Vibration Actuator Control Using Neural Speed and Thrust Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the speed or thrust of vibration-type actuators require sensors for detecting parameters like load torque, temperature, and current, leading to increased size and decreased accuracy due to frequency and temperature variations.
Innovation Solution
A control device for vibration-type actuators uses a neural network to estimate thrust and speed based on phase differences, frequencies, and amplitudes of alternating-current signals, along with vibrational states and admittance characteristics, without the need for thrust or speed detection sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a torque sensor or temperature sensor is added to detect load torque or temperature for speed estimation, then the estimation can be performed, but the device size increases and manufacturing complexity increases
Solution Approach 1:
The vibration-type actuator uses its own operating parameters (driving voltage frequency, current, amplitude) to estimate its own speed through neural network processing, eliminating the need for external sensors. The actuator essentially monitors and processes its own operational data to achieve self-diagnosis and self-control.
Solution Approach 2:
The patent replaces physical sensors (torque sensors, temperature sensors) with a neural network-based estimation system that processes electrical signals and operational parameters. This substitutes mechanical/physical measurement devices with an information-processing system that achieves the same functional goal without adding hardware complexity.
2Measurement precision
If a torque sensor is added to detect load torque for speed estimation, then speed can be estimated, but the device size increases
Solution Approach 1:
The actuator estimates its own speed using its own operational parameters (current, driving voltage frequency, amplitude) processed through a neural network, eliminating the need for external torque sensors that would increase the actuator's volume.
Solution Approach 2:
The patent extracts and utilizes the speed estimation function from a separate torque sensor component and integrates it into the control system through neural network processing of existing operational parameters, thereby removing the need for additional sensing hardware that would increase actuator size.
3Measurement precision
If temperature sensor is added to detect temperature for speed estimation, then speed can be estimated, but the device complexity increases
Solution Approach 1:
The actuator uses its own operational parameters (driving voltage frequency, current, amplitude) to estimate speed through neural network processing, eliminating the need for external temperature sensors. The system achieves self-monitoring using already-available operational data.
Solution Approach 2:
The neural network estimation system serves multiple functions by processing the same set of operational parameters (current, voltage frequency, amplitude) to derive speed information without requiring separate dedicated sensors for each measurement, thereby achieving multi-functionality with minimal hardware.
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 accurate estimation of thrust and speed without sensors, improving size reduction and accuracy by utilizing a neural network to process input values related to the actuator's vibrational state and electrical properties.
Implementation Method 1
the piezoelectric body 202 serves as an electro-mechanical energy converter
Data Source
AI summary
To enable estimating thrust or speed of a vibration-type actuator without use of a thrust or speed detection sensor, a device is provided to control the vibration-type actuator, with a neural network that includes a plurality of input layers which receives at least a first input value and a second input value. The first input value is based on at least one of a phase difference between a first alternating-current signal and a second alternating-current signal, frequencies of the first alternating-current signal and the second alternating-current signal, and an amplitude of the first alternating-current signal or the second alternating-current signal. The second input value is at least one of a measured value of a vibrational state in a vibrating body and a measured value corresponding to an admittance characteristic of the vibrating body.