Electrical Machine Reliability via Current-Based Thermal Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for determining the reliability of electrical machines, such as MTBF and MTTF, are computationally expensive, require additional sensors, and are not generalized for various types of failures, making them impractical for aerospace applications with space and weight constraints.
Innovation Solution
Estimate the operating temperature of electrical machines using current measurements and existing sensors, then use temperature-based models like the Arrhenius model to determine reliability parameters like MTBF, considering both thermal and electrical stresses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional thermocouple methods are used to determine MTBF, then reliability estimation can be obtained, but additional sensors are required which increases device complexity and is not always feasible
Solution Approach 1:
The electrical machine's existing current sensors serve dual purposes: monitoring current for operational control and providing data for temperature estimation and reliability determination. The system uses self-generated electrical signals and existing sensor infrastructure to determine reliability parameters without requiring separate dedicated sensors.
Solution Approach 2:
Existing current measurement infrastructure is made multi-functional by using it not only for operational control but also for estimating operating temperature and determining reliability parameters. This universal use of existing sensors eliminates the need for additional thermocouples or temperature sensors.
2Reliability
If large databases of historical or accelerated aging data are used for MTBF calculation, then reliability parameters can be determined, but computational cost and time increase significantly
Solution Approach 1:
Reliability parameters are determined continuously during normal operation using real-time current data and temperature models, rather than waiting for failure events to occur and then performing retrospective analysis on large historical databases. This proactive approach provides up-to-date reliability information without requiring extensive post-processing computation.
Solution Approach 2:
The method transitions from using large static databases of historical failure data to using dynamic real-time parameters (current, estimated temperature) combined with predictive models. This changes the input parameters from extensive historical records to minimal real-time measurements, dramatically reducing computational requirements.
3Temperature
If thermocouples are placed in particularly hot or cool regions of the winding, then temperature measurement is obtained, but MTBF may be over- or under-predicted reducing measurement precision
Solution Approach 1:
Electrical current measurements serve as an intermediary parameter that indirectly reflects the operating temperature of the windings. By using current data combined with thermal models, the system estimates temperature without direct physical contact with the windings, avoiding the placement issues of thermocouples and achieving more representative temperature measurement.
Solution Approach 2:
The mechanical/physical thermocouple measurement system is replaced with an electrical-based temperature estimation system. Instead of physically inserting thermocouples into the windings, the method uses electrical current measurements and thermal modeling to determine operating temperature, eliminating placement-related inaccuracies.
4Reliability
If specific standard rated life equations for bearings are used, then bearing reliability can be determined, but the method is not generalized for other types of failures
Solution Approach 1:
A single unified methodology based on operating temperature estimation from current measurements is applied to determine reliability for multiple failure modes including winding insulation failure, bearing failure, and other thermal-related failures. This universal approach replaces the need for separate specialized equations for different component types.
Solution Approach 2:
The reliability determination is segmented by failure mode-specific models that all use the same fundamental input (operating temperature from current measurements). Each failure type has its own reliability model, but they all benefit from the common temperature estimation approach, enabling both specialization and generalization.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Provides accurate and efficient determination of reliability parameters without additional sensors, improving health monitoring and maintenance planning for electrical machines in harsh environments.
Implementation Method 1
The electrical machine may be a rotating electrical machine, and may be a motor or generator
Implementation Method 2
The method may comprise determining one or more power losses of at least a part of the electrical machine... at least one of a resistive loss, such as a winding loss
Data Source
AI summary
A computer-implemented method of determining a value of a reliability parameter of an electrical machine, the method includes: estimating an operating temperature of at least a part of the electrical machine based on a measure of a current drawn or supplied by the electrical machine; and determining the value of the reliability parameter based on the estimated operating temperature.