Motor control circuit
AssigneeInfineon Technologies AG
InventorsFlorian Hauer, Marko Gecic, Kajetan Nürnberger, Jürgen Schafer
Life of this patent
Estimating the term…
Abstract
According to various embodiments, a motor control circuit is described having a controller configured to determine values of a plurality of control voltages for a motor. The motor control circuit includes one or more current sensors configured to measure a plurality of operation currents of the motor and a neural network having a multi-layer perceptron architecture. The neural network is trained to estimate a rotor position of the motor for a current control cycle. The controller is configured to determine values of the plurality of control voltages for the current control cycle using the estimate of the rotor position.
Claims20
A motor control circuit comprising:
- a controller configured to determine values of a plurality of control voltages for a motor;
- one or more current sensors configured to measure a plurality of operation currents of the motor;
- and
- a neural network having a multi-layer perceptron architecture, wherein the neural network is trained to estimate a rotor position of the motor for a current control cycle from an input including,
- for each control cycle of one or more previous control cycles,
- values of the plurality of control voltages supplied to the motor in the control cycle;
- and
- values of the plurality of operation currents measured by the one or more current sensors in the control cycle;
- and
- for the current control cycle, values of the plurality of operation currents measured by the one or more current sensors in the current control cycle;
- wherein the controller is configured to determine values of the plurality of control voltages for the current control cycle using the estimate of the rotor position.
A method for controlling a motor, the method comprising:
- measuring a plurality of operation currents of the motor;
- estimating a rotor position of the motor for a current control cycle by a neural network having a multi-layer perceptron architecture, wherein the neural network is trained to estimate the rotor position of the motor for the current control cycle from an input including,
- for each control cycle of one or more previous control cycles,
- values of a plurality of control voltages supplied to the motor in the control cycle;
- and
- values of the plurality of operation currents measured by one or more current sensors in the control cycle;
- and
- for the current control cycle, values of the plurality of operation currents measured by the one or more current sensors in the current control cycle;
- and
- determining values of the plurality of control voltages for the current control cycle using the estimate of the rotor position.
A motor control circuit comprising:
- a controller with a voltage output, a current input, and an estimated position input;
- a micro-controller peripheral with a voltage input coupled to the voltage output of the controller over a voltage interface, and a current output coupled to the current input of the controller over a current interface;
- and
- a neural network with a voltage input coupled to the voltage interface, a current input coupled to the current interface, and an estimated position output, wherein
- the estimated position output of the neural network is coupled to the estimated position input of the controller.
Dependencies are inferred from the claim text.