Prospective aircraft design accounting for technology maturing

from jax import vmap
from matplotlib.lines import Line2D
from matplotlib.patches import Patch
from matplotlib.pyplot import subplots
from numpy import array
from numpy import linspace

from noads.application.base_objects import categories_mission
from noads.application.base_objects import category_conso
from noads.application.base_objects import propulsion_architectures
from noads.application.base_objects import propulsion_mission
from noads.application.base_objects import tech_params_lower_mid_upper_2020_2040_2060
from noads.core.models.fleet.aircraft_design import AircraftDesign
from noads.core.models.fleet.aircraft_operation import AircraftOperation
from noads.core.models.fleet.aircraft_tech_parameter import AircraftTechParameter

Aircraft Technology Evolution

Let’s start by initializing the time evolution of some key aircraft technology components, according to several literature sources.

data_sources = {
    "NASA": {
        "battery_specific_energy": array([[2035, 2040], [500, 750]]),
        "emotor_specific_power": array([[2022, 2035, 2035], [2, 13, 16]]),
        "lh2tank_gravimetric_index": array([[], []]),
        "fuelcell_specific_power": array([[], []]),
        "fuelcell_efficiency": array([[2035], [55.2]]),
        "electronics_specific_power": array([[2035, 2040, 2050], [13, 19, 26]]),
        "struct_weight_factor": array([[], []]),
    },
    "IATA": {
        "battery_specific_energy": array([[2020, 2030, 2035], [200, 225, 400]]),
        "emotor_specific_power": array([[], []]),
        "lh2tank_gravimetric_index": array([
            [2025, 2035, 2035, 2050],
            [20, 35, 50, 70],
        ]),
        "fuelcell_specific_power": array([[2025, 2040, 2050], [4, 8, 10]]),
        "fuelcell_efficiency": array([[], []]),
        "electronics_specific_power": array([[], []]),
        "struct_weight_factor": array([
            [2030, 2030, 2030, 2030, 2030, 2050, 2050, 2050, 2050, 2050, 2050, 2050],
            [86, 80, 88, 83, 90, 80, 72, 68, 70, 83, 73, 61],
        ]),
    },
    "ATI": {
        "battery_specific_energy": array([[2025], [220]]),
        "emotor_specific_power": array([[2026, 2035, 2050], [13, 23, 25]]),
        "lh2tank_gravimetric_index": array([
            [2035, 2035, 2040, 2040, 2040, 2050, 2050, 2050],
            [47, 58, 61, 66, 72, 64, 69, 75],
        ]),
        "fuelcell_specific_power": array([
            [2025, 2030, 2035, 2050],
            [1.5, 2.0, 2.5, 5.0],
        ]),
        "fuelcell_efficiency": array([[2025, 2030, 2035, 2050], [60, 65, 70, 75]]),
        "electronics_specific_power": array([[2026, 2030, 2050], [8.92, 20, 30]]),
        "struct_weight_factor": array([[2025, 2030, 2035, 2050], [80, 75, 70, 60]]),
    },
    "ICCT": {
        "battery_specific_energy": array([[2025, 2030, 2050], [250, 300, 500]]),
        "emotor_specific_power": array([[], []]),
        "lh2tank_gravimetric_index": array([[2035, 2035], [20, 35]]),
        "fuelcell_specific_power": array([[2035, 2035], [1, 2]]),
        "fuelcell_efficiency": array([[2035, 2035], [45, 65]]),
        "electronics_specific_power": array([[2025], [8]]),
        "struct_weight_factor": array([[], []]),
    },
    "EASA": {
        "battery_specific_energy": array([[2022], [210]]),
        "emotor_specific_power": array([[2022, 2022], [57.6 / 22.7, 73.5 / 56.6]]),
        "lh2tank_gravimetric_index": array([[], []]),
        "fuelcell_specific_power": array([[], []]),
        "fuelcell_efficiency": array([[], []]),
        "electronics_specific_power": array([[], []]),
        "struct_weight_factor": array([[], []]),
    },
    "Adler et al. (2025)": {
        "battery_specific_energy": array([[2023, 2030], [300, 600]]),
        "emotor_specific_power": array([[2025], [8]]),
        "lh2tank_gravimetric_index": array([[2035], [50]]),
        "fuelcell_specific_power": array([[2035, 2050], [2, 6]]),
        "fuelcell_efficiency": array([[2035, 2050], [60, 75]]),
        "electronics_specific_power": array([[], []]),
        "struct_weight_factor": array([[], []]),
    },
}
markers_sources = {
    "NASA": "o",
    "IATA": "s",
    "ATI": "d",
    "ICCT": "X",
    "EASA": "*",
    "Adler et al. (2025)": "P",
}
units = {
    "battery_specific_energy": "Wh/kg",
    "emotor_specific_power": "kW/kg",
    "electronics_specific_power": "kW/kg",
    "fuelcell_specific_power": "kW/kg",
    "lh2tank_gravimetric_index": "%",
    "fuelcell_efficiency": "%",
    "struct_weight_factor": "%",
}
name_to_fullname = {
    "battery_specific_energy": "Battery Specific Energy",
    "emotor_specific_power": "E-motor Specific Power",
    "electronics_specific_power": "Power electronics Specific Power",
    "fuelcell_specific_power": "Fuel cell Specific Power",
    "lh2tank_gravimetric_index": "LH2 tank Gravimetric Index",
    "fuelcell_efficiency": "Fuel cell Efficiency",
    "struct_weight_factor": "Structural weight reduction",
}
propulsion_colors = {
    "JetA-GasTurbine": "maroon",
    "Battery-Electric": "limegreen",
    "lH2-FuelCell": "royalblue",
    "lH2-GasTurbine": "orangered",
}

Now let’s visualize how the Technology Evolution scenarios compare with these sources.

years = linspace(2020, 2060, 41)
fig1, axes1 = subplots(
    2,
    4,
    figsize=(12, 7),
    layout="constrained",
)
tech_params = {}
for i, (name, values) in enumerate(tech_params_lower_mid_upper_2020_2040_2060.items()):
    lower, mid, upper = values
    lower_param = AircraftTechParameter(name, (lower[0], lower[1], lower[2]))
    mid_param = AircraftTechParameter(name, (mid[0], mid[1], mid[2]))
    upper_param = AircraftTechParameter(name, (upper[0], upper[1], upper[2]))
    tech_params[name] = (lower_param, mid_param, upper_param)

    lower_values = lower_param.value_at_entry_into_service(years)
    mid_values = mid_param.value_at_entry_into_service(years)
    upper_values = upper_param.value_at_entry_into_service(years)

    ax = axes1.flat[i]
    ax.fill_between(
        years,
        lower_values,
        upper_values,
        alpha=0.2,
        color="k",
        label="Upper-to-Lower",
    )
    ax.plot(years, lower_values, "k-", linewidth=2, label="Lower")
    ax.plot(years, mid_values, "k:", linewidth=2, label="Mid")
    ax.set_title(name_to_fullname[name], fontsize="medium")
    ax.set_ylabel(units[name])
    ax.set_xlabel("Year")
    for source, source_data in data_sources.items():
        x_data = source_data[name][0]
        y_data = source_data[name][1]
        ax.plot(
            x_data,
            y_data,
            markers_sources[source],
            label=source,
            markersize=8,
        )

# axes1[0, 0].legend(loc="upper left", framealpha=0.4)
axes1[-1, -1].clear()
axes1[-1, -1].set_axis_off()

legend_handles = [
    Line2D([0], [0], color="k", ls="-", lw=2, label="Lower"),
    Line2D([0], [0], color="k", ls=":", lw=2, label="Mid"),
    Patch(alpha=0.2, facecolor="k", label="Upper-to-Lower"),
]
legend_handles.extend([
    Line2D([0], [0], ls="None", marker=marker, markersize=8, label=name, color=f"C{j}")
    for j, (name, marker) in enumerate(markers_sources.items())
])

axes1[-1, -1].legend(handles=legend_handles, loc="center")


fig1.suptitle(
    "Aircraft Technology Parameters evolution",
    fontsize="large",
)
fig1.savefig("./aircraft_technology.pdf")
print(
    "Aircraft technology parameters evolution figure saved as 'aircraft_technology.pdf'"
)
# show()

Aircraft technology parameters evolution figure saved as 'aircraft_technology.pdf'

Prospective Aircraft Design

Now for each year of Entry-Into-Service we’ll design an airplane using GAM and compare their overall empty weight and mission energy consumption.

Unfeasible designs may yield infinite mass and energy, therefore the (mass/energy) metric is calculated on a per (seat-km) basis and then inverted. They are therefore a measure of traffic per (mass/energy). The highest the metric the lower the (mass/energy).

fig2, axes2 = subplots(2, 3, layout="constrained", figsize=(12, 8))
fig2.suptitle("Aircraft Energy Efficiency\n[seat km / MJ]", fontsize="x-large")

fig3, axes3 = subplots(2, 3, layout="constrained", figsize=(12, 8))
fig3.suptitle("Aircraft Empty-Mass Efficiency\n[seat km / kg]", fontsize="x-large")

categories = categories_mission.keys()
ymax = 4.0

for cat, category in enumerate(categories):
    max_range = 1e-3 * categories_mission[category]["range"]
    seat = categories_mission[category]["npax"]

    ax_e = axes2.flat[cat]
    ax_m = axes3.flat[cat]

    current_energy_per_ask = category_conso[category]
    reference_aircraft = AircraftOperation(
        name=f"{category}_RECENT",
        propulsion=None,
        energy_per_ask=current_energy_per_ask[1],
        recent=True,
    )
    ax_e.hlines(
        y=1.0 / current_energy_per_ask[0],
        xmin=years[0],
        xmax=years[-1],
        colors="dimgray",
        linestyles="-",
        linewidth=1.5,
    )
    ax_e.hlines(
        y=1.0 / current_energy_per_ask[1],
        xmin=years[0],
        xmax=years[-1],
        colors="dimgray",
        linestyles=":",
        linewidth=1.5,
    )
    ax_e.fill_between(
        [years[0], years[-1]],
        [1.0 / current_energy_per_ask[0], 1.0 / current_energy_per_ask[0]],
        [1.0 / current_energy_per_ask[2], 1.0 / current_energy_per_ask[2]],
        alpha=0.4,
        color="dimgray",
    )
    for _prop, propulsion in enumerate(propulsion_architectures.keys()):
        ask_per_energy_low_to_up = []
        ask_per_owe_low_to_up = []
        for tech_idx in range(3):
            aircraft = AircraftDesign(
                name=f"{category}_{propulsion}",
                propulsion=None,
                mission={
                    **categories_mission[category],
                    **propulsion_mission[propulsion],
                    "category": category,
                },
                power_system=propulsion_architectures[propulsion],
                aircraft_tech_params=[
                    tech_param[tech_idx]
                    for tech_name, tech_param in tech_params.items()
                ],
                reference_aircraft=reference_aircraft,
            )
            model = aircraft.design_model()
            model.discipline.jax_out_func = vmap(model.discipline.jax_out_func)
            outputs = model.discipline.execute({
                f"{category}_{propulsion}.entry_into_service": years
            })
            ask_per_energy_low_to_up.append(
                1.0 / outputs[f"{category}_{propulsion}.energy_per_ask"]
            )
            ask_per_owe_low_to_up.append(
                seat * max_range / outputs[f"{category}_{propulsion}.owe"]
            )

        ax_e.fill_between(
            years,
            ask_per_energy_low_to_up[0],
            ask_per_energy_low_to_up[-1],
            alpha=0.2,
            color=propulsion_colors[propulsion],
        )
        ax_e.plot(
            years,
            ask_per_energy_low_to_up[0],
            color=propulsion_colors[propulsion],
            linestyle="-",
            linewidth=3,
        )
        ax_e.plot(
            years,
            ask_per_energy_low_to_up[1],
            color=propulsion_colors[propulsion],
            linestyle=":",
            linewidth=3,
        )

        ax_m.fill_between(
            years,
            ask_per_owe_low_to_up[0],
            ask_per_owe_low_to_up[-1],
            alpha=0.2,
            color=propulsion_colors[propulsion],
        )
        ax_m.plot(
            years,
            ask_per_owe_low_to_up[0],
            color=propulsion_colors[propulsion],
            linestyle="-",
            linewidth=3,
        )
        ax_m.plot(
            years,
            ask_per_owe_low_to_up[1],
            color=propulsion_colors[propulsion],
            linestyle=":",
            linewidth=3,
        )

    ax_e.set_ylim(ymin=0.0, ymax=ymax)
    ax_e.set_title(
        f"{category.replace('_', ' ')}\n({seat} seat, {max_range} km)",
        fontsize="large",
    )
    ax_m.set_ylim(ymin=0.0)
    ax_m.set_title(
        f"{category.replace('_', ' ')}\n({seat} seat, {max_range} km)",
        fontsize="large",
    )

axes2[-1, -1].clear()
axes2[-1, -1].set_axis_off()
axes3[-1, -1].clear()
axes3[-1, -1].set_axis_off()

# Manually add legend

handles = [
    Patch(color=color, label=prop_name)
    for prop_name, color in propulsion_colors.items()
]
handles.append(Patch(color="dimgray", label="Current Technology"))

axes2[-1, -1].legend(
    handles=handles,
    loc="center",
)
axes2[-1, 0].set_xlabel("Entry-Into-Service")

axes3[-1, -1].legend(
    handles=[
        Patch(color=color, label=prop_name)
        for prop_name, color in propulsion_colors.items()
    ],
    loc="center",
)
axes3[-1, 0].set_xlabel("Entry-Into-Service")

fig2.savefig("./aircraft_prospective_energy.pdf")
fig3.savefig("./aircraft_prospective_mass.pdf")
print(
    "Figures saved as aircraft_prospective_energy.pdf and aircraft_prospective_mass.pdf"
)

# show()

• 
• 

Figures saved as aircraft_prospective_energy.pdf and aircraft_prospective_mass.pdf