Rotor Shaft Position Detection Using Kalman Filter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for torque control in electrical machines, such as those in hybrid drive technology, face inaccuracies due to manufacturing tolerances in angle sensors, leading to errors in signal extrapolation and affecting control functionality and quality.
Innovation Solution
The use of a Kalman filter to determine the position of the rotor shaft between measurement times, allowing for high accuracy and robustness by eliminating the need for linear extrapolation, and enabling the use of low-resolution angle sensors, thereby reducing manufacturing costs and complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If linear extrapolation of past signals is used to supply estimated signals, then the control system can operate with low-resolution angle sensors, but large errors occur between individual measured values due to manufacturing tolerances
Solution Approach 1:
The Kalman filter acts as an intermediary between the low-resolution angle sensor and the control system. It processes the sensor signals and input signals through a mathematical model to generate accurate position information, effectively mediating the trade-off between sensor resolution and measurement accuracy. The filter combines measurements with a dynamic model to produce optimal estimates that would otherwise require higher-resolution sensors.
2Measurement precision
If high-resolution angle sensors are used to improve position information accuracy, then control quality improves, but manufacturing costs and device complexity increase
Solution Approach 1:
The patent replaces the mechanical solution of using high-resolution physical sensors with a computational approach. Instead of relying on the physical resolution of the angle sensor, the system uses a Kalman filter with a mathematical model of the electrical machine to compute accurate position information from lower-resolution measurements, substituting mechanical precision with algorithmic processing.
3Reliability
If high-resolution angle sensors are used to reduce errors in signal extrapolation, then control functionality improves, but manufacturing costs increase
Solution Approach 1:
The patent changes the parameters used in the estimation process from direct sensor readings to filtered estimates that incorporate a dynamic model of the system. By transforming the raw sensor signals through the Kalman filter, the system achieves reliable control functionality with lower-cost sensors that have lower resolution parameters.
Data Source
Figure 1~2
Figure 3a~3c
Figure 4~5
AI summary
The invention relates to a method and a device for acquiring information on the position of the rotor shaft of an electrical machine (12), using a sensor signal that can be associated with an angle of rotation and at least one input signal (13, 14) of the electrical machine (12). The method comprises the following steps: acquiring a sensor signal that can be associated with the angle of rotation and at least one input signal (13, 14) at a time tk; acquiring information on the position of the rotor shaft at a time tk, using the acquired sensor signal and the at least one input signal (13, 14); acquiring a sensor signal that can be associated with the angle of rotation and the at least one input signal (13, 14) at a time tk+1; acquiring information on the position of the rotor shaft at a time tk+1, using the acquired sensor signal and the at least one input signal (13, 14); the information on the position of the rotor shaft being acquired between the times tk and tk+1 using a Kalman filter (11, 44), based on the acquired sensor signal that can be associated with the angle of rotation and the at least one input signal (13, 14).