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

24050 results about "Calculation methods" patented technology

Calculation methods. A calculation method implements a piece of the calculation framework. Different calculation method classes are used for the different operations that are required to complete a calculation within the calculation framework.

Pre-control and monitoring method and system for whole process of actual grouting engineering based on digital geological model

The invention provides a pre-control and monitoring method for a whole process of actual grouting engineering based on digital geological model, comprising: extracting discontinuous fracture surfaces in a geologic body structure, to build a underground geologic body structure model; build a multi-source geologic body attribute model, optimizing and solving grouting simulations of a disaster high-risk region by using multiphase flow calculation method, and controlling grouting equipment by using optimized grouting parameters to carry out an actual grouting engineering; obtaining a multi-factor diffusion range of variable coefficients by adjusting physical values of the parameters in the grouting simulation and attribute data structure model by using a control variable method, learning and capturing complex mapping relationship between different parameters by using a neural network, optimizing the grouting parameters in the actual engineering in real-time, so as to realize the pre-controlling, monitoring and optimization of the whole process of the actual grouting engineering.
Owner:SHANDONG UNIV

Quadruped robot autonomous navigation method and system for special environment

The invention relates to the technical field of navigation, and particularly provides a quadruped robot autonomous navigation method and system for a special environment, and the method comprises the steps: generating a semantic tag for depth point cloud data corresponding to a depth image through employing a recognition result of a target detection network for the depth image; calculating local odometer information according to the laser point cloud data and the inertial sensor data by using a laser radar mileage calculation method; fusing the depth point cloud data with the semantic label, the laser point cloud data with the semantic label and the local speedometer information to obtain positioning data in a pre-constructed point cloud map; constructing a multi-layer cost map according to the pre-constructed point cloud map, the newly acquired depth image, the laser point cloud data and the inertial sensor data by using a sensing method of a region planner; and generating navigation path information according to the positioning data and the multi-layer cost map by using a global path planning algorithm and a local path planning algorithm. The method improves the navigation precision.
Owner:SHANDONG ENERGY GRP CO LTD +1

Edge calculation method based on AI

The invention discloses an edge computing method based on AI (artificial intelligence), relates to the technical field of AI, and aims to solve the problems of insufficient real-time performance, difficulty in meeting real-time performance requirements, poor privacy protection and large bandwidth occupation existing in fusion of existing edge computing and AI technologies. According to the technical scheme, the method is characterized by comprising an improved real-time data processing framework, a high-efficiency privacy protection computing framework, a dynamic AI model updating mechanism, an intelligent resource scheduling system, a high-reliability system framework and intelligent energy consumption management, and through an improved streaming processing algorithm, data processing delay is reduced by more than 50%; the model precision is improved by 15% by introducing incremental learning and federated learning, the model updating bandwidth occupation is reduced by 70%, the data leakage risk is reduced by 90% by applying homomorphic encryption and differential privacy, the calculation efficiency is only reduced by 10%, the data transmission quantity is reduced by 60% by the adaptive compression algorithm, and the information loss does not exceed 5%.
Owner:LIAONING SHAGUO CLOUD TECHNOLOGY CO LTD

Method and system for realizing Text2SQL (Structured Query Language)

The invention discloses a Text2SQL (Structured Query Language) implementation method and system, and relates to the field of data processing, and the method comprises the following steps: firstly, receiving a natural language query, and analyzing a query intention, field classification and a key entity through a planner; the searcher obtains domain knowledge, entity information, a database table structure and a historical query mode in a multi-path parallel mode based on the planning result; the generator constructs an SQL framework according to the retrieval result and generates an initial statement; the verifier carries out grammar, table field, authority and logic multi-dimensional verification on the SQL, and if the verification fails, iteration adjustment is carried out to generate logic; when the SQL is executed, the result is formatted and a natural language explanation containing query logic, a data source and a calculation method is generated if the SQL is executed successfully, and a diagnosis and error correction mechanism is started for correction and then rechecking is performed if the SQL is executed unsuccessfully. According to the method, through deep fusion of domain knowledge, whole-process verification error correction and interpretability enhancement, the accuracy, robustness and user interaction experience of SQL conversion in a professional scene are improved.
Owner:XUNTU TECH (SHANGHAI) CO LTD

Calculation method and system for unstructured text data

The invention provides an unstructured text data calculation method and system, and relates to the technical field of natural language processing, and the method comprises the steps: carrying out the multi-level processing of input unstructured text data, dynamically adjusting the word segmentation granularity according to the word frequency distribution, constructing a part-of-speech co-occurrence matrix in combination with context semantic information, and extracting entity information. And fusing the part-of-speech co-occurrence matrix and the entity information to generate a hierarchical semantic tag sequence. Feature extraction units with different convolution kernel sizes are adopted to extract feature representation, and cosine similarity between different semantic hierarchies is calculated to establish an association weight matrix. A semantic enhancement vector is constructed based on entity information, and adversarial training is carried out to obtain a multi-modal semantic feature matrix. And calculating the semantic similarity between the fusion feature vectors for clustering, setting difficulty weight sorting according to the complexity, consistency and ambiguity of class clusters, then inputting the class clusters into a classifier, and iteratively optimizing a classification result to obtain a calculation result.
Owner:北京科杰科技有限公司

Industrial equipment real-time monitoring system based on edge computing

The invention discloses an industrial equipment real-time monitoring system based on edge computing, and the system comprises the following steps: collecting, processing and analyzing the operation data of industrial equipment through distributed edge computing nodes, constructing a node cooperation mechanism, and completing the processing of the data; the dynamic data scheduling module is used for dynamically optimizing a data transmission path and task allocation according to node performance; the intelligent model dynamic migration module is used for adjusting lightweight model deployment in combination with node resource conditions; the multi-modal fault sensing module is used for extracting and analyzing multi-modal features by using a cross-node collaborative algorithm; the edge node self-adaptive management module is used for adjusting node operation parameters in real time and repairing an abnormal state; and the equipment relevance reasoning module is used for constructing an equipment relevance model through a graph calculation method and generating a real-time relevance graph. According to the method, efficient monitoring and anomaly detection of industrial equipment are realized through resource optimization and intelligent analysis of collaborative edge computing nodes.
Owner:ZHEJIANG XIAOQU INFORMATION TECH CO LTD

Large-scale UAV mission planning method and system

A large-scale UAV mission planning method includes constructing an objective function and its constraints; initializing a mission sequence and inserting customers into the mission sequence with the smallest objective function value to obtain a plurality of initial mission sequences; iteratively performing the update operations on the plurality of initial mission sequences. Fast acquisition of optimal missions is achieved by the invention. This invention solves the existing problem of not being able to obtain the optimal mission due to the complex and time-consuming calculation method and the difficulty of convergence of the algorithm.
Owner:BEIHANG UNIV

Scheduling calculation method of production plan scheduling platform

The invention provides a production scheduling calculation method of a production plan scheduling platform. The method comprises the steps of data collection, demand prediction intelligent analysis, productivity planning and bottleneck analysis, production scheduling plan making, real-time monitoring and dynamic adjustment, plan execution and feedback analysis and report generation and decision support. Through distributed scheduling and real-time dynamic adjustment, the production plan can respond to emergencies such as equipment failure and order change in production in time, the continuity and stability of the production process can be ensured, production stagnation or delay caused by the emergencies is avoided, meanwhile, the production scheduling plan can be rapidly adjusted through real-time data feedback, and the production efficiency is improved. Resource utilization is optimized, waste is reduced, production efficiency is improved, intelligent productivity analysis and bottleneck identification are utilized, an enterprise can comprehensively know the use condition of each production resource, and possible bottleneck problems are found and solved in advance.
Owner:BEIJING NORTH VEHICLE GROUP CORP

Die structure strength calculation method and system based on datamation design

The invention provides a die structure strength calculation method and system based on datamation design, and relates to the technical field of die design, and the method comprises the steps: obtaining a die assembly, converting the die assembly into a three-dimensional digital model, carrying out adaptive segmentation, generating a semantic feature tree, and setting a parameterization mapping table, generating a sub-region finite element model and constructing a mold finite element model; constructing a material microstructure evolution model, extracting equivalent mechanical performance parameters, establishing a stress-strain analysis model, solving the stress-strain analysis model to obtain full-field dynamic response, and constructing a sub-region stress characteristic parameter database; constructing a mold structure strength evaluation model, generating a mold structure strength knowledge graph, carrying out multi-index fusion calculation, obtaining a mold strength score, judging whether the mold strength meets the use requirement or not, if not, carrying out global sensitivity analysis, determining a key sub-region, generating a topological optimization scheme and a size optimization scheme, and carrying out association modification; and iterating repeatedly, and outputting the three-dimensional digital model.
Owner:NINGBO LONGYUE TECH CO LTD

Multimodal large language model training method, correlation calculation method, and label generation method

The present disclosure relates to the technical field of artificial intelligence, and provides a multimodal large language model training method, a correlation calculation method, and a label generation method. The multimodal large language model training method comprises: on the basis of a sample text feature vector, a sample image feature vector, and a sample first multimodal feature vector which are obtained by processing image information and text description information of a sample commodity by a pre-trained multimodal large language model, training the pre-trained multimodal large language model to obtain a multimodal large language model; and processing a sample search word, the image information and the text description information on the basis of the multimodal large language model to obtain a sample search word feature vector, a first multimodal feature vector and a sample second multimodal feature vector, and training the multimodal large language model to obtain a trained multimodal large language model. The trained multimodal large language model of the present disclosure can simultaneously learn image features related to the search word and the text description information, and thus the generated second multimodal feature vector is more accurate.
Owner:HANGZHOU ALIBABA INT INTERNET IND CO LTD

Radar target analytic calculation method based on multi-dimensional data fusion and radar device

The invention relates to the technical field of radar signal processing, in particular to a radar target analytical calculation method based on multi-dimensional data fusion and a radar device. Comprising the following steps: deploying a multi-band radar sensor array comprising an X band, a C band and a Ku band in a radar monitoring area; performing pulse compression and Doppler processing on the time domain echo signal, and extracting a time domain feature; spectral analysis is carried out on the frequency domain signals, and frequency domain features are extracted; performing angle estimation on the spatial signals, and extracting spatial features; a dynamic weight distribution model is constructed, a fusion weight is calculated through an adaptive algorithm based on three-dimensional quality indexes of a real-time signal-to-noise ratio (SNR), feature stability (SI) and data integrity (CI), and a joint representation vector containing time domain, frequency domain and space multi-dimensional information is generated. According to the invention, by deploying the multi-band radar sensor array, the recognition capability of the subtle feature difference of the target is improved.
Owner:SHANDONG EAGLE INFORMATION ENG CO LTD

Power grid multi-agent large model safety evaluation index calculation method

The invention provides a power grid multi-agent large model safety evaluation index calculation method, and relates to the technical field of intelligent control. The method comprises the following steps: acquiring a voltage change rate and a frequency change rate, dividing four stages according to disturbance starting and extreme points, pairing curve trends, judging delay and synchronization characteristics, extracting a continuous change structure, and generating a disturbance subsection evolution structure; according to the method, through staged analysis of node voltage and frequency change trends and combination of trend consistency judgment and abnormal segment identification, dynamic structure reconstruction and trend mapping of a disturbance process are realized, and coherence and precision of index response state identification are improved through trend boundary collection and frequency classification; cooperative characteristic identification is enhanced through the coupling direction and delay analysis of inter-node multi-parameter change, the node adjustment priority is defined through regulation and control response sorting, an adjustment path under multi-dimensional linkage is constructed, and the refined management and control capacity of disturbance response and the pertinence of control decision are improved.
Owner:UNIV OF SCI & TECH BEIJING +5

Method for solving conductor galloping trajectory equation of high-voltage transmission line

The present invention relates to the technical field of power transmission conductor galloping, and in particular to a method for solving a conductor galloping trajectory equation of a high-voltage transmission line. The method comprises the following steps: establishing a coordinate system; analyzing a nonlinear motion of any point on a conductor; solving a conductor galloping trajectory equation; solving a torsion angle and an acceleration under an arc length constraint; and carrying out convex optimization on an analytic expression of the torsion angle. According to the present invention, an inter-station RTK differential mode is used, and Beidou observation data analyzed by a monitoring station and Beidou observation data analyzed by a reference station are combined, thereby improving the accuracy of a monitoring result; a finite element analysis method is used to carry out discretization processing on the problem of conductor motion integral, so that conductor displacement can be effectively restored to obtain a good curve fitting effect; and a calculation method for the torsion angle under the limitation of the arc length of the conductor is provided, overcoming the error of calculating a trajectory integral by using the acceleration, and an optimized calculation method is provided.
Owner:CHUXIONG POWER SUPPLY BUREAU OF YUNNAN POWER GRID CO LTD

Foundation deformation calculation method based on numerical model stress extraction

Disclosed in the present invention is a foundation deformation calculation method based on numerical model stress extraction. First, a numerical model is established to calculate the vertical additional stress distribution and horizontal additional stress distribution of foundation soil, wherein the elastic modulus of the foundation soil is selected within a range of two to five times the compression modulus on the basis of engineering experience; a layer-wise summation method is used, in which layered settlement is directly calculated for a natural foundation, settlement in a reinforced zone and settlement in an underlying layer are both calculated for a composite foundation, and layered settlement amounts are superimposed and multiplied by a settlement empirical correction coefficient to obtain a final settlement amount; the projection of a horizontal displacement calculation point on the ground and a vertical soil column below the horizontal displacement calculation point are taken as a vertically placed finite elastic long beam, the cross section of the soil column is square, the side length of the soil column is preferably 0.1 m or below, the modulus of the beam is the same as that of the soil, one end is located in a ground surface plane, and the other end is located in a plane where a calculation depth is located; and a distributed load is dispersed into several concentrated loads, a foundation bed coefficient is determined on the basis of a specification, and the horizontal displacement of the calculation point is obtained by means of a beam-on-elastic-foundation method.
Owner:CHINA RAILWAY DESIGN GRP CO LTD

High and cold wetland dynamic monitoring and health evaluation method and system based on multi-source remote sensing

The invention relates to a high and cold wetland dynamic monitoring and health evaluation method and system based on multi-source remote sensing. The method comprises the following steps: step 1, data acquisition and preprocessing; step 2, multi-modal data feature extraction; step 3, performing multi-modal data fusion; step 4, dynamic monitoring; step 5, evaluating wetland health; according to the invention, the calculation method of the SAR polarization decomposition model and the optical wetland index is optimized due to the unique seasonal change and earth surface complexity of the high and cold wetland, a scientific basis is provided for the management of the wetland protection area, and the ecological restoration efficiency is improved; alpine region water resource management and ecological protection policy making are supported; and data support is provided for researching the wetland carbon sink function and the climate change influence.
Owner:INST OF EXPLORATION TECH OF CHINESE ACAD OF GEOLOGICAL SCI

Carbon emission calculation method based on deep learning

The invention relates to the technical field of computers, and discloses a carbon emission calculation method based on deep learning, which comprises the following steps: S1, collecting multi-source data through a sensor, satellite remote sensing, an intelligent electric meter and a monitoring system, carrying out cleaning, denoising and missing value filling on the collected data, and in addition, carrying out data processing on the collected data; the structured data and the unstructured data are integrated through a multi-source data fusion technology; s2, performing feature extraction on the original data by adopting a deep feature learning model, and screening out key features contributing to carbon emission prediction; and S3, performing real-time decomposition and anomaly detection on the data by combining an autoregressive integral moving average model and an online learning algorithm. By combining an autoregressive integral moving average model and an online Bayesian learning method, carbon emission calculation can dynamically adjust parameters according to real-time data to adapt to environmental changes, seasonal fluctuations and emergencies, so that the flexibility and real-time performance of a prediction model are remarkably improved.
Owner:STATE GRID NINGXIA ELECTRIC POWER CO LTD ECO TECH RES INST +1

Rock burst early warning method and system based on data-mechanism dual drive

The invention discloses a data-mechanism dual-drive-based rock burst early warning method and system, and the method comprises the following steps: deploying a multi-modal sensor network to collect coal and rock stratum data, building a rock burst disaster precursor information sample database, providing a rock burst disaster multi-modal data precursor feature recognition algorithm, and carrying out the recognition of rock burst disaster multi-modal data precursor features. Mining the relevance between the multi-modal data and disaster-causing key risk indexes, and establishing a rock burst disaster multi-modal data prediction model; establishing a three-dimensional geological geometric model, fusing a multi-field coupling dynamics constitutive model and a catastrophe criterion, constructing a PINN physical information neural network prediction model of the rock burst disaster, and obtaining a time-space evolution rule of an energy field of a target area; providing a loss function coupling calculation method of a multi-modal data driving sample error and a physical driving control equation residual error, dynamic data and mechanism prediction result weight, comprehensively calculating a risk score, and accurately judging a top disaster danger level.
Owner:CHINA UNIV OF MINING & TECH

Plate surface defect detection system and method based on machine vision

InactiveCN119048472AReduce reusabilityImprove the optimization spaceImage analysisNeural learning methodsVisual technologyAlgorithm
The invention discloses a plate surface defect detection system and method based on machine vision, and particularly relates to the technical field of machine vision, and the method specifically comprises the following steps: adaptively adjusting the illumination condition of a detection environment, and collecting plate images in a multi-element imaging mode; the method comprises the following steps: dynamically adjusting an image acquired in real time by using an illumination compensation algorithm, preprocessing plate image data, establishing a convolutional neural network model to carry out defect detection and classification on the plate image data, and improving the adaptability of the convolutional neural network model to an illumination condition through a multi-task learning method. The transmission delay and calculation time of image data are reduced through an edge calculation method, a model is optimized by adopting a model compression technology, decision support is provided according to the production state of a plate in combination with the performance of a convolutional neural network model and the processing efficiency of edge calculation, the calculation efficiency and the detection capability can be effectively and comprehensively evaluated, and the detection efficiency is improved. And the control problems of excessive fitting and insufficient fitting are avoided.
Owner:JIANGYIN RUILIN PRECISION MASCH MFG CO LTD

Key value cache compression and sparse attention calculation method and system for large language model reasoning

The invention relates to the technical field of artificial intelligence and natural language processing, in particular to a key value cache compression and sparse attention calculation method and system for large language model reasoning, and the method comprises the steps: an offline calibration stage; the online reasoning stage comprises the following steps: a pre-filling step; an autoregression generation step: for each newly generated lexical element, projecting a current query vector Q and a key vector K in a key cache to a low-dimensional space to obtain Q'and K '; calculating an approximate attention score based on Q'and K ', and selecting an index I of the first k most relevant lexical elements which are ranked from high to low; and calculating an accurate attention score based on Q and K [I], and calculating with the value vector V [I] to obtain the output of the current lexical element. According to the scheme, the memory and calculation bottleneck of large model reasoning in a scene of long text sequence input are solved, and the method has the advantages of reducing video memory occupation and calculation complexity at the same time.
Owner:HANGZHOU INTERNATIONAL INNOVATION INSTITUTE OF BEIHANG UNIVERSITY

Goods volume calculation method and system based on panoramic video

The invention discloses a cargo volume calculation method and system based on a panoramic video, and belongs to the technical field of machine vision, and the method specifically comprises the steps: obtaining a warehouse panoramic video through a panoramic camera, and carrying out the real-time detection and segmentation of a two-dimensional contour of a cargo through the deep learning and time-space context analysis technology; a binocular camera and an active projection structured light technology are adopted to extract cargo depth information, the cargo depth information is converted into three-dimensional point cloud data, and a high-quality three-dimensional model is constructed after radius filtering and optimization processing; optimizing the three-dimensional model by using an adaptive camera calibration and point cloud registration technology, and accurately calculating the cargo volume through volume integration and a physical simulation model; in addition, the system also integrates a multi-dimensional time sequence analysis and machine learning model and an automatic learning measurement error mode, and introduces an intelligent error correction mechanism, so that the dynamic adjustment of the volume measurement value is realized, the measurement precision and reliability are improved, and the automation and intelligence level of cargo volume measurement is improved.
Owner:JIANGSU LINGHAO NETWORK TECH CO LTD +1

Coastal port pile foundation structure load calculation method and system based on coupling mode

The invention discloses a coastal port pile foundation structure load calculation method based on a coupling mode, and the method comprises the steps: a hydrodynamic module which evaluates the load and mechanical properties of a port pile foundation structure under different environment conditions; the water power module at least comprises an atmospheric mode module, a sea wave mode module, an ocean mode module and a coupler module; wherein the coupler module exchanges and synchronizes data among the atmosphere mode module, the sea wave mode module and the ocean mode module; the data conversion module is used for converting the wind field data, the wave field data and the flow field data obtained by the hydrodynamic force module into aerodynamic force and hydrodynamic force input files of the pile foundation structure; the finite element analysis module is used for receiving the aerodynamic force and hydrodynamic force input files, establishing a port pile foundation structure model and calculating stress distribution and displacement response of a pile foundation structure, and simulation results output by the finite element analysis module include but not limited to a stress nephogram, displacement field distribution and a buckling phenomenon of a pile foundation.
Owner:CHANGSHA UNIVERSITY OF SCIENCE AND TECHNOLOGY

Unmanned aerial vehicle flight path positioning method and system based on multi-source information fusion

The invention relates to the technical field of unmanned aerial vehicle positioning, in particular to an unmanned aerial vehicle flight path positioning method and system based on multi-source information fusion. The method comprises the following steps: acquiring an original observation data stream of an unmanned aerial vehicle in real time through a sensor array, and preprocessing to output a multi-source data sample set; calculating relative motion increment between adjacent key frames through an IMU pre-integration model, and extracting local pose estimation through a visual geometric calculation method; evaluating the confidence of each sensor in real time to dynamically generate an adaptive weight coefficient; and constructing a tight coupling fusion filter taking position, speed and attitude errors as core state quantities, and outputting a global optimal trajectory of the unmanned aerial vehicle. According to the invention, the weight is dynamically distributed through the real-time confidence of the sensor, the multi-source data association is fully mined, and the flight path positioning precision and environmental adaptability of the unmanned aerial vehicle are improved.
Owner:YANTAI XINFEI INTELLIGENT SYST CO LTD

Attention state recognition neural network modeling and reasoning method based on electroencephalogram sequence

The invention discloses an attention state recognition neural network modeling and reasoning method based on an electroencephalogram sequence. The core is to construct and train a deep neural network model suitable for electroencephalogram signals so as to realize intelligent recognition and classification. Firstly, wavelet transformation and time-frequency feature extraction are carried out on electroencephalogram time sequence signals, and a multi-dimensional input tensor is generated in combination with channel position information; and inputting the feature into a deep network fusing spatial convolution, gating circulation and a residual connection structure, and extracting spatio-temporal joint features. A cross-time-step attention mechanism and a dynamic loss adjustment strategy are introduced in a training stage, so that the discrimination capability of the model on an alertness state transition region is improved. The final model can conduct reasoning on electroencephalogram data of any length, and a classification label and a confidence score are output and used for measuring classification reliability. The method focuses on construction and optimization of a specific calculation model, reflects application characteristics of an intelligent algorithm in cognitive state recognition, and belongs to an intelligent calculation method with a neural network as a core.
Owner:GUANGDONG UNIV OF TECH

Seawall overtopping amount and overtopping risk calculation method considering mangrove forest wave dissipation effect

The invention discloses a seawall overtopping amount and overtopping risk calculation method considering a mangrove forest wave dissipation effect, and relates to the technical field of coast engineering. The method comprises the following steps: acquiring typhoon path data, inputting the typhoon path data into a Holland typhoon empirical model, and accurately generating a typhoon wind field; calculating coastal site tide level data in combination with a coastal storm surge numerical model; obtaining wave data during a typhoon by using a storm surge typhoon wave forecasting model based on the SWAN; historical typhoon data and relevant information of the seawall and the mangrove forest are collected, a water level-wave-vegetation-seawall coupling model is constructed, the wave dissipation effect of the mangrove forest is accurately simulated, and wave characteristic calculation is optimized; constructing a database, training the database through a CNN neural network model, and improving the prediction precision of seawall front wave height and wave force parameters; the wave overtopping amount is calculated based on the prediction parameters, a scientific basis is provided for wave overtopping risk assessment, it is ensured that real-time assessment of the wave overtopping risk is more accurate and efficient, and the disaster prevention and reduction capacity is effectively improved.
Owner:HOHAI UNIV

High-performance calculation method suitable for multi-scale simulation

The invention provides a high-performance calculation method suitable for multi-scale simulation. The high-performance calculation method comprises the steps that time and space feature scales existing in a multi-scale physical process are judged through a convolutional neural network; according to the obtained time and space characteristic scales, using a depth deterministic strategy gradient algorithm to establish an automatic time step size regulation and control system; adopting an octree adaptive grid realized by a tree-shaped data structure to represent dynamic geometry; capturing the evolution of the geometric structure through an ICP algorithm, and constructing a deformation prediction model; analyzing time scale characteristics of different physical quantities by means of clustering and pattern recognition; designing a coupling step length selection strategy by integrating the numerical stability index and the calculation cost; evaluating a time discretization error by using an error integration method, and predicting numerical stability; on the basis of an error integration method and a parameter identification method, online evaluation of the component is realized; and the next time step value is adjusted through gradient iteration, and the calculation precision is optimized.
Owner:GUANGXI POWER GRID CORP

Geothermal field parameter inversion calculation method, device and system and storage medium

The invention discloses a geothermal field parameter inversion calculation method, device and system, and a storage medium. The method comprises the following steps: establishing a geophysical stratified model; through Monte Carlo sampling of preset parameter spaces of the crustal heat generation rate and the heat conductivity, a temperature value is calculated in combination with a heat conduction equation, and a Gaussian likelihood function evaluation model is constructed to predict the matching degree of the temperature and the actually measured temperature; obtaining the optimal estimation of the thermal parameters, a confidence interval and a correlation matrix among the parameters based on the posterior probability distribution; and according to the correlation matrix, taking a high-confidence result generated by Monte Carlo inversion as priori knowledge of a physical guidance neural network PINNs, and finally outputting a thermal parameter spatial distribution prediction result of the target area through parameter initialization constraint, output layer range limitation and a physical regularization loss item. By adopting the technical scheme of the invention, the defects that the existing actually measured geothermal field parameters are rare and the regional characteristics cannot be described and the geophysical joint inversion of the geothermal field parameters cannot be realized in the prior art are overcome.
Owner:INST OF GEOMECHANICS

Cloud-native application programming interface (API) recommendation method fusing data augmentation and contrastive learning

Disclosed is a cloud-native application programming interface (API) recommendation method fusing data augmentation and contrastive learning. Service information is included on the basis of a service information double-graph structure, and a mutual attention mechanism is designed to compute an importance degree of each layer of information. A data optimization method for sequence information based on functional similarity and a computation method for similarity between services based on two parts of information are provided; on this basis, data of a service invocation sequence is augmented with the idea of contrastive learning to form an augmented sequence pair; a computational contrastive loss function is combined with a pair-wise recommendation loss function to optimize an overall model, thereby improving the effect of a service recommendation model; and according to a feature embedding representation result of a service, pair-wise recommendation scores are computed to complete service recommendation.
Owner:CHINA JILIANG UNIV

Multi-sensor fusion slope stability intelligent analysis system and edge calculation method

The invention relates to the technical field of slope stability data identification, and particularly provides a multi-sensor fusion slope stability intelligent analysis system and an edge calculation method.The system comprises an environment sensing subsystem, a data processing subsystem and an edge calculation subsystem, the environment sensing subsystem monitors the change of the soil moisture content through a soil humidity sensor, and micro movement data of a slope are obtained through a displacement sensor; using a pressure sensor to sense the change of the ground pressure; meanwhile, pre-screening out abnormal data to form a fused data set; the data analysis subsystem is used for converting abnormal data of the fusion data set into a quantitative index of slope stability, and deducing a potential sliding risk of the slope by analyzing the relevance of the soil water content, the tiny movement data or the ground pressure; and the decision support subsystem provides specific reinforcement countermeasures according to the analysis result of the potential sliding risk. According to the method, the potential sliding risk of the side slope is comprehensively deduced by analyzing the relevance of the soil water content, the tiny movement data and the ground pressure.
Owner:CHENGDU UNIVERSITY OF TECHNOLOGY +2

Government affair processing method of big language model based on knowledge graph and electronic equipment

The invention discloses a government affair processing method based on a big language model of a knowledge graph and electronic equipment. The method comprises the steps of performing named entity and relation extraction on government affair related texts by utilizing the big language model; performing named entity classification and entity disambiguation on the named entities and the relationships, and constructing a knowledge graph; performing community detection on the constructed knowledge graph by using a Leiden graph calculation method to generate a hierarchical community structure, and generating a community abstract from bottom to top by using a large language model; and extracting entities from government affair questions input by a user by using the large language model, performing mixed retrieval on the entities by using the constructed knowledge graph and the retrieval model to obtain retrieval results related to the entities, and enhancing the retrieval results by using the large language model to generate result answers. According to the method, information extraction is carried out by utilizing the large language model, so that the information of the constructed knowledge graph is more complete, and the knowledge graph is combined with a retrieval enhancement generation method, so that the interpretability and comprehensiveness of content generated by the large language model in the field of government affairs are improved.
Owner:XIAMEN MEIYABAIKE INFORMATION SECURITY RES INST CO LTD

Method for calculating rheological numerical value of deep covering layer panel rock-fill dam under consideration of seepage action

The invention discloses a deep covering layer panel rock-fill dam rheological numerical calculation method considering the seepage effect. The method comprises the steps that firstly, contact deformation between a panel and a dam body is simulated; then simulating the unstable seepage process of the rock-fill dam and the foundation, and solving the seepage problem with a free surface; carrying out seepage-stress coupling calculation on the dam body material, and establishing a seepage-stress coupling control equation of the dam body material; carrying out seepage-stress coupling calculation on the overburden foundation, establishing a seepage-stress coupling control equation of the foundation, and finally obtaining a relationship among stress, strain and pore pressure to realize the seepage-stress coupling calculation; numerical simulation is conducted, and a concrete faced rockfill dam rheological numerical simulation result considering the seepage effect is obtained. The problems that in the prior art, the stress deformation rule of the concrete faced rockfill dam under the seepage-rheology coupling effect is not clear yet, and meanwhile, a seepage-rheology coupling calculation method is not perfect enough are solved.
Owner:CCCC HEHAI ENG CO LTD +2