Current Variance Arc Detection for High-Impedance DC Circuits

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional arc detection systems struggle to effectively detect current arcs in high impedance automotive electrical distribution systems, particularly in direct current (DC) circuits, due to the difficulty in distinguishing arcing currents from normal operating conditions and the complexity of time-current curves caused by pulse-modulated loads and switching converters.

Innovation Solution

The implementation of a current variance arc fault detector that measures periodic current flow between a battery and loads or a source and loads, calculates variance, and uses a machine learning model to identify outliers and determine if the variance exceeds a threshold, thereby detecting arc faults in electrical distribution systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional arc detection systems are used in high impedance automotive electrical distribution systems, then the system structure remains simple, but the detection precision deteriorates due to difficulty in distinguishing arcing currents from normal operating conditions

Engineering Contradiction:
Improvearc detection precisionVSAvoiddetection system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the detection approach by changing from direct current magnitude analysis to statistical parameter analysis. It calculates mean, standard deviation, and coefficient of variation of current values over time windows, converting raw current signals into statistical parameters that reveal arc fault patterns hidden in high impedance systems.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent adds temporal dimension to current analysis by implementing sliding time windows and calculating statistical parameters across multiple time points. This transforms one-dimensional current magnitude data into multi-dimensional statistical feature space, enabling discrimination of arc faults from normal variations.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If pulse-modulated loads and switching converters are used to improve system functionality, then the adaptability of the electrical distribution system improves, but the difficulty of detecting and measuring arcs increases due to complex time-current curves

Engineering Contradiction:
Improvesystem functionalityVSAvoidarc detection difficulty
Core Design Contradiction:
Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent extracts statistical features (mean, standard deviation, coefficient of variation) from the complex current signals generated by pulse-modulated loads and switching converters. By separating these statistical parameters from raw current data, the system can identify arc fault patterns independent of the complex switching waveforms.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements dynamic detection by using sliding time windows that continuously update statistical parameters as current conditions change. This allows the detection system to adapt to varying load conditions and switching frequencies while maintaining arc detection capability.

Inventive Principle:
Principle #15Dynamics

3Power

If the voltage in the distribution system is increased to meet modern automotive power requirements, then the power delivery capability improves, but the object-affected harmful factors increase due to increased potential for arcing and associated fire risks

Engineering Contradiction:
Improvepower delivery capabilityVSAvoidarcing and fire risk
Core Design Contradiction:
PowerVSObject-affected harmful factors

Solution Approach 1:

The patent implements feedback-based arc detection by continuously monitoring current statistical parameters and comparing them against threshold values. When the coefficient of variation exceeds a threshold, the system triggers arc fault detection and can initiate protective actions, creating a closed-loop safety mechanism for high voltage systems.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3695476B1Arc detection based on variance of current flow
Publication Date: 2025.01.08 LITTELFUSE INC
  • EP3695476B1 patent drawingFigure 1
  • EP3695476B1 patent drawingFigure 2
  • EP3695476B1 patent drawingFigure 3

AI summary

Arc fault detection devices and methods are described current between a source and a load is periodically measured. A variance of the periodically measured current values is derived and an arc fault can be detected abased on the derived variance. A variance interval signal can be incremented based on the derived variance increasing above a threshold level and a low pass filter arranged to detect an arc based on the incremented variance interval signal.