Electric Motor Temperature Plausibility Check Using Virtual Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional temperature measurement methods in electric motors fail to accurately determine the maximum temperature due to sensor placement discrepancies and require complex physical modeling, leading to potential overheating and performance degradation.
Innovation Solution
A data-driven temperature model is employed to validate the plausibility of temperature measurements by monitoring steady-state operations, using a probabilistic regression model to compare actual and modeled temperature values, with adjustments for ambient conditions and current load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If temperature sensors are placed on stator coils to monitor temperature, then temperature monitoring is enabled, but the measured temperature does not correspond to the maximum temperature since sensor position differs from maximum temperature position
Solution Approach 1:
The patent introduces a virtual temperature sensor that acts as an intermediary between the physical temperature sensor and the actual maximum temperature. The virtual sensor calculates the maximum temperature by combining the measured temperature from the physical sensor with modeled temperature differences, thereby indirectly obtaining the maximum temperature without placing a physical sensor at the exact location where it occurs.
Solution Approach 2:
The patent creates a virtual copy of the temperature measurement system that simulates the temperature distribution throughout the motor. Instead of placing multiple physical sensors to capture all temperature points, a virtual model replicates the temperature field based on limited sensor data and thermal modeling, providing temperature information at locations where no physical sensor exists.
2Measurement precision
If physical modeling is used to determine temperatures in moving components like the rotor, then temperature estimation is enabled, but the model requires complex simulation and test bench measurements for calibration
Solution Approach 1:
The patent performs preliminary calibration of the thermal model during the manufacturing phase using controlled test bench measurements. The model parameters are pre-adjusted to match the specific motor's thermal characteristics before the motor enters service. This preliminary action eliminates the need for complex ongoing calibration and allows the model to provide accurate temperature estimates during normal operation.
Solution Approach 2:
The patent adjusts key thermal model parameters such as thermal conductivity, heat capacity, and heat transfer coefficients to match the actual motor's thermal behavior. By changing these parameters based on initial characterization measurements, the model accurately reflects the specific motor's thermal properties without requiring complex simulation for each operating condition.
3Reliability
If conventional validation methods monitor temperature changes over time or detect implausible values, then anomaly detection is enabled, but the methods cannot detect sensor drift during steady-state operation
Solution Approach 1:
The patent implements a feedback mechanism where the virtual temperature sensor continuously compares its calculated maximum temperature with the actual measured temperature from the physical sensor. This feedback loop detects discrepancies that indicate sensor drift, allowing the system to identify validation issues during steady-state operation when conventional methods would fail to detect problems.
Solution Approach 2:
The patent applies validation not only during transient conditions but also during steady-state operation by continuously comparing virtual and actual temperatures. This excessive validation approach ensures that sensor drift is detected even when temperature changes are minimal, going beyond the conventional practice of only monitoring during dynamic conditions.
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
Enables timely detection of sensor drift and prevents overheating by adjusting power usage based on validated temperature readings, improving motor reliability and performance.
Implementation Method 1
Conventional solutions involve arranging one or more temperature sensors, usually temperature-sensitive resistors such as NTC (Negative Thermal Coefficient) sensors, on stator coils
Implementation Method 2
During the operation of electric motors, current flows in the stator and/or rotor coils generate power losses, which lead to heating of the motor's components. The amount of heat generated depends, among other things, on the motor current flowing into the motor
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for checking the plausibility of a temperature measurement of a temperature sensor (4) at a component (21, 22) of an electrical machine (2) in a technical device (1), comprising the following steps: monitoring (S1) a current in the electrical machine (2); having established an operation with a stationary load, providing (S2) a first temperature measurement value of a temperature of the component (21, 22) of the electrical machine (2), the duration of the operation with a stationary load and an ambient temperature; detecting (S7) a second temperature measurement value at the end of the operation with a stationary load; determining (S8) a modelled temperature value (Tmod) using a data-based temperature model according to the first temperature measurement value, the duration and the ambient temperature, wherein the temperature model is trained to assign the first temperature measurement value, the duration and the ambient temperature to a modelled temperature value (Tmod); checking the plausibility (S9) of the second temperature measurement value with the modelled temperature value.