Object Use (MPCC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl1.apm|Piecewise Linear Function with PWL Object (MPCC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl2.apm|Piecewise Linear Function with LOOKUP and PWL Objects (MPCC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=sign_mpec.apm|SIGN: Signum Operator (MPCC Form)]]
April 11, 2017, at 10:31 AM by 10.5.113.121 -
Changed line 37 from:
* Mathematical Programs with Equilibrium Constraints (MPECs)
to:
* Mathematical Programs with Equilibrium/Complementarity Constraints (MPECs/MPCCs)
August 02, 2016, at 05:49 PM by 10.5.113.135 -
Added line 29:
* [[https://apmonitor.com/me575/index.php/Main/MiniMax|Minimax and Maximin]]
February 15, 2013, at 07:43 PM by 128.187.97.21 -
Changed lines 33-34 from:
Two popular methods for reformulation of the conditional statements or operators is to either use:
to:
Two popular methods for reformulation of the conditional statements is to either use:
Changed lines 38-39 from:
Each of these techniques are described below with a number of examples. The approach taken above with the ABS function used a single binary variable. The ABS function can also be expressed as an MPEC as shown below.
to:
Each of these techniques are described below with a number of examples.
Deleted lines 41-42:
* [[https://apmonitor.com/online/view_pass.php?f=abs4.apm|ABS: Absolute Value Operator (Integer Form)]]
Added lines 44-49:
The approach taken above with the ABS function used a single binary variable.

* [[https://apmonitor.com/online/view_pass.php?f=abs4.apm|ABS: Absolute Value Operator (Integer Form)]]

----

Added lines 68-81:
Piece-wise Linear (PWL) functions are one method to approximate any nonlinear function. The following example programs demonstrate a piece-wise linear function with binary decision variables. The sum of the binary decision variables is required to be one, meaning that only one of the linear approximations can be active at a time.

* [[https://apmonitor.com/online/view_pass.php?f=pwl3.apm|Simulate Piecewise Linear Function with Binary Decision Variables]]

The value of ''x'' is now declared as a decision variable and is adjusted by the optimizer to minimize the objective function that is equal to ''y''. By inspection, the optimal value is at ''x=2'' giving a result of ''y=0''. 

* [[https://apmonitor.com/online/view_pass.php?f=pwl4.apm|Optimize Piecewise Linear Function with Binary Decision Variables]]

Attach:pwl_function.png

The example shown above demonstrates a PWL function with one input and one output but it can also be extended to cases with multiple inputs and multiple outputs.

----

Added lines 89-91:
* [[https://apmonitor.com/online/view_pass.php?f=pwl0.apm|Piecewise Linear Function without Object Use (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl1.apm|Piecewise Linear Function with PWL Object (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=pwl2.apm|Piecewise Linear Function with LOOKUP and PWL Objects (MPEC Form)]]
February 15, 2013, at 07:05 PM by 128.187.97.21 -
Changed lines 44-45 from:
Attach:if_else_2_functions.png
to:
----

The following examples demonstrate the use of conditional statements to switch between different
functions. In this case, the function ''y=6*exp(-x)+1'' if the value of ''x<3'' else the function ''y=x'' is chosen when ''x>=3''. The simulate example verifies that the correct value is reported with a fixed value of ''x''. The second example shows how ''x'' can be adjusted to find a minimum value of ''y'' or ''f(x)''.
Changed lines 51-52 from:
Attach:if_elseif_else_3_functions.png
to:
Attach:if_else_2_functions.png

----

With three functions an if...elseif...else structure may be used. The following two examples show how to implement these conditional statements to achieve continuous first and second derivatives.

Added lines 60-62:
Attach:if_elseif_else_3_functions.png

----
February 15, 2013, at 06:57 PM by 128.187.97.21 -
Changed lines 15-16 from:
The solver can still converge because the solution is away from the discontinuity. In Example 2, the value of ''x'' is now determined by the optimizer. The objective function to minimize (y+3)'^2^' means that the optimal solution is at ''x=0'' and ''y==0''.  However, when the solution is at the discontinuity, a gradient based solver will often fail to converge.
to:
The solver can still converge because the solution is away from the discontinuity. In Example 2, the value of ''x'' is now determined by the optimizer. The objective function to minimize (y+3)'^2^' means that the optimal solution is at ''x=0'' and ''y=0''.  However, when the solution is at the discontinuity, a gradient based solver will often fail to converge.
Changed lines 40-42 from:
!!!! Logical Conditions with Integer Variables

to:
!!!! Logical Conditions with Binary Variables

* [[https://apmonitor.com/online/view_pass.php?f=abs4.apm|ABS: Absolute Value Operator (Integer Form)]]

Attach:if_else_2_functions.png

* [[https://apmonitor.com/online/view_pass.php?f=if1a.apm|IF: Conditional Statement with Two Functions - Simulate (Integer Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=if1b.apm|IF: Conditional Statement with Two Functions - Optimize (Integer Form)]]

Attach:if_elseif_else_3_functions.png

* [[https://apmonitor.com/online/view_pass.php?f=if2a.apm|IF: Conditional Statement with Three Functions - Simulate (Integer Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=if2b.apm|IF: Conditional Statement with Three Functions - Optimize (Integer Form)]]
February 15, 2013, at 06:47 PM by 128.187.97.21 -
Changed line 30 from:
* [[https://en.wikipedia.org/wiki/Signum_function|Signum Function]]
to:
* Signum Function ([[https://en.wikipedia.org/wiki/Signum_function|What is Signum?]])
February 15, 2013, at 06:46 PM by 128.187.97.21 -
Changed lines 5-6 from:
!!!! Conditional Statements in Optimization
to:
!!! Conditional Statements in Optimization
Changed lines 13-14 from:
* [[https://apmonitor.com/online/view_pass.php?f=abs1.apm|Example 1: Absolute Value Operator]]
to:
* [[https://apmonitor.com/online/view_pass.php?f=abs1.apm|Example 1: Absolute Value Operator Away from the Discontinuity]]
Changed lines 17-22 from:
* [[https://apmonitor.com/online/view_pass.php?f=abs1.apm|Example 2: Absolute Value Operator in Optimization]]




to:
* [[https://apmonitor.com/online/view_pass.php?f=abs2.apm|Example 2: Absolute Value Operator at the Discontinuity]]

In this case, the IPOPT solver reaches the default maximum number of iterations (100) and reports a failure to converge. Now if the problem is reformulated with an additional variable ''intb'', the optimizer can find a solution to the same problem that failed with the abs(x) operator. The value of ''intb'' is a binary variable that can be either ''0'' or ''1''. The value of ''intb'' is zero when ''x<0'' and is one when ''x>=0''. The following two examples show that this reformulation allows the solver to quickly find a solution either away from or at the discontinuity.

* [[https://apmonitor.com/online/view_pass.php?f=abs3.apm|Example 3: Absolute Value away from the Discontinuity]]
* [[https://apmonitor.com/online/view_pass.php?f=abs4.apm|Example 4: Absolute Value at the Discontinuity]]

Added line 25:
Changed lines 29-62 from:
* Sign Operator
* Most other discontinuous
functions





!!!! MPEC: Mathematical Programs with Equilibrium Constraints

Mathematical Programs with Equilibrium Constraints (MPECs) are formulations that can be used to model certain classes of discrete events. 
MPECs can be more efficient than solving mixed integer formulations of the optimization problems because it avoids the combinatorial difficulties of searching for optimal discrete variables.

----

!!! SIGN Operator (MPEC Form)

* %list list-page% [[Attach:sign.apm | SIGN Operator Example]]

----

!!! Absolute Value (ABS) Operator

* %list list-page% [[Attach
:abs.apm | ABS Operator Example]]

----

!!! Minimum Selector (MIN) Operator

* %list list-page% [[Attach:min.apm | MIN Operator Example]]

----

!!! Maximum Selector (MAX) Operator

* %list list-page% [[Attach:max.apm | MAX Operator Example
]]
to:
* Piece-wise Linear Functions
* [[https://en.wikipedia.org/wiki/Signum_function|Signum Function]]
* Other discontinuous
functions

Two popular methods for reformulation of the conditional statements or operators is to either use:

* Binary or integer variables
* Mathematical Programs with Equilibrium Constraints (
MPECs)

Each of these techniques are described below with a number
of examples. The approach taken above with the ABS function used a single binary variable. The ABS function can also be expressed as an MPEC as shown below.

!!!! Logical Conditions with Integer Variables




!!!! MPECs
: Mathematical Programs with Equilibrium Constraints

Mathematical Programs with Equilibrium Constraints (MPECs) are formulations that can also be used to model certain classes of discrete events.  MPECs can be more efficient than solving mixed integer formulations of the optimization problems because it avoids the combinatorial difficulties of searching for optimal discrete variables.

* [[https://apmonitor.com/online/view_pass.php?f=abs_mpec.apm|ABS: Absolute Value Operator (MPEC Form)
]]
* [[https://apmonitor.com/online/view_pass.php?f=max_mpec.apm|MAX: Maximum Operator (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=min_mpec.apm|MIN: Minimum Operator (MPEC Form)]]
* [[https://apmonitor.com/online/view_pass.php?f=sign_mpec.apm|SIGN: Signum Operator (MPEC Form)]]

February 15, 2013, at 06:28 PM by 128.187.97.21 -
Changed lines 11-30 from:
Minimizing this function can cause a problem for solvers that rely on first and second derivative information to determine search directions for each iteration.
to:
Minimizing this function can cause a problem for solvers that rely on first and second derivative information to determine search directions at each iteration. As an exercise, solve the following model that uses the absolute value operator.

* [[https://apmonitor.com/online/view_pass.php?f=abs1.apm|Example 1: Absolute Value Operator]]

The solver can still converge because the solution is away from the discontinuity. In Example 2, the value of ''x'' is now determined by the optimizer. The objective function to minimize (y+3)'^2^' means that the optimal solution is at ''x=0'' and ''y==0''.  However, when the solution is at the discontinuity, a gradient based solver will often fail to converge.

* [[https://apmonitor.com/online/view_pass.php?f=abs1.apm|Example 2: Absolute Value Operator in Optimization]]





The above examples demonstrate this concept with the absolute value operator. The same techniques can be applied to:
* If...Else Statements
* Max Function
* Min Function
* Sign Operator
* Most other discontinuous functions

February 15, 2013, at 06:13 PM by 128.187.97.21 -
Added lines 1-60:
(:title Logical Conditions in Optimization:)
(:keywords if conditions, maximum, minimum, max operator, min operator, sign operator, APMonitor, summation:)
(:description Formulate conditional statements in numerical optimization so that the problem can be solved with gradient based approaches.:)

!!!! Conditional Statements in Optimization

Conditional statements can cause problems for gradient-based optimization algorithms because they are often posed in a form that gives discontinuous functions, first derivatives, or second derivatives. For example, the absolute value operator is a continuous function but gives a discontinuous first derivative at the origin.

Attach:absolute_value_1st_deriv.png

Minimizing this function can cause a problem for solvers that rely on first and second derivative information to determine search directions for each iteration.



!!!! MPEC: Mathematical Programs with Equilibrium Constraints

Mathematical Programs with Equilibrium Constraints (MPECs) are formulations that can be used to model certain classes of discrete events.  MPECs can be more efficient than solving mixed integer formulations of the optimization problems because it avoids the combinatorial difficulties of searching for optimal discrete variables.

----

!!! SIGN Operator (MPEC Form)

* %list list-page% [[Attach:sign.apm | SIGN Operator Example]]

----

!!! Absolute Value (ABS) Operator

* %list list-page% [[Attach:abs.apm | ABS Operator Example]]

----

!!! Minimum Selector (MIN) Operator

* %list list-page% [[Attach:min.apm | MIN Operator Example]]

----

!!! Maximum Selector (MAX) Operator

* %list list-page% [[Attach:max.apm | MAX Operator Example]]

----

(:html:)
 <div id="disqus_thread"></div>
    <script type="text/javascript">
        /* * * CONFIGURATION VARIABLES: EDIT BEFORE PASTING INTO YOUR WEBPAGE * * */
        var disqus_shortname = 'apmonitor'; // required: replace example with your forum shortname

        /* * * DON'T EDIT BELOW THIS LINE * * */
        (function() {
            var dsq = document.createElement('script'); dsq.type = 'text/javascript'; dsq.async = true;
            dsq.src = 'https://' + disqus_shortname + '.disqus.com/embed.js';
            (document.getElementsByTagName('head')[0] || document.getElementsByTagName('body')[0]).appendChild(dsq);
        })();
    </script>
    <noscript>Please enable JavaScript to view the <a href="https://disqus.com/?ref_noscript">comments powered by Disqus.</a></noscript>
    <a href="https://disqus.com" class="dsq-brlink">comments powered by <span class="logo-disqus">Disqus</span></a>
(:htmlend:)