hotfire-data

Tools and notebooks for preprocessing, analyzing, and visualizing engine hotfire thrust, pressure, temperature, and event data.

Python modules

• filtering.py - Defines an LPF helper that builds and applies a SciPy FIR low-pass filter to smooth sensor signals.
• preprocessing.py - Converts raw hotfire data into processed data.npz/data.csv files, parses controller logs into events.csv, and imports processed data/events for analysis notebooks.
• visualization.py - Provides utilities for finding samples by time and plotting sensor data around logged events with optional filtering and multi-curve overlays.

Notebooks

Each analysis-*.ipynb notebook is a hotfire-specific analysis workspace. Most follow the same rough pattern: define the data directory, map raw sensor names to readable labels, map driver IDs to valve/event names, preprocess raw logs/data, import processed data/events, then create ignition-centered plots.

Usage

Dependencies

• Install the packages listed in requirements.txt
• If you do not wish to affect existing installations on your system, create and run a Virtual Environment

Data requirements

• Data file pulled from the RPi must be named raw.csv
• Log file must be named console.log
• Place these two files in a data-raw folder which is placed in a folder as shown below

<hotfire-dir>/
  data-raw/
    raw.csv
    console.log

TODO: Add preprocessing example.

import preprocessing as prep

DATA_DIR = "TODO"
SENSORS = {
    # "raw_sensor_name": "Readable Sensor Name",
}
DRIVERS = {
    # driver_id: {"name": "Valve Name", "False": "Close", "True": "Open"},
}

prep.process_events_quonkboard(DATA_DIR, DRIVERS)
prep.process_data(DATA_DIR, SENSORS)

TODO: Add analysis/plotting example.

import preprocessing as prep
import visualization as vis
from filtering import LPF

events = prep.import_events(DATA_DIR)
labels, data = prep.import_data(DATA_DIR, events)

plotter = vis.EventPlotter(data, events, dpi=100)
filter = LPF(fs=300, length=101, cutoff=10, window="blackman")

plotter.plot(
    sensor_id=1,
    event_id=0,
    duration=40,
    filter=filter,
    title="TODO",
    ylabel="TODO",
)