LOAD ARENA / SAMPLE RESULTS
SIMULATION POST-PROCESSING
Load Arena
One workflow for statistics, extreme loads, fatigue analysis, and result exploration.
ENGINEERING · PYTHON · DATA
Arash Atasen
Engineering · Python development · Data analysis
I combine engineering knowledge, Python development, and data analysis to solve practical problems.
Copenhagen, Denmark
Selected projects
Software and data analysisWIND ANALYZER / MEASURED WIND
SENSOR DATA ANALYSIS
Wind Analyzer
Explore wind measurements across heights, turbulence, vertical shear, and agreement between instruments.
Previous workplace projects
Engineering methods and Python developmentEARLY DESIGN · LOAD ESTIMATION
FAST Load Estimator
A Python tool for estimating wind turbine blade loads during early design, when running full aeroelastic simulations for every design change can slow down development.
Developed in Python.
Read the approach and inspect the results
The approach
Combine steady aerodynamic loads from HAWCStab2 with simplified relationships for turbulence, rotor speed variation, and gravity. Estimate extreme flapwise and edgewise blade loads, with a later extension for flapwise fatigue loads.
The result
Results were compared with HAWC2 simulations for two turbine models. The comparisons showed useful agreement, while identifying differences across wind speeds and methods. The tool supports early design assessment; detailed simulations remain necessary for final load verification.
Agreement varies across wind speeds, with noticeable underestimation and overestimation in some conditions. These differences are important when assessing where the method is useful.
The lifetime comparison reported closer agreement with HAWC2 for the tested cases, despite larger differences at some individual wind speeds. This finding applies to the evaluated turbine models and wind distributions.
EXTREME LOADS · STATISTICAL EXTRAPOLATION
Load Extrapolation
Extrapolation of wind turbine extreme loads during normal operation. A Python tool that uses simulation peaks to estimate rare loads for a chosen return period, beyond the range observed in the simulations.
Developed in Python.
Read the approach and inspect the results
The approach
Estimate rare extreme loads during normal turbine operation using statistical extrapolation. The method extracts simulation peaks, fits load distributions at each wind speed, and combines them to estimate loads for a chosen return period.
The result
The study evaluated the method on the DTU 10 MW and NREL 5 MW turbines and examined sensitivity to block count and turbulence assumptions. Two and three blocks gave similar extrapolated loads in the tested cases, with better distribution fits than using one global extreme per simulation. Differences varied by load channel and turbulence condition.
Divide simulation time series & extract peaks
Divide each simulation time series into blocks and collect the maximum and minimum load in each block. Repeat this for the simulations and channels being analysed. The dashed red lines mark block boundaries, and the red circles mark the selected peaks.
Fit short-term exceedance distributions using three Weibull parameters
Gather and rank the peaks at each wind speed. Fit a three-parameter Weibull distribution, with shape, scale, and location parameters, using least squares. Use the fitted peak distribution and the number of peaks per simulation to calculate the short-term exceedance distribution. The blue circles show ranked peak probabilities; the orange curve shows the fitted cumulative distribution.
Combine the long-term exceedance curve to estimate the load for a chosen return period
Weight the short-term exceedance distributions by the probability of each wind speed and sum them into a long-term exceedance curve. Find the load corresponding to the target exceedance probability for the chosen return period. The lower plot shows the combined curve and the dashed target-probability line.
ABOUT ME
Engineering experience applied to software and data
I’m an engineer with more than 10 years of experience working with simulation and measurement data. I develop Python tools and automated workflows for analysis, model validation, visualisation, and reporting. I’m interested in roles where I can build useful software, automate analysis, and help teams work with technical data.
CONTACT
Let’s talk about your team.
I’m looking for opportunities to apply my engineering, Python, and data experience.
