Difference between revisions of "Category:Muscod"

From mintOC
Jump to: navigation, search
m (Text replacement - "<bibreferences/>" to "<biblist />")
(Added description)
 
(One intermediate revision by one other user not shown)
Line 1: Line 1:
[insert description of muscod here]
+
This category lists all problems that have a description of differential equations in MUSCOD-format.
 +
 
 +
MUSCOD is an abbreviation for MUltiple Shooting COde for Direct Optimal Control. The original version was developed by Bock and Plitt in the 1980s and used in the seminal paper <bib id="Bock1984" /> which is often cited as the origin of the Direct Multiple Shooting method. The reimplementation MUSCOD-II by Daniel Leineweber in the 1990s is still the basis for ongoing development, in particular in the work groups of Georg Bock, Christian Kirches, Andreas Potschka (all Heidelberg), and Sebastian Sager (Magdeburg). MUSCOD-II is based on the direct multiple shooting method to treat multistage optimal control problems with DAE constraints. The extension MS-MINTOC by Sager is able to treat integrality requirements in an efficient way, compare <bib id="Sager2009" />.
 +
 
 +
For a standard MIOC written as
 +
 
 +
<p>
 +
<math>
 +
\begin{array}{llcl}
 +
\displaystyle \min_{x(\cdot), u(\cdot), v(\cdot), q, \rho} & \phi(x(t_f), q, \rho) \\[1.5ex]
 +
\mbox{s.t.} & \dot{x}(t) & = & f(x(t), u(t), v(t), q, \rho), \\
 +
& 0 &\le& c(x(t),u(t),v(t), q, \rho), \\[1.5ex]
 +
& 0 &=& r^{\text{eq}}(x(t_0),x(t_1), \dots, x(t_m), q, \rho), \\
 +
& 0 &\le& r^{\text{ieq}}(x(t_0),x(t_1), \dots, x(t_m), q, \rho), \\[1.5ex]
 +
& v(t) &\in& \Omega := \{v^1, v^2, \dots, v^{n_\omega} \},\\
 +
& \rho &\in& \Rho := \{\rho^1, \rho^2, \dots, \rho^{n_\Rho} \},
 +
\end{array}
 +
</math>
 +
</p>
 +
 
 +
the format looks, e.g., something like this:
 +
<source lang="cpp">
 +
rhs[0] =  xd[0] - xd[0]*xd[1] - p[0]*u[0]*xd[0];
 +
rhs[1] = - xd[1] + xd[0]*xd[1] - p[1]*u[0]*xd[1];
 +
</source>
 +
 
 +
for the special case
 +
 
 +
<p>
 +
<math>
 +
\begin{array}{llcl}
 +
& \dot{x}_0(t) & = & x_0(t) - x_0(t) x_1(t) - \; c_0 x_0(t) \; w(t), \\
 +
& \dot{x}_1(t) & = & - x_1(t) + x_0(t) x_1(t) - \; c_1 x_1(t) \; w(t).
 +
\end{array}
 +
</math>
 +
</p>
 +
 
 +
Here the vector entries ''rhs[i]'' contain the values of the right hand side function <math>f_i(\cdot)</math>. The vector entries ''xd[i]'' contains the differential variables <math>x_i(t)</math>, ''u[i]'' the control variables <math>u_i(t)</math> or <math>v_i(t)</math>, and ''p[i]'' contains the model parameters <math>p</math>or control differential variables <math>q_i(t)</math> resp. <math>\rho_i(t)</math>.
  
 
== References ==
 
== References ==
 
<biblist />
 
<biblist />
  
 
+
[[Category: Implementation]]
<!--List of all categories this page is part of. List characterization of solution behavior, model properties, ore presence of implementation details (e.g., AMPL for AMPL model) here -->
+
[[Category:Problem characterization]]
+

Latest revision as of 10:09, 28 January 2016

This category lists all problems that have a description of differential equations in MUSCOD-format.

MUSCOD is an abbreviation for MUltiple Shooting COde for Direct Optimal Control. The original version was developed by Bock and Plitt in the 1980s and used in the seminal paper [Bock1984]Address: Budapest
Author: H.G. Bock; K.J. Plitt
Booktitle: Proceedings of the 9th IFAC World Congress
Pages: 242--247
Publisher: Pergamon Press
Title: A Multiple Shooting algorithm for direct solution of optimal control problems
Url: http://www.iwr.uni-heidelberg.de/groups/agbock/FILES/Bock1984.pdf
Year: 1984
Link to Google Scholar
which is often cited as the origin of the Direct Multiple Shooting method. The reimplementation MUSCOD-II by Daniel Leineweber in the 1990s is still the basis for ongoing development, in particular in the work groups of Georg Bock, Christian Kirches, Andreas Potschka (all Heidelberg), and Sebastian Sager (Magdeburg). MUSCOD-II is based on the direct multiple shooting method to treat multistage optimal control problems with DAE constraints. The extension MS-MINTOC by Sager is able to treat integrality requirements in an efficient way, compare [Sager2009]Author: Sager, S.; Reinelt, G.; Bock, H.G.
Journal: Mathematical Programming
Number: 1
Pages: 109--149
Title: Direct Methods With Maximal Lower Bound for Mixed-Integer Optimal Control Problems
Url: http://mathopt.de/PUBLICATIONS/Sager2009.pdf
Volume: 118
Year: 2009
Link to Google Scholar
.

For a standard MIOC written as


\begin{array}{llcl}
 \displaystyle \min_{x(\cdot), u(\cdot), v(\cdot), q, \rho} & \phi(x(t_f), q, \rho) \\[1.5ex]
 \mbox{s.t.} & \dot{x}(t) & = & f(x(t), u(t), v(t), q, \rho), \\
 & 0 &\le& c(x(t),u(t),v(t), q, \rho), \\[1.5ex]
 & 0 &=& r^{\text{eq}}(x(t_0),x(t_1), \dots, x(t_m), q, \rho), \\
 & 0 &\le& r^{\text{ieq}}(x(t_0),x(t_1), \dots, x(t_m), q, \rho), \\[1.5ex]
 & v(t) &\in& \Omega := \{v^1, v^2, \dots, v^{n_\omega} \},\\
 & \rho &\in& \Rho := \{\rho^1, \rho^2, \dots, \rho^{n_\Rho} \},
\end{array}

the format looks, e.g., something like this:

rhs[0] =   xd[0] - xd[0]*xd[1] - p[0]*u[0]*xd[0];
rhs[1] = - xd[1] + xd[0]*xd[1] - p[1]*u[0]*xd[1];

for the special case


\begin{array}{llcl}
 & \dot{x}_0(t) & = & x_0(t) - x_0(t) x_1(t) - \; c_0 x_0(t) \; w(t), \\
 & \dot{x}_1(t) & = & - x_1(t) + x_0(t) x_1(t) - \; c_1 x_1(t) \; w(t).
\end{array}

Here the vector entries rhs[i] contain the values of the right hand side function f_i(\cdot). The vector entries xd[i] contains the differential variables x_i(t), u[i] the control variables u_i(t) or v_i(t), and p[i] contains the model parameters por control differential variables q_i(t) resp. \rho_i(t).

References

There were no citations found in the article.