Electrical Power Fault Detection Using Current Harmonic Anomalies

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fault detection methods for electrical power systems in aerospace applications require additional sensors, which are not feasible due to space and weight constraints, and lack a unified approach to detect both power converter and electrical machine faults effectively.

Innovation Solution

A method utilizing existing current measurements to transform time domain data into frequency bands, calculate statistical measures, apply principal components analysis (PCA), and determine Mahalanobis distances to detect anomalies, enabling unified fault detection in power converters and electrical machines.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If additional sensors are installed for fault detection, then measurement precision is improved, but weight and space constraints are worsened

Engineering Contradiction:
Improvefault detection precisionVSAvoidsystem weight
Core Design Contradiction:
Measurement precisionVSWeight of moving object

Solution Approach 1:

The system uses existing current measurements from the power electronics converter to detect faults, making the existing measurement system serve dual purposes. The current measurements originally intended for control are repurposed for fault detection through advanced signal processing techniques including frequency domain transformation and statistical analysis

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

Physical sensors are replaced with signal processing algorithms that analyze existing electrical measurements. The fault detection function is achieved through mathematical transformations (FFT, wavelet transform) and statistical methods rather than additional physical sensing hardware

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If multiple separate detection methods are used for power converter and electrical machine faults, then reliability is improved, but device complexity is worsened

Engineering Contradiction:
Improvefault detection reliabilityVSAvoiddetection system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

A unified fault detection framework is implemented that can detect multiple types of faults (power converter faults, electrical machine faults, bearing faults) using a single integrated approach. The same signal processing pipeline handles different fault types by analyzing characteristic frequencies and statistical features in the current measurements

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

Separate fault detection methods are merged into a single comprehensive system. The approach combines frequency domain analysis, time domain statistical measures, and pattern recognition algorithms into one unified process that simultaneously monitors for multiple fault types across different system components

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4597131A1Fault detection in an electrical power system
Publication Date: 2025.08.06 ROLLS ROYCE PLC
  • EP4597131A1 patent drawingFigure 1
  • EP4597131A1 patent drawingFigure 2
  • EP4597131A1 patent drawingFigure 3a~3b

AI summary

The disclosure relates to a fault detection method for an electrical power system (100) comprising a power electronics converter (102) connected to an electrical machine (101), the method comprising: i) measuring an electric current (la, lb) through a winding of the electrical machine over a measurement time period to acquire a first set of time domain current data; ii) transforming the first set of time domain current data to the frequency domain to provide a set of frequency domain current data; iii) dividing the set of frequency domain current data into a plurality of frequency bands, each frequency band containing a harmonic frequency of operation of the electrical machine; iv) transforming each of the plurality of frequency bands into the time domain to provide a second set of time domain current data for each frequency band; v) calculating a plurality of statistical measures of each second set of time domain current data; vi) applying principal components analysis to the plurality of statistical measures for each second set of time domain current data; vii) calculating a Mahalanobis distance for each second set of time domain data; and viii) determining a fault if any of the calculated Mahalanobis distances is anomalous compared to a baseline measure.