9results about How to "Effective decoupling" patented technology

This invention provides a fault detection method and system based on low-rank decomposition and common trend decoupling. The method first acquires historical process data under normal operating conditions of an industrial process as training data. Then, it decomposes the training data into clean data representing the main structural information and sparse noise data representing noise interference using low-rank decomposition. Next, it constructs a common trend learning model based on the clean data, reconstructs the non-stationary common trend, and uses the residual between the clean data and the non-stationary common trend as a stationary feature. Decoupling is achieved by applying stationarity constraints. Further, a fault detection model is obtained through joint optimization. In the detection stage, the process data to be detected is input into the model, and monitoring statistics based on the reconstruction error of the stationary feature and the common trend are constructed respectively. The result of the control limit comparison determines whether a fault has occurred, thereby improving the robustness of fault detection in non-stationary industrial processes.

The application relates to the technical field of electronic component reliability, and discloses a flat transformer voltage withstand performance sure reliability modeling and evaluation method, which comprises the following steps: firstly, a constitutive model fusing a physical mechanism is constructed, a time degradation model and a temperature influence model are established based on the geometric size, material properties and environmental stress of an insulating layer, and a process capability parameter is introduced to represent manufacturing fluctuation; secondly, a reliability evaluation model is established, a failure threshold and a margin equation are defined, and physical parameters are identified as random variables to quantify uncertainty; then, Monte Carlo algorithm is used to sample random variables, virtual samples are generated, and voltage withstand margin is calculated on an evaluation time sequence; finally, the number of samples meeting a safety criterion is counted, instantaneous reliability is calculated, and a full-life-cycle reliability curve is generated. The application effectively decouples and quantifies the influence of manufacturing processes, temperature stress and time degradation on voltage withstand performance, and realizes sure reliability evaluation with high confidence.

This invention belongs to the field of vehicle control technology and relates to a vehicle motion control system based on angular module dynamics. The system constructs a two-degree-of-freedom driver control system including steering wheel input and vehicle posture lever input, used to represent path curvature tracking requirements and vehicle posture adjustment requirements, respectively. Based on the driver control input, ideal yaw rate control targets and ideal center-of-gravity sideslip angle control targets are constructed, and the four-wheel steering angle feedforward control quantities are determined through vehicle kinematics methods of pure yaw motion and pure lateral motion. Through feedback optimization control, the additional yaw moment and four-wheel steering angle control quantities are coordinated under composite operating conditions. Based on adaptive adjustment of control weights and constraints under operating conditions, the distribution control of four-wheel drive / braking torque and four-wheel steering angle is achieved. This invention achieves decoupled control of vehicle yaw motion and lateral motion by constructing a two-degree-of-freedom driver control system and combining feedforward kinematics and feedback dynamics control.

This invention discloses an underwater wireless optical communication alignment method and system based on spatial coherence adaptive differential observation. The method includes: constructing a photodetector array in the receiver, constructing a normalized differential observation structure, and through differential observation and dynamic spacing adjustment scheme, a laser system transceiver alignment method that can effectively suppress turbulent multiplicative scintillation and adapt to different water dynamic conditions can be developed, significantly improving the alignment accuracy and robustness between the receiver and the transmitter in complex turbulent environments.

This invention discloses a method for measuring Rydberg atoms using micro-Doppler agility under a polarization-multiplexed microwave field, belonging to the field of quantum precision measurement technology. The method includes generating a dynamic coded sequence and an interleaved coded microwave field by aperiodically pairing and jumping the microwave polarization state with the carrier frequency; acquiring the target echo signal and applying it to the Rydberg atom ensemble with a reference microwave field to generate a quantum coherent state; applying a synchronous laser pulse to acquire a time-series optical response signal; obtaining the signal's fundamental frequency, envelope spectrum, and initial phase component through time-frequency transformation and feature separation; and performing parameter mapping and synthesis to generate three-dimensional micro-motion parameters. This invention uses an aperiodically agile polarization-frequency-multiplexed microwave field for active target detection and utilizes the Rydberg atom ensemble for quantum coherent detection and optical readout of the echo signal, enabling high-sensitivity, high-precision, and high-anti-interference measurement of the target's three-dimensional micro-motion parameters.

This application relates to an AI model image generation method, system, device, and medium, belonging to the technical field of image processing. It involves collecting multimodal data of the target garment, including mannequin images, flat static images, and material close-up images; selecting a reference image from a target pose library whose pose matches the desired pose, and extracting the corresponding pose skeleton image as the target pose skeleton image; inputting the flat static image and the material close-up image into a visual language model to generate a text description; and inputting the target pose skeleton image, mannequin image, and text description into a pre-trained LoRA diffusion model to generate a preliminary dressed image. The LoRA diffusion model's input layer includes an independent pose condition injection channel, through which the target pose skeleton image is input into the LoRA diffusion model. This application has the beneficial technical effect of avoiding stiff and unnatural AI model poses, thereby improving the aesthetics of the generated images.

A SiC MOSFET gate oxide layer degradation comprehensive diagnosis method and system, the method comprises the following steps: constructing a data set, and training to obtain a junction temperature estimation function; obtaining actual precursor parameters under actual operating conditions through a measurement circuit, substituting the actual precursor parameters into the junction temperature estimation function to obtain a current actual junction temperature estimation value, substituting the current actual junction temperature estimation value into a linear relationship between the precursor parameters and the junction temperature before the gate oxide layer degrades to obtain the flat band voltage and the threshold voltage of the gate oxide layer without degradation at the junction temperature; calculating the difference between the actual precursor parameters and the precursor parameters at the junction temperature without degradation of the gate oxide layer to obtain a precursor parameter drift amount, and performing comprehensive diagnosis of the gate oxide layer degradation according to the precursor parameter drift amount to obtain a comprehensive diagnosis result. The present application solves the misdiagnosis problem caused by temperature drift in the prior art, improves the accuracy and robustness of the degradation diagnosis, and realizes real-time monitoring and thermal state management of the device working junction temperature state.

This invention relates to a UAV target recognition method based on ISAC signal decomposition and background contrast attention fusion, belonging to the field of radar signal processing. The method includes: configuring the transmitted signal of an ISAC base station based on a 5G NR frame structure; receiving and establishing a CSI mixed signal model; constructing a parameterized operator; using the rmD-NSP algorithm to iteratively decompose the mixed signal, separating the rotor micro-Doppler component, and converting the separated signal into a micro-Doppler time-frequency spectrum; using the RMBL module to model multi-scale background features of the time-frequency spectrum; using the BCA module to calculate the difference weights between target features and background features, generating a background-suppressed micro-motion feature vector; generating an HRRP sequence using the CSI mixed signal; using an LSTM network to extract the target's geometric structure feature vector; and weighted concatenating and fusing the micro-motion feature vector and the geometric structure feature vector and inputting them into a classifier to output the UAV target's category probability.

The application provides a 4D video generation method, device and equipment with controllable camera and illumination, relates to the technical field of computer vision and computer graphics, and aims to solve the problem that the existing model cannot simultaneously control the camera trajectory and the illumination condition. The method comprises the following steps: based on an input video, generating a dynamic point cloud and a sparse heavy-light point cloud, the dynamic point cloud is used for explicitly representing the geometric structure and motion information of the scene in the input video, and the sparse heavy-light point cloud is used for providing the illumination prior of the scene under the target illumination condition; according to a target camera trajectory, the dynamic point cloud and the sparse heavy-light point cloud are processed respectively to generate geometric prior information and illumination prior information aligned with a target view angle; the geometric prior information and the illumination prior information are encoded into visual tokens, and illumination query tokens are extracted from heavy-light key frames; denoising generation is performed by using the visual tokens and the illumination query tokens, and a target video consistent with the target camera trajectory and the target illumination condition is output.