Motor control apparatus and method using machine learning

The motor control device addresses discrepancies in existing technologies by using a data-driven approach to classify and update control settings, ensuring precise and stable motor operation with reduced complexity and cost in applications like electric vehicles and UAVs.

KR102993122B1Active Publication Date: 2026-07-21NUVINDA CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
NUVINDA CO LTD
Filing Date
2025-11-27
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing motor control technologies, such as Pulse Width Modulation (PWM) and Field Oriented Control (FOC), struggle with discrepancies between actual operating environments and theoretical design values, leading to system instability and increased complexity and cost in applications like electric vehicles and UAVs, requiring precise and accurate control without complex algorithms.

Method used

A motor control device and method that includes a data collection module, state classification module, parameter calculation module, and correction module to generate accurate motor control signals by analyzing motor sensing data, classifying state variables, and updating control settings in real-time based on calculated parameters.

Benefits of technology

The solution ensures precise motor control with reduced errors, improved stability, and simplified algorithms, maintaining optimal performance under varying conditions by accurately reflecting the motor's actual operating state and adapting control settings in real-time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112025133476397-PAT00002_ABST
    Figure 112025133476397-PAT00002_ABST
Patent Text Reader

Abstract

The present disclosure relates to a motor control device and a method. The motor control device of the present disclosure comprises a data collection module that generates state data based on motor sensing data received from an inverter of a motor, a state classification module that derives state variable data of a motor based on the state data, a parameter calculation module that calculates parameters of a motor based on the state data and state variable data, a correction module that updates control settings of a motor based on the calculated parameters, and an output module that provides a motor control signal generated based on the updated control settings to an inverter. By using a classifier and a regression model optimized according to the type of state variable, level data of the state variable, or change pattern, parameters that change in real time can be precisely calculated while minimizing the computational burden on a processing device.
Need to check novelty before this filing date? Find Prior Art