Sensorless Electric Machine Control for Non-Periodic Signal Injection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensorless control methods for electric machines fail when the supplementary excitation is of a general nature, such as caused by pulse-width modulation with varying periods or direct torque control, or is non-periodic, limiting their applicability.
Innovation Solution
A method involving the injection of a high-frequency supplementary excitation into the drive voltage, measuring the drive current, and estimating the instantaneous state variable using a procedure that includes generating a modulation basis, applying finite-length filters, and solving a system of linear equations to determine the modulating signal accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing sensorless control methods are used with periodic supplementary excitation, then control can be achieved under specific conditions, but the method fails when the supplementary excitation is of a general nature (non-periodic or varying period PWM)
Solution Approach 1:
The patent develops a signal processing method that works universally for diverse supplementary excitation signals including periodic, non-periodic, and varying period PWM signals. The method uses finite-length filters and moment calculations that are not dependent on signal periodicity, enabling the same control approach to function across multiple excitation types without requiring separate specialized methods for each signal characteristic.
Solution Approach 2:
The patent changes the fundamental parameters of the signal processing approach by using finite-length filters with moment calculations instead of traditional periodicity-based methods. This parameter change allows the system to handle general excitation signals where the frequency and period may vary over time, rather than requiring fixed periodic excitation signals.
2Measurement precision
If signal processing methods are designed for specific periodic excitations, then measurement precision is maintained for those signals, but the method cannot handle general or non-periodic excitations
Solution Approach 1:
The patent segments the signal processing into distinct stages: applying finite-length filters to extract relevant components, calculating moments of the filter outputs, and using these moments to determine the modulating signal. This segmentation allows each stage to be optimized independently and ensures that the overall method remains robust across different excitation signal types without relying on global periodicity assumptions.
Solution Approach 2:
The patent introduces finite-length filters and moment calculations as intermediary processing steps between the raw current measurements and the final state variable estimation. These intermediaries transform the diverse excitation signals into a standardized form that can be accurately processed regardless of the original signal characteristics, bridging the gap between different PWM schemes and the estimation algorithm.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of determining the instantaneous operating state of an electric machine for its sensorless control with signal injection, comprising: injecting a HF supplementary excitation into the drive voltage of the electric machine, thereby modulating the drive current (y) of the electric machine by a modulating signal (z), measuring the drive current (y) of the electric machine, and estimating a state variable (x) of the electric machine using the measured drive current (y), wherein the estimation step includes: generating a modulation basis (s) from the excitation, multiplying (P1) the measured drive current (y) by a demodulation basis (r), which is correlated with the modulation basis (s), to obtain a first intermediate signal (ry), multiplying (P2) the transpose (sT) of the modulation basis (s) by the demodulation basis (r) to obtain a second intermediate signal (rsT), applying (F1) a set of finite-length filters to the first intermediate signal (ry), and applying (F2) the same set of finite-length filters and their moments to the second intermediate signal (rsT), to obtain a linear equation system (L), solving the linear equation system (L) to obtain the modulating signal (z), and estimating the state variable (x) based on the modulating signal (z)