ortools近似实现标准差作为目标函数

from ortools.sat.python import cp_model

model = cp_model.CpModel()

# relations = model.NewIntVar(0, 99999, 'relations')
# 初始化
a=model.NewIntVar(0, 99999, 'a')
b=model.NewIntVar(0, 99999, 'b')
# 初始化自变量
x = model.NewIntVar(0, 100, 'x')
y = model.NewIntVar(0, 200, 'y')
z = model.NewIntVar(0, 300, 'z')
l=[x,y,z]
gap_lst=[]
# 初始化平均值变量
avg=model.NewIntVar(0, 99999, 'avg')
# 初始化加和变量
sum_l=model.NewIntVar(0, 99999, 'sum_l')
# 对x,y,z求和，赋值给sum_l
model.Add(sum_l==sum(l))
# 对x,y,z求平均，赋值给avg
model.AddDivisionEquality(avg,sum_l,len(l))
error_lst=[]
for i in range(len(l)-1):
    # 计算每个自变量对平均的差值
    error=model.NewIntVar(0, 99999, 'error_%s'%i)
    t = model.NewIntVar(0, 99999, 't_%s' % i)
    model.Add(t==l[i]-avg)
    #对差值取绝对值并赋值给error
    model.AddAbsEquality(error,t)
    error_lst.append(error)

print(gap_lst)
​​​​​​​#最小化error之和
model.Minimize(sum(error_lst))

solver = cp_model.CpSolver()
status = solver.Solve(model)
print(solver.StatusName(status))
print('x=',solver.Value(x))
print('y=',solver.Value(y))
print('z=',solver.Value(z))
print('b=',solver.Value(b))
print('a=',solver.Value(a))