RESOLVE Capacity Expansion Model

User Manual

November 2019

RESOLVE

Capacity Expansion Model

Energy and Environmental Economics, Inc.

44 Montgomery Street, Suite 1500

San Francisco, CA 94104

415.391.5100

www.ethree.com

User Manual

November 2019

RESOLVE

Capacity Expansion Model

 
 
 
Table of Contents 
Disclaimer ............................................................................................................................ i 
Introduction ................................................................................................................. 1 
1  Overview ....................................................................................................................... 1 
2  Structure of This Document ....................................................................................... 5 
Setting Up RESOLVE ................................................................................................. 6 
1  System Requirements ................................................................................................ 6 
1.1  Operating System ................................................................................ 6 
1.2  Python ................................................................................................. 6 
1.3  Solvers ................................................................................................ 7 
2  Installing Python and RESOLVE ............................................................................... 7 
2.1  Basic Installation & Updating Packages .............................................. 8 
2.2  Advanced Installation (Conda Virtual Environment) ............................ 9 
3  RESOLVE Package Organization .......................................................................... 10 
3.1  RESOLVE Package File Hashes ....................................................... 12 
Interacting with RESOLVE ....................................................................................... 14 
1  RESOLVE Scenario Tool ......................................................................................... 14 
1.1  Running Pre-Defined Scenarios ........................................................ 14 
1.2  Creating New Scenarios .................................................................... 18 
1.3  Input Data Worksheets ...................................................................... 20 
2  RESOLVE Results Viewer ....................................................................................... 22 

 
2.1  Dashboard ......................................................................................... 22 
2.2  Portfolio Analytics .............................................................................. 23 
2.3  Scenario Comparison ........................................................................ 24 
2.4  Raw Summary Results ...................................................................... 25 
2.5  Lists ................................................................................................... 26 
RESOLVE Model Details .......................................................................................... 27 
1  Raw Input Files .......................................................................................................... 27 
2  Python Scripts ............................................................................................................ 33 
3  Raw Output Files ....................................................................................................... 35 
 
 

Disclaimer

The core of the RESOLVE model is written in the Python scripting language. The E3 RESOLVE Model is free

software under the terms of the GNU Affero General Public License as published by the Free Software

Foundation, either version 3 of the License, or (at your option) any later version.

NO WARRANTY OF ANY KIND, IMPLIED, EXPRESSED, OR STATUTORY, INCLUDING BUT NOT LIMITED TO

THE WARRANTIES OF NON-INFRINGEMENT OF THIRD PARTY RIGHTS, TITLE, MERCHANTABILITY, FITNESS

FOR A PARTICULAR PURPOSE, AND FREEDOM FROM COMPUTER VIRUS, IS GIVEN WITH RESPECT TO THE

RESOLVE SOFTWARE INCLUDING ITS PYTHON SCRIPTS, THE WEB PAGE HOSTING THE RESOLVE SOFTWARE

OR HYPERLINKS TO OTHER INTERNET RESOURCES. E3 bears no responsibility for the consequences of any

modifications to the model, including its Python scripts, whether intentional or unintentional.

Introduction

1 Introduction

1.1 Overview

The purpose of this document is to provide users with the guidance needed to set up and run the RESOLVE

model and to analyze the results of scenarios once they have been completed. RESOLVE is a linear

program written in Python with Excel-based interfaces for scenario development and results processing.

RESOLVE is an optimal investment and operational model designed to inform long-term planning

questions around renewables integration in systems with high penetration levels of renewable energy.

The model is formulated as a linear optimization problem that co-optimizes investment and dispatch for

a selected set of days over a multi-year horizon in order to identify least-cost portfolios for meeting

renewable energy targets and other system goals.

RESOLVE incorporates a representation of neighboring regions in order to characterize transmission flows

into and out of a main zone of interest endogenously. As shown in Figure 1.1, RESOLVE can solve for the

optimal investments in renewable resources, energy storage technologies, demand response, new gas

plants, and retention of existing thermal plants. These least-cost decisions are subject to annual

constraints on delivered renewable energy that reflects the RPS policy, annual constraints on greenhouse

gas emissions, capacity adequacy constraints to maintain reliability, and constraints on the ability to

develop specific renewable resources. RESOLVE simulates operations on sampled days and includes a

linearized version of the unit commitment problem. An overview of constraint groups is shown in Table

1.1.

Page | 2 |

Introduction

Figure 1.1. RESOLVE Least-Cost Capacity Expansion Model

Introduction

Table 1.1. Overview of Key Constraint Groups in RESOLVE

Read more about conda environments here

Page | 10 |

Setting Up RESOLVE

1. If you have not yet installed Anaconda, follow the first step in the Basic Installation

instructions.RESOLVE Scenario Tool, users will find two options for controlling how the VBA

buttons call RESOLVE from the Scenario Tool:

a. Python Path:

Copy the path you found in step (2) into the yellow input cell. By default, the cell is

blank, and RESOLVE will use the default Python installed on your computer. By inputting

a path here, you are telling RESOLVE to use this specific version of Python.

b. Solver:

By default, this is set to cbc (which is a free open-source solver that can be used with

RESOLVE, see Section 2.1.3). Other options (if the user has correctly installed them) are

cplex and gurobi, and advanced users could modify the Scenario Tool to use additional

solvers if desired.

2.3 RESOLVE Package Organization

While users may wish to review the raw input and output files and Python scripts that constitute the core

of RESOLVE, the RESOLVE package is designed to allow users to run scenarios and analyze results using

only the Excel-based user interfaces for developing scenarios and viewing results. A schematic of the

RESOLVE environment is shown in Figure 2.1.

Setting Up RESOLVE

Figure 2.1. Schematic of RESOLVE Modeling Components

The individual components of the RESOLVE modeling environment are described below:

1. RESOLVE Resource Costs and Build Workbook

Excel workbook containing upstream information on baseline resources, candidate resource

cost and potential, and the Pro Forma financial model. Data updates can be implemented in this

workbook and manually copied over to corresponding sheets in the Scenario Tool.

2. RESOLVE Scenario Tool

Excel workbook that includes a scenario management dashboard and input data worksheets.

The Scenario Tool provides a simple interface to develop and run RESOLVE scenarios after

setting up data inputs to the model.

3. Input Files: inputs/

RESOLVE accepts scenario inputs as tab-delimited (.tab) text files. These are created by the

Scenario Tool and contain the minimal set of data needed to define scenario assumptions. The

inputs directory contains scenario subdirectories with input files for each scenario.

4. RESOLVE Python Scripts: resolve_code/

RESOLVE is written in Python and consists of multiple scripts that (1) read the raw input files, (2)

formulates a linear programming optimization model to minimize total incremental system

Page | 12 |

Setting Up RESOLVE

costs, and (3) exports the resulting optimal portfolio of investments and operations across the

analysis time horizon. The resolve_code directory contains the RESOLVE Python scripts.

5. Output Files: results/

RESOLVE prints out results in CSV files. These results files contain all results read by the Results

Viewer, as well as other files that report detailed information. The results directory and

scenario-specific subdirectories are created automatically upon running the RESOLVE code.

6. Results Viewer

Excel workbook that is the interface through which users may review results of a completed

RESOLVE run and compare results across runs. This workbook can be used to import and view

summaries of the raw Output Files for a specific model. Section 3.2 provides a summary of

functionality included in the Results Viewer. Scenario-specific subdirectories in the

results_summaries/ directory containing result summaries for each scenario are created

automatically when results are loaded into the Results Viewer.

7. setup.py

Setup script that installs relevant dependencies to configure RESOLVE.

8. environment.yml

Optional file used by conda to define a new virtual environment.

If the user is interested in reviewing inputs and outputs to the pre-built RESOLVE scenarios, the Resource

Cost and Build Workbook, Scenario Tool, and Results Viewer provide can be used directly in Excel without

installing Python or setting up RESOLVE.

2.3.1 RESOLVE PACKAGE FILE HASHES

File hashes for the current public release of RESOLVE are listed in Table 2.1.

Setting Up RESOLVE

Table 2.1. RESOLVE File Hashes

RESOLVE File

SHA-256 Hash

setup.py

771da4bba9eeaeba06f05ebaa92bd970c3250477dac9d21265c7b0a280ccc0b6

environment.yml

8614b3e68e5fbe57ffee555b3d75bd3e422af89f4884f3092e6bae35907da733

resolve_code/create_results_summary.py

bf678e70dc7a638a7faea2f279010a6f55750e53167b3412e283779ab1874920

resolve_code/export_results.py

32e2adda33cd56c0c552f4a9d841de1e4593fc3190fcfacf851493921886aae3

resolve_code/fileio.py

6b9ffba288f440a5670bd0376b0c5635fc44efaef2e62da135e23b3e938653b5

resolve_code/load_data.py

83bea04d476e43140a9461b7312c4a28874438bfdc56191d88fd90a99b117c68

resolve_code/model_formulation.py

0777225ef7f91fb1150bc8ac53d53a978e7ca3b2f8aa3df887ca52f9a42c4be9

resolve_code/run_opt.py

698871f7e374f1353f2e0c5b30c8786137c853421d0b4e8dce4c90ce533398c2

resolve_code/runbatch.py

4c9a34b949b424e5baa545bfef87154d08aa55c8e71a5ab504b1669d2ff55d61

Page | 14 |

Interacting with RESOLVE

3 Interacting with RESOLVE

3.1 RESOLVE Scenario Tool

Cells throughout the Scenario Tool are color-coded to indicate how users should interact with them.

Figure 3.1. RESOLVE Scenario Tool Dashboard

3.1.1 RUNNING PRE-DEFINED SCENARIOS

The RESOLVE Scenario Tool comes with a set of predefined scenarios. Users can load and run a predefined

scenario as follows:

1. The dropdown menu in cell D6 of the Dashboard lists all predefined scenarios (stored on the

Scenario Settings tab). You can use the Single Scenario Selection dropdown and 1a. Load

Cell Type

Notes

Inputs

Yellow cells indicate user-defined inputs. These can be changed by the user, although it is advised to

choose from any of the existing options through the scenario toggles in the Dashboard tab.

Dropdown

Orange cells indicate values. These can be changed by the user but must adhere to a set of pre-defined

options listed in the dropdown list. For example, True/False toggles or the names of available zones in

the pre-defined cases.

(Note: True/False toggles will change color from green to red to help indicate the toggle setting)

Results

Green cells indicate cells linked to other inputs via formulas. These should automatically update, and it is

advised not to change the formulas to ensure links stay intact.

Fixed or

Inactive

Grey cells indicate calculations that should not be changed or values that are inactive.

Labels

These are hard-coded labels indicating what the cell(s) adjacent to it should be used for.

Interacting with RESOLVE

Selected Scenario Settings button (shown in Figure 3.2) to load a scenario’s settings into the

Dashboard for viewing.

Figure 3.2. RESOLVE Single Scenario Selection Interface

2. To run scenarios, use the Batch Scenario Run section of the Dashboard to add/remove named

scenarios (listed in the Saved Scenarios list). You may need to press the 2. Refresh Saved

Scenarios List button to update the list. To add selected scenarios, select the cells with the

scenario names you are interested in running, then press the Add Selected button. The selected

scenarios should appear in the Scenarios to be Run list to the right. The Add All button will add

all scenarios in the Saved Scenarios list to the right side. The Remove Selected/All buttons

works similarly.

3. Press the 3. Create Input Files button, which will invoke VBA code that prints out the input data

into the appropriate scenario-specific input files in an input subdirectory with the scenario’s

name.

Page | 16 |

Interacting with RESOLVE

Figure 3.3. RESOLVE Scenario Run Interface

4. Press the 4. Run Scenarios button to run RESOLVE. This will open a command prompt window

and show text output as RESOLVE runs. RESOLVE runs can take anywhere between 10 minutes

to many hours depending on specific scenario settings and which solver is being used. Note that

the command prompt window will stay open even after RESOLVE has completed running.

3.1.1.1 Running RESOLVE via Command Line

There are a few reasons you may choose to run RESOLVE via the command line. For example, if you are

running on a Linux or macOS machine, a machine without Excel installed, or a machine where you cannot

use VBA macros, you will not be able to use the Scenario Tool to run RESOLVE. Additionally, if you want

to use a non-default solver (as described in Section 2.1.3), the easiest way to specify the name of the

solver is as a command line argument.

To run RESOLVE via the command line, navigate in your selected command line terminal (e.g., Command

Prompt, PowerShell, bash, etc.) to the resolve_code directory. This is generally done using the cd

Interacting with RESOLVE

command (for change directory). For example, to switch to the directory where you have stored your

resolve_code folder, you would use the command once, as shown below:

>> cd resolve_code

From within the resolve_code directory, you have two options for running RESOLVE cases.

1. The first is to run a single case using the run_opt.py script. There are two command line

arguments that follow. The first argument is required, [scenario name], and is the name of the

single scenario you plan to run. This must correspond to the name of a directory in the inputs

folder. The second argument is optional, [optional: solver name], and is used to specify a

different solver than the default CBC solver:

>> python run_opt.py [scenario name] [optional: solver name]

For example, to run the scenario full_run with and without the solver Gurobi, use the

commands:

>> python run_opt.py full_run

>> python run_opt.py full_run gurobi

2. The second option is to run a batch of scenarios using the runbatch.py script by populating the

CSV file called cases_to_run.csv with the list of scenarios you wish to run. runbatch.py runs all of

the scenarios cases_to_run.csv in serial. The command line input is:

>> python runbatch.py [optional: solver name]

To run a batch of scenarios with and without the solver Gurobi, use the commands:

Page | 18 |

Interacting with RESOLVE

>> python runbatch.py gurobi

>> python runbatch.py

3.1.2 CREATING NEW SCENARIOS

Users may wish to create and run new scenarios based on existing input data, rather than the pre-defined

scenarios. This can be done by adjusting the dropdowns in the Scenario Definition box on the RESOLVE

Dashboard.

To create a new scenario:

0. (Optional) Select and load a scenario of interest in the Single Scenario Selection box

1. Customize the dropdown toggles in the Single Scenario Definition box (as shown in Figure 3.4).

As indicated by the cell color, most of the inputs are dropdowns, but some (such as discount

rate) are direct user inputs.

Interacting with RESOLVE

Figure 3.4. RESOLVE Dashboard Single Scenario Selection and Scenario Definition Interface

2. Save the new custom scenario by pressing the 1b. Save Current Inputs as New Scenario button.

A macro will prompt the user to enter a new scenario name and saves the scenario settings in

the Scenario Settings tab. If a user would like to overwrite an existing scenario, they can do so

by entering an existing scenario name in the dialog box. The macro will prompt you to confirm

overwriting the scenario.

Note that if a user changes any scenario toggles, they must be saved via the macro button for

them to take effect in subsequent runs.

3. Press the 3. Create Input Files button, which will invoke VBA code that prints out the input data

into the appropriate scenario-specific input files in an input subdirectory with the scenario’s

name.

Page | 20 |

Interacting with RESOLVE

4. Press the 4. Run Scenarios button to run RESOLVE. This will open a command prompt window

and show text output as RESOLVE runs. RESOLVE runs can take anywhere between 10 minutes

to many hours depending on specific scenario settings and which solver is being used. Note that

the command prompt window will stay open even after RESOLVE has completed running.

3.1.3 INPUT DATA WORKSHEETS

The input data worksheets in the Scenario Tool are categorized into three high-level themes: system,

loads, and resources. Brief summaries of each tab are included in Table 3.1. The cells in these data

worksheets also adhere to the color-coding described at the start of Section 3.1.

Additional documentation of the contents of the data worksheets is provided in the Inputs & Assumptions

documentation.

Interacting with RESOLVE

Table 3.1. Description of RESOLVE Scenario Tool Worksheets

Section

Tab

Description

DASHBOARD

Dashboard

Contains a user interface to prepare and run scenarios. This is the main tab the user will interact with

SCENARIOS

Scenario Settings

Database with scenario settings that were saved through the user interface

SYSTEM

Sys - Fuels

Fuel cost assumptions and calculations

Sys - PRM

Planning reserve margin assumptions and calculations

Sys - RPS GHG

Renewable Portfolio Standard and greenhouse gas target assumptions

Sys - Reserves

Operating reserves requirement assumptions

Sys - Tx

Regional assumptions, such as transmission limits, hurdle rates, etc.

Sys - Inputs Passthrough

Inputs to passthrough to the results (e.g., baseline costs)

LOADS

Loads - Profiles

Hourly load profiles by end-use

Loads - Forecasts

Database of load forecasts

Loads - Hydrogen

Inputs for hydrogen electrolysis feature

Loads - EE

Inputs for EE investment feature

Loads - EV

Electric vehicles assumptions and calculations

Loads - DR

Assumptions for candidate demand response resources

Loads - Flexible

Assumptions for flexible loads

SUPPLY RESOURCES

Technologies

Technology characteristics

Resources - All

Resource inputs and mapping to technologies

Resources - Candidate

Inputs for resources that can be expanded

Resources - Variable

Shapes for variable resources

Resources - Hydro

Energy budgets and constraints for hydro resources

Resources - Maintenance

Maintenance/derate schedules (if not defined, resource will be rated at 100%)

Resources - Scenarios

Scenario combinations of cost and planned installed capacities

Resources - Scenario Costs

Database of resource all-in fixed investment costs for expansion

Resources - Scenario Build

Database of resource planned installed capacities

RAW INPUTS

Inputs2Write

Calculation worksheet that converts inputs from all other tabs into model-ready input tables. Do not edit

Inputs2Write_Hourly

Calculation worksheet that converts inputs from all other tabs into model-ready input tables. Do not edit

LISTS

Lists

Calculation worksheet with supporting lists and tables. Do not edit

Page | 22 |

Interacting with RESOLVE

3.2 RESOLVE Results Viewer

The Results Viewer allows the user to look at the summary results of a scenario of interest. It contains

four main worksheets: Dashboard, Portfolio Analytics, Scenario Comparison, and Lists. In addition, it

contains a number of “raw” summary results files for the scenario of interest.

3.2.1 DASHBOARD

The Dashboard worksheet is the main worksheet the user will interact with to look at the results of a

single scenario.

Figure 3.5. Results Viewer Dashboard

The Dashboard contains the following macro buttons:

4. The “Refresh List of Scenarios” macro lists all subdirectories that exist in the results directory.

Note that if a RESOLVE run is interrupted mid-run, a results folder will be created but no results

Interacting with RESOLVE

will be available. Consequently, the user will not be able to load results for that scenario until the

simulation has completed running.

5. After selecting one of the scenarios from the list under “Scenarios with Results”, the “Retrieve

Results of Selected Scenario” macro will load all summary results files into the appropriate

worksheets (named raw_ + file name) for the selected scenario.

A common cause of errors is the fso.GetFolder() function in the VBA macro. If this function raises an

error, go to Tools > References > find and tick 'Microsoft Scripting Runtime'.

The Dashboard worksheet contains key summary results for the CAISO zone, such as the resource

buildout, portfolio costs, etc. The worksheet also includes graphs on the right side of the tables.

The year columns are grouped using Excel’s grouping functionality (see Data > Outline > Group), and can

be expanded and minimized by clicking on the “+” or “-“ signs in the columns sidebar, or by clicking on the

numbers (1,2) on the top left of the spreadsheet. Note that expanding the grouped columns will interfere

with the formatting of the charts. If the user has created a RESOLVE scenario that looks at a different set

of years than the default case (2020, 2021, 2022, 2023, 2024, 2026, 2030), the “Regroup Columns” macro

will regroup the columns to show the representative set of years.

3.2.2 PORTFOLIO ANALYTICS

This worksheet contains more detailed summary tables that are pulled from the raw summary results

worksheets and processed where necessary.

The results are grouped using Excel’s grouping functionality (see Data > Outline > Group), and can be

expanded and minimized by clicking on the “+” or “-“ signs in the rows/columns sidebar, or by clicking on

the numbers (1,2) on the top left of the spreadsheet. If the user has created a RESOLVE scenario that looks

at a different set of years than the default case, the “Regroup Columns” macro will regroup the columns

to show the representative set of years.

Page | 24 |

Interacting with RESOLVE

3.2.3 SCENARIO COMPARISON

This worksheet is set up so the user can compare summary results of multiple scenarios. It allows the user

to select scenarios of interest and to compare the summary results of these scenarios for a year of interest.

The summary results are the same as those shown on the Dashboard for an individual scenario.

Figure 3.6. Results Viewer - Scenario Comparison Worksheet

To compare a set of scenarios, follow the steps below:

1. Refresh the list of available scenarios by pressing “Refresh List of Scenarios”. This macro lists all

subdirectories that exist in the results directory. Results files must be present for all scenarios for

the scenario comparison function to work.

2. [Optional] If necessary, remove any scenarios you don’t want to compare from the “Scenarios to

Compare” list by using the “Remove Selected” or “Remove All” Button.

a. The “Remove Selected” macro will remove the selected scenario from the list of scenarios

listed under “Scenarios to Compare”. Note that in this context, selected scenario means

the cell that is selected within the “Scenarios to Compare”. If a cell outside of this box is

selected, a warning will pop up and the macro will stop.

Interacting with RESOLVE

b. The “Remove All” macro will remove all scenarios listed under “Scenarios to Compare”.

3. Add scenarios of interest to the “Scenarios to Compare” list using the “Add Selected” or “Add All”

buttons.

a. The “Add Selected” macro will add the selected scenario to the list of scenarios listed

under “Scenarios to Compare”. Note that in this context, selected scenario means the cell

that is selected within the “Saved Scenarios Menu” box (not the value in cell D8). If a cell

outside of this box is selected, a warning will pop up and the macro will stop.

b. The “Add All” macro will add all scenarios listed under “Saved Scenarios” to the list of

scenarios listed under “Scenarios to Compare”.

4. Select a year of interest in cell I8 (shaded yellow). Please ensure that this is a year for which there

are RESOLVE results.

5. Compare all selected scenarios by pressing the “Compare” macro button. This macro will load the

summary results for each of the scenarios listed under “Scenarios to Compare” to the Dashboard,

and then copy the results for the year of interest to the Scenario Comparison table. It can take a

few minutes to compare the results of many scenarios because the “Retrieve Results of Selected

Scenario” macro on the Dashboard is called upon many times in a row.

3.2.4 RAW SUMMARY RESULTS

The set of worksheets that start with “raw_” contain a copy of the raw summary results files for the

scenario of interest. Whenever the macro “Retrieve Results of Selected Scenario” is run, these

worksheets are updated. Note that while the user should be able to find all the information needed in the

first three tabs, if there is interest in investigating the “raw_” tabs to filter or edit any specific data,

rerunning the “Retrieve Results of Selected Scenario” macro will overwrite any changes that have made.

In that case, make sure to press the “Retrieve Results of Selected Scenario” button twice to ensure all

the “raw_” tabs are fully refreshed.

Page | 26 |

Interacting with RESOLVE

3.2.5 LISTS

This worksheet contains a set of lists to support the functions in this workbook. The user should not

change anything in this worksheet.

RESOLVE Model Details

4 RESOLVE Model Details

Advanced users may want to review the raw input and output files directly for their own analyses. This section provides a description of the raw

files.

4.1 Raw Input Files

When the user selects and runs a scenario from the Scenario Tool, RESOLVE will generate a series of text-based input files for the linear program.

While running RESOLVE does not require users to manipulate these files directly, users may wish to review their contents and structure. Each of

the input files to RESOLVE is described in Table 4.1. Some input files are only written when certain features are enabled.

Page | 28 |

RESOLVE Model Details

Table 4.1. RESOLVE Raw Input Files

Input File

Description

capacity_groups.tab

Specifies which resources are in each annual capacity installation limit group.

capacity_limits.tab

The maximum capacity of each resource that can be built in each period. Resources that do not have capacity

limits are not included.

capacity_limits_local.tab

The maximum capacity of each resource that can be built in each period, specifically in local capacity areas.

conventional_dr_period_limits.tab

The maximum amount of energy that can be dispatched (shed) annually and daily capacity factor limit (i.e., call

hours/day) from conventional (Shed) demand response resources.

day_weights.tab

The weight associated with each day in RESOLVE; should sum up to 365.

elcc_surface.tab

Effective load carrying capability (ELCC) surface facet coefficients for wind and solar power.

ev_params.tab

The charging efficiency of each EV fleet.

ev_period_params.tab

The total battery capacity of each EV fleet in each period, and the minimum energy that must always be

available in each fleet’s battery.

ev_timepoint_params.tab

The amount of demand from each EV fleet in each timepoint.

feature_toggles.tab

Toggles to enable/disable optional features. Some optional features may not work without input data.

flexible_load_capacity_period_params.tab

Flexible load (shift) minimum and maximum resource potential limits for each period.

flexible_load_cost_curve.tab

Flexible load (shift) supply curve for each period.

flexible_load_cost_curve_index.tab

Indices that define each breakpoint in the flexible load (shift) supply curve.

RESOLVE Model Details

Input File

Description

flexible_load_timepoint_params.tab

The amount of load that can be shifted up or down in each timepoint as a fraction of the total daily flexible

load potential.

flexible_params.csv

The maintenance derate fraction (1 is fully available, 0 is completely unavailable) for each resource that has a

specified maintenance schedule. Other parameter values can also be set using this input file.

fuel_prices.tab

The price of each fuel by period and month.

fuels.tab

Defines the set of fuels and the carbon content of each fuel.

ghg_import_rates.tab

The assumed greenhouse gas (GHG) emissions intensity resulting from imports into the main zone in each

period for each transmission line.

ghg_targets.tab

GHG targets for the main zone in each period.

hurdle_rates.tab

Hurdle rates (cost per MW of energy flow) on each transmission line by period for both flow directions.

hydro_daily_params.tab

The daily energy budget, minimum generation level, and maximum generation level for each hydro resource

and each day.

hydro_ramps.tab

The limits on hydro ramps for each ramp duration for the main zone hydro resource.

hydro_resources_ramp_limited.tab

Hydro resources that are subject to multi-hour ramping constraints.

hydrogen_electrolysis_daily_params.tab

The minimum hourly hydrogen load and daily average hydrogen load for each day.

hydrogen_electrolysis_period_params.tab

The hydrogen electrolysis installed capacity for each period.

inputs_passthrough.csv

Values to calculate total future revenue requirement and rate impacts of RESOLVE investment and operational

decisions.

min_cumulative_new_build.tab

The minimum amount of new capacity of each resource that must be built through each period. The cost of

building these resources is not assumed to be sunk (in contrast to planned_installed_capacities.tab).

Page | 30 |

RESOLVE Model Details

Input File

Description

period_discount_factors.tab

The weight/discount factor applied to costs occurring in each period, and the number of years represented by

each period.

planned_installed_capacities.tab

The planned installed capacity, minimum operational capacity (lower bound on thermal retention), and fixed

O&M cost of each resource in each period (as opposed to varying by vintage for new build resources). The

capital cost of capacity included here is assumed to be sunk and consequently is not included in the

optimization.

planned_storage_energy_capacity.tab

The planned installed energy capacity and fixed O&M cost of each storage resource in each period.

planning_reserve_margin.tab

The planning reserve margin target in each period, and other quantities related to the planning reserve margin.

Also included is the amount of capacity needed in local areas in each period.

renewable_targets.tab

RPS target information by period.

reserve_resources.tab

Determines which resources can provide different reserve types.

reserve_timepoint_requirements.tab

The regulation and load-following reserve requirements in each timepoint.

resource_prm_nqc.tab

The net qualifying capacity (NQC) fraction for firm capacity and storage resources.

resource_tx_zone_map.tab

Specifies the relationship between resources and transmission deliverability zones.

resource_tx_zones.tab

Specifies whether the resource is located inside or outside of the planning reserve margin area (for resource

adequacy accounting).

resource_use_tx_capacity.tab

Allocate dedicated firm transmission rights between zones to certain resources.

resource_variable_renewable

Flags indicating which variable renewable resources are curtailable.

resource_variable_renewable_prm.tab

Parameters related to variable renewable resource participation in the planning reserve margin and local

capacity constraints.

RESOLVE Model Details

Input File

Description

resource_vintage_params.tab

The annual capital cost per unit of capacity ($/kW-yr) and fixed O&M cost by resource (new build resources

only) and vintage.

resource_vintage_storage_params.tab

The annual fixed cost of per unit of energy capacity ($/kWh-yr) and fixed O&M cost for storage resources by

vintage.

resources.tab

Defines for each resource the associated technology, zone, RPS eligibility, and other characteristics.

shapes.tab

The normalized profiles for each variable resource for each day and hour.

simultaneous_flow_group_lines.tab

The line-directions included in each simultaneous flow group.

simultaneous_flow_groups.tab

The names of the groups of lines over which simultaneous flow constraints are enforced.

simultaneous_flow_limits.tab

The limits on flow over each simultaneous flow group by period.

system_params.tab

A range of single-value parameters including penalties for unserved energy, overgeneration, and reserve

violations; the durations of hydro and intertie ramps to constrain; parameterizations of the sub-hourly

behavior when providing regulation and load-following reserves; parameterizations of the ability of variable

generation to provide reserves; whether to require renewable overbuild when satisfying RPS constraints;

whether to allow RPS banking; whether to enforce GHG targets; the number of hours of duration that receives

full ELCC credit; and the assumed timeframe for operational reserves.

tech_dispatchable_params.tab

Parameters associated with each dispatchable thermal technology: minimum stable level as fraction of

capacity, ramp rate as fraction of capacity, startup and shutdown time (integer hours), unit size, and startup

and shutdown costs.

tech_storage_params.tab

Parameters associated with each storage technology: charging and discharging efficiencies, and minimum

storage duration.

tech_thermal_params.tab

Parameters associated with each thermal technology: the fuel used, and the fuel burn slope and intercept.

Page | 32 |

RESOLVE Model Details

Input File

Description

technologies.tab

All technologies modeled, with flags for various operational characteristics.

timepoints.tab

All timepoints modeled with their associated metadata: which period, month, and day the timepoint is in, and

which hour of the day it represents.

transmission_lines.tab

All transmission lines with their origin (from) and destination (to) for the positive flow direction, the minimum

and maximum flow on the line, a flag for whether the line is ramp-constrained, and a flag for whether a hurdle

rate is applied on the line.

transmission_ramps.tab

The up and down ramp limits for each ramp-constrained line for each ramp duration.

tx_zones.tab

The transmission zone aggregations for which energy only or fully deliverable transmission capacity will be built

for new renewable resources. Capacity limits for energy only and zero-cost fully deliverable capacity are

included, along with the cost to build new fully deliverable capacity.

zone_curtailment_costs

The cost of curtailment in each zone in each period.

zone_timepoint_load.tab

The input load in each zone in each timepoint.

zones.tab

The zones modeled, as well as flags that determine if the zone is included or excluded from various constraints

(RPS, GHG, Planning Reserve Margin, Load Following), and the zonal spinning reserve requirements.

RESOLVE Model Details

4.2 Python Scripts

The input files described in Section 4.1 are formatted to be the minimal set of information needed to formulate the linear programming

optimization model written in Python. Users are not expected to interact directly with the Python optimization model or its accompanying

input/output scripts; however, users are welcome to review the structure and logic of RESOLVE’s formulation.

The Python scripts that make up RESOLVE are summarized in Table 4.2

Table 4.2. RESOLVE Python Scripts

Python Script

Description

run_opt.py

This is the ‘main’ script of the RESOLVE model. As described in Section 3.1.1.1, it takes one required argument: the name of the

scenario to run. For example, to run a scenario named ‘full_run,’ the user would need to run the run_opt.py script and give it the

argument full_run. The scenario name must be the same as the name of a subdirectory in the inputs directory.

model_formulation.py

This script contains the RESOLVE problem formulation. RESOLVE is written in Pyomo, a Python-based optimization modeling

language. The model object is defined in model_formulation.py and is called resolve_model. It is a Pyomo AbstractModel object,

which is then assigned various attributes—parameters, sets, variables, and constraints—that describe the RESOLVE linear problem.

load_inputs.py

This script contains the scenario_data function that returns a DataPortal Pyomo object. The DataPortal is a way to load data into a

Pyomo AbstractModel class. The scenario_data function takes the scenario inputs directory as argument, finds the TAB files

containing the scenario data, and initializes the resolve_model class with these data (see the create_problem_instance function in

run_opt.py).

export_results.py

This script contains the export_results function, which is called by run_opt.py when the problem is solved. This function takes the

model instance, results, and scenario results directory as arguments. A final argument, debug_mode, tells the function how to

Page | 34 |

RESOLVE Model Details

Python Script

Description

handle errors that may arise: exit if debug_mode is set to 0, launch the Python debugger if debug_mode is set to 1. The

export_results function calls other functions that export various optimization results, e.g. the build variables, the operations

variables, the transmission flows, etc.

create_summaries.py

Once results are exported, run_opt.py calls the create_summaries function from create_results_summaries.py file. This function

calls various other functions, also in the create_results_summaries.py, that perform various aggregations of the results and write

them to the summary subdirectory in each scenario’s results directory.

runbatch.py

This script simply runs the run_opt.py file in series for each of the scenarios listed in ‘cases_to_run.csv’. It allows users to run a

batch of scenarios simply, rather than waiting for each scenario to finish before running the next one. Note: advanced users may

wish to run many scenarios in parallel by opening multiple command prompt windows. Runbatch.py runs each scenario in series,

not in parallel.

fileio.py

Stores functions related to file input and output.

RESOLVE Model Details

4.3 Raw Output Files

Output files in the root scenario results directory (e.g., results/full_run/) are created by export_results.py. Files in the root results directory are not

directly used in the results viewer but are available when more detailed analysis of results is required.

Output files in the summary subdirectory within each scenario results directory (e.g., results/full_run/summary/) are created by

create_results_summary.py. The summary directory contains various aggregations and combinations of the data in the results files described below.

Only data in the summary directory is imported into the results viewer.

Note that the term ‘dual’ is used frequently to refer to the shadow price of a constraint. Technical note on dual values: the reported values reflect

real dollars in either hourly or annual quantities.

Table 4.3. RESOLVE Raw Output Files

Output File

Description

curtailment.csv

This file contains hourly and sub-hourly variable renewable curtailment decisions for each zone in RESOLVE.

Blank cells indicate that RESOLVE does not make this decision.

elcc_surface_facets.csv

This file contains results for each facet of the effective load carrying capability (ELCC) surface in each period. These

results can show which (if any) of the ELCC facets is active in each period.

Page | 36 |

RESOLVE Model Details

Output File

Description

fuel_burn_by_resource.csv

This file contains the hourly fuel burn and greenhouse gas (GHG) emissions by resource.

ghg.csv

This file contains the input GHG emissions target in each period, and the shadow price of meeting that target.

Blank cells indicate that a GHG emissions target was not modeled.

ghg_imports.csv

This file contains the hourly GHG emissions imported along transmission lines.

loads_and_power_balance.csv

This file contains hourly loads for each zone and timepoint that were input into the RESOLVE optimization, as well as the

amount of overgeneration and unserved energy for each zone and timepoint. It also contains the shadow price of the

zonal power balance constraint for each timepoint, which is analogous to the hourly energy price. Care should be taken

interpreting this energy price as the RESOLVE investment framework differs in several fundamental ways from a

conventional production simulation.

local_capacity_resources.csv

This file contains the local capacity investment decisions made by RESOLVE for each local capacity resource in each

period.

objective_function_value.txt

The final value of the objective function for each RESOLVE run.

operations.csv

This file contains the hourly operational decisions made by RESOLVE for each resource on all days modeled in RESOLVE

(currently 37 days per year). Types of operational decisions included in this file are:

• unit commitment

• power production

• reserve commitment

• flexible load dispatch

• storage charging

These operational decisions are also converted to operational costs, including variable O&M costs, fuel costs,

startup/shutdown costs, and curtailment costs.

Blank cells indicate that RESOLVE does not make this decision.

RESOLVE Model Details

Output File

Description

planning_reserve_margin.csv

This file contains the input planning reserve margin (PRM) and local capacity targets in each period, as well as the

shadow price of meeting those targets. A summary of PRM contributions by resource type is also included.

ramping_duals.csv

This file contains the hourly values relating to ramp constraints of dispatchable thermal generation.

reserve_timepoints.csv

This file contains the hourly reserve commitment shortfalls (or violations) and shadow prices for each reserve product.

resource_build.csv

This file contains the investment and economic thermal retention decisions made by RESOLVE for each candidate

resource in each period. This file also contains capacities of resources for which no investment decisions were made (e.g.

existing resources and contracted resources that come online at some point in the future). The fully deliverable/energy

only status of new renewable resources is included.

The associated capital and fixed O&M costs for these build and retirement decisions is also included.

Blank cells indicate that RESOLVE does not make this decision.

rps.csv

This file contains the input RPS target level in each period, as well as the shadow price of the RPS constraint, the RPS

credits banked, and a high-level summary of the components of the RPS constraint.

sim_flow_group_duals.csv

This file contains the hourly shadow prices of constraints that limit the sum of flows on groups of transmission lines.

storage_build.csv

This file contains investment decisions made by RESOLVE in every period for the energy capacity of each storage

resource.

Blank cells indicate that RESOLVE does not make a decision.

transmission_costs.csv

This file contains the cost of building new transmission in each period triggered by new renewable resource investment

decisions made by RESOLVE. Also included is the breakdown of fully deliverable/energy only capacity in each

transmission zone in each period.

transmit_power.csv

This file contains the hourly transmission dispatch decisions made by RESOLVE for each transmission line, as well as the

shadow price of flow limits on each line.

Page | 38 |

RESOLVE Model Details