459 results about "Tensor decomposition" patented technology

The invention provides a self-adaptive network topology dynamic reconstruction method and system based on deep reinforcement learning, and relates to the technical field of deep reinforcement learning, and the method comprises the steps: obtaining the topology state information, service flow distribution information and historical reconstruction records of a current network; extracting topological correlation characteristics among nodes through graph convolution operation, and generating fusion state representation in combination with service flow information; inputting the fusion state representation into a deep reinforcement learning model to identify bottleneck nodes and redundant links, and outputting a reconstruction action candidate set; searching and evaluating the long-term cumulative income of the candidate actions through a Monte Carlo tree, and screening an optimal reconstruction action sequence; a graph coloring algorithm is utilized to allocate time slots and process resource conflicts, and a resource-feasible topology adjustment scheme is generated; and extracting a network evolution rule through tensor decomposition, and constructing a topological optimization association mapping graph. According to the method, the network bottleneck can be intelligently identified, the network topology structure is dynamically optimized, and the network performance and the resource utilization rate are effectively improved.

The invention discloses a precise interference avoidance method and device in a radio system, and relates to the field of signal processing, and the method comprises the steps: constructing a sensing matrix through sensing node data, and capturing a transient interference signal through aperiodic scanning; performing tensor decomposition on the signal data to extract time domain, frequency domain, space domain and modulation domain features, constructing a dual-mode spectrum analysis model, reconstructing an instantaneous spectrogram by using compressed sensing, and predicting an interference mode through LSTM; after the instantaneous spectrogram and the predicted interference graph are fused, threat assessment is carried out through a multi-stage interference classification model; according to the interference category and the threat level, beam forming is optimized, adaptive null is generated, a power density optimization model is constructed, and the transmitting power is dynamically adjusted; an anti-interference frequency hopping sequence is generated based on a chaotic mapping algorithm, and spectrum camouflage and tracking interference resistance are realized. The method has the advantages that accurate identification and dynamic avoidance of interference are realized through multi-dimensional perception, intelligent prediction and adaptive beam forming, and the interference avoidance capability of a wireless system is improved.

The invention discloses a multi-target task and resource intelligent modeling method, particularly relates to the field of complex adversarial simulation, is used for solving the problems of dynamic constraint optimization and robustness improvement in multi-dimensional task planning, and aims at realizing multi-dimensional coupling of space-time resource parameters by constructing a three-dimensional hypergraph model, mining a parameter association rule by means of tensor decomposition, and realizing multi-dimensional optimization of the space-time resource parameters. Dynamic constraint quantization is supported, multi-dimensional index priority evaluation is fused in a constraint layered injection stage, hard constraints are recognized, a solution domain is compressed, a hybrid optimization strategy regulates and controls balance between global exploration and local optimization, and after annealing is simulated to jump out of a local extreme value, a multi-target particle swarm algorithm is used for screening a space-time resource equilibrium solution in a trimming solution domain. Digital twinborn verification promotes physical and virtual space interaction data closed loop, a parameter correlation degree matrix is corrected, scheme robustness is enhanced, efficient generation and adaptive optimization of a task planning scheme under complex constraints are realized, and system stability and multi-target cooperation capability under sudden disturbance are improved.

The invention relates to the field of equipment state monitoring in industrial Internet of Things, and discloses an equipment state monitoring, analysis and evaluation method and system based on big data, and the method comprises the steps: carrying out the adaptation of a multi-source heterogeneous data protocol, and carrying out the cleaning of a dynamic mask, and generating a standardized data stream; constructing a dynamic hypergraph of an embedded constraint equation based on physical topology; combining incremental tensor decomposition with manifold constraint to update a core tensor; abnormal association is positioned based on singular value distribution and a hyperedge propagation algorithm; cross-equipment model migration is realized through topological optimal transmission and knowledge distillation, and a target equipment evaluation model is generated; the system comprises a data preprocessing module, a hypergraph modeling module, a tensor analysis module, a state evaluation module, a transfer learning module and a dynamic tuning module. According to the method, through multi-source data dynamic cleaning, physical constraint hypergraph modeling, incremental tensor decomposition and manifold constraint, and in combination with an abnormal positioning closed loop and cross-equipment topology migration, equipment state monitoring and rapid model adaptation are realized.

The invention discloses a concrete strength remote monitoring method and system suitable for a complex environment, and belongs to the technical field of civil engineering structure health monitoring. The remote monitoring method comprises the following steps: step 1, acquiring performance data of a concrete structure and related environmental factor data, and transmitting multi-source heterogeneous data to a data processing center in real time through a preset wireless communication protocol; step 2, constructing a five-dimensional tensor data structure, and realizing accurate mathematical expression of a complex coupling relationship between environmental factors and material characteristics through tensor decomposition; step 3, capturing nonlinear time-varying characteristics of concrete strength evolution; 4, quantifying the age effect through an intensity development rate index; step 5, based on the intensity development rate change trend, adaptively adjusting the data sampling frequency and monitoring the environmental condition fluctuation; and step 6, evaluating the safety state of the concrete structure in real time, and ensuring safe and reliable operation of the concrete structure in a complex environment.

The invention discloses a big data-driven hydrogen fluoride purification risk entropy assessment and multi-parameter monitoring system, which relates to the technical field of chemical process monitoring, comprises a multi-mode heterogeneous sensor network, and realizes high-precision detection and three-dimensional monitoring by using a nanowire FET (Field Effect Transistor) and the like. The space-time tensor decomposition module extracts features and identifies a causal relationship; the risk entropy manifold learning module evaluates the risk; the self-adaptive twin network monitors the abnormity and simulates and disposes the abnormity; and a knowledge graph-reinforcement learning system decision making subsystem and a plurality of subsystems such as an optical monitoring subsystem and a multi-scale modeling subsystem are also provided, so that whole-process risk management and control are realized. The hydrogen fluoride purification monitoring level is greatly improved, the detection sensitivity and the anomaly detection accuracy are remarkably improved, risk early warning is more timely, decision response is accelerated, the production efficiency is improved, energy consumption is reduced, the system has self-repairing and self-power-supply capabilities, data are safe and traceable, reliable operation of the system is guaranteed, and economic losses and potential safety hazards are reduced.

The invention provides a chronic disease risk assessment and intervention strategy generation system based on data analysis. According to the system, multi-source heterogeneous information including clinical examination, behavior records, environment data and the like is collected, key features are extracted through a data fusion technology, and time and space features of data are enhanced through a space-time weighted tensor decomposition method. And in combination with a causal reasoning technology, the system can accurately evaluate the chronic disease risk of an individual, eliminate confounding factors and provide more reliable risk prediction. In addition, the system dynamically generates a personalized intervention strategy through a reinforcement learning algorithm, adjusts intervention measures according to real-time health data, and ensures accurate chronic disease management. The method has an efficient risk prediction capability and a personalized intervention scheme, and is helpful for improving the accuracy and effect of chronic disease management.

The invention provides an ocean wind field prediction method based on a neural network, and belongs to the technical field of ocean wind field prediction.The method comprises the steps that sparse ocean observation data are collected, a spatial covariance matrix is established, the spatial covariance matrix is converted into a graph structure, and then multi-hop neighborhood feature aggregation is conducted through a graph convolutional network; a tensor decomposition algorithm is combined for modeling high-order feature interaction to generate a gridding wind field, a bidirectional long-short-term memory network encoder is used for extracting space-time invariant features, a multi-layer perceptron predictor is used for directly mapping a future multi-step wind field, and a course learning strategy and a Shenchang differential equation boundary layer are matched for correction. The technical problem that sparse ocean observation data are difficult to accurately reconstruct into a high-resolution gridding wind field is solved.

The invention provides a dynamic scene reconstruction method and system based on spatial decomposition and Gaussian splashing, and belongs to the technical field of computer vision and graphics. The method comprises the following steps: decomposing a dynamic scene into a static standard space and a dynamic playground; reconstructing the standard space by using three-dimensional Gaussian splashing to obtain Gaussian primitives of the standard space; decomposing the characteristics of the motion field into standard space characteristics and a plurality of motion subspace characteristics based on tensor decomposition; fusing the standard space feature and the multiple motion subspace features through a motion information decoder, and decoding motion information of Gaussian primitives in the standard space; and applying the motion information to Gaussian primitives of a standard space, generating any moment representation of a dynamic scene, and rendering a target view angle image through a snowball throwing method to realize reconstruction of the dynamic scene. Through the method, the reconstruction precision and the rendering quality are effectively improved, and the real-time rendering capability is kept.

The invention relates to the technical field of pumped storage, and discloses a state detection method for a water energy storage unit, which comprises the following steps: S1, deploying vibration, temperature, electrical and pressure sensors on the water energy storage unit, collecting multi-dimensional operation data, and carrying out noise reduction, standardization and time sequence alignment; s2, multi-dimensional features of the data are extracted through tensor decomposition, local space information is acquired by using a convolutional neural network, and a time sequence relation is modeled in combination with a Transform model; and S3, constructing a sensor topological structure through the graph convolutional network. The multi-source data fusion technology is adopted, data of vibration, temperature, electrical and pressure sensors are processed in a unified mode, time sequence prediction and an intelligent fault detection algorithm are combined, comprehensive health monitoring of the water energy storage unit is achieved, the detection precision is improved, meanwhile, potential faults can be found in time, and compared with data monitoring of a single sensor in a traditional method, the method has the advantage that the monitoring accuracy is improved. And the problem of inaccurate identification caused by large and complex data volume is obviously solved.

The invention discloses a self-adaptive test parameter optimization method, and relates to the technical field of parameter optimization, and the method comprises the steps: collecting an original sensing signal in a mechanical test system, and carrying out the preprocessing of the original sensing signal; based on the preprocessed data set, constructing a four-dimensional space-time tensor, and executing improved parallel factor tensor decomposition to obtain a decoupled core factor and space-time feature component matrix; performing cross-modal alignment on the decoupled core factor and the time-space feature component matrix through a wear feature channel and an acoustic emission feature channel to obtain a fused cross-modal feature vector; performing crack growth rate prediction on the fused cross-modal feature vectors to obtain a crack risk level, and adjusting a strategy through dynamic parameters to obtain an optimized parameter set; the problem of signal noise and time mismatch is solved through multi-mode signal preprocessing, and high signal-to-noise ratio input is provided for subsequent analysis in combination with wavelet denoising, space-time alignment and double-domain feature extraction.

The invention relates to the technical field of data processing, in particular to a confidential propaganda and education system based on multi-modal interaction, which comprises a data acquisition module, a data processing module, a training module and an optimization module. The data collection module collects user eye movement tracks, voice instructions, operation behaviors and physiological signal data, noise is eliminated through time synchronization and self-adaptive filtering, and then a three-dimensional data cube is constructed through a tensor decomposition algorithm. And reversely adjusting knowledge graph nodes and scene parameters by using a genetic algorithm to form a closed-loop feedback link. And the cross-module cooperation unit aggregates a global cognitive feature model through a federated learning framework, and realizes privacy protection and resource dynamic allocation in combination with differential privacy protection and an ant colony optimization algorithm. Through multi-modal data fusion, dynamic strategy generation and closed-loop evaluation optimization, the problem of unidirectional interaction and feedback lag of a traditional system is solved, and user confidential knowledge internalization efficiency and complex scene coping capacity are improved.

The invention discloses a control method and system of a multi-source intelligent power manager based on 5G communication, and the method comprises the steps: obtaining multi-source heterogeneous power data of the multi-source intelligent power manager, constructing a dynamic energy topology through employing a multi-scale pulse fusion tensor decomposition algorithm, and generating a power state joint tensor with time-space alignment; inputting the power supply state joint tensor into a physically constrained adversarial prediction network, and outputting an energy dynamic balance vector including power supply margin prediction; performing multi-target dynamic game optimization on the energy dynamic balance vector, and outputting an anti-interference multi-source cooperative power supply strategy matrix; and inputting the multi-source cooperative power supply strategy matrix into a distributed cooperative control framework driven by edge computing, and finally outputting a multi-source intelligent control instruction set meeting low delay and high reliability. According to the embodiment of the invention, the flexibility and response speed of power management can be improved, and the stability and reliability of power supply are ensured.

The invention relates to the technical field of power data fusion, in particular to a data fusion mining method and system based on multi-modal power cross-domain. The method comprises the following steps: acquiring a multi-source heterogeneous power data set; performing tensor decomposition on the multi-source heterogeneous power data set to obtain a power core feature tensor set; performing feature selection and reconstruction on the power core feature tensor set to obtain a reconstructed power feature data set; performing domain adaptive feature mapping on the reconstructed power feature data set to obtain a cross-domain power feature mapping data set; performing domain difference elimination on the cross-domain power feature mapping data set to obtain a domain alignment power feature data set; performing attention weight calculation on the domain alignment power feature data set to obtain a cross-domain fusion power feature data set; according to the method, the utilization efficiency of multi-source data in the power system can be remarkably improved.

The invention discloses an on-load tap-changer vibration fault diagnosis algorithm based on tensor feature and adaptive weighted Stacking integration, relates to the technical field of on-load tap-changer fault diagnosis, and is used for improving the fault diagnosis precision. Comprising the following steps: S1, data acquisition; s2, feature extraction; the method comprises the following steps: extracting multi-scale time-frequency characteristics of an on-load tap-changer vibration signal by using wavelet scattering transform WST, and realizing low-rank decomposition and dimensionality reduction characterization of high-dimensional characteristics by combining a non-negative tensor decomposition model NTF; s3, fault diagnosis; a multi-base learner Stacking integration framework is adopted, and a prediction matrix is generated through K-fold cross validation; through a swarm intelligent optimization algorithm SRA, hyper-parameters and fusion weights of all base learners are adjusted, L2 regularization suppression over-fitting is introduced, and finally fault classification is realized by adopting a logic regression element learner with Softmax cross entropy. According to the invention, through fault diagnosis of multi-model adaptive fusion and optimization, the fault identification precision, stability and on-line monitoring capability are improved.

The invention discloses a remote proving method for a trusted data space, which relates to the technical field of computer and data security, and comprises the following steps: initializing a trusted environment and generating a dynamic identity key for a heterogeneous terminal containing a CPU (Central Processing Unit), a GPU (Graphic Processing Unit) and an FPGA (Field Programmable Gate Array) based on a hardware security module and a physical unclonable function, and generating a challenge value by a verifier by using a quantum random number; the method comprises the following steps of: firstly, transmitting to a proving party through quantum encryption and a neuromorphic photonic network, constructing proving by the proving party by adopting a tensor decomposition zero-knowledge proving protocol, finishing multi-stage verification on a verification party in combination with a space-time cause and effect graph and quantum signature aggregation, and finally realizing dynamic trust evaluation and adaptive strategy adjustment through a quantum Bayesian network, reinforcement learning and biological feedback. According to the method, the remote attestation performance of the trusted data space is improved, dynamic trust evaluation is realized by means of the quantum Bayesian network and biological feedback, attacks are effectively resisted, real-time requirements are met, trust is accurately evaluated, and the development of the trusted data space is promoted.

The invention relates to the technical field of hoisting equipment monitoring, and discloses a hoisting equipment digital real-time monitoring system based on the Internet of Things. A dynamic load analysis module of the system collects multi-dimensional operation parameters in real time through distributed edge computing nodes; the risk situation assessment module executes tensor decomposition operation on the parameters, extracts feature vectors and generates a three-dimensional risk map; the self-adaptive safety control module dynamically adjusts the working state of the equipment according to the risk map; the digital twin mapping module is used for realizing time-space synchronous mapping of real-time parameters and a three-dimensional model and outputting holographic running state projection; and the cloud collaborative diagnosis module fuses the historical fault case library to generate a preventive maintenance strategy and returns the preventive maintenance strategy. The system can comprehensively monitor the equipment state, accurately assess the risk, realize dynamic safety control and preventive maintenance, and improve the safety and reliability of the operation of the hoisting equipment.

The invention belongs to the technical field of smart city safety management, and discloses a smart park safety shunting control method, which comprises the steps of collecting data through a multi-modal sensor network to construct a space-time tensor, predicting people flow distribution through tensor decomposition and calculating a residual error, generating a path strategy by adopting a multi-agent game, and performing intelligent park safety shunting control. A shunting instruction is dynamically adjusted in combination with Lyapunov optimization, and closed-loop control is realized by cooperatively updating model parameters through residual feedback; the system comprises a multi-modal sensor network module, a spatio-temporal data fusion module, a tensor decomposition and prediction module, a multi-agent game planning module, a Lyapunov optimization control module and a parameter collaborative updating module. According to the method, the space-time tensor is constructed through multi-modal sensing and data fusion, people flow prediction is realized in combination with tensor decomposition, individual and system targets are balanced and coordinated by using a game, and based on a Lyapunov dynamic optimization shunting strategy, parameters are adaptively adjusted through a residual feedback closed loop, so that the stability of park safety management and control is improved.

The invention relates to the technical field of display screen detection, and discloses a liquid crystal display screen color difference detection system and method.The system comprises a multispectral imaging module, a curved surface deformation modeling module, a coupling tensor decomposition module, a biological excitation feature extraction module, a dynamic compensation control module and a technological parameter interface module; the method comprises the following steps of synchronously collecting a multispectral image and screen deformation data, constructing and decomposing a multidimensional tensor containing geometric constraints, extracting biological excitation visual features, and generating and applying voltage compensation through model prediction control so as to dynamically correct the color difference of the flexible screen. According to the method, through multispectral and geometry synchronous perception, multidimensional tensor constraint decomposition, biological excitation and deformable convolution feature extraction, time-varying model prediction control and dynamic weight adjustment of fused manufacturing parameters, high-precision capture and personalized compensation of the dynamic chromatic aberration of the flexible screen are realized. The defects of dynamic perception, quick response and individual adaptability are overcome.

The invention relates to the field of computer vision and deep learning, and discloses an image big data classification and recognition method and system based on deep learning, and the method comprises the steps: generating a gating matrix through the extraction of an image frequency domain energy coefficient, compressing a convolution kernel through the combination of asymmetric tensor decomposition, and carrying out the self-adaptive training through cross-modal semantic alignment and a meta-learning task. Efficient classification reasoning of dynamic path selection is realized, and the precision and the calculation efficiency are improved; the system comprises a frequency domain analysis module, a dynamic sparse gating module, an asymmetric tensor decomposition module, a meta-learning task generation module, a cross-modal alignment module and a dynamic inference engine module. According to the method, through cross-modal semantic alignment and meta-learning task optimization, in combination with lightweight parameter storage and edge calculation path selection, fine-grained classification precision improvement, model compression and high-efficiency reasoning are realized, and the calculation efficiency and generalization ability in a complex scene are remarkably enhanced.

The invention provides a spatio-temporal data fusion-based power distribution network load prediction method and system, and relates to the technical field of power distribution network load prediction, and the method comprises the steps: collecting related data of a power distribution network, carrying out the wavelet transform decomposition of historical load data, obtaining a load feature matrix, constructing a hierarchical graph convolution network based on topological structure data, and extracting topological correlation features; and generating a spatial-temporal feature tensor through tensor decomposition fusion, training a depth probability prediction model adopting a variational auto-encoder structure, and adjusting prediction probability distribution in combination with environmental data. According to the method, the prediction precision is improved, a complex space-time dependency relationship can be captured, and reliable uncertainty quantization is provided.

The invention provides a multi-modal data situation intelligent arrangement system and method, and relates to the technical field of data arrangement. The system comprises a dynamic heterogeneous data fusion module, an intelligent situation arrangement module and an incremental situation analysis module. And the dynamic heterogeneous data fusion module maps the multi-modal data flow into a hypergraph structure with dynamic weight through a semantic hypergraph fusion method of spatio-temporal context perception. And the situation intelligent arrangement module generates an evolution path of the cross-modal situation according to the node association strength and the semantic similarity in the hypergraph structure. And the incremental situation analysis module performs multi-granularity situation prediction and anomaly detection on an evolution path through hypergraph tensor decomposition and an attention mechanism, and dynamically optimizes an arrangement strategy. The method can be used for rapidly generating a large-screen situation display page of a data track and a situation thermodynamic diagram, high-speed situation arrangement and updating are achieved, and visual and dynamic support is provided for commanding and decision making.

The invention relates to the field of training simulation, and discloses a public security AI-driven data generation type investigation teaching practical training system, which comprises the following modules: a case dynamic generation module, which is used for receiving original data of a public security case database, generating a logically closed virtual case through tensor decomposition, and sending the virtual case to the public security case database; transmitting the generated case data to a multi-modal virtual-real fusion training module; and the multi-modal virtual-real fusion training module is used for receiving the virtual case data, realizing interactive training of multi-modal clues through a mixed reality technology, dynamically injecting interference clues, and transmitting student operation data to the causal reinforcement learning evaluation module. Through the collaborative technical scheme of tensor decomposition and nonlinear constraint optimization, the integrity and space-time consistency generation of the virtual case evidence chain is realized, and compared with a traditional case generation method based on a fixed template, the problem of contradiction between material evidence association fracture and behavior logic is effectively solved.

The invention provides an intelligent collaborative innovation incubation and campus security fusion management platform, which relates to the technical field of fusion management and comprises a multi-modal data acquisition and fusion system, an intelligent decision and strategy optimization engine, an edge calculation and security communication architecture, a scenarized analysis and prediction module and a security assurance and model adaptation system. The multi-modal data acquisition and fusion system comprehensively collects multi-source data of a campus environment, personnel behaviors and equipment states through a multi-sensor array, realizes efficient fusion of different modal features by using a feature fusion algorithm based on tensor decomposition, provides a comprehensive and refined data basis for subsequent analysis, and improves the accuracy of data fusion. And the intelligent decision-making and strategy optimization engine learns and generates an optimal decision-making strategy for various scenes of campus safety management by virtue of a plurality of advanced algorithms such as a consensus double-Q network algorithm, an artificial fish swarm optimization algorithm based on Levy flight, a particle swarm optimization algorithm based on quantum behaviors and the like.

The invention provides an infrared small target detection method and system based on depth-guided low-rank sparse decomposition, and relates to the technical field of remote sensing image processing, and the method comprises the steps: obtaining all original infrared images shot by remote sensing equipment, and sequentially stacking the original infrared images according to an obtaining time sequence, and obtaining an infrared original tensor; performing low-rank background and sparse target decomposition processing based on the infrared original tensor to obtain a low-rank sparse tensor decomposition model; a low-rank background tensor containing nonlinear transformation is obtained through processing of a constructed hierarchical nonlinear tensor ring background module; processing through a sparse target module fused with an attention mechanism to obtain a sparse feature tensor of the infrared small target area; and reconstructing a low-rank sparse tensor decomposition model guided by the deep neural network, and carrying out solving processing to obtain a final infrared small target detection result. According to the invention, accurate, robust and rapid detection can be carried out on a small target under a complex background.

The invention belongs to the technical field of software development and testing, and discloses an artificial intelligence-based software development automatic test case generation system, which is characterized in that an immune heuristic case self-repairing module is adopted, defects are regarded as antigens, antibody cases capable of being self-updated are generated by using a clone selection algorithm, and a gene rearrangement mechanism is automatically triggered when an interface is changed, so that the test efficiency is improved. The details of the use case are adjusted while the core detection logic is reserved; compared with a traditional method, the mechanism can realize use case dynamic adaptation without manual intervention, the maintenance workload is remarkably reduced, and the method is particularly suitable for a complex software system with frequent iteration; the space-time coupling test scene generation engine fuses dynamic scenes such as interaction and state transition of a module and short-time operation after precise coverage login by using a space-time convolutional network; the cross-dimension holographic use case synthesis module integrates multi-source data such as codes, hardware and user behaviors through tensor decomposition to generate a composite use case; functions and performance of software in a complex scene can be comprehensively verified, and test blind areas are remarkably reduced.

The invention relates to the technical field of enterprise data analysis, and discloses an enterprise data clustering processing method and system based on NLP and machine learning, and the method comprises the steps: S1, obtaining structured data and unstructured text data of an enterprise; s2, carrying out standardization processing on the structured data, extracting key business indexes, and forming structured feature vectors; the system comprises a data acquisition and preprocessing module, a multi-modal feature construction module, a causal semantic alignment module, a tensor decomposition module, a graph modeling module, a clustering module and an anti-fact reasoning module. The comprehensiveness and accuracy of enterprise behavior analysis are improved through multi-modal data fusion, a graph neural network and a causal reasoning technology; modeling by utilizing a dynamic graph and a causal relationship, and deeply mining an enterprise transaction relationship; and more accurate enterprise risk early warning is realized through anti-fact analysis and a risk scoring mechanism.

The invention provides a multi-modal data feature alignment and optimization system and method based on an FPGA, and the system comprises an input buffer module which is used for receiving multi-modal data, carrying out the data processing, and optimizing the buffer depth configuration based on the modal number and the data rate; the space-time perception scheduler is used for extracting space-time characteristics of the multi-modal data and fusing the space-time characteristics according to dynamic weights; the self-adaptive matching kernel is used for mapping the multi-modal feature codes to a unified semantic space and ensuring semantic consistency; the reconfigurable feature compression unit is used for dynamically adjusting a compression ratio through tensor CP decomposition and controlling a reconstruction error based on a rank parameter; a self-calibration optimization mechanism is adopted, feature alignment errors are monitored in real time, mapping matrix parameters are dynamically adjusted through a gradient descent algorithm, and calibration frequencies are switched according to data distribution changes. Through the FPGA pipeline architecture and self-adaptive resource allocation, the calculation delay is reduced, and the throughput is improved.

The invention relates to the technical field of intelligent construction site safety management and Internet of Things, and discloses a construction personnel safety protection system and method for intelligent construction site safety management, and the system comprises a data collection module, a four-dimensional space-time tensor construction module, an incremental tensor decomposition module, a dynamic risk field modeling module, and a federated learning parameter updating module. A multistage safety response module; the method comprises the steps of collecting personnel data through intelligent equipment and constructing a four-dimensional tensor, extracting core features through dynamic decomposition to establish a risk propagation model, optimizing parameters in combination with federal learning, and finally triggering three-level response according to a risk field gradient to realize full-process closed-loop management and control of construction safety. According to the method, multi-modal data are fused to construct a four-dimensional tensor, incremental decomposition is carried out to extract dynamic features, a physical field model is coupled to predict risks, federated learning is used for collaborative privacy optimization, hierarchical response is carried out for closed-loop linkage control, and construction safety and precise protection and efficient emergency are realized.

The invention relates to the technical field of data compression and storage, and discloses a monitoring data compression and storage method driven by a time sequence database, comprising the following steps: S1, preprocessing multi-source monitoring data, and constructing a dynamic tensor comprising a timestamp, a numerical index and a multi-dimensional label; s2, performing dynamic dimension reduction processing on the dynamic tensor to generate a core tensor and a multi-dimensional factor matrix; s3, based on the core tensor and the multi-dimensional factor matrix, determining a compression parameter, a decompression parallelism degree and an index granularity through a joint optimization model; and S4, performing hierarchical coding on residual data generated by the dynamic dimension reduction processing to generate a lightweight residual coding result. Through a dynamic tensor decomposition and incremental updating technology, low storage overhead and real-time dimension expansion capability of streaming monitoring data are realized, the problems of calculation redundancy and storage expansion caused by the fact that the streaming monitoring data cannot adapt to dynamic newly-added tags are solved, and meanwhile, frequent reconstruction cost caused by data dynamic expansion is avoided.