Pyomo报错：ScalarParam未构造即求值，求解决方案

错误信息 #

执行代码时触发如下错误：

ValueError: Evaluating the numeric value of parameter 'ScalarParam' before the Param has been constructed (there is currently no value to return).

可复现代码片段 #

本人正在优化一个大型模型，以下是能复现该问题的代码：

from pyomo.environ import ConcreteModel, Set, RangeSet, Param, Var, Objective, Constraint, SolverFactory

model = ConcreteModel()

model.days = Set(initialize=RangeSet(31))
T = model.days

model.Y = Param(initialize=1) 
model.L= Param(initialize=330)  
model.LL= Param(initialize=3300) 
model.LU= Param(initialize=3300) 
model.U= Param(initialize=250) 
model.X0= Param(initialize=0) 

model.M= Var(T,bounds=(0, None), initialize = 10) 
model.X= Var(T,bounds=(0, None), initialize = 10) 

def obj_rule(model):
  return sum(model.X[i] for i in model.days)

model.funcobj = Objective(rule=obj_rule, sense=maximize)

def C1A_rule(model,t):
  return model.Y*model.L <= model.M[t]

def C1B_rule(model,t):
  return model.M[t] <= model.Y*model.U

model.C1A = Constraint(model.days, rule=C1A_rule)
model.C1B = Constraint(model.days, rule=C1B_rule)

def C14_rule(model,i): # 物料平衡约束（b1部门）
  if i == T[1]:  # Pyomo集合为1索引
    return (model.X[i] == model.M[i] + model.X0)
  if i != T[1]:   
    return (model.X[i] == model.M[i] + model.X[i-1])

model.C14 = Constraint(model.days, rule=C14_rule)

def C15A_rule(model,t):
  return model.LL <= model.M[t] / 1.15741e-5
def C15B_rule(model,t):
  return model.M[t] / 1.15741e-5 <= model.LU

model.C15A = Constraint(model.days, rule=C15A_rule)
model.C15B = Constraint(model.days, rule=C15B_rule)

SolverFactory('glpk', executable='/usr/bin/glpsol').solve(model).write()

解决方法 #

错误根源是约束C15A和C15B中直接对变量进行了浮点数除法操作，Pyomo在构建约束时会尝试提前计算参数的数值，导致参数未完成初始化时触发报错。解决方式是将除法转换为乘法，避免直接数值运算，让Pyomo正确处理表达式逻辑：

1. 修改C15A_rule和C15B_rule的约束逻辑：

   • 原C15A的model.LL <= model.M[t] / 1.15741e-5等价于model.LL * 1.15741e-5 <= model.M[t]
   • 原C15B的model.M[t] / 1.15741e-5 <= model.LU等价于model.M[t] <= model.LU * 1.15741e-5

2. 修改后的约束代码：

def C15A_rule(model,t):
  return model.LL * 1.15741e-5 <= model.M[t]
def C15B_rule(model,t):
  return model.M[t] <= model.LU * 1.15741e-5

验证效果 #

替换修改后的约束代码后，模型可以正常完成构建与求解，不会再触发该数值计算错误。

内容的提问来源于stack exchange，提问作者Alex