MethodError: no method matching (:: LinearInterpolation {Vector {Real}, Vector {Real}, true, Real}) (:: AffExpr) occurs when DataInterpolation is called in the optimization objective

[chooron]

Hello, I am using JuMP for reservoir optimization scheduling. Because the reservoir capacity changes, I need to obtain the reservoir water level changes using interpolation methods. Therefore, I used DataInterpolation.jl. However, when building the optimization problem, the variable x generated is of type Vector{VariableRef}, which does not match the type required by DataInterpolation, resulting in this error. Is there a solution for this issue? Here is my code:

# 构建水库调度的优化函数基于JuMP
using DataInterpolations, Statistics, JuMP
import HiGHS

# 定义reservoir对象
default_levels = Float32.([22, 24, 26, 28, 30, 32])
default_storages = Float32.([0.61e8, 0.694e8, 0.876e8, 1.133e8, 1.450e8, 1.852e8])
inflow = Float32.([28, 69, 105, 367, 1320, 2440, 2760, 3020, 3140, 2900, 2750, 1870, 1300, 1100, 980, 820])
dt = 1

# 约束条件设置
min_outflow = 100.0;
max_outflow = 2000.0;

max_level = 26.0;
min_level = 22.0;


struct Curve
    levels::Vector{Real}
    storages::Vector{Real}
end

curve = Curve(default_levels, default_storages)

function interp(c::Curve, input_series, series_name)
    interp_linear = Nothing
    if series_name == "storages"
        interp_linear = LinearInterpolation(c.levels, c.storages)
    else
        interp_linear = LinearInterpolation(c.storages, c.levels)
    end
    interp_series = interp_linear.(input_series)
    return interp_series
end

# test interp
# curve = Curve(default_levels, default_storages)
# interp_res = interp(curve, [23,25,26], "levels")

function water_balance(inflow, outflow, dt)
    Δstorage = (inflow - outflow) * dt * 3600
    return Δstorage
end

function objective(u)
    tmp_storages = water_balance.(inflow, u, dt)
    tmp_levels = interp(curve, tmp_storages, "storages")
    minimum(tmp_levels)
end

function min_level_constraint(u)
    min_storage = minimum(water_balance.(inflow, u, dt))
    interp(curve, min_storage, "storages")
end

function max_level_constraint(u)
    max_storage = maximum(water_balance.(inflow, u, dt))
    interp(curve, max_storage, "storages")
end


n = size(inflow, 1)
model = Model(HiGHS.Optimizer)
@variable(model, min_outflow <= x[1:n] <= max_outflow, Int)
@constraint(model, min_level_constraint(x) >= min_level)
@constraint(model, max_level_constraint(x) <= max_level)
@objective(model, Min, objective(x))

[odow]

Hi @chooron,

HiGHS is a linear programming solver. You cannot use regular Julia functions (like interp) to define your problem. You must formulate it as a mixed-integer linear program with linear constraints and an objective.

If you need help with the formulation, please post on the community forum: https://discourse.julialang.org/c/domain/opt/13

[odow]

Closing this as won't-fix.

@chooron if you post on the https://discourse.julialang.org/c/domain/opt/13 I'll be happy to discuss further 😄

We try to keep the GitHub issues for bug reports and feature requests.