Page 42 - tech fest 2025 אשדוד חוברת תקצירים

42
Application of Regenerative Braking Energy in Rail Transport
EEE-A-13
Ben Dahan; benda4@ac.sce.ac.il
Advisor: Dr. Peter Beker
SCE - Shamoon College of Engineering, Ashdod
Rail transport loses significant energy during braking, which could be recovered and reused to improve efficiency. This project focused on utilizing regenerative braking energy (RBE) by returning it directly to the train. We explored different onboard storage solutions, including batteries (like the Siemens Mireo Plus B) and flywheel systems, which can store and rapidly release energy to support acceleration. The feasibility of using hybrid solutions like the Siemens Desiro HC in trains was also analyzed. Through simulations (via Simulink, the project evaluated the practicality, efficiency, and economic benefits of these technologies to optimize energy recovery in modern rail systems.
Keywords: electricity generation, rail transportation, regenerative braking
Enhancing Gas Detection Under Thermal Variations Using Spectral Absorption and Machine Learning
EEE-A-14
Ninoy Glam; ninoyglam@gmail.com Lior Zaguri; liorzaguri22@gmail.com
Advisor: Dr. Tom Trigano
SCE - Shamoon College of Engineering, Ashdod
This project focused on enhancing a UV-based gas detection system developed at Emerson Spectronics for hazardous gases such as hydrogen sulfide (H2S), ammonia (NH3), and sulfur dioxide (SO2), using a xenon light source and spectral analysis. Gas molecules absorb UV light at characteristic wavelengths, and the transmitted signal is analyzed to determine gas type and concentration. However, thermal variations may weaken the signal by up to 90%, reducing detection accuracy. To overcome this, we developed a three-stage approach: (1) analyzing spectral behavior under thermal changes, (2) conditioning the signal with amplification and normalization, and (3) implementing a machine learning-based autonomous correction model. Integrating optical sensing with this algorithmic processing was shown to enhance the system's performance in dynamic environments.
Keywords: control systems, machine learning, sensors and instrumentation, signal processing