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

87results about How to "Realize deep integration" patented technology

Group behavior analysis method for multi-modal information fusion and dynamic updating

The invention discloses a multi-modal information fusion and dynamic updating group behavior analysis method. The method comprises the steps of information integration and dynamic updating, multi-modal information obtaining and information preprocessing, feature extraction, feature representation conversion, alignment and fusion, and multi-modal information real-time processing and dynamic updating through a stream processing framework. A self-learning mechanism and a self-optimization mechanism are introduced; group portrait generation: extracting population statistical characteristics, geographic position distribution, behavior and emotional psychological characteristics, social relationships, social media and economic characteristics from the multi-modal data to generate a group portrait; analyzing group behaviors; and outputting visualization, and outputting group behavior prediction results in a visualized manner, including a behavior trend graph, a group clustering graph, an association rule graph and an abnormal behavior detection graph. According to the method, multi-modal data from various sources and forms are effectively integrated and analyzed, self-learning and self-optimization mechanisms are introduced, and finally group behavior analysis is performed.
Owner:GUANGZHOU UNIVERSITY

Fault diagnosis method and system based on multi-scale adaptive hypergraph convolutional neural network

PendingCN119622459AEfficient, fast and accurate fault diagnosisRealize deep integrationBiological modelsData setEngineering
The invention discloses a fault diagnosis method and system based on a multi-scale adaptive hypergraph convolutional neural network, and belongs to the technical field of fault diagnosis. The method comprises the following steps: acquiring original vibration data of mechanical equipment, and carrying out data preprocessing; constructing a fault diagnosis model based on the multi-scale adaptive hypergraph convolutional neural network; and inputting the vibration data after data preprocessing into the constructed fault diagnosis model, and outputting a fault diagnosis result. According to the method, a vibration signal is decomposed into different frequency bands, features under various scales are extracted respectively, prior knowledge is fused, local topological graphs are constructed through various weighting measures, and a plurality of local topological graphs jointly form a spatial hypergraph to serve as input of a model; a dual-channel graph convolutional network is introduced to extract features, weights of different channels are adaptively distributed through an attention mechanism, and deep fusion of the features is realized. Experiments on two reference data sets show that the method can effectively improve the fault diagnosis accuracy, and has good stability and generalization.
Owner:LANZHOU UNIVERSITY OF TECHNOLOGY

5G message video monitoring method and system based on artificial intelligence

The invention discloses a 5G message video monitoring method based on artificial intelligence. The method comprises the following steps: S1, constructing a multi-modal data set; s2, multi-modal fusion features are generated by using a joint convolutional neural network; s3, generating a target behavior trend curve and a target behavior prediction result by using the 3D convolutional neural network; s4, generating a target threshold curve by using an adaptive algorithm based on variable domain search, and obtaining behavior abnormality warning information; s5, generating comprehensive alarm information by using a method of combining fuzzy logic reasoning and a Bayesian network; s6, cross-camera data space-time correlation reasoning is carried out by using the graph neural network; s7, constructing an event severity curve by using a multivariable regression analysis and analytic hierarchy process, and generating a severity level and an event response measure of the event; and S8, pushing alarm information and event response measures to related personnel or equipment through the 5G message platform. The invention provides an efficient video monitoring method in combination with artificial intelligence and a 5G network.
Owner:JIANGSU YIQIYUN DIGITAL TECHNOLOGY CO LTD

Intelligent gastrointestinal tract postoperative intestinal obstruction diagnosis and intervention decision-making system and method thereof

The invention relates to the technical field of medical systems, in particular to an intelligent gastrointestinal postoperative intestinal obstruction diagnosis and intervention decision-making system and method, and the system comprises a data preprocessing module, a multi-modal data fusion module, a gastrointestinal function recovery model construction module, a self-adaptive progressive reinforcement learning optimization module AHRL and a diagnosis and intervention recommendation module. The self-adaptive progressive reinforcement learning optimization module optimizes an intervention strategy layer by layer through a five-layer progressive algorithm from the steps of dynamic adjustment of feature weights, recursive state propagation, layered accumulation optimization, symptom similarity matrix generation, progressive weight reverse feedback and the like, deep fusion of key symptoms by multi-modal data is achieved, and the accuracy of the key symptoms is improved. And the intervention scheme is updated in real time according to the patient state change. In the intervention decision recommendation module, the system generates personalized diagnosis suggestions based on the optimized strategy, carries out real-time risk assessment, and provides continuous and dynamic intervention support for the patient.
Owner:THE SECOND AFFILIATED HOSPITAL OF KUNMING MEDICAL UNIV (YUNNAN PROVINCIAL UROLOGY HOSPITAL YUNNAN PROVINCIAL HEPATOBILIARY & PANCREATIC SURGERY HOSPITAL)

Multi-modal recommendation method based on information bottleneck principle

The invention relates to the technical field of data processing, in particular to a multi-modal recommendation method based on an information bottleneck principle. The method comprises the following steps: constructing a data set # imgabs0 # according to a user-article historical interaction record in an Amazon data set; constructing a user-article graph for each mode according to the data set, and performing multi-layer graph convolution calculation on the embedding of the user and the article to obtain the final expression of the user and the fusion embedding of the article; an article-article similarity graph is constructed for each mode of the articles, neighbors of the first k articles with the highest similarity are reserved, article semantic signals are grouped according to the contribution degree of the article modes to downstream recommendation tasks, and final expressions of the articles are obtained; and performing optimization through denoising, loss calculation and an Adam optimizer. According to the method, key information can be extracted from multi-modal data, redundancy and noise can be removed, and the contribution degree of article characterization to downstream recommendation tasks in different modals can be calculated.
Owner:QILU UNIVERSITY OF TECHNOLOGY (SHANDONG ACADEMY OF SCIENCES) +1

Task planning method of intelligent world model with body based on visual reasoning

The invention provides a task planning method for an intelligent world model with a body based on visual reasoning, and the method combines the language processing capability of a large language model and an advanced visual perception technology, so that an intelligent agent can pre-reasone an environment state and simulate possible actions through the world model under the condition of obtaining a small amount of environment visual information; therefore, effective decision making and rapid adaptation to constantly changing environment states are realized in a complex environment, and the task execution efficiency and the task completion success rate of the intelligent agent in a complex scene are remarkably improved.
Owner:XIANGTAN UNIV +1

Integrated sensing module for dynamic weighing system

The invention relates to the technical field of measurement, in particular to an integrated sensing module for a dynamic weighing system, which comprises a physical domain, an environmental domain and a virtual domain, and realizes dynamic adjustment of parameters through cross-domain interaction so as to deal with high-frequency vibration and rapid temperature difference change in a complex environment. The physical domain is responsible for collecting multi-source data, the environment domain senses external interference in real time, and the virtual domain realizes time sequence feature extraction and dynamic model updating by combining the optimization capability of a recurrent neural network and reinforcement learning. According to the technical scheme, the real-time performance and adaptability of the system are remarkably improved, the environment change trend can be accurately predicted, the weighing strategy is adjusted in advance, and the error rate is controlled to be at an extremely low level. Meanwhile, by combining a negative feedback mechanism and objective function optimization, the sampling frequency is dynamically adjusted, and the energy consumption is reduced on the basis of guaranteeing the data processing efficiency. And deep fusion of wavelet decomposition, a dynamic normalization algorithm and a neural network model is adopted, so that the accuracy and stability of data processing are greatly enhanced.
Owner:HUNAN HAIDEWEI TECH

Multi-source heterogeneous instrument data fusion and anomaly detection method and system

The invention belongs to the technical field of graph structure data processing, and discloses a multi-source heterogeneous instrument data fusion and anomaly detection method and system. The method comprises the following steps: defining a heterogeneous instrument data network diagram, and representing data of different instruments as a diagram structure; and performing instrument node data similarity learning, representation learning of instrument time series data and anomaly detection, calculating the probability of anomaly detection by transforming an output vector of the graph attention network, and performing model training by adopting a cross entropy loss function. The method is used for detecting abnormal data of instruments and meters, can effectively capture complex relations between data, and has high data expression ability; according to the invention, deep fusion of multi-source heterogeneous data is realized; according to the method, the multi-source heterogeneous data can be automatically processed, the dependence on manual operation is reduced, and the accuracy and efficiency of anomaly detection are improved.
Owner:QINGDAO ZHONGYU ELECTRONIC TECHNOLOGY CO LTD

Digital planting GAP system and secondary node combined tracing system and method

The invention provides a digital planting GAP system and secondary node combined traceability system and method, solves the problems of digital planting traceability tracking and the like, and comprises that a digital planting system performs data transmission with a secondary node through a traceability system, and the traceability system comprises a production link, a processing link, a circulation link and a sales link. The method has the advantages of good traceability effect, high response rate and the like.
Owner:JIANGSU KANION PHARMA CO LTD

Multi-source heterogeneous data integration system, method and equipment and storage medium

The invention discloses a multi-source heterogeneous data integration system, method and device and a storage medium, and relates to the technical field of data processing, and the system comprises an information analysis module which is used for scanning each accessed data source through a target adapter so as to extract target information, analyzing the target information through a machine learning model, and obtaining an analysis result; the data source adaptation module is used for generating adaptation logic of each data source based on the analysis result and completing adaptation operation on the data sources according to the adaptation logic; the item set and rule determination module is used for cleaning and standardizing the data of each data source to obtain processed data, and determining a frequent item set and an association rule between the processed data based on an association rule mining algorithm; and the data association determination module is used for determining association of the processed data by utilizing the frequent item set, the association rule and the long-short-term memory network so as to analyze the processed data to complete data integration. According to the invention, intelligent integration of data is realized.
Owner:INSPUR YUNZHOU (SHANDONG) IND INTERNET CO LTD

Rail transit vehicle multi-monitoring system fusion integrated system

The invention provides a rail transit vehicle multi-monitoring system fusion integrated system, which comprises fusion hosts, fusion slaves and a switch module, and is characterized in that the fusion hosts are respectively arranged on a head train and a tail train and are used for carrying out data acquisition, data management and monitoring analysis on all front-end equipment of the whole train; the fusion slave is arranged in a middle vehicle and is connected with the fusion host through a gigabit Ethernet; and the switch modules are respectively configured in the fusion host and the fusion slave and are used for constructing a vehicle monitoring operation and maintenance backbone network. Platform and modular system design is mainly adopted, deep fusion of multiple monitoring systems of the vehicle is achieved, on the premise of multiple front-end devices, the number of detection hosts is reduced, the integration difficulty of the vehicle to the monitoring systems is reduced, data alignment of the multiple monitoring systems is achieved through the fusion mode, and the system reliability is improved. And multi-system data fusion and correlation analysis are realized.
Owner:CRRC DALIAN CO LTD

Two-wheel wheel core driving clutch disc integrated type self-adaptive automatic speed changing electric driving system

The invention discloses a two-wheel wheel core driving clutch disc integrated type self-adaptive automatic speed changing electric driving system. The system comprises a box body, a main shaft, a motor assembly, a speed reducing assembly, a frame clutch disc mechanism and a self-adaptive transmission sensing mechanism, wherein the frame clutch disc mechanism and the self-adaptive transmission sensing mechanism are mainly arranged in a rotor. Each clutch unit axially moves back and forth in the frame clutch disc mechanism in an elastic pressure-equalizing sorting mode under the combined acting force of traction force and driving resistance, and work tasks such as power transmission, non-power transmission, transmission direction change and power distribution are completed. The inner friction plates and the outer friction plates are alternately arranged, and each equidistant space is always kept in a pressure-equalizing sequence under the elastic action of the outer elastic rings and the inner elastic rings, so that when the movable pressing disc releases pressure, each transfer synchronous check ring synchronously pushes the inner friction plates at the first time, and each inner elastic ring and each outer elastic check ring synchronously generate elastic release; gaps are synchronously formed between all the outer friction plates and all the inner friction plates, and the abrasion resistance, stability and reliability of the friction clutch are improved.
Owner:CHONGQING ZHIZHU TRANSMISSION IND TECH RES INST CO LTD

Converter smelting equipment and method for recovering coal gas throughout the process

ActiveCN114807500BTo meet the gas demand of the rear processemission reductionManufacturing convertersExhaust gas handlingThermodynamicsFlue gas
The present invention discloses a converter smelting equipment and method for recovering coal gas in the whole process, which belongs to the field of iron and steel metallurgy. The equipment includes a converter furnace body, a flue gas evaporative cooling system, a dust removal system, a flue gas detection system and a gas holder connected in sequence, and also includes an oxygen-coal gun and a coal injection control system. The oxygen-coal gun is deep inside the converter furnace body. The oxygen-coal gun includes a gun body, and an oxygen gun nozzle and a coal powder nozzle are arranged at the bottom. The upper part of the oxygen-coal gun body is provided with a coal powder and carrier gas pipe, an oxygen pipe, a cooling water inlet pipe and an outlet pipe, which are respectively used for conveying coal powder, conveying oxygen, and cooling water inlet and outlet. The gun body is provided with a coal powder channel, an oxygen channel, a cooling water inlet channel and a cooling water outlet channel. The coal powder and carrier gas pipe are connected to the powder making and injection system. The comprehensive use of this process and equipment can realize the recovery of coal gas in the whole process, reduce CO emission, and the equipment runs smoothly, which greatly reduces the production cost.
Owner:SHENYANG DONGDASHANHUI ENVIRONMENTAL TECH CO LTD

Power module, circuit system and vehicle

The utility model discloses a power module, a circuit system and a vehicle, and the power module comprises a substrate, at least one surface of the substrate is provided with a conductive layer, and the conductive layer comprises a plurality of conductive regions which are arranged at intervals; the plurality of power chips are respectively arranged in the plurality of conductive regions, and the power chips comprise at least one first chip, at least one second chip, at least one third chip and at least one fourth chip; the first chip is electrically connected with a first conductive region in the plurality of conductive regions to form a first circuit; the second chip is electrically connected with a second conductive region in the plurality of conductive regions to form a second circuit; the third chip is electrically connected with a third conductive region in the plurality of conductive regions to form a third circuit, and the third circuit is used for supplying power to the high-voltage electric equipment; and the fourth chip is electrically connected with a fourth conductive region in the plurality of conductive regions to form a fourth circuit, and the fourth circuit is used for supplying power to the low-voltage electric equipment.
Owner:JINAN BYD SEMICON TECH CO LTD

Nondestructive testing method and system for deep compaction degree of large-thickness water-stable base

ActiveCN121856533BBreak the limitations of single dataComprehensive engineering scenario data supportData sourceStructural engineering
The present application relates to the field of engineering detection technology, in particular to a deep compaction degree nondestructive testing method and system for large-thickness water-stable base course. A multi-source heterogeneous dataset of a to-be-detected area is obtained using an integrated detection device; a construction process compliance index and a process uniformity index of the to-be-detected area are obtained; each data source in the multi-modal nondestructive testing data is deeply corrected to obtain a deep correction compaction degree estimation set of each data source; the deep correction compaction degree estimation set of each data source is input into an adaptive weight fusion model to obtain a preliminary compaction degree distribution map; a fusion compaction degree distribution map of the to-be-detected area and a dynamic confidence score of each spatial point are calculated; the compaction quality of the to-be-detected area is determined, and a detection strategy optimization instruction is generated for the area with a confidence score lower than a confidence threshold. The present application can realize high-precision, high-confidence and engineering-actual nondestructive testing of deep compaction degree of large-thickness water-stable base course.
Owner:YUNNAN CONSTR INVESTMENT PAVEMENT ENG CO LTD +2

A personnel trajectory drawing method and system based on an AI algorithm

The application discloses a personnel trajectory drawing method and system based on an AI algorithm, and the method comprises the following steps: collecting positioning device signals and monitoring video data; performing dynamic Kalman filtering denoising processing on the positioning device signals to obtain first trajectory data, and performing self-adaptive light compensation enhancement processing on the monitoring video data to obtain first video data; performing trajectory prediction compensation on a signal interruption section based on an LSTM neural network; extracting personnel motion posture features and scene semantic features; constructing a space-time correlation graph by using a graph convolutional neural network to obtain enhanced trajectory data; performing multi-scale trajectory reconstruction based on an improved Transformer architecture to generate final personnel trajectory information; and outputting a visual trajectory report containing trajectory precision evaluation, abnormal behavior marking and electronic fence triggering events. The application realizes deep fusion of multi-source heterogeneous data and accurate reconstruction of trajectories, and significantly improves the continuity and reliability of personnel trajectory drawing in complex scenes.
Owner:HENGFENG INFORMATION TECH CO LTD

Perception-aided predictive handover method in high-speed railway tunnel scenario

The perception-assisted predictive handover method in the high-speed railway tunnel scene of the application belongs to the technical field of wireless communication and intelligent transportation, and the steps include: constructing a bistatic communication and perception integrated architecture, establishing a communication and perception signal model considering double selective channels, deducing a lower bound of Cramer-Rao based on joint time delay-Doppler parameter estimation to establish a perception-assisted location prediction model, designing a three-state predictive handover mechanism based on perception information, and proposing a perception-assisted dual deep Q network algorithm, jointly optimizing handover decision, clamp antenna configuration and power allocation. By integrating perception capability into handover decision, the application realizes predictive switching before signal degradation, significantly improving the handover success rate and communication throughput in the high-speed railway tunnel scene.
Owner:JILIN UNIVERSITY

Automatic segmentation and recognition method and system for cancer image fused with multi-modal images

PendingCN122510567AReduce feature interferenceRealize deep integration
This invention relates to the field of medical image processing technology, specifically to an automatic segmentation and recognition method and system for cancer images based on the fusion of multimodal images. The method includes: acquiring a multimodal medical image set containing computed tomography (CT) images, magnetic resonance imaging (MRI) images, and positron emission tomography (PET) images; generating spatially aligned data blocks through three-dimensional spatial registration and voxel alignment; extracting fusion features using an improved encoder-decoder network optimized based on the characteristics and spatial dependencies of different modal channels; obtaining enhanced semantic feature maps through multi-level context-aware processing; identifying suspected cancer regions through a region candidate generation network; and simultaneously outputting pixel-level segmentation masks and cancer type probabilities through a segmentation network, ultimately generating a three-dimensional segmentation and recognition result. This method can accurately capture multimodal-specific features, clearly distinguish the boundaries between lesions and normal tissue, simplify the processing flow, reduce errors, and achieve accurate and efficient segmentation and recognition of cancer lesions.
Owner:SHIYAN CITY PEOPLES HOSPITAL (PEOPLES HOSPITAL AFFILIATED TO HUBEI UNIV OF MEDICINE)

Physical logic driven intelligent groundwater reserve prediction method and system

ActiveCN122088798AEnsure physical self-consistencyHigh precisionForecastingBiological modelsData setObservation data
The invention belongs to the technical field of groundwater reserve prediction, and particularly relates to a physical logic driven groundwater reserve intelligent prediction method and system. Comprising the following steps: performing complementation and physical verification on multi-source heterogeneous observation data related to groundwater reserves in a target area to obtain a physical verification complete data set conforming to physical logic; performing space-time decoupling causal contribution analysis on the physical verification complete data set to generate a causal contribution matrix; according to the causal contribution matrix, constructing and training a physical constraint emergence type space-time prediction model; applying a physical anchoring adversarial migration strategy to adapt the trained emergence type prediction model to the target new region to obtain a region adaptive prediction model; and inputting the future scene conditions into the regional adaptation prediction model for simulation, and generating a groundwater reserve prediction result. According to the method, intelligent prediction with high precision, high robustness and physical interpretability on groundwater reserves can be realized, and scientific decision support is provided for water resource management.
Owner:SHANDONG UNIV

Wind power blade multi-robot task scheduling method based on multi-layer stackelberg game model

PendingCN122713625ASolve the problem of poor process connectionRealize the craft
This invention discloses a multi-robot task scheduling method and apparatus for wind turbine blades based on a multi-level Stackelberg game model, relating to the field of wind turbine blade manufacturing technology. The method includes: constructing a Stackelberg game framework, assigning roles according to process dominance, with the edge-cutting robot as the first-level leader, the grinding robot as the second-level leader, and the roller-coating robot as the follower; the edge-cutting robot formulates an initial scheduling strategy based on an auction mechanism, and the grinding robot optimizes the strategy through a same-level auction, the combination of these two strategies forming the first-level game; the roller-coating robot determines the optimal solution based on the combined strategy of the first two, forming a second-level game with the grinding robot; a multi-level Stackelberg game model is established based on work efficiency, energy consumption, load balance, and conflict penalty terms; the game equilibrium is solved using backward induction, finally obtaining the optimal scheduling scheme for each robot. Using this invention can improve the overall work efficiency of wind turbine blade post-processing.
Owner:UNIV OF SCI & TECH BEIJING

Four-axis manual-control reversing clutch plate type self-adaptive automatic variable-speed electric driving system

PendingCN119037127AAdaptive mechanical shiftingMeet the requirements for efficient and energy-saving useElectric propulsion mountingControl devicesElectric machineryElectric drive
The invention discloses a four-shaft manual control reversing clutch sheet type self-adaptive automatic speed change electric driving system, which comprises a motor, a speed change assembly and a differential mechanism, not only realizes a reverse gear function, but also is quick in gear shifting response, reduces one layer of transmission in the radial direction, simplifies a transmission path and improves the response speed of gear shifting. The self-adaptive mechanical transmission has the advantage of self-adaptive mechanical transmission, switching and maintaining of a low-speed gear and a high-speed gear are achieved under the combined action of traction force and running resistance, high driving torque can be provided in a constant torque area, and high rotating speed can be provided in a constant power area; and low-speed, large-torque and high-efficiency requirements of various complex working conditions of vehicle acceleration, climbing and high-speed driving can be met. Each clutch unit axially moves back and forth in the frame clutch disc mechanism in an elastic pressure-equalizing sorting mode under the combined acting force of traction force and driving resistance, and work tasks such as power transmission, non-power transmission, transmission direction change and power distribution are completed.
Owner:CHONGQING ZHIZHU TRANSMISSION IND TECH RES INST CO LTD

A three-dimensional packing method based on a graph neural network, a medium and an equipment

The present application relates to the technical field of three-dimensional packing, and more particularly to a three-dimensional packing method based on a graph neural network, a medium and an equipment, which solves the constraint perception defects of relying only on the spatial geometric state by constructing a heterogeneous relationship graph containing multiple complex constraints such as mutual exclusion, pressure bearing and temperature control, and generating relationship perception embedding vectors and global graph embedding vectors by using a graph neural network coding; the deep fusion of space and constraint information is realized by simultaneously fusing container space occupation, global constraint profiles and personalized constraint features of goods through decision state representation; the latest constraint topology and space conditions are ensured for each packing action by cyclically executing the packing action and dynamically updating the state, so that a complete packing scheme meeting multiple constraints is finally output, and the compliance and space utilization of the packing scheme are improved.
Owner:TIANJIN ANXIN DIGITAL TECHNOLOGY CO LTD

Simulation test system and method for aircraft obstacle avoidance algorithm

The invention belongs to the technical field of aircraft control, and particularly relates to a simulation test system and method for an aircraft obstacle avoidance algorithm. The invention discloses a simulation test system and method for an aircraft obstacle avoidance algorithm. The system comprises a virtual environment modeling device, a dynamics resolving device, an obstacle avoidance logic processing device, a physiological feature modeling device, a stress parameter mapping device, a comprehensive evaluation feedback device and a data monitoring device. According to the method, kinematic parameters are mapped into physiological stress parameters such as gravity overload and motion sickness, and an obstacle avoidance track is fed back and corrected according to a physiological comfort threshold, so that an obstacle avoidance path conforms to the tolerance limit of a human body on the premise of ensuring safety. According to the method, deep fusion of control engineering and aeronautical medicine is realized, obstacle avoidance assessment extends to the field of physiological tolerance from physical safety, the scientificity of simulation assessment of the manned aircraft is improved, and the risk of personnel disability caused by violent maneuver in real flight is reduced.
Owner:SHANXI ZHENGHETIAN TECH CO LTD

Cloud edge-end cooperative intelligent control method and platform for chemical water treatment black light factory

The invention relates to the technical field of chemical water treatment, in particular to a cloud side-end cooperative intelligent control method and platform for a chemical water treatment black light factory, and the method comprises the steps: constructing a digital twin model fusing equipment topology and operation parameters, and simulating and optimizing the optimal operation state of a system in an unattended and multi-unit cooperative scene; the edge sensing terminal is used for collecting water quality, energy consumption and equipment operation parameters of each process unit under dynamic working condition disturbance in real time, and the actual operation state of the system is accurately recognized through feature extraction and data fusion analysis; by comparing actual and optimal operation states, extracting deviation parameters, performing multi-dimensional analysis, identifying an operation instability mode and key influence factors, generating a self-adaptive cloud side-end cooperative control strategy and driving an intelligent execution device to realize closed-loop regulation, the response capability, the operation stability and the energy efficiency level of the system to complex disturbance are improved, and the energy efficiency of the system is improved. And unmanned, intelligent and collaborative optimization operation of the chemical water treatment process is effectively supported.
Owner:NAT ENERGY CHANGYUAN HANCHUAN POWER GENERATION CO LTD +1

Pathological section image segmentation method and device based on multi-modal supervised contrast learning

PendingCN122289169ARealize deep integrationOvercome the problem of insufficient feature abstractionPattern recognitionMedicine
This invention belongs to the interdisciplinary field of medical image analysis and deep learning, and discloses a method and apparatus for pathological slide image segmentation using multimodal supervised contrastive learning. The method generates biological prior feature maps through decoupling, extracts visual and semantic features by combining the master encoder and prior encoder, and achieves feature alignment using a cross-modal attention mechanism. Supervised contrastive learning is introduced to enhance intra-class compactness and inter-class separability. A multi-scale decoupled decoder is designed, with structural similarity loss and boundary-aware loss respectively supervising structural and detail branches. Finally, dynamic weighted fusion and pseudo-label calibration strategies are used to optimize segmentation performance, significantly improving the accuracy and robustness of pathological image segmentation.
Owner:XINYI CITY PEOPLES HOSPITAL

An ai service collaborative optimization method and system based on zero-cost agent

This invention relates to a method and system for collaborative optimization of AI services based on zero-cost agents. The method includes: acquiring the model structural features of the AI ​​service to be evaluated; using a pre-set non-trained evaluation model to quantitatively analyze the information transmission characteristics of the model structural features to obtain performance scores for each candidate model version; based on the performance scores and combined with the real-time status of computing resources, determining the deployment strategy of the AI ​​service instance on the target computing node through a deployment algorithm that balances queuing latency benefits and communication latency benefits; monitoring the service request load in real time; selecting the target version from the candidate model versions based on load changes and the performance scores; and performing online replacement of the deployed AI service. This invention, through a proxy model constructed with quantitative encoding, communication-queuing collaborative awareness deployment, and load awareness model replacement, realizes a link collaboration mechanism, which can effectively reduce end-to-end latency and improve resource utilization efficiency and service stability.
Owner:HUBEI DATA GROUP CO LTD +1

Multifunctional in-memory computing circuit and chip based on 9T1C-SRAM cell

This invention relates to an in-memory computing circuit in the field of integrated circuits, and particularly to a multifunctional in-memory computing circuit and chip based on a 9T1C-SRAM cell. The multifunctional in-memory computing circuit includes five NMOS transistors N1-N5, four PMOS transistors P1-P4, and a capacitor C0. P1, P2, N1, and N2 form a latch structure; N3 controls row-direction read / write operations via word line WL and bit line BL; P3, P4, N4, and N5 are used to implement Boolean logic write-back, row-direction read / write, and multiply-accumulate functions. This invention's multifunctional in-memory computing circuit simultaneously supports Boolean logic write-back, signed multiply-accumulate (MAC) operations, and bidirectional row and column read / write, achieving deep integration of logical operations, arithmetic operations, and data movement within the same memory cell.
Owner:ANHUI UNIV

Supply chain multimodal transport intelligent data system and method

PendingCN121961369AImprove risk management and control capabilitiesAchieve unified convergenceAlarmsFeature extractionData view
The invention provides a supply chain multimodal transport intelligent data system and method, and belongs to the technical field of supply chain management. The invention provides a supply chain multimodal transport intelligent data system. The system comprises a unified data base module, a core calculation engine module, an intelligent decision engine module, a hierarchical alarm response module and a closed-loop optimization module. The unified data base module is used for accessing multi-source heterogeneous data and cleaning, standardizing and fusing the multi-source heterogeneous data to form a global unified data view; the core calculation engine module is used for performing data processing and feature extraction on the global unified data view; the intelligent decision engine module is used for carrying out prediction analysis on the extracted feature data and generating a prediction result and a risk score; the grading alarm response module is used for comparing the risk score with a preset threshold value, triggering an alarm and executing grading response; and the closed-loop optimization module is used for returning feedback data generated by execution and intervention to the unified data base module.
Owner:贵州诚睿信数智科技有限公司

Power terminal data authentication method and system based on network identification dynamic binding

This invention provides a power terminal data authentication method and system based on dynamic binding of network identifiers. First, it acquires a set of terminal operation and associated network topology data of the target power terminal within a preset time window. Dynamic network identifier generation parameters, including characteristics such as terminal state fluctuations, are generated from the terminal operation data. Based on these dynamic network identifier generation parameters and the network topology data, dynamic network identifier allocation is performed, generating a dynamic network identifier sequence associated with a sub-time period. This dynamic network identifier sequence is dynamically bound to the terminal operation data, generating a power terminal data set with time-series authentication tags. The tags are used to verify data integrity and source legitimacy. The tagged data is dynamically authenticated and matched according to preset rules, generating authentication results and policy adjustment parameters to adjust the generation parameter logic. Finally, the generation parameters are updated based on the results and parameters and synchronized to the terminal, triggering the data authentication iteration for the next time window.
Owner:STATE GRID SHANDONG ELECTRIC POWER CO

Geological disaster slope reinforcing structure with high-stability load

The invention discloses a geological disaster slope reinforcing structure with a high-stability load, and belongs to the technical field of geological disaster protection. The structure comprises a protective net arranged at the slope bottom, and a plurality of cement beads, fixed pulley assemblies and slope top fixing columns which are arranged between the protective net and the slope top; the cement ribs are mounted through the elastic anchoring assemblies and can generate elastic displacement along the slope surface when being pulled; the suspended rope is connected with the top of the protective net, bypasses the fixed pulley and is fixed on the cement rib; the buried rope is connected with the top of the protective net, penetrates through the bottom of the cement rib from the lower portion and is finally connected to the fixing column. When the protective net bears loads and tends to turn over, the buried ropes are tensioned and drive the cement protruding edges to move, the force direction is converted through the fixed pulleys, the suspended ropes are promoted to generate reverse tension, and therefore the rollover force is counteracted. Dynamic dispersion and feedback compensation of the load on the whole slope are achieved through linkage cooperation of micro-displacement of the cement protruding edges and the rope system.
Owner:JIANGXI SHANGYE CONSTRUCTION ENGINEERING CO LTD