Sensorless Motor Torque And Velocity Estimation At Low Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating motor drive torque and velocity are inaccurate, especially at low speeds or zero speed, requiring position and torque sensors, which are costly and time-consuming to install, and result in significant inaccuracies due to changes in motor direction and filtering issues.
Innovation Solution
A method and apparatus that estimate motor torque and velocity profiles based on line-to-line voltages and phase currents without a position or velocity input, using a processor and programmable filter to downsample and filter signals, allowing for accurate estimation of velocity and torque without the need for sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If position and torque sensors are installed to improve measurement accuracy, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces mechanical sensors (position sensors, torque sensors) with an electrical measurement system that uses voltage and current signals already present in the motor drive circuit. The processor calculates torque and velocity by processing these electrical signals through mathematical models, eliminating the need for physical sensors and their associated mounting hardware.
Solution Approach 2:
The patent creates a virtual model of motor operation by calculating torque and velocity from electrical signals. Instead of directly measuring physical quantities with sensors, the system computes equivalent values through signal processing and mathematical relationships between voltage, current, and motor parameters, producing accurate estimates without physical contact with moving parts.
2Measurement precision
If filtering is applied to reduce noise in sensor signals, then measurement precision is improved, but time delay and response speed worsen
Solution Approach 1:
The patent replaces physical signal filtering hardware with computational signal processing. The processor applies digital filtering algorithms to the voltage and current signals, achieving noise reduction without the inherent time delays associated with analog RC filters or other physical filtering mechanisms. The computational approach allows for adaptive filtering that can be optimized for minimal latency.
3Measurement precision
If sensors are installed to capture motor direction changes accurately, then measurement precision is improved, but device complexity and installation time increase
Solution Approach 1:
The patent replaces mechanical direction sensors with computational detection of direction changes from voltage and current signal patterns. The processor identifies reversals by analyzing the signs and transitions of electrical signals, eliminating the need for physical sensors mounted on rotating components. This approach captures direction changes instantaneously without mechanical lag or installation complexity.
Data Source
AI summary
For estimating motor torque and velocity, a method estimates a velocity profile for a motor based on line-to-line voltages and phase currents for the motor. The velocity profile is estimated without a position input and a velocity input. The method further estimates a torque profile for the motor based on the line-to-line voltages, the phase currents, and a time interval of the velocity profile of motor velocities greater than a velocity threshold, and wherein the motor is operating over the time interval.