update style to match codebase

README.md

@@ -19,6 +19,14 @@ model, vars, cons = opf_model(
     form = :polar
 )
 result = madnlp(model; tol=1e-6)
+
+
+#Alternatively, solve using DC linearization
+model, vars, cons = dcopf_model(
+    "pglib_opf_case118_ieee.m";
+    backend = CUDABackend()
+)
+result = madnlp(model; tol=1e-6)
 ```
 
 ### Security-constrained optimal power flow

src/constraint.jl

@@ -151,12 +151,13 @@ function c_comp(pstd, pstc)
 end
 
 # DCOPF
-function c_ohms_law_dcopf(br, pf, va_f, va_t, b_val)
+function c_ohms_law_dcopf(br, pf, va_f, va_t)
+    r2_x2 = br.br_r^2 + br.br_x^2
+    b_val = -br.br_x / r2_x2
     return -b_val * (va_f - va_t) - pf
 end
 
-function dcopf_branch_b(br)
-    r2_x2 = br.br_r^2 + br.br_x^2
-    b_val = -br.br_x / r2_x2
-    return b_val
-end
+function c_active_power_balance_dc(b)
+    return b.pd + b.gs
+end
+

src/dcopf.jl

@@ -1,12 +1,15 @@
-function build_dcopf(data; backend = nothing, T = Float64, core = nothing, kwargs...)
+function dummy_extension(core, vars, cons)
+    return (;), (;)
+end
+
+function build_dcopf(data, user_callback; backend = nothing, T = Float64, core = nothing, kwargs...)
+
     core = isnothing(core) ? ExaCore(T; backend = backend) : core
     T, backend = typeof(core).parameters[1], core.backend
 
     va = variable(core, length(data.bus))
     pg = variable(core, length(data.gen); lvar = data.pmin, uvar = data.pmax)
 
-    pd = parameter(core, map(b->b.pd, data.bus))
-    bs = parameter(core, map(dcopf_branch_b, data.branch))
 
     pf = variable(
         core,
@@ -21,7 +24,7 @@ function build_dcopf(data; backend = nothing, T = Float64, core = nothing, kwarg
 
     c_ohms_law = constraint(
         core,
-        c_ohms_law_dcopf(br, pf[br.i], va[br.f_bus], va[br.t_bus], bs[br.i])
+        c_ohms_law_dcopf(br, pf[br.i], va[br.f_bus], va[br.t_bus])
         for br in data.branch
     )
 
@@ -34,7 +37,8 @@ function build_dcopf(data; backend = nothing, T = Float64, core = nothing, kwarg
 
     c_active_power_balance = constraint(
         core,
-        pd[b.i] + b.gs for b in data.bus
+        c_active_power_balance_dc(b) for b in data.bus
+
     )
     constraint!(core, c_active_power_balance, g.bus => -pg[g.i] for g in data.gen)
     constraint!(core, c_active_power_balance, br.f_bus => pf[br.i] for br in data.branch)
@@ -53,24 +57,47 @@ function build_dcopf(data; backend = nothing, T = Float64, core = nothing, kwarg
         c_active_power_balance = c_active_power_balance,
     )
 
-    params = (
-        pd = pd,
-        bs = bs,
-    )
 
-    model = ExaModel(core; kwargs...)
+    vars2, cons2 = user_callback(core, vars, cons)
+    model =ExaModel(core; kwargs...)
+
+    vars = (;vars..., vars2...)
+    cons = (;cons..., cons2...)
 
-    return model, vars, cons, params
+    return model, vars, cons
 end
 
+
+"""
+    dcopf_model(filename; backend, T, user_callback)
+
+Return `ExaModel`, variables, and constraints for a static linearized DC Optimal Power Flow (DCOPF) problem from the given file.
+
+# Arguments
+- `filename::String`: Path to the data file.
+- `backend`: The solver backend to use. Default if nothing.
+- `T`: The numeric type to use (default is `Float64`).
+- `user_callback`: User function that extends the model
+- `kwargs...`: Additional keyword arguments passed to the model builder.
+
+# Returns
+A vector `(model, variables, constraints)`:
+- `model`: An `ExaModel` object.
+- `variables`: NamedTuple of model variables.
+- `constraints`: NamedTuple of model constraints.
+"""
+
 function dcopf_model(
     filename;
     backend = nothing,
     T = Float64,
+
+    user_callback = dummy_extension,
     kwargs...,
 )
     data = parse_ac_power_data(filename)
     data = convert_data(data, backend)
 
-    return build_dcopf(data; backend = backend, T = T, kwargs...)
+    return build_dcopf(data, user_callback; backend = backend, T = T, kwargs...)
+
 end