[Tex/LaTex] How to move the matrix on the left side

matricespositioning

I would like to move the matrix on the left side, alligned with the text. Do you know how to make it?

\documentclass[10pt,a4paper,titlepage]{book} 
\usepackage[paperwidth=165mm, paperheight=238mm, left=2.5cm,right=2cm,top=2cm,bottom=2cm]{geometry}


%\renewcommand{\baselinestretch}{1.8}
%\smartqed  % flush right qed marks, e.g. at end of proof
%
\usepackage[graphicx]{realboxes}
\usepackage{graphicx}
\usepackage{epstopdf}
%\usepackage[english]{babel}
\usepackage{lineno}
\linenumbers
\usepackage{amsmath}

\usepackage{amssymb}
\usepackage[justification=centering]{caption}

\usepackage{pdflscape}
\usepackage{nomencl}
\makenomenclature

\usepackage{etoolbox}
\renewcommand\nomgroup[1]{%
  \item[\bfseries
  \ifstrequal{#1}{A}{Symbols}{%
  \ifstrequal{#1}{B}{Subscripts}{%
  \ifstrequal{#1}{C}{Superscripts}{%
  \ifstrequal{#1}{D}{Acronyms{}}}}}%
]}

 % This will add the units
%----------------------------------------------
\newcommand{\nomunit}[1]{%
\renewcommand{\nomentryend}{\hspace*{\fill}#1}}
%----------------------------------------------


\begin{document}
%\fboxsep=0pt
%\vspace*{\fill}
 % 


      % \centering
        %\rotatebox{90}{\fbox{%

\section{APPENDIX  }
\label{sec:5}
Here below the matrix when Dirichlet and Neumann boundary condition are set.

%\fboxsep=0pt

%\par\noindent


\begin{minipage}{\linewidth}

%\hspace*{30 cm}
 \vspace*{1 cm}

 \begin{minipage}{0.9\textheight}
               %\hspace*{\fill}
               % \vspace*{\fill}
                \footnotesize

\begin{align*}
\begin{bmatrix}
T_{1} \\
T_{2} \\
T_{3} \\
... \\
T_{n-1}  \\
T_{n}
\end{bmatrix}^{v+1}
\quad
=&
\begin{bmatrix} 
1+2b+2H  & -2b  &   0  &  0   & 0    &  0   \\
-b       & 1+2b &  -b  &  0   & 0    &  0   \\
0        &  -b  & 1+2b &  -b  & 0    &  0   \\
...      & ...  & ...  & ...  & ...  & ...  \\
0        &   0  &  0   &  -b  & 1+2b &  -b  \\
0        &   0  &  0   &   0  &  0   &  1
\end{bmatrix}^{inv}
\quad
\begin{bmatrix} 
T_{1} \\
T_{2} \\
T_{3} \\
...   \\
T_{n-1}\\
T_{n}
\end{bmatrix}^{v}
\\ &\quad+
\begin{bmatrix} 
1+2b+2H  & -2b  &   0  &  0   & 0    &  0  \\
-b       & 1+2b &  -b  &  0   & 0    &  0  \\
0        &  -b  & 1+2b &  -b  & 0    &  0  \\
...      & ...  & ...  & ...  & ...  & ... \\
0        &   0  &  0   &  -g  & 1+2g &  -g  \\
0        &   0  &  0   &   0  &  0   &   1
\end{bmatrix}^{inv}
\quad
\begin{bmatrix} 
\frac{2h\Delta \theta Tair}{\rho c \Delta x} \\
0 \\
0 \\
...   \\
0\\
60
\end{bmatrix}^{v}
\end{align*}
                %\vspace*{\fill}
            \end{minipage}
       % }
        %}
        %\captionof{figure}{Some Matrix}
       % \label{eq:somematrix}
    \end{minipage}
\vspace{1 cm}

$ H $ is a term included in the convective matrix coming from the convective boundary condition and $b$, $g$ are the Fourier numbers for different materials.
\begin{equation}
H=\frac{ h \cdot \Delta \theta }{\rho \cdot c \cdot \Delta x}
\end{equation}
\nomenclature[A]{$b,g$}{ Fourier numbers for different materials \nomunit{}}
\end{document}

Best Answer

With a light (local) squeezing of the space between columns and some additional negative spacing where appropriate. I removed the wrong \quad spaces you are using.

\documentclass[10pt,a4paper,titlepage]{book} 
\usepackage[paperwidth=165mm, paperheight=238mm, left=2.5cm,right=2cm,top=2cm,bottom=2cm]{geometry}


\usepackage[graphicx]{realboxes}
\usepackage{graphicx}
\usepackage{epstopdf}
%\usepackage[english]{babel}
\usepackage{lineno}
\linenumbers
\usepackage{amsmath}

\usepackage{amssymb}
\usepackage[justification=centering]{caption}

\usepackage{pdflscape}
\usepackage{nomencl}
\makenomenclature

\usepackage{etoolbox}
\renewcommand\nomgroup[1]{%
  \item[\bfseries
  \ifstrequal{#1}{A}{Symbols}{%
  \ifstrequal{#1}{B}{Subscripts}{%
  \ifstrequal{#1}{C}{Superscripts}{%
  \ifstrequal{#1}{D}{Acronyms{}}}}}%
]}

 % This will add the units
%----------------------------------------------
\newcommand{\nomunit}[1]{%
\renewcommand{\nomentryend}{\hspace*{\fill}#1}}
%----------------------------------------------


\begin{document}

\section{APPENDIX}\label{sec:5}

Here below the matrix when Dirichlet and Neumann boundary condition are set.
\begingroup\addtolength{\arraycolsep}{-1.3pt}
\begin{align*}
\begin{bmatrix}
T_{1} \\
T_{2} \\
T_{3} \\
\dots \\
T_{n-1}  \\
T_{n}
\end{bmatrix}^{\!v+1}
&=
\begin{bmatrix} 
1+2b+2H  & -2b  &   0  &  0   & 0    &  0   \\
-b       & 1+2b &  -b  &  0   & 0    &  0   \\
0        &  -b  & 1+2b &  -b  & 0    &  0   \\
\dots      & \dots  & \dots  & \dots  & \dots  & \dots  \\
0        &   0  &  0   &  -b  & 1+2b &  -b  \\
0        &   0  &  0   &   0  &  0   &  1
\end{bmatrix}^{\!\mathrm{inv}}
\begin{bmatrix} 
T_{1} \\
T_{2} \\
T_{3} \\
\dots   \\
T_{n-1}\\
T_{n}
\end{bmatrix}^{\!v}
\\ &+
\begin{bmatrix} 
1+2b+2H  & -2b  &   0  &  0   & 0    &  0  \\
-b       & 1+2b &  -b  &  0   & 0    &  0  \\
0        &  -b  & 1+2b &  -b  & 0    &  0  \\
\dots      & \dots  & \dots  & \dots  & \dots  & \dots \\
0        &   0  &  0   &  -g  & 1+2g &  -g  \\
0        &   0  &  0   &   0  &  0   &   1
\end{bmatrix}^{\mathrm{inv}}
\begin{bmatrix} 
\frac{2h\Delta \theta Tair}{\rho c \Delta x} \\
0 \\
0 \\
\dots   \\
0\\
60
\end{bmatrix}^{\!v}
\end{align*}\endgroup
$H$ is a term included in the convective matrix coming from the convective boundary 
condition and $b$, $g$ are the Fourier numbers for different materials.
\begin{equation}
H=\frac{ h \cdot \Delta \theta }{\rho \cdot c \cdot \Delta x}
\end{equation}
\nomenclature[A]{$b,g$}{ Fourier numbers for different materials \nomunit{}}
\end{document}

Related Solutions

[Tex/LaTex] How to have a brace on the left and the top of a matrix without overlapping

Here’s one way you could do it:

\documentclass{article}

\usepackage{amsmath}
\usepackage{rotating}

\begin{document}

\begin{equation*}
  %
  % the vertical brace
  %
  \rotatebox[origin=c]{90}{\scriptsize{
    $n$ parts
  }}
  %
  % an invisible matrix for height
  %
  \left\{\begin{matrix}
     \vphantom{} \\ \vphantom{} \\ \vphantom{} \\
     \vphantom{} \\ \vphantom{} \\ \vphantom{} \\
  \end{matrix}\right.
  %
  % the horizontal brace and the visible matrix
  %
  \overbrace{
    \begin{matrix}
      \bullet & \bullet & \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \cdots & \bullet\\
      \bullet & \cdots & \bullet\\
    \end{matrix}
  }^{\text{
    even number
  }}
\end{equation*}

\end{document}

which looks like

enter image description here

This is very similar to your examples, with only one difference: the two braces are now wrapping around different matrices.

The matrix whose entries are all \vphantom{} doesn’t show up in the printed document, but it has the same height as the matrix of dots. If you made the dot matrix bigger or smaller, you’d need to add or remove entries from this matrix to compensate. This matrix provides the size for the vertical brace.

Then the horizontal brace wraps around the original dot matrix, and never sees the space occupied by the vertical brace.

Alternatively, here’s a fairly simple TikZ-based solution that uses the matrix library to lay out the entries, and then the decorations library to draw the braces.

\documentclass{article}

\usepackage{amsmath}
\usepackage{rotating}
\usepackage{tikz}

\usetikzlibrary{matrix, decorations.pathreplacing}

\begin{document}

\begin{tikzpicture}[decoration={brace, amplitude=6pt}]
    \matrix (m) [matrix of math nodes] {
      \bullet & \bullet & \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \bullet & \bullet & \cdots & \bullet\\
      \bullet & \cdots & \bullet\\
      \bullet & \cdots & \bullet\\
    };
    \draw [decorate, transform canvas={xshift=-0.3em}, thick]
           (m-6-1.south west) -- node [left=6pt]
           {\rotatebox[origin=c]{90}{\scriptsize\text{$n$ parts}}} (m-1-1.north west);
    \draw [decorate, transform canvas={yshift=0.5em, xshift=0.5em}, thick]
          (m-1-1.north west) -- node [above=6pt]
          {\scriptsize\text{even number}} (m-1-6.north east);
\end{tikzpicture}

\end{document}

This is what the TikZ solution looks like:

enter image description here

Move one column in a matrix to the right

Just add another column just for the \ddots. BTW, you don't need to pad all the &s to the right. OTOH, if you don't load amsmath, some of the asterisks will disappear. \null * or {*} will work though.

\documentclass{article}
\usepackage{mathtools}
\begin{document}

\[
A=\left[\begin{array}{ccccccccccccc}
    * & *& *&  & & & & & & & &\\
    * & *& *&  & & & & & & & &\\
    * & *& *& *&*& & & & & & &\\
      &  & *& *&*& & & & & & &\\
      &  & *& *&*& & & & & & &\\
      &  &  &  & &\ddots &  & & & & & \\
    & & &  &  &  &*&* & * & *& & & \\
    & & &  &  &  &*& *& *&*& & & \\
    & & &  &  &  &*&* & * & *& & & \\
    & & &  &  &  &*& *& *&*& *&* &* \\
    & & &  &  &  & & & & * & *& * & * \\
    & & &  &  &  & & & &* & * & * & * \\
    & & &  &  &  & & & &* & * & * & *
\end{array}\right].
\]
% nested arrays
\[
A=\left[\begin{array}{ccc}
  \begin{array}{ccccc}
    * & *& * &\\
    * & *& * &\\
    * & *& *& *&*\\
      &  & *& *&*\\
      &  & *& *&*
  \end{array}\\
   &\ddots \\
   & & \begin{array}{ccccccc}
    *& *& *&* \\
    *&* & * & *\\
    *&* & * & *\\
    *& *& *&*& *&* &* \\
     & & & * & *& * & * \\
     & & &* & * & * & * \\
     & & &* & * & * & *
  \end{array}
\end{array}\right].
\]
\end{document}

Best Answer

Related Solutions

[Tex/LaTex] How to have a brace on the left and the top of a matrix without overlapping

Move one column in a matrix to the right

Related Question