我在两列环境中遇到了一个非常简单的问题，即公式和大型矩阵。我也在网站上进行了搜索，但我的问题的正确答案不在网站上找到的其他问题和答案中。我已附上书面文本。如何解决公式或矩阵落在下一列文本上的问题？您的解决方案是什么？
\documentclass[twocolumn]{article}
\usepackage{amsmath}
\begin{document}
Heretofore various and fascinating articles and books on numerical methods
\begin{eqnarray}\label{Main}
\min\hspace{0.5cm} J&=&\frac{1}{2}x^T(t_f)Sx(t_f)+\frac{1}{2}\int_0^{t_f}\{
x^T(t)Qx(t)+u^T(t)Ru(t)\}dt\\
s.t\ \ \ \ \ \ {\dot x}(t)&=&A_1x(t)+ A_2x(t-\tau)+B_1u(t)+B_2u(t-\tau),\ \ 0\leq t \leq L,\\
x(0)&=&x_0,\\
x(t)&=&\theta(t), \ \ -\tau \leq t<0,\\
u(t)&=&\psi (t),\ \ -\tau \leq t<0,
\end{eqnarray}
So we can have following matrix
\begin{align*}
\renewcommand{\arraystretch}{1.3}
{\tilde A}=\left(
\begin{array}{ccc|c|ccc}
a_0 & \frac{a_{-1}+a_1}{2} & \frac{a_{-2}+a_2}{2} & a_{-3} & \frac{a_{-2}-a_2}{2} &
\frac{a_{-1}-a_1}{2} & 0 \\
\frac{a_{-1}+a_1}{2} & a_0 & \frac{a_{-3}+a_{-1}+a_1-a_3}{2} & a_{-2} & \frac{a_{-3}+a_{-1}-
a_1+a_3}{2} & 0 & \frac{a_{1}-a_{-1}}{2} \\
\frac{a_{-2}+a_2}{2} & \frac{a_{-3}+a_{-1}+a_1-a_3}{2} & \frac{a_{-2}+2a_0-a_2}{2} & a_{-1} &
\frac{a_{-2}+a_2}{2}& \frac{a_{-3}-a_{-1}+a_1+a_3}{2} & \frac{a_{2}-a_{-2}}{2} \\ \hline
a_{-3} & a_{-2} & a_{-1} & a_0 & a_1 & a_2& a_{3}\\ \hline
\frac{a_{-2}-a_2}{2} & \frac{a_{-3}+a_{-1}-a_1+a_3}{2} & \frac{a_{-2}+a_2}{2} & a_1& \frac{a_2+2a_0-
a_{-2}}{2}& \frac{a_3+a_{-1}+a_1-a_{-3}}{2} & \frac{a_{-2}+a_2}{2}\\
\frac{a_{-1}-a_1}{2} & 0 &\frac{a_{-3}-a_{-1}+a_1+a_3}{2} & a_2& \frac{a_{3}+a_1+a_{-1}-a_{-3}}{2} &
a_0& \frac{a_{-1}+a_1}{2}\\
0& \frac{a_{1}-a_{-1}}{2} & \frac{a_{2}-a_{-2}}{2}& a_3& \frac{a_{-2}+a_2}{2} & \frac{a_{-1}+a_1}
{2} &a_0
\end{array}
\right).
\end{align*}
The Fourier trigonometric functions are the very most straightforward example of an orthogonal
system of eigenfunctions of a self-adjoint boundary value problem. The key to the efficacy of
Fourier series rests on the orthogonality properties of the trigonometric functions, which is a
direct consequence of their status as eigenfunctions of the most basic self-adjoint boundary value
problem. In this manner, the Fourier series can also be viewed as a function space version of the
finite-dimensional spectral theory of symmetric matrices and orthogonal eigenvector bases. The main
problem is that we must deal with infinite series rather than finite sums; therefore, convergence
subject that does not appear in the finite-dimensional situation, become of paramount importance.\\
The aim of present paper is to use the Hartley series to solve the optimal quadratic time-
independent delay optimal control problem
The Fourier trigonometric functions are the very most straightforward example of an orthogonal
system of eigenfunctions of a self-adjoint boundary value problem. The key to the efficacy of
Fourier series rests on the orthogonality properties of the trigonometric functions, which is a
direct consequence of their status as eigenfunctions of the most basic self-adjoint boundary value
problem. In this manner, the Fourier series can also be viewed as a function space version of the
finite-dimensional spectral theory of symmetric matrices and orthogonal eigenvector bases. The main
problem is that we must deal with infinite series rather than finite sums; therefore, convergence
subject that does not appear in the finite-dimensional situation, become of paramount importance.\\
The aim of present paper is to use the Hartley series to solve the optimal quadratic time-
independent delay optimal control problem\\
\end{document}
输出如下：
答案1
解决方案是在条带环境中添加最小化问题。无论如何，相应的方程太宽，无法放在单个列中。这里有一个工作代码，希望它符合您的要求。我冒昧地用包eqnarray中的一个环境替换了间距不好的环境optidef。方程和约束被编号为子方程，而不是独立方程，布局是自动完成的。我还用包中的中等大小的分数改进了矩阵nccmath。
\documentclass[twocolumn]{article}
\usepackage{array, amsmath}
\usepackage{optidef, nccmath}
\usepackage{cuted}
\begin{document}
Heretofore various and fascinating articles and books on numerical methods
\begin{strip}
\setlength{\abovedisplayskip}{0pt}
\setlength{\belowdisplayskip}{0pt}
\begin{gather*}
\begin{mini!}|s|
{} {J=\frac{1}{2}x^T(t_f)Sx(t_f)+\frac{1}{2}\int_0^{t_f}\{
x^T(t)Qx(t)+u^T(t)Ru(t)\}dt}{\label{Main1}}{}
\addConstraint{\dot x(t)}{=A_1x(t)+ A_2x(t-\tau)+B_1u(t)+B_2u(t-\tau),\quad 0\leq t \leq L,}
\addConstraint{x(0)}{=x_0,}
\addConstraint{x(t)}{=\theta(t), \quad -\tau \leq t<0}
\addConstraint{u(t)}{=\psi (t),\quad -\tau \leq t<0.}
\end{mini!}
\intertext{So we can have following matrix}
\renewcommand{\arraystretch}{1.3}\setlength{\arraycolsep}{2pt}\setlength{\extrarowheight}{3pt}
{\tilde A}=\left(
\begin{array}{ccc|c|ccc}
a_0 & \mfrac{a_{-1}+a_1}{2} & \mfrac{a_{-2}+a_2}{2} & a_{-3} & \mfrac{a_{-2}-a_2}{2} &
\mfrac{a_{-1}-a_1}{2} & 0 \\
\mfrac{a_{-1}+a_1}{2} & a_0 & \mfrac{a_{-3}+a_{-1}+a_1-a_3}{2} & a_{-2} & \mfrac{a_{-3}+a_{-1}-
a_1+a_3}{2} & 0 & \mfrac{a_{1}-a_{-1}}{2} \\
\mfrac{a_{-2}+a_2}{2} & \mfrac{a_{-3}+a_{-1}+a_1-a_3}{2} & \mfrac{a_{-2}+2a_0-a_2}{2} & a_{-1} &
\mfrac{a_{-2}+a_2}{2}& \mfrac{a_{-3}-a_{-1}+a_1+a_3}{2} & \mfrac{a_{2}-a_{-2}}{2} \\[1ex] \hline
a_{-3} & a_{-2} & a_{-1} & a_0 & a_1 & a_2& a_{3}\\ \hline
\mfrac{a_{-2}-a_2}{2} & \mfrac{a_{-3}+a_{-1}-a_1+a_3}{2} & \mfrac{a_{-2}+a_2}{2} & a_1& \mfrac{a_2+2a_0-
a_{-2}}{2}& \mfrac{a_3+a_{-1}+a_1-a_{-3}}{2} & \mfrac{a_{-2}+a_2}{2}\\
\mfrac{a_{-1}-a_1}{2} & 0 &\mfrac{a_{-3}-a_{-1}+a_1+a_3}{2} & a_2& \mfrac{a_{3}+a_1+a_{-1}-a_{-3}}{2} &
a_0& \mfrac{a_{-1}+a_1}{2}\\
0 & \mfrac{a_{1}-a_{-1}}{2} & \mfrac{a_{2}-a_{-2}}{2}& a_3 & \mfrac{a_{-2}+a_2}{2} & \mfrac{a_{-1}+a_1}
{2} & a_0
\end{array}
\right).
\end{gather*}
\end{strip}
The Fourier trigonometric functions are the very most straightforward example of an orthogonal system of eigenfunctions of a self-adjoint boundary value problem. The key to the efficacy of Fourier series rests on the orthogonality properties of the trigonometric functions, which is a direct consequence of their status as eigenfunctions of the most basic self-adjoint boundary value problem. In this manner, the Fourier series can also be viewed as a function space version of the finite-dimensional spectral theory of symmetric matrices and orthogonal eigenvector bases. The main problem is that we must deal with infinite series rather than finite sums; therefore, convergence subject that does not appear in the finite-dimensional situation, become of paramount importance.
The aim of present paper is to use the Hartley series to solve the optimal quadratic time-independent delay optimal control problem.
The Fourier trigonometric functions are the very most straightforward example of an orthogonal system of eigenfunctions of a self-adjoint boundary value problem. The key to the efficacy of Fourier series rests on the orthogonality properties of the trigonometric functions, which is a direct consequence of their status as eigenfunctions of the most basic self-adjoint boundary value problem. In this manner, the Fourier series can also be viewed as a function space version of the finite-dimensional spectral theory of symmetric matrices and orthogonal eigenvector bases. The main problem is that we must deal with infinite series rather than finite sums; therefore, convergence subject that does not appear in the finite-dimensional situation, become of paramount importance.
The aim of present paper is to use the Hartley series to solve the optimal quadratic time-independent delay optimal control problem.
\end{document}
