Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

High-Endurance UCAV Propulsion System: A 1-D CNN-Based Real-Time Fault Classification for Tactical-Grade IPMSM Drive

Tahmin Mahmud

Abstract

High-performance propulsion for mission-critical applications demands unprecedented reliability and real-time fault resilience. Conventional diagnostic methods (signal-based analysis and standard ML models) are essential for stator/rotor fault detection but suffer from high latency and poor generalization across variable speeds. This paper proposes a 1-D Convolutional Neural Network (CNN) framework for real-time fault classification in the HPDM-350 interior permanent magnet synchronous motor (IPMSM). The proposed architecture extracts discriminative features directly from high-frequency current and speed signals, enabling sub-millisecond inference on embedded controllers. Compared to state-of-the-art long short term memory (LSTM) and classical ML approaches, the 1-D CNN achieves a superior weighted F1-score of 0.9834. Validated through high-fidelity magnetic-domain MATLAB/Simscape models, the method demonstrates robust performance across a +-2700 RPM envelope, providing a lightweight solution for mission-critical electric propulsion systems.

High-Endurance UCAV Propulsion System: A 1-D CNN-Based Real-Time Fault Classification for Tactical-Grade IPMSM Drive

Abstract

High-performance propulsion for mission-critical applications demands unprecedented reliability and real-time fault resilience. Conventional diagnostic methods (signal-based analysis and standard ML models) are essential for stator/rotor fault detection but suffer from high latency and poor generalization across variable speeds. This paper proposes a 1-D Convolutional Neural Network (CNN) framework for real-time fault classification in the HPDM-350 interior permanent magnet synchronous motor (IPMSM). The proposed architecture extracts discriminative features directly from high-frequency current and speed signals, enabling sub-millisecond inference on embedded controllers. Compared to state-of-the-art long short term memory (LSTM) and classical ML approaches, the 1-D CNN achieves a superior weighted F1-score of 0.9834. Validated through high-fidelity magnetic-domain MATLAB/Simscape models, the method demonstrates robust performance across a +-2700 RPM envelope, providing a lightweight solution for mission-critical electric propulsion systems.
Paper Structure (37 sections, 23 equations, 4 figures, 4 tables)

This paper contains 37 sections, 23 equations, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Honeywell TPE331-10GD Turboprop and its Challenges
  4. From Legacy ICE to Next-Gen All-Electric Propulsion
  5. HPDM-350 IPMSM Modeling
  6. Machine Fault Types and Mathematical Characterization
  7. Demagnetization Fault
  8. Ground Fault (Coil-to-Ground Short)
  9. Isolated Turn Fault (Inter-Turn Short Circuit)
  10. Open-Circuit Fault
  11. DL Pipeline: 1-D CNN Fault Classifier
  12. Dataset Preparation
  13. HPDM-350 IPMSM Operating Principle
  14. Fault Class Definitions
  15. Simulated Signal Set
  16. ...and 22 more sections

Figures (4)

  • Figure 1: MQ-9 Reaper Powertrain. (a) Three-view dimensional layout of the atomic airframe, (b) Close-up $2\times$ stacked HPDM-350 IDS.
  • Figure 2: System-level integration of the proposed UCAV propulsion. (a) Honeywell TPE331-10GD turboprop engine anatomy, (b) Propulsion system evolution: legacy vs next-gen, (c) HPDM-350's conceptual machine cross-section and phasor analysis, (d) Double-layer unfolded winding layout for the 12S stator.
  • Figure 3: Signal-to-decision fault diagnostic chain. (a) Fault signatures & frequency spectra, (b) FaultCNN1D architecture, and (c) Hardware specifications.
  • Figure 4: Dynamic performance and thermal characteristics of the HPDM-350 IPMSM. (a) DQ current vs. EM torque, (b) Output power profile, (c) Torque-speed envelope, (d) Thermal gradient $\Delta T_\mathrm{RW}$, (e) Speed and torque response, (f) Temp. distribution, (g) Speed profile, (h) 3P currents, (i) IOC & (j) ITSC-core loss.