simulated annealing

src/EvoLP.jl

@@ -20,6 +20,7 @@ include("testfunctions.jl")
 
 include("algorithms/ga.jl")
 include("algorithms/ea.jl")
+include("algorithms/sa.jl")
 include("algorithms/swarm.jl")
 
 include("deprecated.jl")
@@ -32,6 +33,7 @@ export Particle, normal_rand_particle_pop, unif_rand_particle_pop  # Particles
 
 # Algorithms
 export GA, GA!
+export SA
 export oneplusone, oneplusone!
 export PSO, PSO!
 

src/algorithms/sa.jl

@@ -0,0 +1,54 @@
+"""
+    SA(f, pop, k_max, S, C, M; T_init)
+
+Simulated annealing.
+
+## Arguments
+- `f::Function`: objective function to **minimise**.
+- `population::AbstractVector`: a list of vector individuals.
+- `k_max::Integer`: number of iterations.
+- `S::ParentSelector`: one of the available [`ParentSelector`](@ref).
+- `C::CrossoverMethod`: one of the available [`CrossoverMethod`](@ref).
+- `M::MutationMethod`: one of the available [`MutationMethod`](@ref).
+- `T₀::Real=100.0`: initial temperature
+- `α::Real=0.99`: cooling rate
+
+Returns a [`Result`](@ref).
+"""
+function SA(
+    f::Function,
+    population::AbstractVector,
+    k_max::Integer,
+    S::ParentSelector,
+    C::Recombinator,
+    M::Mutator;
+    T₀::Real=100.0,
+    α::Real=0.99
+)
+    fx = Inf
+    best_ind = copy(population)
+    T = T₀
+
+    runtime = @elapsed begin
+        for _ in 1:k_max
+            c = mutate(M, population)
+            fc = f(c)
+            ΔE = fc - fx  # Calculate the difference in objective function values
+
+            # Acceptance probability function
+            if ΔE < 0 || rand() < exp(-ΔE / T)
+                population = copy(c)
+                fx = fc
+            end
+
+            fx < f(best_ind) && (best_ind = copy(population))
+
+            T *= α
+        end
+    end
+
+    n_evals = k_max
+
+    return Result(fx, best_ind, [best_ind], k_max, n_evals, runtime)
+end
+