New Analyzer for Measuring Hydrocarbons
Author Information
Author(s): Radosaw Bandomir, Mariusz Krawczyk, Jacek Namiesnik
Primary Institution: Gdańsk University of Technology
Hypothesis
Can a new type of analyzer based on a pellistor sensor and neural network data postprocessing effectively measure hydrocarbons in the lower explosive limit range?
Conclusion
The new analyzer shows promise in discriminating hydrogen from hydrocarbons, but further development is needed for better accuracy among individual hydrocarbons.
Supporting Evidence
- The analyzer was tested with 552 data sets to evaluate its performance.
- The neural network used had 3 layers and was trained on the data collected.
- Results indicated that while hydrogen could be discriminated, individual hydrocarbon discrimination was inadequate.
Takeaway
This study created a new tool to measure gases that can explode, which is important for safety in places like oil rigs.
Methodology
The study involved preparing gas mixtures and measuring their concentrations using a pellistor sensor and neural networks for data analysis.
Limitations
The analyzer's ability to discriminate between individual hydrocarbons is still low.
Want to read the original?
Access the complete publication on the publisher's website