The present disclosure provides a permanent-magnet fault-tolerant in-wheel motor based on an active sensorless strategy and drive and design methods thereof. The present disclosure proposes the permanent-magnet fault-tolerant in-wheel motor drive system based on an active sensorless strategy by considering sensorless operation performance in advance in a motor design stage. The present disclosure adopts fractional-slot concentrated windings, and ingeniously arranges alternating poles, a multi-layer magnetic barrier, and auxiliary permanent magnets, thus improving a sensorless operation accuracy of the motor while ensuring fault tolerance of the motor. The present disclosure proposes a frequency-band-adaptive secondary harmonic suppression strategy at a control layer to suppress an influence of a secondary salient harmonic on position observation and improve dynamic response performance of a system.

Described herein is an approximate dynamic programming (ADP) vector controller for control of a permanent magnet (PM) motor. The ADP controller is developed using the full dynamic equation of a PM motor and implemented using an artificial neural network (ANN). A feedforward control strategy is integrated with the ANN-based ADP controller to enhance the stability and transient performance of the ADP controller in both linear and over modulation regions. Simulation and hardware experiments demonstrate that the proposed ANN-based ADP controller can track large reference changes with high efficiency and reliability for PM motor operation in linear and over modulation regions.

A motor control unit (10) includes, for example, a motor control block (11) that performs feedback control of a drive current that flows through a motor (20) and a machine learning block (14) that analyzes input data including at least the drive current so as to detect a failure level of the motor (20). The motor control block (11) could be configured to dynamically switch a control parameter or a control method in accordance with the failure level. The input data may further include, for example, a drive voltage applied to the motor (20). Furthermore, the input data may further include, for example, at least one of vibrations and a temperature of the motor (20) or a motor device (1) mounting the motor (20) therein.

A motor controller is provided with degradation estimation circuitry configured to calculate an estimated degradation level obtained by estimating a degradation level of a device on which a motor is mounted or a degradation level of an inverter; operation decision circuitry configured to compare the estimated degradation level and a reference degradation level, which is predetermined, and set an operation mode of the motor, in a case in which the estimated degradation level is lower than the reference degradation level, to an normal operation, and set the operation mode, in a case in which the estimated degradation level is higher than or equal to the reference degradation level, to a low-noise pulse operation; and control circuitry configured to control the inverter according to the operation mode. The low-noise pulse operation is designed to reduce switching loss to be less than does the normal operation.

In accordance with a trained deep learning model, the data processing system is configured to estimate a temperature of the stator windings of the electric machine based on the following input data: observed current into direct-axis current, observed quadrature-axis current, observed or estimated direct-axis voltage, observed or estimated quadrature-axis voltage, observed direct-current bus voltage, estimated torque of the rotor of the electric machine, estimated speed of the rotor of the electric machine, sensed coolant inlet temperature, and sensed coolant flow rate, wherein the trained deep learning model is trained in accordance with a truncated back propagation through time technique.

A method for detecting a structural fault of an electric motor. The method includes: (i) acquiring a measurement signal of a physical parameter representative of the rotation of the electric motor, (ii) obtaining the high frequency component of the measurement signal, resulting in a corrected signal, (iii) applying a set of band-pass filters to the corrected signal resulting in a set of filtered corrected signals, (iv) determining a stator frequency and a rotor frequency, (v) computing a frequency signature vector, (vi) computing a temporal symptom vector from the frequency signature vector and the set of filtered corrected signals, and (vii) detecting a structural fault of the electric motor from the determined temporal symptom vector and from a classifier model.

A brake system includes a sensor module including a motor current sensor and a force sensor, electric mechanical brake units mounted to wheels of a vehicle and including motors, respectively, and a controller configured to control one or more of the electric mechanical brake units, and the controller predicts states one of the motors of the electric mechanical brake units based on current signals of the motors detected by the motor current sensor, when at least one of the predicted states of the motors indicates that at least one of the motors fails, determines a failure level of the failed at least one of the motors based on sensor data obtained from the sensor module, calculates a requested torque of each of the wheels based on the determined failure level of the failed at least one of the motors, and controls a torque of each of the wheels based on the calculated requested torque of each of the wheels.

A control parameter generation method used in a production device, the method comprising: acquiring measurement data indicating a position from a sensor that measures the position of an object to be driven; and generating, based on the measurement data, evaluation index data indicating vibration of the object to be driven after a time at which the object to be driven has arrived. In a case where the vibration indicated by the evaluation index data is equal to or less than a predetermined threshold, based on the evaluation index data, the control parameter is updated using a machine learning model that learns the relationship between the evaluation index data and the control parameter, and in a case where the vibration indicated by the evaluation index data exceeds the predetermined threshold, the update of the control parameter is interrupted.

A method for optimizing operating accuracy of a brushless winch motor includes outputting a d-axis reference current and a rotor position according to a winch motor startup signal using a flux linkage observer-based sensorless observer model and a motor model; obtaining a real-time rotational speed and a real-time winding/unwinding position of the winch motor according to an operating parameter of the winch motor using the flux linkage observer-based sensorless observer model and the motor model; and obtaining a first offset between the real-time winding/unwinding position of the winch motor and a preset position and a second offset between the real-time rotational speed of the winch motor and a preset rotational speed, performing closed-loop control of the operating parameter with the first offset and the second offset as feedback parameters, and adjusting the operating parameter to optimize operating accuracy of the winch motor.

Examples described herein provide a system that includes a controller operable to execute controller operations that include processing raw sensor output (RSO) to generate an error-compensated version of the RSO; and generating state information based at least in part on the error-compensated version of the RSO. The state information is operable to convey a state of a target.