General-purpose improvements from the WECC branch; more exports, utility function upgrades like automatically pickling the an entire solution object and reloading it, optionally plotting results, updates to a few examples to make them consistent with the current code.

Author: josiahjohnston

.gitignore

@@ -4,3 +4,4 @@
 .DS_Store
 gurobi.log
 switch_model.egg-info/
+venv

AUTHORS

@@ -4,11 +4,13 @@
 # contributions of individuals can be tracked via git history.
 
 Matthias Fripp <mfripp@hawaii.edu>
+Rodrigo Henriquez <rhenriqueza@uc.cl>
+Patricia Hidalgo-Gonzalez <patricia.hidalgo.g@berkeley.edu>
 Josiah Johnston <siah@berkeley.edu>
 Benjamin Maluenda <bmaluend@uc.cl>
 Ana Mileva
 Mark Seaborn
-Rodrigo Henriquez <rhenriqueza@uc.cl>
+
 
 # Organizations
 

DEV_INSTALL.txt

@@ -0,0 +1,70 @@
+ADVANCED INSTALLATION NOTES FOR DEVELOPERS AND ADVANCED USERS
+
+If you are prone to work with more than one version of switch, you will
+greatly benefit from python virtual environments which isolate your project's
+python software from other versions that may be installed on the computer.
+Software requirements and compatibilities change over time, sometimes in
+incompatible ways. Virtual environments also help developers ensure they have
+included all necessary dependencies in setup.py and aren't relying on
+mannually installed system libraries. Using virtual environments to
+encapsulate each project's customized software allows active and archived
+projects to coexist on a single computer. These instructions assume you have
+read INSTALL, have installed Anaconda (or equivalent), pip and solver
+prerequisites.
+
+INSTALLATION WITH PYTHON VIRTUAL ENVIRONMENTS FOR DEVELOPMENT
+
+1. Open either a terminal or Anaconda command prompt (Start -> Anaconda ->
+   Anaconda Prompt)
+
+2a. DEVELOPERS
+   Developers often work with switch example directories and don't necessarily
+   have a separate folder with inputs for analysis. In this case, the source
+   directory doubles as your project directory. Execute these commands, but
+   set PROJECT_DIR to the actual path you want to use.
+        PROJECT_DIR=/path/to/new/switch/checkout
+        SRC_DIR="$PROJECT_DIR"
+
+2b. ANALYSTS
+   Analysts typically have a directory per project that stores data for their
+   input scenarios as well as a copy of the code they are using. Execute these
+   commands, but set PROJECT_DIR to the actual path you want to use. 
+        PROJECT_DIR=/path/to/project/directory
+        SRC_DIR="$PROJECT_DIR"/switch_source
+
+   Note: If you are using one of Matthias's packaged scenarios with a
+   submodule or subtree, set SRC_DIR to that subfolder instead of doing a
+   fresh git checkout. In this case, skip step 3.
+
+3. Download a copy of switch
+        git clone https://github.com/switch-model/switch.git "$SRC_DIR"
+
+4. Create a virtual environment in your project directory, and install switch
+   and its prerequisites.
+        cd "$PROJECT_DIR"
+        pip install virtualenv
+        virtualenv venv
+        source venv/bin/activate
+        pip install --editable "$SRC_DIR"
+
+   Note: The --editable flag ensures that any changes to your downloaded copy
+   of switch will be immediately reflected to your installed copy in the
+   virtual environment.
+
+5. Record the exact version of every library you are using so you can 
+   precisely replicate this setup in the future. 
+        pip freeze > requirements.txt
+   Without this step, you are at higher risk of not being able to readily
+   repeat or extend this analysis, especially if future releases of libraries
+   are not completely backwards compatible (this has happened a few times in
+   the past few years).
+
+
+EXECUTION
+
+1. Open either a terminal or Anaconda command prompt (Start -> Anaconda ->
+   Anaconda Prompt) and navigate to your project directory.
+2. Activate your virtual environment with the following command:
+       source venv/bin/activate
+3. Execute switch like normal
+       switch solve ...

examples/3zone_toy_stochastic_PySP/PySPInputGenerator.py

@@ -56,8 +56,8 @@
 # Stage names. Can be any string and must be specified in order.
 stage_list = ["Investment", "Operation"]   
 stage_vars = {
-    "Investment": ["BuildProj", "BuildLocalTD", "BuildTrans"],
-    "Operation": ["DispatchProj", "ProjFuelUseRate"]
+    "Investment": ["BuildGen", "BuildLocalTD", "BuildTx"],
+    "Operation": ["DispatchGen", "GenFuelUseRate"]
 }
 # List of scenario names
 scenario_list = [

examples/3zone_toy_stochastic_PySP/README.md

@@ -13,7 +13,7 @@ Available at: http://mpc.zib.de/index.php/MPC/article/viewFile/85/39
 Pyomo Online Documentation:
 https://software.sandia.gov/downloads/pub/pyomo/PyomoOnlineDocs.html
 
-If any concepts on stochastic programming need to be adquired or refreshed, a
+If any concepts on stochastic programming need to be acquired or refreshed, a
 good reference is:  Birge, J. R. & Louveaux, F. (2011). Introduction to
 Stochastic Programming. 2nd Edition. Springer Science+Business Media.
 

examples/3zone_toy_stochastic_PySP/ReferenceModel.py

@@ -53,14 +53,13 @@
 # be called from this script.
 
 def calc_tp_costs_in_period(m, t):
-        return sum(
-            getattr(m, tp_cost)[t] * m.tp_weight_in_year[t]
-            for tp_cost in m.cost_components_tp)
-
+	return sum(
+		getattr(m, tp_cost)[t] * m.tp_weight_in_year[t]
+		for tp_cost in m.Cost_Components_Per_TP)
 def calc_annual_costs_in_period(m, p):
-        return sum(
-            getattr(m, annual_cost)[p]
-            for annual_cost in m.cost_components_annual)
+	return sum(
+		getattr(m, annual_cost)[p]
+		for annual_cost in m.Cost_Components_Per_Period)
 
 # In the current version of Switch-Pyomo, all annual costs are defined 
 # by First Stage decision variables, such as fixed O&M and capital 
@@ -74,12 +73,13 @@ def calc_annual_costs_in_period(m, p):
 # Further comments on this are written in the Readme file.
 
 model.InvestmentCost = Expression(rule=lambda m: sum(
-                calc_annual_costs_in_period(m, p) * financials.uniform_series_to_present_value(
-                m.discount_rate, m.period_length_years[p]) * financials.future_to_present_value(
-                m.discount_rate, (m.period_start[p] - m.base_financial_year)) for p in m.PERIODS))
-model.OperationCost = Expression(rule=lambda m: sum(
-                sum(calc_tp_costs_in_period(m, t) for t in m.PERIOD_TPS[p]) * financials.uniform_series_to_present_value(
-                m.discount_rate, m.period_length_years[p]) * financials.future_to_present_value(
-                m.discount_rate, (m.period_start[p] - m.base_financial_year)) for p in m.PERIODS))
+                calc_annual_costs_in_period(m, p) * m.bring_annual_costs_to_base_year[p]
+                for p in m.PERIODS))
+
+model.OperationCost = Expression(rule=lambda m: 
+	sum(
+		sum(calc_tp_costs_in_period(m, t) for t in m.TPS_IN_PERIOD[p]
+		   ) * m.bring_annual_costs_to_base_year[p]
+	for p in m.PERIODS))
 
 print "model successfully loaded..."