Merge Stochastic implementation of Energy Communities

Project.toml

@@ -6,31 +6,36 @@ version = "0.3.1"
 [deps]
 CSV = "336ed68f-0bac-5ca0-87d4-7b16caf5d00b"
 DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0"
+Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 ExportAll = "ad2082ca-a69e-11e9-38fa-e96309a31fe4"
 FileIO = "5789e2e9-d7fb-5bc7-8068-2c6fae9b9549"
 Format = "1fa38f19-a742-5d3f-a2b9-30dd87b9d5f8"
 InteractiveUtils = "b77e0a4c-d291-57a0-90e8-8db25a27a240"
 JuMP = "4076af6c-e467-56ae-b986-b466b2749572"
 MathOptInterface = "b8f27783-ece8-5eb3-8dc8-9495eed66fee"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
+PointEstimateMethod = "f5532426-4504-4a0a-b768-abad0680b22a"
 SankeyPlots = "8fd88ec8-d95c-41fc-b299-05f2225f2cc5"
 StatsPlots = "f3b207a7-027a-5e70-b257-86293d7955fd"
+StochasticPrograms = "8b8459f2-c380-502b-8633-9aed2d6c2b35"
 TheoryOfGames = "eb50afb4-6f20-4b37-9b66-473e668300bf"
 XLSX = "fdbf4ff8-1666-58a4-91e7-1b58723a45e0"
 YAML = "ddb6d928-2868-570f-bddf-ab3f9cf99eb6"
 
 [compat]
 CSV = "0.10"
 DataFrames = "1"
+Distributions = "0.25"
 ExportAll = "0.1"
 FileIO = "1"
 Format = "1"
 JuMP = "1"
-MathOptInterface = "1.0"
+MathOptInterface = "1"
 Plots = "1, 2"
+PointEstimateMethod = "0.1"
 SankeyPlots = "0.2.2, 0.3"
 StatsPlots = "0.15"
-TheoryOfGames = "0.1, 0.2"
-XLSX = "0.9, 0.10"
-YAML = "0.4"
-julia = "1"
+StochasticPrograms = "0.6"
+TheoryOfGames = "0.1"
+XLSX = "0.10"
+YAML = "0.4"

README.md

@@ -56,4 +56,4 @@ optimize!(ECModel)
 
 # create some plots
 plot(ECModel, output_plot_combined)
-```
+```

examples/RunStochModel(TBD).jl

@@ -0,0 +1,178 @@
+#using StochasticPrograms#master
+#using JuMP
+#using Base.Threads
+#using DataStructures
+#using LinearAlgebra
+#using Parameters
+#using Distributions
+#using Random
+#using JLD2
+#using FileIO
+#using PointEstimateMethod
+#using YAML
+#using DataFrames
+#using CSV
+#using XLSX
+#using Formatting
+# Useful package to built plot
+#using Makie
+#using CairoMakie
+#using ColorSchemes
+#using StochasticPrograms
+
+#import CPLEX
+
+# # Run this script from EnergyCommunity.jl root!!!
+# using Pkg
+# Pkg.activate("examples")
+
+using EnergyCommunity, JuMP
+using HiGHS, Plots
+
+# Data extraction
+file_name = "./src/stochastic/stoch_data/energy_community_model.yml"
+data = read_input(file_name)
+
+(gen_data,
+  users_data,
+  market_data) = explode_data(data)
+
+
+n_users = length(user_names(gen_data, users_data))
+init_step = field(gen_data, "init_step")
+final_step = field(gen_data, "final_step")
+n_steps = final_step - init_step + 1
+project_lifetime = field(gen_data, "project_lifetime")
+peak_categories = profile(market_data, "peak_categories")
+
+# Set definitions
+user_set = user_names(gen_data, users_data)
+year_set = 1:project_lifetime
+time_set = 1:n_steps
+peak_set = unique(peak_categories)
+
+# Number of scenarios to be extracted
+scen_s_sample = field(gen_data, "n_s")
+scen_eps_sample = field(gen_data, "n_eps")
+n_scen_sample = scen_s_sample * scen_eps_sample
+
+isdet = false
+if scen_eps_sample == 1 && scen_s_sample == 1
+  isdet = true
+end
+
+scen_s_set = 1:scen_s_sample
+scen_eps_set = 1:scen_eps_sample
+
+# Standard deviation associated with load and renewable production in long period uncertainty
+
+sigma_load = 0.3
+
+mean_pv = 1.0
+sigma_pv = 0.1
+
+mean_wind = 0.95
+sigma_wind = 0.15
+
+# Extract Uncertain Variable
+unc_var = field(gen_data, "uncertain_var")
+
+# Extraction of the point used to sample the distributions associated to the long period uncertainty
+(point_s_load,
+  point_s_pv,
+  point_s_wind,
+  scen_probability) = pem_extraction(scen_s_sample, 
+                        sigma_load, 
+                        mean_pv,
+                        sigma_pv,
+                        mean_wind,
+                        sigma_wind,
+                        unc_var)
+
+# Include the sampler for distributions associated to short period uncertainty and a function to generate scenarios (new version: sample the normalized distributions)
+
+# include("point_Scen_eps_sampler.jl")
+
+# To define an empty stochastic model we have to declare previously the scenarios
+
+# OUTPUT: sampled_scenarios: array containing all the scenarios created for the first phase
+#         point_eps_load_sampled: extracted points for the normalized distributions associated with load demand
+#         point_eps_ren_sampled: extracted points for the normalized distributions associated with renewable production
+
+sampled_scenarios = scenarios_generator(data,
+                              point_s_load,
+                              point_s_pv,
+                              point_s_wind,
+                              scen_s_sample,
+                              scen_eps_sample,
+                              unc_var,
+                              point_probability=scen_probability,
+                              first_stage=true,
+                              deterministic=isdet)
+
+# Initialize the empty non-cooperative version of a EC
+
+EC_NonCooperative = StochasticEC(file_name,GroupNC(),CPLEX.Optimizer,sampled_scenarios,scen_s_sample,scen_eps_sample)
+
+# Build the NC model
+
+build_base_model!(EC_NonCooperative,CPLEX.Optimizer)
+
+# set the technical paraters for the NC optimization
+time_lim = 60 * 60 * 10  # max time in second
+primal_gap = 1e-2 # primal gap (1e-4 = 1%)
+n_threads = 64 # number of threads to be used
+
+set_parameters_ECmodel!(EC_NonCooperative,primal_gap,time_lim,n_threads,1)
+
+optimize_deterministic_ECmodel(EC_NonCooperative) # optimize the deterministic equivalent version and store the results
+
+# save the data of the first stage model
+output_file_NC = "first_stage_output_NC_($scen_s_sample,$scen_eps_sample)"
+
+print_first_stage(output_file_NC * ".xlsx",EC_NonCooperative)
+save(output_file_NC * ".jld2", EC_NonCooperative)
+
+# get the number of installed resource by users
+x_NC_fixed = EC_NonCooperative.results[:x_us].data
+
+# add the installed capacity of the entire EC
+x_tot_NC = calculate_x_tot(EC_NonCooperative)
+
+#Free memory
+EC_NonCooperative = StochasticEC();
+
+GC.gc() # garbage collector
+
+# Initialize the cooperative version of a EC
+
+EC_Cooperative = StochasticEC(file_name,GroupCO(),CPLEX.Optimizer,sampled_scenarios,scen_s_sample,scen_eps_sample)
+
+# Build the CO model
+
+build_specific_model!(GroupCO(),EC_Cooperative,CPLEX.Optimizer)
+
+# set the technical parameters for the CO optimization
+time_lim = 60 * 60 * 24 # max time in second
+primal_gap = 1e-2 # primal gap (1e-4 = 1%)
+n_threads = 64 # number of threads to be used
+
+set_parameters_ECmodel!(EC_Cooperative,primal_gap,time_lim,n_threads,1)
+
+optimize_deterministic_ECmodel(EC_Cooperative) # optimize the deterministic equivalent version and store the results
+
+# save the data of the first stage model
+output_file_CO = "first_stage_output_CO_($scen_s_sample,$scen_eps_sample)"
+print_first_stage(output_file_CO * ".xlsx",EC_Cooperative)
+save(output_file_CO * ".jld2", EC_Cooperative)
+# get the number of installed resource by users
+x_CO_fixed = EC_Cooperative.results[:x_us].data
+
+# Plot some useful image of the installed capacity
+colors = Makie.wong_colors()
+
+# add the installed capacity of the entire EC
+x_tot_CO = calculate_x_tot(EC_Cooperative)
+
+plot_resource("installed_capacity1_($scen_s_sample,$scen_eps_sample).png",["PV","wind"],users_data,x_tot_CO,x_tot_NC,colors[1:2]) # renewable asset
+plot_resource("installed_capacity_($scen_s_sample,$scen_eps_sample).png",["batt"],users_data,x_tot_CO,x_tot_NC,[colors[3]]) #battery