Electric Drive Unit Temperature Estimation During Sensor Faults
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating temperatures in electric drive units, especially in vehicles with electric motors, face challenges when direct measurement is impossible due to space constraints or functional limitations, leading to incomplete temperature monitoring and potential demagnetization of permanent magnets or insulation melting.
Innovation Solution
A method using a state space model and a Luenberger observer to calculate non-measurable temperatures, allowing for continuous temperature estimation even with faulty temperature detection elements by switching between control and throughput modes, ensuring accurate temperature calculation and accounting for significant heat inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct temperature measurement is implemented using temperature sensors, then measurement precision is improved, but device complexity and space requirements increase
Solution Approach 1:
The patent replaces physical temperature sensors with a mathematical model-based temperature estimation system. The state space model and Luenberger observer calculate temperature based on electrical measurements (current, voltage) and thermal parameters, eliminating the need for direct physical temperature sensing in the stator or rotor, thus reducing device complexity while maintaining measurement capability
Solution Approach 2:
The patent introduces intermediate electrical measurements (stator current, voltage) as mediators to indirectly determine temperature. By measuring electrical parameters and using thermal models to translate them into temperature estimates, the system achieves temperature monitoring without direct thermal contact, resolving the contradiction between measurement accuracy and installation complexity
2Measurement precision
If temperature sensors are installed in inaccessible positions, then measurement precision is improved, but ease of manufacture deteriorates
Solution Approach 1:
The patent substitutes physical sensors in inaccessible locations with a virtual sensing system using electrical measurements. The state space model estimates temperatures in the stator and rotor based on easily accessible electrical parameters, eliminating the need to physically access difficult-to-reach areas for sensor installation while maintaining precise temperature monitoring capability
Solution Approach 2:
The patent makes the electrical measurement system serve multiple functions: it simultaneously controls the motor operation and provides temperature estimation data. The same current and voltage sensors used for motor control also feed the thermal model, eliminating the need for separate temperature sensing infrastructure and simplifying manufacturing
3Reliability
If temperature calculation is suspended during detection element faults, then reliability of temperature data is improved, but productivity deteriorates
Solution Approach 1:
The patent prepares for sensor failures in advance by having a model-based temperature estimation system ready to take over when sensor faults occur. The state space model continuously runs in parallel with sensor measurements, and upon detecting sensor failure, seamlessly transitions to using model-based estimates, ensuring uninterrupted temperature monitoring and preventing productivity loss while maintaining data reliability through fallback mechanisms
Solution Approach 2:
The patent changes the input parameters of the temperature calculation system dynamically: under normal conditions it uses direct sensor measurements, but upon detecting sensor faults, it switches to using electrical parameters (current, voltage) and thermal model parameters. This parameter switching allows the system to maintain temperature estimation capability despite sensor failures, balancing reliability and productivity
4Device complexity
If model-based temperature estimation is used without direct measurement, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent implements feedback through the Luenberger observer, which continuously compares model predictions with actual sensor measurements and adjusts the state estimates accordingly. When sensors are available, the observer uses measurement feedback to correct model errors, significantly improving estimation precision. The feedback mechanism allows the system to maintain high accuracy while using the simpler model-based approach
Solution Approach 2:
The patent merges direct sensor measurement with model-based estimation into a hybrid system. The state space model and observer combine electrical parameter measurements, thermal model calculations, and sensor feedback to produce a unified temperature estimate. This merging leverages the accuracy of direct measurement when available while maintaining the simplicity and continuity of model-based estimation, achieving both reduced complexity and high precision
Data Source
AI summary
A method for calculating a temperature of an electric drive unit includes calculating a temperature model output variable (ym) using an input variable (u) and a temperature model, assigning the temperature model output variable (ym) to a system observer, and measuring an actual temperature (y) using a temperature detection element. A model state variable (xm) is regulated and outputted based on an output variable difference (Δy) between the actual temperature (y) and the output variable (ym) using the system observer. A fault mode or a normal mode of the temperature detection element is indicated using a detection state signal (s1). Depending on the signal (s1), the system observer is switched between a control mode, in which the model state variable (xm) is controlled depending on the output variable difference (Δy), and a throughput mode, in which the model state variable (xm) is calculated independently of the actual temperature (y).