Flow curve fitter

Software that fits rheological models to viscometer measurements of stress vs. strain rate. Code is written in Python language. Demo video: https://www.youtube.com/watch?v=0Tc-vnOoss0

1. Background

Shear viscosity is an important measurement technique for several industries: food, chemicals, concrete are common examples. To quantify the material properties several models have been developed, from a simple power law to more complex equations that are virtually impossible to insert into common spreadsheets software. This software makes the above task intuitive, fast and easy.

2. The data

Synthetic data representative of a shear thinning fluid such as mayonnaise or chocolate spread.

Only two columns (features), shear strain rate and shear stress, are given as input in the file "input_data.csv". To fit your data simply update the file pasting the corresponding values from your measurement.

You obtain and Excel file as output. It contains a sheet with experimental and calculated data from each fit, and a sheet with constants calculated values and model scores.

3. The models

• Ostwald – de Waele power law
  
• Carreau
  
• Carreau-Yasuda
  
• Cross
    
• Bingham
   
• Herschel-Bulkley
   
• Casson

Details for each can be easily found online. Equation are written into decorators and fitted to the experimental values using the Scikit-Learn curve-fitting package.

The whole code can be found in the following Jupyter notebook

Data from a selected model are presented as experimental fit and plotted with Matplotlib.

4. Conclusions

• Open the file "input_data.csv" and paste the values of strain rate and stress you would like to model. Several theoretical models are quickly fitted to experimental data. Constant and curvefits are calculated for each.

• You can get all results in a convenient format so that can be further analysed with any spreadsheet software. Output file is named at your wish using a simple input-name function in the code.