Runout Measurements using Multi-Probe Separation

Overview This dataset contains time-series data from runout measurements. The processed data is provided in a single, highly compressed HDF5 file named runout_measurements.hdf5. File Format and Compression The data is stored in the widely supported HDF5 binary format. To ensure efficient storage, maximum lossless gzip compression (level 9) alongside byte shuffling is used. Data Structure The HDF5 file contains a flat structure of datasets, where each dataset represents a single measurement run. Global Attributes labels: Attached to the root of the file, this attribute contains a list of strings defining the variable names for the columns in the data arrays. The channels are ordered as follows: ["time", "angle", "s_1", "s_2", "s_3", "s_4"] . The channels s_1 through s_4 represent the individual sensors utilized during the application of the multi-probe separation method for runout measurement. Datasets Individual measurements are stored as 2D arrays (where rows represent time steps and columns represent the aforementioned channels).Datasets are named sequentially using the format meas_0001, meas_0002, meas_0003, and so on . Dataset Attributes speed: Each individual dataset has an attached float attribute representing the rotational speed of the shaft in degrees per second during that specific measurement run . Loading the Data To facilitate easy access to the data, a Python script named load_runout_measurements.py is provided alongside the dataset. This script uses the h5py library to efficiently read the HDF5 file, extract the global channel labels, and load all measurement arrays along with their corresponding speed parameters into memory as a list of dictionaries.