Aerosol research requires the use of different measuring devices for different particle size ranges. Also, devices from multiple manufacturers, or self-built devices are used in parallel. Even though manufacturers provide software solutions for their own devices, the software is usually not applicable for devices of other manufacturers. For that, we built a library of import functions in Python based on basic functionalities and packages such as numpy and pandas.

The import functions read the measured data from txt-files exported from the measurement devices, or the manufacturer software, into a common data structure. For concentration data, as measured by condensation particle counters, the data structure is shown in the left figure above. From every imported data file, a dictionary is created containing arrays filled with all number concentration values and the elapsed time since the start of the measurement. All additional information recorded by the measuring device is stored in another array and a "results" array is produced to store the results of calculations performed with the data. Particle size distribution data is imported in a similar structure. However, instead of the elapsed time, the concentration data is correlated with the midpoint diameters of each size bin of a measurement.

Additionally, a set of basic functions for the data processing and evaluation of the aerosol data is available. The data can be cut to a specific time, or size range and multiple data files can be merged into one. For concentration data, a mean concentration per measurement can be calculated and the data can be plotted per measurement or as a time series of means. For size distribution data, the count median diameter, the geometric standard deviation and the diameters accessible via Hatch-Choate relationships can be calculated. The size distributions can be plotted as single distributions, mean distributions with error bars and cumulative distributions. A multi-modal fit based on a linear combination of lognormal distributions is implemented as shown in the right figure above.

The code can be run in the python console using IPython, or in a Jupyter notebook. It is accessible on GitHub (https://github.com/KevinRMaier/py_particle_analysis/).