EMI-Based Power Electronics Degradation Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for a system and method to detect the degradation of power electronic components in critical systems, such as inverter systems used in electric vehicles, elevators, and process controls, to prevent failures that could result in injury or loss.
Innovation Solution
The system monitors electromagnetic interference (EMI) produced by a three-phase inverter and applies machine learning algorithms to identify EMI characteristics indicative of component degradation or failure. Additionally, failure modes due to aging, abuse, or defects are modeled to enhance the detection capabilities, and genetic algorithms may be used in a 'virtual twin' of power electronics to detect component age or failure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional monitoring methods are used for power electronic components, then the system structure remains simple, but the ability to detect component degradation is insufficient
Solution Approach 1:
The patent converts electromagnetic interference (EMI), traditionally considered a harmful byproduct of power electronic operation, into a useful diagnostic signal. By analyzing EMI characteristics, the system detects component degradation without requiring additional sensors or measurement equipment, thus improving reliability while avoiding increased system complexity
Solution Approach 2:
The monitoring system utilizes the power electronic system's own EMI emissions as the measurement signal, eliminating the need for external diagnostic equipment. The existing EMI filter and control circuitry are repurposed for diagnostic functions, allowing the system to self-monitor its health status without adding complexity
2Measurement precision
If EMI monitoring with machine learning is implemented, then component degradation detection accuracy improves, but computational requirements and algorithm complexity increase
Solution Approach 1:
The patent applies machine learning algorithms selectively to analyze specific EMI characteristics that are most indicative of component degradation. Rather than processing the entire EMI spectrum, the system focuses on key frequency ranges and signal features, achieving high detection accuracy while managing computational complexity through targeted analysis
3Reliability
If continuous monitoring is implemented, then early detection of degradation is achieved, but energy consumption and system resource usage increase
Solution Approach 1:
The system continuously monitors component health by analyzing EMI signals that are already present during normal operation. Since EMI is an inherent byproduct of power electronic switching, the monitoring process requires minimal additional energy, achieving continuous early detection without significant energy overhead
Solution Approach 2:
The monitoring system uses the existing operational signals and control circuitry for diagnostics, requiring minimal additional power. The EMI filter and control processor that already exist in the system are utilized for monitoring functions, avoiding the need for separate powered sensors or measurement equipment
Data Source
AI summary
A method of measuring electromagnetic interference (EMI) to noninvasively identify component degradation or failure in power electronics circuitry. The method involves characterizing the degradation or failure characteristics of the component and modeling those characteristics to enable a machine learning algorithm to identify EMI frequency distribution characteristics that correspond to the degradation or failure. The EMI frequency distribution is measured and the data provided to the machine learning algorithm whereupon the algorithm identifies degradation or failures indicated by the measured data.