Luchsinger Power Model

A configurable power model for pumping kite airborne wind energy (AWE) systems based on the Luchsinger model [1] [2]. Takes awesIO-format YAML configuration files as input and always computes power curves using segmented wind shear profiles.

Overview

This repository provides a standalone power model for AWE pumping kite systems. It calculates power curves from cut-in to cut-out wind speed, accounting for:

• Aerodynamic forces on the kite
• Ground station generator and storage efficiency losses
• Optimal reeling velocity control
• Three operating regions (below force limit, force-limited, power-limited)

Purpose

This repository is designed to be:

1. Used locally via the scripts in scripts/ with configuration from data/
2. Integrated into a larger toolchain where PowerModel is instantiated directly with paths to awesIO YAML files

Features

• Reeling-factor optimization: Optimal reel-out and reel-in velocity factors calculated per operating region
• awesIO format: Input and output follow the awesIO data schema; optional validation on load and export
• Configurable: All physical parameters via awesIO-format YAML files

Installation

Requirements

• Python >= 3.8
• NumPy >= 1.20.0
• SciPy >= 1.7.0
• PyYAML >= 5.4.0
• Matplotlib >= 3.4.0

Install dependencies

pip install -r requirements.txt

Or using conda:

conda install numpy scipy pyyaml matplotlib

Quick Start

Using the Script

python scripts/calculate_power_curves.py

Using as a Library

import numpy as np
from power_luchsinger import PowerModel

model = PowerModel(
    system_config_path='data/soft_kite_pumping_ground_gen_system.yml',
    wind_resource_path='data/profileed_profiles_wind_resource.yml',
    simulation_settings_path='data/simulation_settings_config.yml',
    validate_file=True,
)

# Generate power curves for all wind profiles in the wind resource file
data = model.generate_power_curves(
    wind_speeds=np.arange(4, 25, 1),   # optional; overrides num_points from settings
    output_path='results/luchsinger_power_curves.yml',
    verbose=True,
    show_plot=True,
    save_plot=True,
    validate_file=True,
)

print(f"Generated {len(data['profiles'])} power curves")

Configuration Files

Three YAML files are required, all following the awesIO schema:

File	Description
soft_kite_pumping_ground_gen_system.yml	System configuration (kite, tether, ground station, operational envelope)
profileed_profiles_wind_resource.yml	Wind resource with profileed normalized wind shear profiles
simulation_settings_config.yml	Simulation settings (e.g. number of power curve points)

Physical Model

Luchsinger Pumping Cycle

The model implements a ground-based pumping kite system with two phases:

1. Reel-out phase: Kite flies crosswind generating high tether force, pulling out the tether and driving the generator
2. Reel-in phase: Kite is depowered, tether is reeled back in consuming energy from storage

Operating Regions

Region	Description
Region 1	Below force limit — reel-out and reel-in factors jointly optimized
Region 2	Force-limited — tether force held at nominal; reel-in factor optimized
Region 3	Power-limited — generator power held at nominal; reel-in factor optimized

Model Variants

The simulation setting in data/simulation_settings_config.yml selects one of two model formulations:

Model key	Description
luchsinger_original	Original Luchsinger-style pumping-cycle formulation with reel-in force represented using a reel-in elevation angle term. For fixed wing kites.
luchsinger_extended_const_lod_in	Extended formulation with explicit lift-to-drag-based reel-in force treatment and tether drag contribution in reel-out force factor. For soft wing kites.

Practical guidance

• Keep all other inputs identical and change only model to compare formulations.
• The same three operating regions are used in both models; the main differences are in force-factor definitions and reel-in/reel-out aerodynamic expressions.
• The comparison workflow in scripts/plot_power_curves.py is intended for side-by-side evaluation of outputs from the two model variants.

Project Structure

├── src/
│   └── power_luchsinger/
│       ├── __init__.py
│       ├── power_model.py          # Main PowerModel class
│       ├── calculations.py         # Pure aerodynamic calculation functions
│       ├── config_loader.py        # awesIO YAML loading and parsing
│       └── plotting.py             # Plotting utilities
├── scripts/
│   ├── __init__.py
│   └── calculate_power_curves.py   # Script for generating power curves
├── data/
│   ├── soft_kite_pumping_ground_gen_system.yml
│   ├── profileed_profiles_wind_resource.yml
│   └── simulation_settings_config.yml
├── results/                        # Generated outputs (YAML + plots)
├── requirements.txt
├── README.md
└── LICENSE

References

1. R.H. Luchsinger: "Pumping cycle kite power". In Airborne Wind Energy, Springer, 2013. https://doi.org/10.1007/978-3-642-39965-7_3
2. R. Schmehl, M. Rodriguez, L. Ouroumova, and M. Gaunaa: "Airborne Wind Energy for Martian Habitats". In Adaptive On- and Off-Earth Environments, Springer, 2024. https://doi.org/10.1007/978-3-031-50081-7_7

License

This project is licensed under the Apache License - see the LICENSE file for details.