python两种方法实现线性规划

Python实现线性规划的计算

线性规划的目标函数既可以是求最大值，也可以是求最小值，约束条件可以是小于等于也可以是大于等于，但为了方便起见，MATLAB和python都规定线性规划标准型为：
$$\begin{gathered} z=min{\mathbf{f}}^T\mathbf{x}, \\ s.t.=\{\begin{array}{l}\mathbf{A}\cdot \mathbf{x}\le \mathbf{b},\\ Aeq\cdot \mathbf{x}=beq,\\ lb\le \mathbf{x}\le ub\end{array} \end{gathered}$$

转换为标准型的方法

• 目标函数$max\to min$
  若目标函数 $z_1=max{\mathbf{f}}_1^T\mathbf{x},$则可以通过加负号来实现变成$z_2=min-{\mathbf{f}}_1^T\mathbf{x}$
• 约束条件：大于等于$\to$小于等于
  $$\mathbf{A}\cdot \mathbf{x}\ge \mathbf{b}\Rightarrow -\mathbf{A}\cdot \mathbf{x}\le -\mathbf{b}$$

官方示例及代码（不用第三方库）

$$\begin{gathered} z=min-x_0+4x_1 \\ s.t.=\{\begin{array}{l}-3x_0+x_1\le 6,\\ -x_0-2x_1\ge -4,\\ x_1\ge -3\end{array} \end{gathered}$$
与MATLAB中$linprog(c,A,b,Aeq,beq,lb,ub)$ 求解函数类似，python中可以调用scipy库中的optimize.linprog()函数

The problem is not presented in the form accepted by `linprog`. This is
easily remedied by converting the "greater than" inequality
constraint to a "less than" inequality constraint by
multiplying both sides by a factor of :math:`-1`. Note also that the last
constraint is really the simple bound :math:`-3 \leq x_1 \leq \infty`.
Finally, since there are no bounds on :math:`x_0`, we must explicitly
specify the bounds :math:`-\infty \leq x_0 \leq \infty`, as the
default is for variables to be non-negative. After collecting coeffecients
into arrays and tuples, the input for this problem is:

>>> c = [-1, 4]
>>> A = [[-3, 1], [1, 2]]
>>> b = [6, 4]
>>> x0_bounds = (None, None)
>>> x1_bounds = (-3, None)
>>> from scipy.optimize import linprog
>>> res = linprog(c, A_ub=A, b_ub=b, bounds=[x0_bounds, x1_bounds])

Note that the default method for `linprog` is 'interior-point', which is
approximate by nature.

>>> print(res)
     con: array([], dtype=float64)
     fun: -21.99999984082494 # may vary
 message: 'Optimization terminated successfully.'
     nit: 6 # may vary
   slack: array([3.89999997e+01, 8.46872439e-08] # may vary
  status: 0
 success: True
       x: array([ 9.99999989, -2.99999999]) # may vary

If you need greater accuracy, try 'revised simplex'.

>>> res = linprog(c, A_ub=A, b_ub=b, bounds=[x0_bounds, x1_bounds], method='revised simplex')
>>> print(res)

• 运行结果：

     con: array([], dtype=float64)
     fun: -22.0 # may vary
 message: 'Optimization terminated successfully.'
     nit: 1 # may vary
   slack: array([39.,  0.]) # may vary
  status: 0
 success: True
       x: array([10., -3.]) # may vary

使用pulp库求解（推荐，功能更强大）

from pulp import *

# 1. 建立问题
prob = LpProblem("Problem", LpMinimize)

# 2. 建立变量
x1 = LpVariable("x1")
x2 = LpVariable("x2", -3)

# 3. 设置目标函数 z
prob += -x1 + 4*x2, 

# 4. 施加约束
prob += -3*x1 + x2 <= 6.0, "constraint1"
prob += -x1 - 2*x2 >= -4, "constraint2"

# 5. 求解
prob.solve()

# 6. 打印求解状态
print("Status:", LpStatus[prob.status])

# 8. 打印最优解的目标函数值
print("z= ", value(prob.objective))

# 7. 打印出每个变量的最优值
for v in prob.variables():
    print(v.name, "=", v.varValue)

• 运行结果：

Status: Optimal
x1 = 10.0
x2 = -3.0
z=  -22.0

小结

• 就以上代码来看，scipy.optimize.linprog()会比较简便，但是pulp库的代码适用性更强，没有$\ge$或$\le$的限制，同时目标函数还支持直接设为最大优化。
• 具体参数设置可以参照pulp求解线性规划问题或者官方文档

注释

• 代码参考自官方文档
• 部分内容参考自数学建模算法与应用
• 觉得有用的不妨点赞支持一下