← Back to Publications List

MLP-Based Classification of Buffer Tank Pressure States in Hydrogen Refueling Stations

Students & Supervisors

Student Authors
Md. Asif Khan
Bachelor of Science in Electrical & Electronic Engineering, FE
Rajin Al Fatemi
Bachelor of Science in Electrical & Electronic Engineering, FE
Supervisors
Abu Shufian
Lecturer, Faculty, FE

Abstract

This paper presents an MLP-based classification framework for hydrogen refueling station buffer tank pressure, capable of identifying Low, Medium, and High states using real operational data. The proposed model achieves high classification accuracy while accounting for practical misclassification scenarios, supported by comprehensive visualization including training curves, correlation heatmaps, confusion matrices, and predicted probabilities. The framework provides a complete workflow from preprocessing and feature scaling to evaluation, enabling reliable real-time monitoring and operational decision support. While the framework is deployment-ready, its real-time performance may be affected by high-frequency data acquisition or limited computational resources on edge devices, which should be considered during station implementation. Results demonstrate that the MLP-based approach enhances situational awareness, improves system safety, and facilitates efficient hydrogen dispensing.

Keywords

Hydrogen fueling infrastructure buffer tank pressure monitoring Multi-Layer Perceptron Operational Safety.

Publication Details

  • Type of Publication:
  • Conference Name: IEEE Region 10 TENSYMP 2026
  • Date of Conference: 29/06/2026 - 29/06/2026
  • Venue: IEEE Malaysia Section, Penang, Malaysia
  • Organizer: IEEE Malaysia Section, Penang, Malaysia
View on Publisher