Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

35results about How to "Reduce computational efficiency" patented technology

Time sequence anomaly detection method and system based on frequency enhanced convolution Transform

The invention belongs to the field of time sequence anomaly detection, and discloses a time sequence anomaly detection method and system based on frequency enhanced convolution Transform, and the method comprises the steps: carrying out the sequence-level preprocessing of an input sequence, including normalization and block generation; encoding the input sequence by adopting a multi-head self-attention mechanism; extracting interrelationships in the sequence by a full convolution module having a plurality of stacked 1D convolution layers of different kernel sizes; s2 and S3 are repeated, and multi-layer coding is carried out on the sequence; converting the encoded embedding into a frequency signal by using a Fourier transform method; the KL divergence and the absolute error are respectively used for calculating the consistency loss of the time domain and the frequency domain, calculating the point-by-point anomaly score, and identifying whether the point is abnormal or not. Wide experiments on five common data sets show that the method is superior to an existing method in the aspect of abnormal recognition.
Owner:UNIV OF CHINESE ACAD OF SCI

Water quality detection method and system based on artificial intelligence

The invention discloses a water quality detection method and system based on artificial intelligence. The method comprises the steps of data collection, data preprocessing, water quality pre-classification, water quality trending analysis, pollution traceability prediction and water quality detection. The invention relates to the technical field of water quality detection, in particular to a water quality detection method and system based on artificial intelligence, and the scheme adopts an artificial intelligence water quality detection method including water quality pre-classification, trend analysis and pollution tracing; performing water quality pre-classification by adopting a deep neural network method combining a genetic algorithm and a random forest model; performing water quality pre-classification and water quality trend analysis by adopting a deep neural network method combining a genetic algorithm and a random forest model; according to the method, data are classified in advance through a water quality pre-classification task, unpolluted water sources are eliminated, and then the calculation speed and the calculation effect of pollution tracing are improved through a causal enlightenment operator and an improved graph neural network in combination with a causal enlightenment graph neural network.
Owner:DITIAN ENVIRONMENT TECH (NANJING) CO LTD

Mechanical structure reliability analysis method and system based on active learning agent model

The invention discloses a mechanical structure reliability analysis method and system based on an active learning agent model, and belongs to the technical field of mechanical structure reliability analysis, and the method comprises the steps: analyzing the working condition and load condition of a mechanical structure to be subjected to reliability analysis, and determining a performance function of the mechanical structure in combination with the safety working requirements of the structure; based on the principle of equal probability conversion, the input variables are converted into a standard normal space, and Monte Carlo reliability analysis samples are generated; in a standard normal space, using a center cross method to generate a 4n initial fitting sample, calculating a function value of the sample, obtaining an initial response matrix, and fitting an initial proxy model; and establishing a distance constraint learning function based on the existing agent model and the fitting sample, and iteratively updating the agent model, the fitting sample, the response value matrix and the failure probability in combination with a prior reliable circular ring sample reduction method until a reliability analysis result with enough precision is obtained, and ending the reliability analysis system.
Owner:TAIYUAN UNIVERSITY OF TECHNOLOGY

Real-time orbit determination trigger starting method for low-orbit satellite

ActiveCN119148180AEnsure real-time orbit determination accuracyavoid double countingSatellite radio beaconingError reportingSatellite orbit determination
The invention discloses a low-orbit satellite real-time orbit determination trigger starting method, which belongs to the field of satellite positioning time service and is used for ground data processing, and real-time low-orbit satellite satellite-borne GNSS observation updating conditions are detected, low-orbit satellite satellite-borne GNSS observation and real-time completeness and continuity detection calculation is carried out on GNSS products, so that the real-time low-orbit satellite orbit determination trigger starting is realized. Cyclic standard judgment is carried out among different GNSS systems, orbit determination arc lengths and GNSS products, real-time orbit determination is triggered and started only when observation and product quality reach the standard and orbit determination arc section observation is updated, the real-time orbit determination precision can be guaranteed, repeated calculation when observation is not updated is avoided, and the real-time orbit determination efficiency is improved. In addition, invalid / low-quality orbit determination or frequent error reporting when feeding observation and product quality have problems can be avoided, the calculation efficiency can be saved, and repeated problem tracing during operation and maintenance is avoided.
Owner:NAT TIME SERVICE CENT CHINESE ACAD OF SCI

CAV ecological driving guiding method based on vehicle-road cooperation in mixed traffic environment

The invention discloses a CAV ecological driving guiding method based on vehicle-road cooperation in a mixed traffic environment. The method comprises the following steps: 1, collecting vehicle information at a t0 moment and a real-time state of a signal lamp at a current intersection; 2, judging the position of a traffic scene where the CAV is located; 3, determining a terminal state of CAV ecological driving according to different traffic scenes; 4, determining an optimal speed trajectory curve of the CAV; and 5, dynamically updating and optimizing the track through a trigger type track updating rule. Through information interaction between the CAVs and the roadside facilities and between the CAVs in the vehicle-road coordination environment, the phenomena of starting, stopping, idling, acceleration and deceleration of the CAVs at the intersection can be reduced, so that the traffic efficiency is improved, the energy consumption is reduced, and the sustainable development of traffic is realized.
Owner:HEFEI UNIV OF TECH

Power flow calculation method and system for power system

The invention discloses a power flow calculation method and system for a power system, and aims to relax key variables such as part of generator output and node voltage from equality constraints to inequality constraints in order to solve the problem that part of generator output and node voltage are difficult to give due to large-range change of an operation mode in a novel power system; after the inequality constraint is introduced, the power flow model is changed into a novel power system constraint power flow model from an original nonlinear equation set solving problem; converting an inequality in the constraint power flow model into an equation by adopting a method of constructing an auxiliary function, and then converting the constraint power flow model into a bounded constraint least square problem; aiming at the problem that a traditional optimization algorithm is difficult to converge, a trust region algorithm is adopted for solving, and the adjustment amount of variables such as generator output and node voltage and a final power flow solution are obtained. The method can effectively solve the problem that the power flow has no solution due to the fact that the initial power flow value is difficult to give, and theoretical support is provided for planning and operation of a power system.
Owner:CHINA SOUTHERN POWER GRID COMPANY

Double-optimization efficient learning training method for on-orbit interaction of double spacecrafts

The invention discloses a double-optimization efficient learning training method for on-orbit interaction of double spacecrafts, which aims at a task entering a specified target area, fully considers time-space characteristics of the task, establishes relative orbit dynamics of a tracking star and a target star and a mathematical model of the target area, designs constraint conditions, and establishes win-lose conditions of the tracking star and the target star. In order to improve the training efficiency, a twice optimized efficient learning training method is designed, firstly, a tracking star and a target star respectively adopt a decision-making method based on zero-control miss distance to carry out pursuit interaction, track data in the interaction process of the two parties are collected and analyzed, and a neural network is adopted to fit the action of the tracking star for guiding subsequent training; according to the method, the defect that convergence is difficult due to the fact that training starts from zero can be overcome, the neural network obtained through training can output the optimal thrust of the tracking star, and the tracking star can enter a designated area at the designated moment.
Owner:BEIJING INST OF CONTROL ENG

Distortion image correction method based on fast multi-scale residual network

The invention discloses a distorted image correction method based on a fast multi-scale residual network. Relates to the field of image geometric distortion correction, in particular to the technical field of distortion image correction based on a fast multi-scale residual network. Comprising the following steps: S1, constructing a data set, and generating a paired data set of a distorted image and a standard image and a deformation displacement field file of a corresponding number of the distorted image by using a geometric distortion model; s2, a distorted image correction model based on a fast multi-scale residual network is constructed, the model is divided into a backbone network and a tributary network, the backbone network is used for recovering image features, and the tributary network is used for outputting image classification features; s3, training the distorted image correction model based on the fast multi-scale residual network by using the pairwise data set of the distorted image and the standard image and the deformation displacement field file; and S4, processing the to-be-corrected image by using the trained distortion image correction model based on the fast multi-scale residual network to obtain the corrected image and the corresponding classification thereof.
Owner:JILIN UNIVERSITY +1

Rolling bearing fault method and system based on Riesz atlas sparse decomposition

The invention belongs to the technical field of time-frequency analysis, and particularly relates to a rolling bearing fault method and system based on Riesz atlas sparse decomposition, which utilizes a new time-frequency analysis method-Riesz four-parameter spectrum time-frequency analysis method to obtain four-parameter information of amplitude, frequency, phase and direction of a vibration signal in a time-frequency domain. The time-frequency aggregation and the anti-noise performance of an analysis result are improved; in order to screen out an optimal parameter spectrum and expand the symbol dynamics entropy to a two-dimensional scale, a new entropy calculation method, namely cosine symbol dynamics entropy, is provided; and reconstructing components of the parameter spectrum obtained by entropy screening, and constructing a sparse filter parameter optimization component amplitude, so that a decomposition result has physical significance. The bearing vibration signal under the variable speed working condition and the bearing vibration signal under the constant speed are analyzed, the RSSD method is provided on the basis, and feature extraction and diagnosis of rolling bearing faults under the constant rotating speed working condition and the variable rotating speed working condition are achieved.
Owner:HUNAN UNIV OF SCI & TECH

Hyperspectral anomaly detection method based on background clustering constraint under potential feature separation

The application discloses a hyperspectral anomaly detection method based on background clustering constraint and potential feature separation, and relates to the technical field of hyperspectral remote sensing detection applications. First, band selection and partition generation of a hyperspectral original image are carried out to generate fixed-size image blocks. Then, a basic image reconstruction framework is constructed based on a lightweight spectral-spatial feature extraction encoder and a convolution decoder. Then, a pixel-level deep clustering network is introduced to complete latent feature clustering modeling, calibrate the background clustering label of the image block, and obtain the intra-class RX detection score in the feature space. Then, the intra-class RX score and the reconstructed image are reversely mapped to the original image space to construct a global RX response map and a reconstructed full image. Finally, a weight matrix is generated by using the reconstruction error between the original image and the reconstructed full image, the global RX response map is modulated, and the final anomaly detection result is output. The application fuses the background clustering constraint and the potential feature separation mechanism, effectively improves the detection accuracy of small anomaly targets, and significantly suppresses false alarm interference.
Owner:SHANGHAI INSTITUTE OF TECHNICAL PHYSICS CHINESE ACADEMY OF SCIENCES

A method for anisotropic ionospheric very low frequency scattering at low altitude based on transfer matrix method

The present invention belongs to the field of electromagnetic wave technology, and discloses a very low frequency scattering method of anisotropic ionosphere in medium and low altitude based on the transmission matrix method. The present invention establishes a low ionosphere layered model structure considering the anisotropy of the ionosphere. First, the transmission matrix of the anisotropic layered structure is derived and verified. Then, based on the transmission matrix, the scattering characteristics of the very low frequency electromagnetic waves in the above-mentioned ionosphere layered structure under different conditions are accurately calculated, and the propagation law is summarized. This work provides a reliable calculation model for the propagation of very low frequency in the low and medium ionosphere, and greatly improves the accuracy of its propagation and scattering parameter calculation. This has important guiding and reference value for studying the propagation of very low frequency electromagnetic waves in the atmosphere.
Owner:NAVAL UNIV OF ENG PLA

Inference calculation method and related equipment

PendingCN119598999Areduce computational efficiencyshorten the timeSemantic analysisInference methodsLinguistic modelAlgorithm
The invention discloses a reasoning calculation method and related equipment, and relates to the technical field of artificial intelligence, and the reasoning calculation method comprises the steps of obtaining document data to be subjected to reasoning calculation, segmenting the document data to obtain multiple segments of text data, reasoning and calculating the multiple segments of text data based on a preset large language model to obtain multiple segments of target text data, the preset large language model is a trained current large language model, determining relevance among the multiple segments of target text data based on a preset paragraph relation determination algorithm, and splicing the multiple segments of target text data based on the relevance to obtain document data after reasoning calculation. According to the method, the document data to be subjected to reasoning calculation are segmented, reasoning calculation is performed on each segment of text data, and then each segment of text data after reasoning calculation is spliced to obtain the document data after reasoning calculation, so that the reasoning calculation efficiency is improved.
Owner:CHINA MERCHANTS BANK

Statistical calculation method of harmonic output variable based on three-point estimation method

On the basis of a statistical calculation method of harmonic output variables of a three-point estimation method, harmonic voltage Z of the output variables is respectively expanded at expected values of a fan, a photovoltaic system and a load as input random variables Xi in combination with a formula # imgabs0 #, and position coefficients xii, j and weights wi, j corresponding to each estimation point xi, j are obtained. According to the method, a group of estimation points x'i and j are additionally added on the basis of three-point estimation to construct multiple sampling points, so that accurate solution of random variable statistical characteristics is realized, the calculation accuracy is ensured, introduction of high-order moment can be avoided, and three-point sampling has higher practical value.
Owner:CHINA THREE GORGES UNIV

Impedance detection circuit, method and system

The invention discloses an impedance detection circuit, method and system, and relates to the field of circuit detection. The impedance detection circuit comprises an electrified body, a positive load circuit and a negative load circuit; the anode of the electrified body is connected with the anode load circuit, the cathode of the electrified body is connected with the cathode load circuit, and the series node of the anode load circuit and the cathode load circuit is grounded; the positive load circuit comprises a first capacitor and a first load, and the negative load circuit comprises a second capacitor and a second load; in the charging or discharging process of the first capacitor, the first capacitor is used for providing first voltages corresponding to at least three time nodes respectively; in the charging or discharging process of the second capacitor, the second capacitor is used for providing second voltages corresponding to at least three time nodes respectively; the first voltage and the second voltage are used for determining an insulation resistance detection result of the electrified body to the ground. The circuit can improve the result accuracy when the insulation resistance of a charged body to the ground is detected through a capacitor under the condition of low circuit cost.
Owner:GONEO GRP CO LTD

Method and system for calculating bushing electric field distribution based on two-dimensional axial symmetry model

The invention provides a method and system for calculating bushing electric field distribution based on a two-dimensional axial symmetry model, and belongs to the technical field of power equipment simulation, and the method comprises the steps: 1, carrying out the dimension reduction processing of an actual transformer bushing three-dimensional structure according to the actual geometric structure symmetry of a transformer bushing, and obtaining an actual transformer bushing electric field distribution value; extracting a transformer bushing two-dimensional axisymmetric model capable of reflecting the actual structure of the bushing; 2, establishing a temperature field simulation model based on the two-dimensional axisymmetric model of the transformer bushing extracted in the step 1, and calculating the two-dimensional temperature field distribution of the transformer bushing; 3, establishing an electric field simulation model based on the two-dimensional axisymmetric model of the transformer bushing extracted in the step 1, coupling the temperature distribution of the transformer bushing obtained in the step 2 into the electric field simulation model, and calculating the two-dimensional electric field distribution of the transformer bushing; and 4, performing circumferential scanning on the calculation result of the two-dimensional electric field distribution of the transformer bushing, restoring the three-dimensional electric field distribution of the transformer bushing, and obtaining the three-dimensional electric field distribution of the transformer bushing.
Owner:STATE GRID JIANGSU ELECTRIC POWER CO LTD NANTONG POWER SUPPLY BRANCH

Mammary gland dynamic enhancement MRI intelligent analysis system and method thereof

The invention relates to the technical field of medical information systems, in particular to a breast dynamic enhancement MRI intelligent analysis system and method, and the system comprises a data collection module which is used for obtaining breast dynamic enhancement magnetic resonance imaging (DCE)-MRI data; the image preprocessing module is used for carrying out standardization processing on the DCE-MRI data; the segmentation module is used for segmenting the preprocessed image data into a format suitable for model input; the model reasoning module is used for performing deep learning analysis and focus diagnosis on the segmented image; the report generation module is used for generating a diagnosis report according to the model reasoning result; and the data storage module is used for storing data of each stage and a final diagnosis result, so that the analysis capability of the mammary gland DCE-MRI is comprehensively improved. The accuracy and the efficiency of diagnosis are remarkably improved, particularly, the performance is excellent in the aspect of tiny focus detection, and the clinical credibility is improved by enhancing the interpretability.
Owner:XUZHOU MEDICAL UNIVERSITY

Rotating shaft modal calculation method and device in multi-bearing supporting mode, electronic equipment and storage medium

The invention provides a rotating shaft modal calculation method and device in a multi-bearing supporting mode, electronic equipment and a storage medium, and the method comprises the steps: constructing a one-dimensional simulation model of a gear system; input parameters of the gear system one-dimensional simulation model are determined, and the input parameters of the gear system one-dimensional simulation model comprise real working condition data of a bearing connected with a rotating shaft and gearbox shell parameters; performing system deformation analysis based on the gear system one-dimensional simulation model to obtain the direction connection rigidity of the bearing; the modal of the rotating shaft is calculated based on the three-dimensional finite element model of the rotating shaft, in the three-dimensional finite element model of the rotating shaft, rigid connection is adopted in the connecting position of the bearing, and the other side of the rigid connection is connected with a one-dimensional connection unit; the directional connection stiffness in the one-dimensional connection unit is determined based on the directional connection stiffness of the bearing. The rotating shaft modal can be calculated based on the real working condition data of the bearing, and then the accuracy of rotating shaft modal calculation can be improved.
Owner:GAC AION NEW ENERGY AUTOMOBILE CO LTD

A method for triggering and starting real-time orbit determination of low-orbit satellites

The present invention discloses a method for triggering and starting real-time orbit determination of a low-orbit satellite, which belongs to the field of satellite positioning and timing. The method is used for ground data processing, detects the update status of real-time low-orbit satellite-borne GNSS observations, detects and calculates the real-time integrity and continuity of low-orbit satellite-borne GNSS observations and GNSS products, performs cyclic compliance judgment among different GNSS systems, orbit determination arc lengths, and GNSS products, and triggers and starts real-time orbit determination when the observation and product quality meet the standards and the orbit determination arc segment observation is updated. The method can ensure the accuracy of real-time orbit determination, avoid repeated calculation when the observation is not updated, and avoid invalid / low-quality orbit determination or frequent error reporting when problems occur in the input observation and product quality, thereby saving calculation efficiency and avoiding tracing repeated problems during operation and maintenance.
Owner:NAT TIME SERVICE CENT CHINESE ACAD OF SCI

A water quality detection method and system based on artificial intelligence

ActiveCN119249287BAids in single function maintenancereduce computational efficiencyGeneral water supply conservationTesting waterWater sourceWater quality
The present invention discloses a water quality detection method and system based on artificial intelligence, and the method includes data collection, data preprocessing, water quality pre-classification, water quality trend analysis, pollution source tracing prediction and water quality detection. The present invention relates to the field of water quality detection technology, and specifically refers to a water quality detection method and system based on artificial intelligence. The present scheme adopts an artificial intelligence water quality detection method of water quality pre-classification, trend analysis and pollution source tracing; a deep neural network method combining genetic algorithm and random forest model is used for water quality pre-classification; a deep neural network method combining genetic algorithm and random forest model is used for water quality pre-classification and water quality trend analysis; a water quality pre-classification task is used to first pre-classify the data, exclude unpolluted water sources, and then combine the causal-inspired graph neural network, and through the causal-inspired operator and the improved graph neural network, the calculation speed and calculation effect of pollution source tracing are improved.
Owner:DITIAN ENVIRONMENT TECH (NANJING) CO LTD

An efficient wide-rate domain format construction method based on algebraic form

The present invention relates to the field of aerospace computational fluid dynamics technology, and discloses an efficient wide-speed domain format construction method based on algebraic form, including: generating a simulation discrete grid for the aerospace aircraft configuration to be calculated; decomposing the control equation flux passing through the discrete grid unit into convective flux and viscous flux; discretizing the convective flux using the HLLE++ format, and converting it into a combination of the left and right unit flux means and the dissipative term; decomposing the dissipative term into a combination of multiple groups of vectors, and decomposing each group of vectors into a product combination of HLLE++ format characteristic root terms, density terms, and velocity terms; integrating the convective flux into algebraic operations; iteratively solving the control equation to obtain the flow field information of each discrete grid unit. The present invention can solve the problems of low calculation efficiency and poor scalability of the existing HLLE++ construction, and has great application prospects in wide-speed domain multi-body separation, aircraft-missile separation, and numerical simulation of variant aircraft.
Owner:CALCULATION AERODYNAMICS INST CHINA AERODYNAMICS RES & DEV CENT

A method for azimuthally anisotropic seismic data inversion for dipping fracture rocks

PendingCN122129250ASolve for stabilityGood application resultsSeismic signal processingBorehole/well accessoriesCovarianceLeast squares
This invention is a method for inverting azimuth difference seismic data in inclined fractured rocks. Previous AVAZ inversion methods for fracture parameters were mostly based on the HTI medium assumption, with insufficient consideration for fracture dip angle. The fixed form of the inversion constraint terms made it difficult to adapt to changes in parameter statistical characteristics, and the regularization parameters were all manually adjusted, affecting inversion accuracy and efficiency. This invention establishes an approximate formula for the reflection coefficient of the TTI medium, characterized by the P-wave and S-wave velocities, density, and fracture elastic parameters of the background medium. It utilizes pre-stack azimuth difference data to perform AVAZ inversion of fracture parameters. Dynamic sparsity constraints are introduced, dynamically adjusting the sparsity of the constraints based on data characteristics, and a covariance matrix is ​​introduced to consider the statistical regularity between parameters. Combining iterative reweighted least squares and adaptive regularization parameter acquisition methods, the dynamic constraint inverse problem is transformed into a weighted... L A 2-norm objective function is used to achieve a stable solution. Data testing shows that this invention can effectively improve the inversion accuracy of crack elastic parameters.
Owner:HOHAI UNIV

Epilepsy prediction system based on adjustable suppression emission pulse neural network

The invention relates to the technical field of epilepsy prediction, in particular to an epilepsy prediction system based on an adjustable emission suppression spiking neural network, which specifically comprises a data preprocessing module, a spiking neural network module, a neural network optimization module, an emission mode judgment module and an accelerator module, a data set capable of being provided for neural network training and testing is obtained through the data preprocessing module, and then software optimization and hardware optimization are carried out on a pulse neural network through the pulse neural network module, the neural network optimization module and the emission mode judgment module. And finally, realizing the epilepsy prediction system based on the spiking neural network capable of adjusting suppression emission through an accelerator module. The epilepsy prediction system solves the problems that in the prior art, the calculation speed is low and the requirements of wearable equipment cannot be met due to the fact that the calculation amount of the spiking neural network is huge and the power consumption is large.
Owner:NANJING UNIV

System for constructing high-precision user behavior portrait through improved federal learning algorithm

The technical scheme of the invention discloses a system for constructing a high-precision user behavior portrait through an improved federal learning algorithm, and the system is characterized in that the system comprises a multi-source privacy protection data obtaining module; a federated learning modeling module based on dynamic weighting and contribution proving; a multi-dimensional user behavior portrait module; a personalized vehicle network interaction strategy generation module; a dynamic execution and fair excitation module; and a feedback and model self-optimization module. According to the method, deep analysis and portrait construction can be carried out on historical data of user charging and discharging behavior habits, a high-precision and high-coverage user behavior portrait model is generated through an improved federated machine learning algorithm, the real-time income of users is improved, the user viscosity is enhanced, and the power grid stability and the renewable energy consumption rate are further improved.
Owner:STATE GRID SHANGHAI MUNICIPAL ELECTRIC POWER CO +2

A method, device, equipment and storage medium for calculating diffuse angle coefficient in high temperature engineering field

The present application discloses a method, device, equipment and storage medium for calculating a diffusion angle coefficient, which relates to the field of high temperature engineering. The method comprises: a processing device determines multiple radiation surfaces of a target space, selects an sth radiation surface and a kth radiation surface from the multiple radiation surfaces, determines an hth radiation surface from the multiple radiation surfaces, the hth radiation surface, the sth radiation surface and the kth radiation surface are different radiation surfaces, determines the occlusion between the sth radiation surface and the kth radiation surface by judging the positional relationship between the sth radiation surface, the kth radiation surface and the hth radiation surface, and calculates the diffusion angle coefficient according to the occlusion. The method can reduce the amount of calculation and improve the calculation efficiency.
Owner:BEIJING LINGYUN ZHIQING SOFTWARE CO LTD

Efficient wide speed domain format construction method based on algebraic form

The invention relates to the technical field of aerospace computational fluid mechanics, and discloses an efficient wide speed domain format construction method based on an algebraic form, and the method comprises the steps: generating a simulation discrete grid for the configuration of an aerospace aircraft needing to be calculated; decomposing a control equation flux passing through the discrete grid unit into a convection flux and a viscous flux; discretizing the convection flux by adopting an HLLE + + format, and converting the discretized convection flux into a combination of a left and right unit flux mean value and a dissipation item; decomposing a dissipation item into a combination of multiple groups of vectors, and decomposing each group of vectors into a product combination of an HLLE + + format feature root item, a density item and a speed item; the convection flux is integrated into algebraic operation; and iteratively solving the control equation to obtain the flow field information of each discrete grid unit. The method can solve the problems that an existing HLLE + + structure is low in calculation efficiency and poor in expansibility, and has a large application prospect in wide-speed-range multi-body separation, aircraft-missile separation and variant aircraft numerical simulation.
Owner:CALCULATION AERODYNAMICS INST CHINA AERODYNAMICS RES & DEV CENT

Microgrid privacy protection method and system based on novel homomorphic encryption algorithm

The invention discloses a microgrid privacy protection method and system based on a novel homomorphic encryption algorithm, and relates to the technical field of microgrid control. The method comprises the steps of designing a micro-grid distributed cooperative control method based on a distributed control theory, and realizing frequency recovery of each distributed power supply in a micro-grid; defining a privacy protection connotation and a privacy protection range; according to the defined privacy protection connotation and privacy protection range, a microgrid privacy protection method based on a novel homomorphic encryption algorithm is designed, an initial value and a real-time value of each distributed power supply in the microgrid are protected from being leaked to node neighbors, and the distributed cooperative control target of the microgrid is achieved; and verifying the microgrid privacy protection method based on the novel homomorphic encryption algorithm by using a simulation experiment. According to the invention, the credible perception capability and privacy security of distributed cooperative operation of the micro-grid can be improved.
Owner:ZHEJIANG UNIV +2

A software-defined computable storage method and system

The application relates to the technical field of data storage and edge computing, and discloses a software-defined computable storage method and system, which comprises a front-end computing platform and a rear-end solid-state storage device connected through an NVMe protocol communication connection; the front-end computing platform is externally presented as a standard NVMe storage device, adopts a heterogeneous system on chip, and the heterogeneous system on chip integrates multiple heterogeneous processors; the front-end computing platform is sequentially provided with a PCIe endpoint function layer, an NVMe controller layer and a hardware abstraction layer, a kernel-mode user-mode bridging layer, a user-mode I / O engine layer, a lightweight user-mode FTL layer, a heterogeneous computing power scheduling layer, a scene-based computing command interface layer, a stream data service implementation layer and a communication expansion layer. The application can solve the technical problems of existing computable storage architectures, such as multiple data movements and function solidification.
Owner:CHANGSHA SUNAC FUTURE TECHNOLOGY CO LTD

A weight data processing method, system and application of model inference matrix multiplication calculation

PendingCN122286058Aimprove performancelarge memory footprintComputational scienceRound complexity
This invention discloses a method for processing matrix multiplication weight data in model inference, adapted to large-scale matrix multiplication in BFP format for large models in the Transformer architecture. Addressing the pain points of existing architectures regarding BFP exponent and mantissa splitting, bandwidth, and resource consumption, this method employs (64,64) weight blocks, dual address generator 1:4 scheduling, waterline flow control, and double buffering techniques to achieve full-bandwidth read / write between DDR, BRAM, and computational units. Through multi-core parallelism and multi-round multiplication and accumulation, performance is improved by more than 3 times compared to FP16, reducing memory usage and design complexity while ensuring inference accuracy, providing an efficient hardware acceleration solution for LLM inference in resource-constrained environments.
Owner:SHANGHAI QUSU CHAOWEI TECHNOLOGY CO LTD

A multi-robot path planning method and system based on hierarchical control

This invention discloses a multi-robot path planning method and system based on hierarchical control. The method acquires model information and basic parameters of each robot in the system, and constructs a realistic map of the working environment using the SLAM algorithm. The grids of the constructed realistic map are merged, transforming a fine-grained grid map into a coarse-grained one. Considering the overall movement direction and target position of the robots, a global path input for the navigation system is obtained. This global path input is published to each robot, and robots can subscribe to other robots' paths to capture their global path information. Each robot perceives local environmental information in a local map and obtains information about obstacles around other robots. Based on the global path results from the MAPF algorithm, a local path planning method is used to generate local paths, enabling avoidance between robots. This provides a reliable solution for efficient collaboration in multi-robot systems.
Owner:XI AN JIAOTONG UNIV

Mechanism deep learning dual-drive flow-slip disaster numerical simulation system

The invention provides a mechanism deep learning dual-drive flow-slip disaster numerical simulation system, and relates to the technical field of geological disaster numerical simulation. The system integrates a physical mechanism model and a deep learning method. Based on a high-resolution DEM (Digital Elevation Model), by introducing a rock-soil type self-adaptive substrate entrainment mechanism, utilizing a shallow water wave equation to accurately calculate the terrain elevation, the flow direction, the flow velocity and the accumulation thickness of the first two time steps; and then taking the obtained multivariable field as four-channel input, and driving a CNN-GRU coupling network to perform rolling prediction. According to the method, the final accumulation form, the influence range and the dynamic evolution process can be quickly obtained only through a small amount of initial mechanism calculation, the simulation efficiency is remarkably improved on the premise of ensuring physical rationality, and the defects that a traditional pure mechanism model is long in calculation time consumption and a pure data driving model is physically distorted are overcome; the method is suitable for rapid risk assessment and emergency decision support of flow slip disasters such as landslide and debris flow.
Owner:HEBEI UNIV OF TECH