38results about How to "High engineering practical value" patented technology

The invention discloses a low-overlapping-rate point cloud registration method based on topology-metric decoupling, and belongs to the technical field of computer vision and three-dimensional reconstruction, and the method comprises the steps: scanning an object through a three-dimensional laser scanning collection terminal, collecting point clouds at two different angles as a source point cloud and a target point cloud, and achieving high-robustness registration in a difficult scene with an extremely low overlapping rate; according to the method, through the asymmetric association strategy, the zero solution problem under the low overlapping rate is solved through the one-way matching union set, and the recall rate is remarkably increased; meanwhile, through a DGTG geometric pruning and probability manifold optimization mechanism, a voxel quantization error is eliminated by using anisotropic covariance and Lie algebra iteration, a translation error and a rotation error are greatly reduced, and a precision bottleneck caused by voxelization is effectively broken through. According to the method, the existence of a solution can be ensured under the condition that the overlapping rate is extremely low, voxel quantization errors can be eliminated through manifold optimization based on covariance, and high-precision and high-efficiency registration of low-overlapping-rate point clouds is achieved.

The application belongs to the technical field of antenna in the communication system comprehensive design, and particularly relates to a kind of omnidirectional broadband antenna based on multimode resonance, L-shaped folded dipole, rectangular loop, U-shaped branch and SMA joint are located at the lower surface of dielectric substrate, feed line is located at the upper surface of dielectric substrate, first metallized via hole and second metallized via hole in the up-down direction are opened on dielectric substrate, the outer conductor of SMA joint is connected with the right arm of L-shaped folded dipole, the inner conductor of SMA joint is connected with one end of feed line by passing through the right arm of L-shaped folded dipole and first metallized via hole, the other end of feed line is connected with the left arm of L-shaped folded dipole by passing through second metallized via hole, the application generates new resonance frequency by bending traditional dipole to form L-shaped folded dipole;rectangular loop is added around L-shaped folded dipole, two new resonance modes are introduced, and the bandwidth of antenna is further widened.

This application relates to a clock tree synthesis method and system for reducing the vertical coupling capacitance of the clock tree, belonging to the field of clock tree synthesis technology. The core of the method lies in the fact that, during the clock tree synthesis process in the physical design phase, by using a preset vertical design of the clock tree and a coupling-aware clock tree synthesis optimization cost function, the parasitic coupling capacitance between clock lines and adjacent signal lines in the vertical metal layer is systematically reduced, thereby improving the noise immunity of the clock network and reducing delay uncertainty and dynamic power consumption.

This invention provides an integrated method for pruning and controlling large-scale models of unmanned aerial vehicles (UAVs), belonging to the field of model lightweighting technology. The proposed method significantly reduces model size and computational load, greatly improves inference speed and control response performance, and enables the deployment of complex neural network controllers on resource-constrained airborne platforms. Simultaneously, the lightweight model maintains or approaches the control accuracy and stability of the original uncompressed model. This method has wide applicability and strong scalability, covering various scenarios from basic flight control to complex mission control, providing a feasible, efficient, and cost-effective solution for applying large models to airborne controllers of UAVs, and possesses high engineering practical value.

This invention relates to the technical fields of intelligent diagnostic equipment for traditional Chinese medicine (TCM), medical IoT, and equipment and systems for internet-based TCM hospitals. Specifically, it relates to an integrated terminal and system for the four diagnostic methods (inspection, auscultation, inquiry, and palpation), incorporating a built-in TCM health model, wireless networking capabilities, and access to an internet-based TCM hospital platform for online registration, AI-based diagnosis, physician review, and electronic prescription processing. It aims to address the problems of single-function diagnostic equipment, data disconnect, and lack of a closed-loop diagnosis and treatment system. The terminal includes an integrated shell and integrated multimodal sign acquisition, edge computing core, interactive touch display, and secure communication gateway module. The acquisition module consists of inspection, auscultation, inquiry, and palpation units, synchronously acquiring tongue appearance, voice, chief complaint, and pulse wave features under a unified clock. The edge computing core incorporates a built-in TCM health model, deeply fusing multimodal features through a cross-modal attention mechanism.

The application provides an antenna array surface fiber grating layout method based on a genetic algorithm and a neural network, and relates to the technical field of deformation sensing of an antenna array surface structure and optimization of fiber grating layout. The method analyzes strain sensitive positions of the array surface structure through the genetic algorithm and the neural network, arranges fiber grating sensors at optimal layout positions of the array surface structure, and realizes high-precision deformation sensing of the array surface structure under complex service loads. The method solves the problems of node layout redundancy, high data processing cost and insufficient displacement prediction accuracy in traditional antenna monitoring, and provides technical support for efficient sensing of antenna structure deformation and stable guarantee of electromagnetic performance.

The invention discloses a damper real-time performance evaluation and health diagnosis system and method. Current traditional damper monitoring faces many limitations. The system comprises a sensor array module, a data acquisition and preprocessing module, a data fusion processing module, a performance evaluation module, a health diagnosis module, an early warning module and a communication and storage module. The data fusion processing module is used for performing automatic fusion processing on the preprocessed data to generate optimized fusion data; the performance evaluation module is used for calculating various performance indexes of the damper through the fused data; the health diagnosis module diagnoses the fault type of the damper according to the fused data and the performance evaluation result, analyzes the fault reason and evaluates the risk level of the fault; the early warning module is used for issuing early warning information; and the communication and storage module is used for data transmission and storage. According to the invention, the accuracy, real-time performance and reliability of damper monitoring are improved.

This invention relates to the field of software operation forensics and fault diagnosis technology, specifically to a method for recording a minimum sufficient evidence package for an airbag control unit and an automotive airbag control unit. The method is executed in a Classic AUTOSAR environment and includes: determining whether both anomaly symptom conditions and operational state time gating conditions are simultaneously met; if so, determining that the trigger condition is established; recording the trigger time; and freezing event information, path signatures, and key variable snapshots at the end of a predetermined window after triggering. Combined with the invariant assertion bitmap at the time of triggering, a minimum sufficient evidence package is generated. Compared with existing technologies, this invention can record a minimum sufficient evidence package with predictable and extremely low operational overhead, and has advantages such as not changing the airbag control logic path, not entering the ignition ISR critical segment, and retaining evidence even in power-down scenarios.

This application provides a method and system for stabilizing and tuning parameters of reactive power compensation equipment in a new energy collection station. The method includes: constructing a frequency domain admittance matrix of the grid-connected system of the new energy collection station; determining the dominant characteristic value of the grid-connected system of the new energy collection station based on the frequency domain admittance matrix; calculating the corresponding system damping ratio based on the dominant characteristic value; calculating the sensitivity of the system damping ratio to at least one parameter of reactive power compensation equipment based on the system damping ratio; constructing a parameter optimization model based on the sensitivity and a preset damping ratio adjustment target; and solving the parameter optimization model to obtain the tuning values ​​of at least one parameter of reactive power compensation equipment for coordinated parameter tuning.

This application discloses a machine learning-based method for constructing numerical atlases of marine propellers, relating to the field of marine technology. This method uses a viscous flow numerical calculation model to calculate the performance parameters of a propeller with design parameters for each set of samples at various sample infeed velocities. Then, using the design parameters and infeed velocities of each set of samples as input and the corresponding performance parameters as output, a performance surrogate prediction model is trained using multiple sets of sample design parameters based on a machine learning model. This surrogate prediction model can then be used to obtain the performance parameters corresponding to different propeller schemes within the atlas parameter range, thus constructing a numerical atlas of marine propellers. This method combines the high efficiency and high accuracy of viscous flow numerical methods with the superior nonlinear fitting capability of machine learning models. It does not rely on extensive physical model experiments, has a simple construction method, and produces propeller atlases with high numerical accuracy.

The invention discloses an unmanned aerial vehicle attitude control method based on multi-scene cruise, and particularly relates to the field of unmanned aerial vehicle flight control. Collecting a wind field change deviation rate, a magnetic interference intensity index and a visual occlusion density abnormal value to form a disturbance characteristic parameter set; constructing a disturbance influence index matrix in each preset cruise scene; acquiring a scene sequence in an actual cruise path, calculating a scene switching frequency and constructing a fusion disturbance vector; inputting the fusion disturbance vector into an attitude control model constructed based on a long short-term memory network, and outputting an attitude control parameter group; dynamically adjusting an unmanned aerial vehicle attitude instruction according to the control parameter group; the control strategy can be dynamically adjusted according to environment disturbance change and scene switching characteristics, attitude stability control of the unmanned aerial vehicle in a complex and changeable environment is achieved, and flight safety and task reliability are improved.

This invention relates to the field of semiconductor equipment manufacturing technology, specifically to a method, system, device, and medium for semiconductor equipment fault prediction based on a time-series graph neural network. The method first collects time-series operational data of each component of the equipment to form a data matrix. The data matrix is ​​then preprocessed and segmented to obtain standardized segments. Next, a time-series graph is constructed for each segment: using components as nodes, edge weights are determined based on physical connections and Pearson coefficients to form a time-series graph sequence. This sequence is then divided into training and testing sets, input into a time-series graph neural network model for training, and optimized using a cross-entropy loss function. Finally, real-time data, processed through the above steps, is input into the trained model, outputting the fault type and probability, and displaying and pushing the results. By constructing a time-series graph data structure and designing a dedicated time-series graph neural network model, accurate early fault prediction is achieved.

This invention provides a 100G coherent acquisition closed-loop demodulation method and system integrating TIADC and demodulation, belonging to the field of digital signal processing. The method includes: TIADC mismatch correction, frame synchronization and symbol synchronization, resampling clock recovery, frequency domain full dispersion compensation, three-level polarization demultiplexing (CMA+LMS+DDLMS), joint carrier frequency offset estimation and compensation, Viterbi-Viterbi carrier phase recovery, soft decision and full performance monitoring. This invention is the first to use TIADC front-end correction as the first stage of demodulation, achieving deep integration of acquisition and demodulation. The three-level cascaded equalization architecture balances convergence speed and compensation accuracy. A joint carrier recovery closed loop is established through phase residual feedback, significantly improving demodulation robustness and accuracy. This method can output full test indicators in real time, uses FPGA parallel pipelines to reduce processing latency, and is suitable for engineering applications such as 100G coherent acquisition cards and optical communication test instruments.

This invention discloses a position angle controller and its control method based on an improved fractional-order active disturbance rejection (ADRROC) module. The position angle controller includes a TD module, a FO-ESO module, and a FO-NLSEF module. The output of the TD module is connected to the input of the FO-NLSEF module. The output of the FO-NLSEF module is connected to the controlled object and the FO-ESO module. The input of the FO-ESO module is connected to the controlled object, and its output is connected to the FO-NLSEF module. This invention utilizes improved ADRROC technology to achieve motor position angle control. By improving the convergence, continuity, and differentiability of the nonlinear function and adding a fractional-order module, the performance of the ADRROC position angle controller is improved and enhanced.

This invention discloses a field-circuit collaborative analysis method and system based on reinforcement learning and Bayesian optimization. The invention utilizes a Bayesian optimization algorithm to optimize the parameter values ​​of each electronic component in a selected topology. Based on the optimized parameter values, a reinforcement learning-based parameter search environment for electronic components in the RF device circuit topology is constructed, modeled as a Markov decision process. An equivalent circuit model of the RF device is designed according to the optimal action. The equivalent circuit model of the RF device is connected to other circuit systems for circuit-level simulation analysis. By integrating reinforcement learning and Bayesian optimization, two intelligent algorithms, and fully leveraging their respective advantages in discrete structure search and continuous parameter optimization, a high degree of automation and intelligence is achieved throughout the entire field-circuit collaborative analysis process. While ensuring the physical rationality of the model, the efficiency and accuracy of the collaborative analysis are significantly improved.

The application discloses a kind of real-time evaluation system and method of automobile interior noise based on deep learning model.The system includes audio acquisition module, feature extraction module, deep learning model module and man-machine interaction module;The deep learning model module includes working condition identification submodule and noise evaluation submodule;The working condition identification submodule inputs the pre-trained improved ResNet working condition identification model with the calibrated mel spectrum feature, and outputs the working condition category of current noise;The noise evaluation submodule inputs the pre-trained CRNN noise evaluation model with the calibrated Log mel spectrum feature, and outputs the comprehensive noise quality score;The man-machine interaction module is used to provide intuitive operation and monitoring interface for user.The application can automatically identify the specific test working condition of vehicle, output the comprehensive score highly consistent with the feeling of professional evaluator, while providing intelligent test progress guidance and visual monitoring.

The application relates to the technical field of wireless communication, and particularly discloses an end-to-end wireless communication system and method based on AI, which collects signals and channels and hardware data in a real scene to construct an original data set; based on the data set, a hybrid modeling method of fusing a differentiable physical simulation and data-driven residual learning is used to construct a parameter-adjustable joint distortion representation; the representation is combined with internal states of a communication process, and a reinforcement learning idea is used to dynamically adjust parameters to generate an adversarial distortion mode; the mode is used for multi-round adversarial training of the communication process, communication parameters and distortion generation strategies are alternately optimized until a stable and robust state is reached; finally, the trained communication process is deployed on an actual link, and parameter adaptive matching and online fine-tuning are realized through environment sensing; the application can effectively simulate the coupling distortion of channels and hardware, and improves the communication reliability and generalization performance in a complex dynamic wireless environment.

The invention discloses a permanent magnet synchronous motor predictive current control method for improving a sliding-mode observer and an overmodulation scheme. The method comprises the following steps: S1, establishing a hyper-local mathematical model of a motor under a synchronous rotating coordinate system; s2, constructing a sliding-mode observer SMO, and introducing a sliding-mode reaching law of a variable exponential term to carry out real-time estimation on current and disturbance; s3, extracting and compensating fifth-order and seventh-order harmonics in the motor current, obtaining harmonic current components through multi-synchronous rotating coordinate system transformation, calculating harmonic compensation voltage through closed-loop feedback, and further generating a final modulation signal; and S4, judging whether the modulation signal is in an overmodulation region or not, and outputting a PWM driving signal to an inverter to control the operation of the permanent magnet synchronous motor by adopting an overmodulation strategy based on carrier PWM combined with zero sequence signal injection in the overmodulation region, which is equivalent to a minimum amplitude error method in space vector PWM. The method has comprehensive advantages in the aspects of steady-state precision, dynamic response, parameter robustness and over-modulation region performance.

The present application relates to the technical field of propeller hydrodynamic measurement method, and particularly relates to a propeller lateral excitation force measurement method, which converts the dynamic force signal measured by a sensor in a rotating coordinate system XOY into a propeller lateral excitation force signal in a fixed coordinate system X1OY1 in the form of equal-phase resampling, and can effectively evaluate the propeller lateral line spectrum excitation force. The main steps include test system calibration, signal synchronous acquisition, equal-phase resampling, coordinate transformation and Fourier transformation. The method has clear design process, clear operation and great engineering practical value.

This invention relates to a method for predicting the fracture behavior of the brazing interface of high-nitrogen steel-molybdenum-niobium alloy based on multi-element coupled modeling, comprising: S1, constructing an initial pure metal crystal structure model; S2, optimizing the geometric structure of the model; S3, cutting the model to construct multiple crystal plane models containing different atomic layer thicknesses; S4, performing convergence tests on the multiple crystal plane models respectively; S5, constructing a high-nitrogen steel principal element / brazing filler metal principal element pure metal interface model and a molybdenum-niobium alloy principal element / brazing filler metal principal element pure metal interface model; S6, constructing a realistic brazing interface composite model with multi-element metallurgical interaction characteristics through atomic substitution; S7, calculating interface adhesion work data and electronic structure data to predict the preferential fracture location and fracture mode. The realistic brazing interface model constructed by this invention with multi-element coupled doping can simulate element segregation and phase transformation in actual brazing, quickly infer stress concentration areas and predict fracture modes, significantly shortening the research and development cycle and reducing costs.

The application relates to the technical field of power electronic converter control, in particular to an H-bridge converter control method and system based on dynamic rotation of high-frequency and low-frequency modules. The method combines real-time collection of junction temperatures and switching losses of each H-bridge module; for each H-bridge module, a first absolute difference value of the current junction temperature and the current average junction temperature is calculated, and a second absolute difference value of the cumulative switching loss and the average cumulative switching loss is calculated; according to the first absolute difference value and the corresponding temperature weight coefficient, and the second absolute difference value and the corresponding power loss weight coefficient, the rotation index of each H-bridge module is determined; and according to the rotation index of each module, a corresponding pulse control signal is generated, the high-frequency module to be rotated is switched to a low-frequency module, and the low-frequency module to be rotated is switched to a high-frequency module. Under the premise of ensuring the quality of electric energy output, the method can balance switching loss and thermal stress and other problems, and significantly improve the loss balance and temperature consistency of the H-bridge converter.

The application discloses a risk evolution identification method and device based on continuous plot state migration, and belongs to the technical field of natural language processing and risk control. The method comprises the following steps: acquiring multi-round plot text data; generating a state vector for each round of plot semantic analysis and sequentially storing the state vector into a sliding window to construct a plot state sequence; calculating a risk state migration probability, a plot inertia value and a trend risk value based on the sequence; weighting and fusing a current round risk value, a historical cumulative risk value, the trend risk value and the plot inertia value to obtain a comprehensive risk score; and calculating a dynamic threshold value according to a context length, a risk fluctuation rate and the plot inertia value, and outputting a graded review result of release, early warning or interception after comparing the dynamic threshold value with the comprehensive risk score. The application solves the problem that risks are difficult to be dynamically evolved and identified in continuous plot interaction, and realizes real-time monitoring and graded control of risk evolution.

This application relates to the field of mechanical testing technology, and proposes a shield tail sealing test device for a tunnel boring machine (TBM). The device includes a transparent test chamber, a hydraulic loading device, a gas delivery device, and a water supply device. The transparent test chamber contains a tail brush assembly, which divides the chamber into a hydraulic pressure chamber and multiple grease chambers. The hydraulic loading device includes a first hydraulic loading pipe, a first hydraulic piston and a first water circuit reversing valve disposed within the first hydraulic loading pipe, a second hydraulic loading pipe, and a second hydraulic piston and a second water circuit reversing valve disposed within the second hydraulic loading pipe. The first hydraulic loading pipe is connected to the hydraulic pressure chamber via the first water circuit reversing valve, and the second hydraulic loading pipe is also connected to the hydraulic pressure chamber via the second water circuit reversing valve. The gas delivery device is connected to both the first and second hydraulic loading pipes. The water supply device is connected to both the first and second water circuit reversing valves. This application has the effect of stabilizing the pressure of the test water.

The application discloses a kind of tunnel lining crack detection method and system based on improved YOLO model, constructs the tunnel lining crack detection model based on improved YOLO (including but not limited to YOLOv8-YOLOv12 and its derivative architecture), including in the feature extraction module of main network Backbone, introduction expansion residual network DWR;Variable convolution and channel-space double attention mechanism are introduced in detection head network Head;Optimize bounding box regression loss function, and boundary box regression loss is calculated using Inner-CIoU loss function.The application realizes the high-precision, real-time detection of multi-scale cracks, especially fine cracks, under the premise of not significantly increasing the computing overhead, through the multi-level, collaborative improvement of network structure.This method is not only suitable for tunnel lining crack detection, but also can be widely used in the detection of other fine crack defects such as bridges, roads and building exterior walls.

The application discloses a double-layer shielding type super-deep stratum concrete flow state detection device and method, belongs to the technical field of water conservancy and hydropower foundation construction, and comprises a box body, a pressure sensor and a driving mechanism are arranged in the box body, and further comprises a detection mechanism, the detection mechanism comprises an outer sleeve, an inner push rod, a driving assembly and a detection assembly; the outer sleeve is axially connected with the pressure sensor, the inner push rod and the driving assembly are arranged in the outer sleeve, a detection cavity is surrounded between the inner push rod and the outer sleeve, and the detection assembly is arranged at one end of the inner push rod close to the detection cavity. The double-layer shielding type nested structure formed by the outer sleeve and the inner push rod makes the stress signal collected by the detection assembly only be the pure axial resistance exerted by the concrete to be detected on the end face of the inner push rod, solves the defects of the existing detection device, such as detection data distortion and insufficient precision caused by resistance superposition, eliminates the detection error caused by resistance superposition, and guarantees the reliability and accuracy of the detection data.

The application discloses a speed control method of a permanent magnet synchronous motor system constrained by current and rotating speed. The method provided by the application comprises the following steps: establishing a dynamic model of the permanent magnet synchronous motor system and converting the dynamic model into a state equation; designing a disturbance observer for unmatched disturbance; determining a penalty function according to the state equation of the permanent magnet synchronous motor system constrained by current and rotating speed obtained in step two and state constraints existing in the system; determining a corresponding adaptive speed controller of the permanent magnet synchronous motor according to the dynamic model of the permanent magnet synchronous motor, the state equation of the permanent magnet synchronous motor, the disturbance observer and the penalty function; and constructing a corresponding Lyapunov function according to the adaptive speed controller of the permanent magnet synchronous motor. The rotating speed and current of the permanent magnet synchronous motor can track a reference signal with a given precision without violating the required constraint conditions, and the method has robustness to external disturbance and uncertain parameters.

This invention discloses an improved sliding mode observer for permanent magnet synchronous motor control and a system relating to the field of motor control technology. Addressing the problems of severe chattering and low estimation accuracy in traditional sliding mode observers, the method includes: acquiring the motor voltage and current and transforming them to a stationary coordinate system; using an adaptive sliding mode observer integrating a fuzzy adaptive adjustment unit and a smoothing switching unit for state observation, outputting a smoothed back EMF estimation signal, wherein the fuzzy adjustment unit dynamically adjusts the sliding mode gain based on the current error, and the smoothing switching unit generates a continuous control signal to suppress chattering; inputting the smoothed back EMF signal into a phase-locked loop (PLL) unit to synchronously extract rotor position angle and speed estimates for closed-loop control. The system includes signal detection, an adaptive sliding mode observer, a PLL, and a closed-loop control module. This invention improves the accuracy of speed and position observation and system stability by synergistically suppressing chattering through fuzzy adaptive and smoothing switching, and by avoiding phase delay through a PLL.

The invention provides an antenna array surface fiber grating layout method based on a genetic algorithm and a neural network, and relates to the technical field of deformation perception of an antenna array surface structure and fiber grating layout optimization. According to the method, the strain sensitive position of the array surface structure is analyzed through the genetic algorithm and the neural network, the optical fiber grating sensor is arranged at the optimal layout position of the array surface structure, and high-precision structural deformation sensing of the array surface structure under the complex service load is achieved. The problems of node layout redundancy, high data processing cost, insufficient displacement prediction precision and the like in traditional antenna monitoring are solved, and technical support is provided for efficient sensing of antenna structure deformation and stable guarantee of electromagnetic performance.

This invention discloses a method and system for identifying non-full-phase protection in the main wiring of a three-half circuit breaker. The method includes a branch non-full-phase protection identification method based on the phase position of the side and interrupted circuit breakers; branch non-full-phase protection based on the side and interrupted circuit breakers, current sequence components, and phase current-current-free characteristics; a three-phase inconsistency protection identification method for the side circuit breaker; and a three-phase inconsistency protection identification method for the interrupted circuit breaker. The system includes an information acquisition module, an information calculation module, a debouncing processing module, and a signal identification module. This invention perfectly adapts to the complex structure of the main wiring of a three-half circuit breaker; it provides accurate identification, significantly improving the accuracy of non-full-phase operation identification and effectively reducing misjudgments; it has high reliability, with an innovative time-delay blocking mechanism that reliably avoids protection maloperation caused by three-phase inconsistency contact sticking, effectively ensuring the stable operation of the protection system.