4results about How to "Few adjustable parameters" patented technology

The invention relates to the technical field of artificial intelligence, and provides a medical image domain generalization method based on a diffusion probability model, and the method comprises the following steps: S1, obtaining image data needed by training; s2, extracting image structure information; s3, building a diffusion generation model network suitable for the image data size and characteristics; s4, using the extracted structure information as prompt information, and prompting and controlling a diffusion generation model learning process; and S5, using a diffusion generation model to realize data distribution transfer of structure information maintenance. Through the technical scheme of the invention, the diffusion generation model conforming to the medical image data is designed, and the image data structure information is extracted as a prompt, so that cross-image modal data conversion maintained by the structure information is realized, meanwhile, the diversity of training data is increased, the purpose of domain generalization in a reasoning stage is achieved, and the reasoning efficiency is improved. Therefore, the downstream segmentation task precision for various image modalities is improved, and rapid and accurate diagnosis is assisted.

This invention relates to a vibration suppression method based on high-precision closed-loop synchronous rotating coordinate transformation. First, a dynamic model of the deflection of a magnetically levitated rotor affected by mass imbalance is established. Then, a synchronous vibration controller based on a high-precision closed-loop synchronous rotating coordinate transformation method is used to track the synchronous frequency disturbance component, and a high-precision closed-loop synchronous frequency signal detection method is employed to improve the detection accuracy of the synchronous frequency vibration signal. Finally, the synchronous frequency vibration signal is eliminated by inverting and compensating the original system, thereby eliminating the synchronous frequency disturbance current and the synchronous frequency vibration torque. This method overcomes the problem of low detection accuracy of synchronous frequency signals in traditional synchronous rotating coordinate transformation methods, effectively improving the suppression effect of synchronous frequency disturbance current, and providing a new control method for unbalanced rotor vibration control. This invention belongs to the field of magnetically levitated rotor control and can be used for unbalanced vibration control of two-degree-of-freedom deflection of magnetically levitated rotors.

The invention relates to the technical field of AC motor control, and discloses an AC motor sensorless control method and system. The method comprises the following steps: acquiring output current, output voltage and high-frequency injection switching value information of an alternating current motor; establishing a position estimation model suitable for the full-speed range of the motor according to the active flux linkage model of the AC motor, and estimating the effective voltage vector according to the output current, the output voltage and the high-frequency injection switching value information of the AC motor; extracting an electrical angle estimation error of the motor rotor according to the estimation value of the effective voltage vector; estimating a motor electric angle and an electric angle speed according to the motor rotor electric angle estimation error; and controlling the motor according to the estimated values of the electric angle and the electric angular speed of the motor. According to the method, the position and the speed of the motor in the full-speed range are estimated by utilizing a position estimation model, the calculation amount is small, adjustable parameters are few, and independent NS magnetic pole identification is not needed.

The invention relates to the technical field of AC motor control, and discloses a control method for an AC motor with an LC filter, and the method comprises the steps: obtaining an instruction voltage vector and a current vector of an inverter end; estimating the voltage vector and the current vector of the motor end under the estimation coordinate system according to the instruction voltage vector and the current vector of the inverter end; and calculating a difference value between the corresponding stator flux linkage vectors according to the voltage vector and the current vector of the motor end under the estimation coordinate system, and estimating the position and the rotating speed of the motor rotor according to the calculated difference value. According to the invention, only the current vector at the inverter end needs to be measured, and the sensorless control of the AC motor with the LC filter can be realized without designing a high-order observer or a high-order controller.