reV rep-profiles

Execute the rep-profiles step from a config file.

reV rep profiles compute representative generation profiles for each supply curve point output by reV supply curve aggregation. Representative profiles can either be a spatial aggregation of generation profiles or actual generation profiles that most closely resemble an aggregated profile (selected based on an error metric).

The general structure for calling this CLI command is given below (add --help to print help info to the terminal).

Usage

reV rep-profiles [OPTIONS]

Options

-c, --config_file <config_file>

Required Path to the rep-profiles configuration file. Below is a sample template config

JSON/JSON5YAMLTOML

{
    "execution_control": {
        "option": "local",
        "allocation": "[REQUIRED IF ON HPC]",
        "walltime": "[REQUIRED IF ON HPC]",
        "qos": "normal",
        "memory": null,
        "queue": null,
        "feature": null,
        "conda_env": null,
        "module": null,
        "sh_script": null,
        "keep_sh": false,
        "num_test_nodes": null,
        "max_workers": null
    },
    "log_directory": "./logs",
    "log_level": "INFO",
    "gen_fpath": "[REQUIRED]",
    "rev_summary": "[REQUIRED]",
    "reg_cols": "[REQUIRED]",
    "cf_dset": "cf_profile",
    "rep_method": "meanoid",
    "err_method": "rmse",
    "weight": "gid_counts",
    "n_profiles": 1,
    "aggregate_profiles": false,
    "scale_profiles": null,
    "save_rev_summary": true,
    "scaled_precision": false,
    "analysis_years": null
}

Parameters

execution_controldict

Dictionary containing execution control arguments. Allowed arguments are:

option:

({‘local’, ‘kestrel’, ‘eagle’, ‘awspc’, ‘slurm’, ‘peregrine’}) Hardware run option. Determines the type of job scheduler to use as well as the base AU cost. The “slurm” option is a catchall for HPC systems that use the SLURM scheduler and should only be used if desired hardware is not listed above. If “local”, no other HPC-specific keys in are required in execution_control (they are ignored if provided).

allocation:

(str) HPC project (allocation) handle.

walltime:

(int) Node walltime request in hours.

qos:

(str, optional) Quality-of-service specifier. For Kestrel users: This should be one of {‘standby’, ‘normal’, ‘high’}. Note that ‘high’ priority doubles the AU cost. By default, "normal".

memory:

(int, optional) Node memory max limit (in GB). By default, None, which uses the scheduler’s default memory limit. For Kestrel users: If you would like to use the full node memory, leave this argument unspecified (or set to None) if you are running on standard nodes. However, if you would like to use the bigmem nodes, you must specify the full upper limit of memory you would like for your job, otherwise you will be limited to the standard node memory size (250GB).

max_workers:

(int, optional) Number of parallel rep profile workers. 1 will run serial, while None will use all available. By default, None.

queue:

(str, optional; PBS ONLY) HPC queue to submit job to. Examples include: ‘debug’, ‘short’, ‘batch’, ‘batch-h’, ‘long’, etc. By default, None, which uses “test_queue”.

feature:

(str, optional) Additional flags for SLURM job (e.g. “-p debug”). By default, None, which does not specify any additional flags.

conda_env:

(str, optional) Name of conda environment to activate. By default, None, which does not load any environments.

module:

(str, optional) Module to load. By default, None, which does not load any modules.

sh_script:

(str, optional) Extra shell script to run before command call. By default, None, which does not run any scripts.

keep_sh:

(bool, optional) Option to keep the HPC submission script on disk. Only has effect if executing on HPC. By default, False, which purges the submission scripts after each job is submitted.

num_test_nodes:

(str, optional) Number of nodes to submit before terminating the submission process. This can be used to test a new submission configuration without submitting all nodes (i.e. only running a handful to ensure the inputs are specified correctly and the outputs look reasonable). By default, None, which submits all node jobs.

Only the option key is required for local execution. For execution on the HPC, the allocation and walltime keys are also required. All other options are populated with default values, as seen above. For HPC runs, configured execution parameters are also exported to the job environment. Variable names are uppercase and prefixed with the CLI name; non-alphanumeric characters are replaced by underscores. For example, the walltime parameter for a CLI named reV is available as REV_WALLTIME. These variables can be used by sh_script or by the command running on the node. The submitted job name is also available as REV_JOB_NAME.

log_directorystr

Path to directory where logs should be written. Path can be relative and does not have to exist on disk (it will be created if missing). By default, "./logs".

log_level{“DEBUG”, “INFO”, “WARNING”, “ERROR”}

String representation of desired logger verbosity. Suitable options are DEBUG (most verbose), INFO (moderately verbose), WARNING (only log warnings and errors), and ERROR (only log errors). By default, "INFO".

gen_fpathstr

Filepath to reV generation output HDF5 file to extract cf_dset dataset from.

Note

If executing reV from the command line, this path can contain brackets {} that will be filled in by the analysis_years input. Alternatively, this input can be set to "PIPELINE", which will parse this input from one of these preceding pipeline steps: multi-year, collect, generation, or supply-curve-aggregation. However, note that duplicate executions of any of these commands within the pipeline may invalidate this parsing, meaning the gen_fpath input will have to be specified manually.

rev_summarystr | pd.DataFrame

Aggregated reV supply curve summary file. Must include the following columns:

• res_gids : string representation of python list containing the resource GID values corresponding to each supply curve point.

• gen_gids : string representation of python list containing the reV generation GID values corresponding to each supply curve point.

• weight column (name based on weight input) : string representation of python list containing the resource GID weights for each supply curve point.

Note

If executing reV from the command line, this input can be set to "PIPELINE", which will parse this input from one of these preceding pipeline steps: supply-curve-aggregation or supply-curve. However, note that duplicate executions of any of these commands within the pipeline may invalidate this parsing, meaning the rev_summary input will have to be specified manually.

reg_colsstr | list

Label(s) for a categorical region column(s) to extract profiles for. For example, "state" will extract a rep profile for each unique entry in the "state" column in rev_summary. To get a profile for each supply curve point, try setting reg_cols to a primary key such as "sc_gid".

cf_dsetstr, optional

Dataset name to pull generation profiles from. This dataset must be present in the gen_fpath HDF5 file. By default, "cf_profile"

Note

If executing reV from the command line, this name can contain brackets {} that will be filled in by the analysis_years input (e.g. "cf_profile-{}").

rep_method{‘mean’, ‘meanoid’, ‘median’, ‘medianoid’}, optional

Method identifier for calculation of the representative profile. By default, 'meanoid'

err_method{‘mbe’, ‘mae’, ‘rmse’}, optional

Method identifier for calculation of error from the representative profile. If this input is None, the representative meanoid / medianoid profile will be returned directly. By default, 'rmse'.

weightstr, optional

Column in rev_summary used to apply weights when computing mean profiles. The supply curve table data in the weight column should have weight values corresponding to the res_gids in the same row (i.e. string representation of python list containing weight values).

Important

You’ll often want to set this value to something other than None (typically "gid_counts" if running on standard reV outputs). Otherwise, the unique generation profiles within each supply curve point are weighted equally. For example, if you have a 64x64 supply curve point, and one generation profile takes up 4095 (99.98%) 90m cells while a second generation profile takes up only one 90m cell (0.02%), they will contribute equally to the meanoid profile unless these weights are specified.

By default, SupplyCurveField.GID_COUNTS.

n_profilesint, optional

Number of representative profiles to save to the output file. By default, 1.

aggregate_profilesbool, optional

Flag to calculate the aggregate (weighted meanoid) profile for each supply curve point. This behavior is in lieu of finding the single profile per region closest to the meanoid. If you set this flag to True, the rep_method, err_method, and n_profiles inputs will be forcibly set to the default values. By default, False.

scale_profilesstr, optional

Column in rev_summary containing the target capacity factor for each supply curve point, for example "capacity_factor_ac" or "capacity_factor_dc". Each output profile is multiplied by the target capacity factor divided by its temporal mean. This preserves profile shape while matching capacity factors adjusted after supply curve aggregation (e.g. by NRWAL or manual post-processing). For regions containing multiple supply curve points, targets are averaged using the sum of each point’s weight values, or equally if weight is None. Targets must be finite and nonnegative. A zero target produces a zero profile; a zero-mean profile cannot be scaled to a positive target. Profiles are not clipped, so scaled values may exceed one. By default, None (no scaling).

save_rev_summarybool, optional

Flag to save full reV supply curve table to rep profile output. By default, True.

scaled_precisionbool, optional

Flag to scale cf_profiles by 1000 and save as uint16. By default, False.

analysis_yearsint | list, optional

A single year or list of years to perform analysis for. These years will be used to fill in any brackets {} in the cf_dset or gen_fpath inputs. If None, the cf_dset and gen_fpath inputs are assumed to be the full dataset name and the full path to the single resource file to be processed, respectively. Note that only one of cf_dset or gen_fpath are allowed to contain brackets ({}) to be filled in by the analysis years. By default, None.

You may remove keys with a null value if you do not intend to update them yourself.

Configuration inheritance

JSON, JSON5, YAML, and TOML configurations may set inherit_from to the path of another configuration file. GAPs loads inheritance recursively and deep-merges dictionaries, with values in the current file taking precedence. Lists, scalar values, null, and values that change type replace the inherited value. Parent references and other relative paths are resolved from the file in which they are defined.

Set a key to the exact, case-sensitive string "DELETE" in an inheriting file to remove that key from the effective configuration. This works at any dictionary depth. For example:

inherit_from: base.yaml
logging:
    log_file: DELETE

The inherit_from key and deleted entries are removed before config validation and preprocessing. The value "DELETE" is therefore reserved in inheriting files. CSV batch configurations do not support inheritance.