将MutivariatePolynomials输出作为用户定义的函数导入JuMP

时间:2018-05-24 12:55:38

标签: julia polynomials nonlinear-optimization julia-jump

我正在尝试从MultivariatePolynomials导入函数的输出,并将其作为函数用于JuMp。以下是相关代码,它是NL方程的大型系统的一部分。

using JuMP
using MultivariatePolynomials 
@polyvar k s θ_1k θ_2k θ_3k θ_4k θ_5k θ_6k θ_7k θ_8k θ_9k θ_10k θ_11k θ_12k θ_13k E ρ   
d=....    
j_11=subs(d,k=>q11(θ_1k,θ_2k,θ_3k,θ_4k,θ_5k,θ_6k,θ_7k,θ_8k,θ_9k),s=>f(s,ρ,E))
c_11=subs(j_11, k=>-0.7071067811865476,s=>0)

function v11(θ_1k,θ_2k,θ_3k,θ_4k,θ_5k,θ_6k,θ_7k,θ_8k,θ_9k,ρ,E)
    c_11
end

m=Model(solver=IpoptSolver())

JuMP.register(m, :v11, 27, v11, autodiff=true)

@NLconstraint(m,cons1, ((v11(θ_1k,θ_2k,θ_3k,θ_4k,θ_5k,θ_6k,θ_7k,θ_8k,θ_9k,θ_10k,θ_11k,θ_12k,θ_13k,ρ, E))*w_1+*β*(1-δ+...

solve(m)
status=solve(m)

我收到以下错误:

MethodError: no method matching parseNLExpr_runtime(::JuMP.Model, ::DynamicPolynomials.PolyVar{true}, ::Array{ReverseDiffSparse.NodeData,1}, ::Int64, ::Array{Float64,1})

Closest candidates are:
  parseNLExpr_runtime(::JuMP.Model, ::Number, ::Any, ::Any, ::Any) at C:\Users\s1364322\.julia\v0.6\JuMP\src\parsenlp.jl:196
  parseNLExpr_runtime(::JuMP.Model, ::JuMP.Variable, ::Any, ::Any, ::Any) at C:\Users\s1364322\.julia\v0.6\JuMP\src\parsenlp.jl:202

似乎输出c_11尚未转换为多变量函数类型,因为我尝试定义函数v11,但仍保留为多项式类型。我该如何解决这个问题? 拜托,我非常感谢你的帮助

1 个答案:

答案 0 :(得分:0)

为以后参考,该问题已在discourse上解决。