HDF5 Output Schema¶

File Structure Overview¶

results.h5
├── [Root Attributes]              # Metadata
├── system_configuration/          # Static system definition
│   ├── generators/                # Generator parameters
│   └── batteries/                 # Battery parameters
├── summary_results/               # Per-year summary (expandable)
├── demand/                        # Input demand data
│   └── year_YYYY_base_demand      # Per-year demand
├── cost_breakdown/                # Granular cost decomposition
│   └── year_YYYY/                 # Per-year cost components (attrs)
└── detailed_results/              # Full time-series results
    └── year_YYYY_threshold_0/     # Per-year detailed data
        ├── generation/            # Generator output
        ├── gen_status/            # On/off status (UC mode)
        ├── gen_startup/           # Startup events
        ├── gen_shutdown/          # Shutdown events
        ├── battery_charge/        # Charging power
        ├── battery_discharge/     # Discharging power
        ├── battery_soc/           # State of charge
        ├── battery_spillage/      # Energy spillage
        ├── battery_capacity_factor/ # Battery CF
        ├── battery_lcoe/          # Battery LCOE
        ├── battery_vallcoe/       # Battery VALLCOE
        ├── reservoir_level/       # Reservoir water level
        ├── reservoir_spillage/    # Reservoir overflow
        ├── reservoir_pump/        # Pump-back power
        ├── gen_investment_power/   # Generator investments
        ├── bat_investment_power/   # Battery power investments
        ├── bat_investment_capacity/ # Battery energy investments
        ├── capacity_factor/       # Generator CF
        ├── lcoe/                  # Generator LCOE
        ├── vallcoe/               # Generator VALLCOE
        ├── technology_selling_prices/ # Revenue analysis
        └── [scalar datasets]      # System-level time series

Root Attributes¶

Attribute	Type	Description
`creation_date`	string	ISO 8601 creation timestamp
`num_nodes`	int	Number of network nodes
`num_years`	int	Number of simulation years
`export_type`	string	`"incremental_results"`
`temporal_resolution_hours`	int	Time step (hours)
`simulation_mode`	string	`"development"` or `"unit_commitment"`
`years_range`	string	Format: `"YYYY-ZZZZ"`
`target_re`	float	RE penetration target
`export_complete`	bool	Whether export finished successfully
`last_update`	string	ISO 8601 last update timestamp
`export_timestamp`	string	ISO 8601 export timestamp

system_configuration/¶

Static system configuration stored once at file creation.

system_configuration/generators/generator_{g}/¶

Per-generator attributes stored as HDF5 attributes:

Attribute	Type	Description
`name`	string	Generator name
`type`	string	"Renewable" or "Non-renewable"
`fuel`	string	Fuel type
`rated_power`	float array	Per-node rated power (MW)
`invest_cost`	float array	Per-node investment cost ($/MW)
`life_time`	int array	Per-node lifetime (years)
`initial_age`	int array	Per-node initial age (years)

system_configuration/batteries/battery_{b}/¶

Per-battery attributes (same structure as generators, plus battery-specific fields).

summary_results/¶

Expandable arrays that grow as each year completes. One row per simulated year.

Dataset	Shape	Type	Description
`year`	(num_years,)	int	Simulation year
`threshold`	(num_years,)	int	Threshold iteration
`feasible`	(num_years,)	int	1 if feasible, 0 if not
`total_cost`	(num_years,)	float	Total annual cost ($)
`renewable_penetration`	(num_years,)	float	RE penetration (0-1)
`co2_emissions`	(num_years,)	float	Annual CO2 (tonnes)
`loss_of_load`	(num_years,)	float	Total unserved energy (MWh)

demand/¶

demand/year_YYYY_base_demand¶

Property	Value
Shape	(hours,) or (num_nodes, hours)
Type	float64
Units	MW
Description	Base demand profile for the year

cost_breakdown/¶

Granular decomposition of the annual system cost into 27 individual components. Each year's costs are stored as attributes on a /cost_breakdown/year_YYYY/ group, accumulated across rolling horizon windows.

This group is present when the optimizer exports per-component cost data. The Financial Analysis module reads this group directly; when it is absent, costs are recalculated from generation output and system configuration.

cost_breakdown/year_YYYY/ (Attributes)¶

Attribute	Type	Units	Description
`fuel_cost`	float	$	Fuel consumption cost
`fixed_om_cost`	float	$	Fixed O&M cost
`maintenance_cost`	float	$	Variable maintenance cost
`startup_cost`	float	$	Generator startup cost
`battery_maintenance_cost`	float	$	Battery O&M cost
`battery_degradation_cost`	float	$	Battery degradation cost
`load_shedding_cost`	float	$	Cost of unserved energy
`curtailment_cost`	float	$	Renewable curtailment penalty
`reserve_static_cost`	float	$	Static reserve deficit penalty
`reserve_dynamic_cost`	float	$	Dynamic reserve deficit penalty
`co2_emission_cost`	float	$	Carbon emission cost
`fre_penetration_cost`	float	$	RE penetration shortfall penalty
`inertia_cost`	float	$	System inertia deficit penalty
`soc_violation_cost`	float	$	Battery SOC violation penalty
`transfer_margin_cost`	float	$	Transfer margin deficit penalty
`v2g_compensation`	float	$	V2G compensation (negative = credit)
`flexible_demand_benefit`	float	$	Demand flexibility benefit (negative = credit)
`investment_cost`	float	$	Capital investment cost
`electrolyzer_cost`	float	$	Electrolyzer operation cost
`converter_cost`	float	$	AC/DC converter cost
`spillage_cost`	float	$	Battery spillage penalty
`delay_retirement_cost`	float	$	Delayed retirement cost
`reservoir_spillage_cost`	float	$	Reservoir overflow penalty
`demand_shift_cost`	float	$	Demand shifting cost
`rooftop_curtailment_cost`	float	$	Rooftop solar curtailment penalty
`npv_penalty_cost`	float	$	NPV lifecycle penalty
`reservoir_invest_cost`	float	$	Reservoir capacity investment cost
`total`	float	$	Sum of all cost components

Reading Cost Breakdown¶

import h5py

with h5py.File("results_system.h5", "r") as f:
    if "cost_breakdown" in f:
        for year_key in sorted(f["cost_breakdown"].keys()):
            attrs = dict(f[f"cost_breakdown/{year_key}"].attrs)
            print(f"{year_key}: total=${attrs['total']:,.0f}")
            print(f"  Fuel: ${attrs['fuel_cost']:,.0f}")
            print(f"  CO2:  ${attrs['co2_emission_cost']:,.0f}")

detailed_results/year_YYYY_threshold_0/¶

Full time-series results for each year. The threshold_0 suffix is the convergence threshold iteration.

Attributes (Per-Year Metadata)¶

Attribute	Type	Description
`year`	int	Simulation year
`threshold`	int	Threshold iteration index
`feasible`	int	Feasibility flag
`total_cost`	float	Total annual cost ($)
`renewable_penetration`	float	Final RE penetration
`co2_emissions`	float	Annual CO2 emissions (tonnes)
`solve_time`	float	Total solve time (seconds)
`total_generation`	float	Total energy generated (MWh)
`total_demand`	float	Total energy demand (MWh)
`load_shed`	float	Total unserved energy (MWh)

Investment/retirement attributes (per generator/battery):

Pattern	Description
`investment_gen_{name}_node_{n}`	Generator investment at node n (MW)
`investment_bat_{name}_power_node_{n}`	Battery power investment (MW)
`investment_bat_{name}_capacity_node_{n}`	Battery energy investment (MWh)
`retirement_gen_{name}_node_{n}`	Generator retirement flag
`retirement_bat_{name}_node_{n}`	Battery retirement flag

Generation Data¶

generation/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MW
Description	Generator output power per node per hour

gen_status/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	Binary (0/1) or continuous (0-1)
Description	Generator commitment status (UC mode: binary; ED mode: continuous)

gen_startup/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	Binary (0/1)
Description	Startup events

gen_shutdown/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	Binary (0/1)
Description	Shutdown events

curtailment¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MW
Description	Aggregated renewable energy curtailment

Battery Data¶

battery_charge/{battery_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MW
Description	Charging power

battery_discharge/{battery_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MW
Description	Discharging power

battery_soc/{battery_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MWh
Description	State of charge

battery_spillage/{battery_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MW
Description	Energy spillage (excess energy discarded)

battery_capacity_factor/{battery_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	Fraction (0-1)
Description	Battery capacity factor

battery_lcoe/{battery_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	$/MWh
Description	Levelized cost of storage

battery_vallcoe/{battery_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	$/MWh
Description	Value-adjusted LCOS

Reservoir Data¶

reservoir_level/{generator_name}¶

Property	Value
Shape	(num_nodes, hours+1)
Type	float64
Units	MWh-eq
Description	Reservoir water level (includes initial state, hence hours+1)

reservoir_spillage/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MW-eq
Description	Uncontrolled water overflow

reservoir_pump/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	MW
Description	Pump-back power consumed (demand-side load)

Investment Data¶

gen_investment_power/{generator_name}¶

Property	Value
Shape	(num_nodes,)
Type	float64
Units	MW
Description	New generation capacity installed

bat_investment_power/{battery_name}¶

Property	Value
Shape	(num_nodes,)
Type	float64
Units	MW
Description	New battery power capacity

bat_investment_capacity/{battery_name}¶

Property	Value
Shape	(num_nodes,)
Type	float64
Units	MWh
Description	New battery energy capacity

Performance Metrics¶

capacity_factor/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	Fraction (0-1)
Description	Generator capacity factor (output / rated)

lcoe/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	$/MWh
Description	Levelized cost of energy

vallcoe/{generator_name}¶

Property	Value
Shape	(num_nodes, hours)
Type	float64
Units	$/MWh
Description	Value-adjusted levelized cost of energy

System-Level Time Series¶

Dataset	Shape	Units	Description
`demand`	(num_nodes, hours)	MW	Demand profile
`reserve_static`	(num_nodes, hours)	MW	Static reserve provided
`reserve_dynamic`	(num_nodes, hours)	MW	Dynamic reserve provided
`loss_of_reserve_static`	(num_nodes, hours)	MW	Static reserve deficit
`loss_of_reserve_dynamic`	(num_nodes, hours)	MW	Dynamic reserve deficit
`loss_load`	(num_nodes, hours)	MW	Load shedding
`CO2_emissions`	(num_nodes, hours)	tonnes	Hourly CO2 emissions
`voltage_angle`	(num_nodes, hours)	radians	Bus voltage angles

Economic Data¶

Dataset	Shape	Units	Description
`electricity_prices`	(hours,)	$/MWh	System-average price
`electricity_prices_energy`	(hours,)	$/MWh	Energy component only
`nodal_electricity_prices`	(num_nodes, hours)	$/MWh	Locational marginal prices
`nodal_electricity_prices_congestion`	(num_nodes, hours)	$/MWh	Congestion component

Network Data¶

Dataset	Shape	Units	Description
`power_flow`	(num_nodes, num_nodes, hours)	MW	Power flow between nodes
`transfer_investment`	(num_nodes, num_nodes)	MW	Transmission investment
`transfer_margin`	(num_nodes, num_nodes, hours)	MW	Transfer margin used

EV Data¶

Dataset	Shape	Units	Description
`EV_charging`	(num_nodes, hours)	MW	EV charging demand
`EV_V2G`	(num_nodes, hours)	MW	Vehicle-to-grid power
`EV_soc`	(num_nodes, hours)	MWh	EV fleet SOC
`EV_loss`	(num_nodes, hours)	MW	Unmet EV demand

Other¶

loss_of_inertia¶

Property	Value
Shape	(hours,)
Type	float64
Units	GW*s
Description	System inertia deficit

technology_selling_prices/{technology_name}¶

Per-technology revenue analysis.

Attributes:

Attribute	Type	Description
`total_generation`	float	Total generation (MWh)
`total_revenue`	float	Total revenue ($)
`average_selling_price`	float	Weighted average price ($/MWh)
`technology_type`	string	Generator type

Dataset: prices_weights

Property	Value
Shape	(n_entries, 3)
Type	float64
Columns	[price ($/MWh), generation (MW), timestep (h)]
Description	Price-quantity pairs for revenue decomposition

Virtual Generator and Battery Datasets¶

Virtual generators and batteries from technology investments appear in the HDF5 output after the original physical units.

Ordering Convention¶

Dataset ordering within each group follows the adapter ordering:

Original generators (from sys.generators in config) --- e.g., "Diesel_1", "Solar_PV", "Wind"
Virtual generators (from technology investments) --- e.g., "Investment Solar PV", "Investment Wind"
Original batteries (from sys.batteries in config) --- e.g., "Li_Ion_1"
Virtual batteries (from battery technology investments) --- e.g., "Investment Battery"

Virtual units use the same dataset structure as their physical counterparts.

Identifying Virtual Units¶

Virtual units are identified by:

Name prefix: always starts with "Investment "
The system_configuration/generators/generator_{g}/ attributes: virtual units have initial_age = 0 and their rated_power reflects the cumulative technology investment up to that year

Dataset Dimensions and Data Types¶

Category	Dataset Pattern	Shape	dtype	Units
Generation	`generation/{name}`	(N, H)	float64	MW
Status	`gen_status/{name}`	(N, H)	float64	0-1
Startup	`gen_startup/{name}`	(N, H)	float64	0/1
Shutdown	`gen_shutdown/{name}`	(N, H)	float64	0/1
Curtailment	`curtailment`	(N, H)	float64	MW
Bat. charge	`battery_charge/{name}`	(N, H)	float64	MW
Bat. discharge	`battery_discharge/{name}`	(N, H)	float64	MW
Bat. SOC	`battery_soc/{name}`	(N, H)	float64	MWh
Bat. spillage	`battery_spillage/{name}`	(N, H)	float64	MW
Reservoir level	`reservoir_level/{name}`	(N, H+1)	float64	MWh-eq
Reservoir spill	`reservoir_spillage/{name}`	(N, H)	float64	MW-eq
Reservoir pump	`reservoir_pump/{name}`	(N, H)	float64	MW
Gen. investment	`gen_investment_power/{name}`	(N,)	float64	MW
Bat. invest. power	`bat_investment_power/{name}`	(N,)	float64	MW
Bat. invest. energy	`bat_investment_capacity/{name}`	(N,)	float64	MWh
Capacity factor	`capacity_factor/{name}`	(N, H)	float64	0-1
LCOE	`lcoe/{name}`	(N, H)	float64	$/MWh
VALLCOE	`vallcoe/{name}`	(N, H)	float64	$/MWh
Demand	`demand`	(N, H)	float64	MW
Power flow	`power_flow`	(N, N, H)	float64	MW
Prices	`electricity_prices`	(H,)	float64	$/MWh
Nodal prices	`nodal_electricity_prices`	(N, H)	float64	$/MWh
Loss of load	`loss_load`	(N, H)	float64	MW

Where N = num_nodes, H = total hours in the year (typically 8760).

Derived Metrics Formulas¶

Capacity Factor¶

\[CF_g = \frac{\sum_{t=1}^{H} P_{g,t}}{P_{rated,g} \times H}\]

Where $P_{g,t}$ is the generator output at hour $t$ and $P_{rated,g}$ is the rated power.

LCOE (Levelized Cost of Energy)¶

\[LCOE_g = \frac{C_{capex,g} \times CRF + C_{opex,g} + C_{fuel,g}}{E_{annual,g}}\]

Where:

$C_{capex,g}$ = total capital cost ($)
$CRF = \frac{r(1+r)^n}{(1+r)^n - 1}$ = capital recovery factor
$C_{opex,g}$ = annual O&M cost ($)
$C_{fuel,g}$ = annual fuel cost ($)
$E_{annual,g}$ = annual generation (MWh)

VALLCOE (Value-Adjusted LCOE)¶

\[VALLCOE_g = LCOE_g - \frac{\sum_{t} \lambda_t \cdot P_{g,t}}{\sum_{t} P_{g,t}}\]

Where $\lambda_t$ is the electricity price at hour $t$. Negative VALLCOE indicates profitability at market prices.

MGA/SPORES Results¶

When near-optimal alternative generation is enabled (master_problem.mga.enabled: true), an additional /mga/ group contains every alternative. The same layout serves both methods; the method and objective attrs disambiguate which generator produced each alternative.

results.h5
└── mga/
    ├── @method = "mga" | "spores"     # Generation method (Phase-4 attr)
    ├── @objectives = [...]            # Distinct objectives in display order
    ├── @num_alternatives
    ├── @slack_fraction
    ├── @optimal_cost
    ├── alternative_0/                  # Cost-optimal seed
    │   ├── @is_optimal = 1
    │   ├── @objective = "cost_optimal"
    │   ├── gen_investment              # (years x generators x nodes)
    │   ├── bat_power_investment        # (years x batteries x nodes)
    │   ├── bat_capacity_investment     # (years x batteries x nodes)
    │   ├── transfer_investment         # (years x nodes x nodes)
    │   ├── cumulative_gen_capacity     # (years x generators x nodes)
    │   └── re_penetration              # (years,)
    ├── alternative_1/                  # First non-optimal alternative
    │   ├── @objective = "hsj_diversity"          # under method=mga
    │   │   or                "min_total_build"   # under method=spores
    │   └── ... (same datasets as alternative_0)
    └── alternative_K/
        └── ...

mga/ (root attributes)¶

Attribute	Type	Description
`method`	string	`"mga"` or `"spores"`. Pre-Phase-4 result files default to `"mga"` on read for back-compat
`objectives`	string[]	Distinct objectives in display order. Empty list for legacy MGA runs (the historical `"hsj_diversity"` tag is implicit). Useful as a cache for the viewer's legend
`num_alternatives`	int	Total alternatives (including cost-optimal)
`slack_fraction`	float	Cost slack used (e.g., 0.05 = 5%)
`optimal_cost`	float	Cost-optimal objective value ($)
`export_timestamp`	string	ISO timestamp when the `/mga/` group was written
`years`	int[]	Planning-horizon years (same as the rest of the file)

mga/alternative_N/ (Attributes)¶

Attribute	Type	Description
`alternative_id`	int	0 = cost-optimal, 1..K = non-optimal alternatives
`is_optimal`	int	1 if cost-optimal, 0 otherwise
`cost`	float	Actual system cost ($)
`diversity_objective`	float	Objective value at this alternative. For `method=mga` this is the HSJ diversity score; for `method=spores` it is the value of the SPORES objective named by `attrs["objective"]`. Absent for `alternative_0`
`objective`	string	Phase-4 tag identifying which objective produced this alternative. Values: `"cost_optimal"` (seed), `"hsj_diversity"` (any MGA alt), or one of `"min_total_build"`, `"max_tech_equity"`, `"max_regional_equity"`, `"evolutionary_dist"` (SPORES alts). Pre-Phase-4 files default to `"cost_optimal"` for the seed and `"hsj_diversity"` for the rest on read

mga/alternative_N/ (Datasets)¶

Dataset	Shape	Type	Units	Description
`gen_investment`	(years, generators, nodes)	float64	MW	Generator investment decisions
`bat_power_investment`	(years, batteries, nodes)	float64	MW	Battery power investment
`bat_capacity_investment`	(years, batteries, nodes)	float64	MWh	Battery energy investment
`transfer_investment`	(years, nodes, nodes)	float64	MW	Transmission investment
`cumulative_gen_capacity`	(years, generators, nodes)	float64	MW	Cumulative capacity (with degradation/retirement)
`re_penetration`	(years,)	float64	fraction	RE penetration ratio per year

MGA Output File Naming¶

MGA results are written to mga_{system_name}.h5. The main results_{system_name}.h5 file still contains the cost-optimal solution's full operational results.

Reading HDF5 Results¶

Basic Example with h5py¶

import h5py
import numpy as np

with h5py.File("results_system.h5", "r") as f:
    # --- Metadata ---
    num_years = f.attrs["num_years"]
    num_nodes = f.attrs["num_nodes"]
    print(f"Simulation: {num_years} years, {num_nodes} nodes")

    # --- Summary across all years ---
    years = f["summary_results/year"][:]
    costs = f["summary_results/total_cost"][:]
    re_pen = f["summary_results/renewable_penetration"][:]

    for y, c, r in zip(years, costs, re_pen):
        print(f"  Year {y}: cost=${c:,.0f}, RE={r:.1%}")

    # --- Detailed: generation for a specific year ---
    gen_output = f["detailed_results/year_2030_threshold_0/generation/solar_pv"][:]
    print(f"Solar PV output shape: {gen_output.shape}")  # (num_nodes, 8760)
    print(f"Total solar generation: {gen_output.sum():.0f} MWh")

Reading Investment Decisions¶

import h5py

with h5py.File("results_system.h5", "r") as f:
    year_grp = f["detailed_results/year_2030_threshold_0"]

    # Read all investment attributes
    for attr_name in year_grp.attrs:
        if attr_name.startswith("investment_"):
            value = year_grp.attrs[attr_name]
            if value > 0:
                print(f"  {attr_name}: {value:.1f} MW")

    # Read generator investment arrays
    inv_grp = year_grp["gen_investment_power"]
    for gen_name in inv_grp:
        inv_data = inv_grp[gen_name][:]  # shape: (num_nodes,)
        total = inv_data.sum()
        if total > 0:
            print(f"  {gen_name}: {total:.1f} MW total investment")

Reading Battery SOC Profiles¶

import h5py
import matplotlib.pyplot as plt

with h5py.File("results_system.h5", "r") as f:
    soc = f["detailed_results/year_2030_threshold_0/battery_soc/li_ion"][:]
    # Shape: (num_nodes, 8760)

    # Plot SOC for node 0 over the first week
    hours = range(168)
    plt.figure(figsize=(12, 4))
    plt.plot(hours, soc[0, :168])
    plt.xlabel("Hour")
    plt.ylabel("SOC (MWh)")
    plt.title("Battery State of Charge - Node 0, Week 1")
    plt.grid(True)
    plt.show()

Reading Electricity Prices¶

import h5py
import numpy as np

with h5py.File("results_system.h5", "r") as f:
    year_grp = f["detailed_results/year_2030_threshold_0"]

    # System-average prices
    prices = year_grp["electricity_prices"][:]
    print(f"Average price: ${np.mean(prices):.2f}/MWh")
    print(f"Peak price: ${np.max(prices):.2f}/MWh")

    # Nodal prices (locational marginal prices)
    nodal = year_grp["nodal_electricity_prices"][:]
    print(f"Nodal prices shape: {nodal.shape}")  # (num_nodes, 8760)

    # Price spread between nodes
    price_spread = nodal.max(axis=0) - nodal.min(axis=0)
    print(f"Average nodal spread: ${np.mean(price_spread):.2f}/MWh")

Reading MGA Results¶

import h5py

with h5py.File("mga_system.h5", "r") as f:
    if "mga" in f:
        mga = f["mga"]
        meta = mga["metadata"]
        n_alts = meta.attrs["num_alternatives"]
        optimal_cost = meta.attrs["optimal_cost"]
        slack = meta.attrs["slack_fraction"]

        print(f"MGA: {n_alts} alternatives, {slack:.0%} slack")
        print(f"Optimal cost: ${optimal_cost:,.0f}")

        for k in range(n_alts):
            alt = mga[f"alternative_{k}"]
            cost = alt.attrs["cost"]
            gen_inv = alt["gen_investment"][:]  # (years, gens, nodes)
            bat_inv = alt["bat_power_investment"][:]
            re_pen = alt["re_penetration"][:]

            print(f"\nAlternative {k}:")
            print(f"  Cost: ${cost:,.0f} ({(cost/optimal_cost - 1)*100:+.1f}%)")
            print(f"  Total gen investment: {gen_inv.sum():.0f} MW")
            print(f"  Total bat investment: {bat_inv.sum():.0f} MW")
            print(f"  Final RE penetration: {re_pen[-1]:.1%}")

Reading Virtual Generator Output¶

import h5py

with h5py.File("results_system.h5", "r") as f:
    gen_grp = f["detailed_results/year_2035_threshold_0/generation"]

    # List all generators (original + virtual)
    all_gens = list(gen_grp.keys())
    print(f"All generators: {all_gens}")

    # Virtual generators have "Investment" prefix
    for name in all_gens:
        data = gen_grp[name][:]
        total_mwh = data.sum()
        if "Investment" in name:
            print(f"  [VIRTUAL] {name}: {total_mwh:,.0f} MWh")
        else:
            print(f"  [ORIGINAL] {name}: {total_mwh:,.0f} MWh")

With the CLI¶

# Export to CSV
esfex export results.h5 --format csv --output-dir ./csv_results/

# Export to Excel
esfex export results.h5 --format excel --output results.xlsx

# Export specific year
esfex export results.h5 --format csv --year 2030