124 results about "Network formation" patented technology

A method for quality inspection of an etching paste material based on a spectral response difference, including: identifying a position of density mutation boundary through spectral collection and density gradient analysis. Applying a specific frequency beam to obtain the resonance response parameter, generating stress field reconstruction data. Performing a reverse optical tracking to obtain a defect formation path, generating degradation prediction data. Establishing a self-organizing optical monitoring grid, to form a hierarchical spectral fingerprint library. Constructing a multi-point linked defect blocking network, to form an interconnected optical energy field. Monitoring a cooperative response to obtain network stability data, Identifying abnormal position coordinates based on a material quality grade and the network stability data, obtaining the abnormal position coordinates. Performing a phase adjustment to obtain a phase difference spectral set, conducting a phase comparison marking on the abnormal position coordinates, completing the quality inspection of the etching paste material.

The invention provides a signal optimization transmission system and method based on HPLC and HRF dual-mode communication, a signal acquisition and processing module of the signal optimization transmission system based on HPLC and HRF dual-mode communication is used for acquiring signals of two communication modes of HPLC and HRF in real time, preprocessing the signals and extracting key signal parameters; the dual-mode switching module is used for dynamically switching an HPLC mode and an HRF mode based on channel quality; the access module is used for constructing a longitudinal backbone network and a transverse Mesh network, forming a hybrid topology, optimizing a routing path and coordinating resource allocation and conflict avoidance of HPLC and HRF; the intelligent decision module is used for predicting future interference intensity according to the historical interference data and adjusting channel parameters; the central coordinator module is used for network initialization and resource allocation; according to the system and the method, the HPLC mode and the HRF mode are automatically switched according to the channel quality, the longitudinal HPLC backbone network is combined with the transverse HRF Mesh network to form a four-dimensional communication network, and the robustness in a complex scene is remarkably improved.

The invention provides a multifunctional inspection device detection system based on a distribution network mobile operation terminal, and relates to the technical field of electric power detection, and the multifunctional inspection device detection system comprises a portable main terminal of a dynamic collaborative architecture, a wearable terminal for augmented reality interaction and a distributed heterogeneous sensing cluster, which form a closed-loop data chain through an electric power dedicated low-delay wireless communication network; the portable main terminal is integrated with a heterogeneous computing unit with adaptive computing power distribution, can perform real-time fusion analysis on multi-dimensional sensing data, and outputs a visual result containing fault location and confidence; the wearable terminal superposes fault information to a real scene in a three-dimensional marking form through a virtual-real fusion positioning technology, and supports eye movement and voice collaborative interaction; multi-module collaborative acquisition of the distributed heterogeneous sensing cluster is combined with a nanosecond timestamp synchronization mechanism, so that the limitation of traditional single parameter detection is broken through, rich and synchronous basic data is provided for subsequent analysis, and the comprehensiveness of routing inspection is greatly improved.

The embodiment of the invention provides a closed-loop control method for power driving and protection, and the method comprises the steps: constructing a voltage feedback type circuit through a high-speed operational amplifier, and forming a closed-loop gain structure through a feedback network of the voltage feedback type circuit; the load current of the output end of the voltage feedback type circuit is monitored in real time, and when the load current exceeds a first set threshold value, a protection mechanism is triggered to turn off the output stage; the power-on surge current is limited through the soft start control circuit; phase information of a load reflection coefficient is obtained, an inductance / capacitance combination of the variable dynamic impedance matching network is selected according to phase information table look-up, inductance / capacitance parameters are switched through a relay, and the transmission standing-wave ratio between the power amplifier and a load is optimized; monitoring the temperature of the radiating fin of the power tube, and dynamically adjusting the resistance value of the feedback network; and performing cooperative control of short-circuit protection, over-temperature protection and surge suppression according to the priority. According to the embodiment of the invention, the performance bottleneck in a load fluctuation scene is broken through.

The invention discloses an elastic distributed Nash equilibrium search method and device under FDI attack, and relates to the technical field of non-cooperative game decision, the method comprises the following steps: constructing a multi-agent network, forming a non-cooperative game, and defining an objective function of each agent in the multi-agent network; establishing a malicious FDI attack model; the malicious FDI attack model maps false data injection attack to control input of an agent; an elastic self-adaptive distributed Nash equilibrium search algorithm is designed based on the target function, so that behaviors of other intelligent agents are estimated under the condition that the intelligent agents cannot obtain global information and attack information; the elastic self-adaptive distributed Nash equilibrium search algorithm comprises a leader-follower consistent consensus estimation-based identifier signal designed by an observer and a self-adaptive dynamic gain part. The elastic self-adaptive distributed Nash equilibrium search method solves the problem of elastic self-adaptive distributed Nash equilibrium search when the non-cooperative game is attacked by unknown FDI.

The invention discloses a cross-chain transaction-oriented smart contract collaboration method, and belongs to the technical field of cross-chain contract collaboration, and the method specifically comprises the steps: deploying a log collection assembly at each cross-chain transaction smart contract node, and recording log information in real time; based on a cross-chain collaborative typical risk scene, various risks are converted into quantifiable detection rules, and a risk rule model reads node log streams in real time; when the risk rule model detects an exception, forming a complete cross-chain calling graph according to the transaction association identifier in the node log; extracting operation features of nodes in the cross-chain calling atlas, comparing the operation features with a normal node feature library, and screening out suspected fault nodes of which the operation features deviate from a threshold value; the suspected fault node abnormal data, the graph fragments and the log information are integrated, a unique responsibility affirmation abstract is generated, and the unique responsibility affirmation abstract and complete materials are stored in a shared evidence storage network to form a non-tampering certificate; and after the voucher is generated, calling a responsibility investigation intelligent contract, and executing cross-chain responsibility investigation operation according to the fault node identifier and the responsibility terms.

The invention discloses a category incremental target detection method and related equipment, and the method comprises the steps: firstly extracting the multi-scale features of an input image, coding a known category text label to generate an embedded feature, calculating an average feature, and then generating an unknown category pseudo text embedded feature; a cross-modal feature pyramid is formed through a cross-modal fusion network, so that a known target is predicted and an unknown target is positioned; self-supervised training is carried out on an unknown target to generate a pseudo label to finish category fine tuning; and finally, model expansion is realized through incremental updating of a low-rank matrix and a shared matrix. According to the scheme, through pseudo-text embedding features and cross-modal fusion, the model is endowed with the unknown target recognition capability, and the application limitation of the open world is broken; by means of self-supervised false label fine tuning and matrix increment updating, full-category retraining is avoided, computing resource consumption and training duration are reduced, interference on original category weights is prevented, and stability and reliability of a target detection system during category expansion are guaranteed.

The invention discloses an electric power monitoring system based on artificial intelligence, and relates to the technical field of electric power monitoring, comprising: extracting a behavior disturbance unit from communication, operation and configuration logs of an electric power system; constructing a disturbance sequence by the behavior disturbance units according to the dependency relationship between the time sequence and the power system module; training an anti-fact generation network by using the disturbance sequences to form anti-fact tracks in one-to-one correspondence with the disturbance sequences; performing node-by-node comparison on the disturbance sequence and the anti-fact trajectory, extracting deviation points and combining the deviation points into an attack evolution chain; constructing a multi-dimensional action prediction tensor based on the attack evolution chain, wherein the multi-dimensional action prediction tensor is used for reasoning next-step action probability distribution of an attacker; and dynamically adjusting defense resources of the power monitoring system according to an action prediction tensor output result, and implementing right limiting, cheating or steering operation with specific granularity to interrupt attack evolution chain extension. According to the method, the attack path can be gradually restored from the detected abnormal behaviors, and the possible next action of an attacker is predicted.

The invention discloses a digital media content element accurate screening method based on artificial intelligence image recognition, and relates to the technical field of digital media, and the method comprises the steps: building a distributed capture network to form a dynamic content pool, and building a metadata index database; calling a multi-modal image perception engine to generate a double-layer characteristic spectrum containing dominant and recessive elements; constructing a distributed recognition model cluster based on federated learning; converting the user demand into a screening parameter set and generating a decision tree; screening and secondarily verifying an output result through a double-path matching mechanism; and constructing a reinforcement learning reward function based on user behaviors, and driving the model and the decision tree to co-evolve. Multi-source heterogeneous content full-dimension analysis is achieved, the recognition comprehensiveness and depth are improved, knowledge barriers and privacy risks are solved, the screening accuracy and flexibility are improved, the system is endowed with the continuous optimization capacity, and the method is suitable for efficient and accurate digital media content screening scenes.

The invention discloses a medicine sales management method and system based on information monitoring, and relates to the technical field of medicine sales management, and the method comprises the steps: collecting medicine sales time sequence data, carrying out the cleaning and structural processing, and generating standardized sales data; based on the standardized sales data, constructing a sales behavior time sequence causal network, embedding virtual sales intervention nodes, and generating a time sequence causal graph; the intervention strategy is coded to the virtual intervention node in the time sequence causal atlas, the graph neural network is utilized to simulate time sequence propagation of the intervention strategy, and a sales prediction result is obtained; and combining the sales prediction result with the real-time inventory, constructing a multi-level distributed sales dependency network, and forming the sales dependency network. According to the method, the preset intervention strategy is embedded into the causal structure in the node form, so that dynamic modeling of the multivariable causal relationship in the sales behavior is realized, and the prediction result not only reflects the historical trend, but also can respond to the possible influence of different intervention strategies.

The invention provides an unmanned aerial vehicle distribution scheduling method and system applied to a communication interruption region, and the method comprises the steps: determining unmanned aerial vehicle information used for the communication interruption region, and carrying out the air networking of all unmanned aerial vehicles according to the unmanned aerial vehicle information; after networking succeeds, defining a coordinate set of all the unmanned aerial vehicles participating in networking, constructing a logarithmic path loss model according to the coordinate set, and obtaining the signal propagation intensity of any position point of the communication interruption region according to the path loss model; constructing a signal coverage rate objective function according to the signal propagation intensity, and constructing a network average link quality objective function and a node energy consumption balance objective function to obtain a comprehensive objective function; and taking the maximization of the comprehensive objective function as a target, and solving by adopting an improved particle swarm optimization algorithm to obtain the deployment positions of all the unmanned aerial vehicles in the communication interruption region. According to the invention, the deployment efficiency and reliability of the disaster area communication system can be significantly improved.

The invention belongs to the technical field of computer hardware, and discloses a self-adaptive voltage regulation method for dynamic power consumption management of a notebook computer, which comprises the following steps: collecting multi-source data, generating a state label, and constructing to obtain a four-dimensional situation map; task association strength and load types are analyzed through space-time association, and double tags are generated in combination with user behaviors; component security boundaries are divided, multi-type credits are distributed, and a DAO smart contract set is formed by constructing a component node network; a state circulation rule is formulated in combination with the real-time load characteristics of the components; in combination with the V-F curve, a voltage regulation parameter table is generated in the sub-regulation scene, a synchronous regulation group is reconstructed, a synchronous regulation group configuration table is generated, then component voltage regulation and state circulation are executed, and a state circulation completion signal is generated; generating a parameter optimization scheme through three-level feedback, and deducing and quantifying the expected effect of the scheme to form an evaluation report; and executing a parameter optimization scheme, and synchronously updating to a preorder link as a new reference.

The invention belongs to the technical field of biomedical signal processing, and discloses an SEEG attack detection method based on a time-frequency space-dynamic graph attention network, which comprises the following steps: constructing an epilepsy brain network sequence, defining SEEG lead points as nodes of a graph, and calculating directed connection edge weights among the nodes through a multiband directed transfer function; constructing a node feature matrix based on epilepsy induced epilepsy index features; arranging the multiband epilepsy brain network according to a time sequence to form a dynamic sequence; extracting space-spectrum correlation by using a GATv2 dynamic attention mechanism, and dynamically weighting node connection weight through a learnable attention matrix; and in combination with a time sequence convolutional network and a time attention mechanism, key time sequence features are extracted, identified and focused. According to the method, a good dynamic propagation characteristic learning capability is obtained, an SEEG pathological connection mode is disclosed, a model mechanism corresponds to an epileptic seizure propagation mechanism, and the method has a good application prospect in the aspects of epileptic seizure detection and epileptic seizure auxiliary positioning.

The invention relates to the technical field of data analysis, in particular to a traffic flow prediction method and system based on an interpretable graph and a multi-scale time trunk. The method comprises the following steps: processing historical traffic flow information of a target road network according to a prediction model to obtain traffic flow prediction information of the target road network in a target time period; the prediction model is a composite model formed by fusing a space backbone network and a time backbone network, the space backbone network comprises a graph neural network, the time backbone network comprises an improved PatchTST model, and the improved PatchTST model comprises a multi-time scale patch. The multi-time scale patch comprises a scale patch corresponding to short-term fluctuation, a scale patch corresponding to intra-day rhythm and a scale patch corresponding to cross-day trend. According to the method, the real constraint of each node in the road network can be combined, the complex time sequence characteristics of the historical traffic flow information of the road network under each time dimension can be fully mined, and more accurate and reliable traffic flow prediction information can be obtained.

The invention discloses an unmanned aerial vehicle complex airspace cooperative passage method based on a graph neural network, and the method comprises the steps: collecting the airspace data of an unmanned aerial vehicle complex airspace, and generating standardized airspace data; constructing a mixed queuing network, and forming arrival, queue length and service capability state of each service node and service link; mapping to obtain graph nodes, graph edges and graph features, and constructing an airspace graph; inputting the airspace graph into the graph neural network, and carrying out constraint updating on the mixed queuing network; integrating the service capability state, the queue length state and the congestion propagation parameters to form a network passing cost relationship; and executing a Frank-Wolfe algorithm, and generating a cooperative passing instruction for each unmanned aerial vehicle. According to the invention, by introducing the mixed queuing network and the Frank-Wolfe algorithm, efficient and stable cooperative passage scheduling of multiple unmanned aerial vehicles in a complex airspace environment is realized.

The invention relates to a subgrade slope stability intelligent prediction method based on deep learning, and belongs to the technical field of data processing, and the method comprises the following steps: 1, constructing a multi-field coupling data collection network, and forming a dynamic coupling data set with synchronous time response; 2, a meta learning-dynamic graph Transform hybrid model is constructed, rapid adaptation of a new slope scene is achieved through meta learning, the dynamic graph Transform updates rock mass unit mechanical association in real time, the dynamic graph Transform captures multi-field coupling data features, and a Hoek-Brown criterion and a damage evolution equation are embedded to serve as double physical constraint layers; 3, a causal discovery algorithm is introduced to identify key stability factors, a multi-objective evolutionary algorithm is adopted to optimize meta-learning-dynamic graph Transform hybrid model hyper-parameters, and a dynamic loss function is constructed by using project full life cycle risk cost; 4, inputting real-time monitoring data, outputting data, and cooperatively updating multi-engineering data; the method has the beneficial effects that the causal effect in a group is quantified, hyper-parameter optimization is guided, and the prediction performance is improved.

The present application belongs to the technical field of computer hardware, and discloses a self-adaptive voltage regulation method for dynamic power management of a notebook computer, comprising: collecting multi-source data, generating a state label, and constructing a four-dimensional situation map; analyzing the task correlation strength and load type through space-time correlation, generating double labels in combination with user behavior; dividing the component safety boundary and allocating multiple types of credit, then forming a DAO smart contract set by constructing a component node network; and formulating state flow rules in combination with the real-time load characteristics of the components; generating a voltage regulation parameter table in combination with a V-F curve, reconstructing a synchronous regulation group and generating a synchronous regulation group configuration table, then executing component voltage regulation and state flow, and generating a state flow completion signal; generating a parameter optimization scheme through three-level feedback, and deducing the expected effect of the quantization scheme to form an evaluation report; and executing the parameter optimization scheme and synchronously updating it to the previous link as a new benchmark.

The invention discloses a task-oriented dialogue system reward function optimization method and system, and belongs to the technical field of natural language processing. The method comprises the following steps: acquiring an expert dialogue track from a dialogue system data set, extracting a state-action-reward triple, training an initial reward function by using a maximum entropy inverse reinforcement learning framework, and initializing a strategy network; an Actor-Critic reinforcement learning algorithm is adopted to train a strategy network, and a suboptimal trajectory is collected to dynamically update a reward function; and taking the dynamic reward function as a unified evaluation signal, and optimizing the strategy network to form a dialogue strategy model. Through dynamic reward function optimization, manual rule dependence is reduced, the generalization ability, the task completion rate and the stability of a dialogue system are improved, and the method is suitable for complex dialogue scenes in multiple fields.

This invention relates to the field of segmentation algorithm technology, and in particular to a lung nodule segmentation algorithm based on TransUNet. The steps include: replacing the Transformer layers in the TransUNet network with MaxViT networks to form an MM-TransUNet network as the segmentation model; training the segmentation model using the MM-TransUNet network; after each training round, saving the segmentation model parameters for that round; comparing the detection results obtained by the segmentation model with the actual results to calculate the loss; adjusting the parameters of the segmentation model based on the loss; reading the optimal parameters saved during the training phase; substituting the optimal parameters into the segmentation model to obtain the optimal lung nodule segmentation model; applying the MM-TransUNet network as the segmentation model can effectively improve the segmentation accuracy of lung nodules; MM-TransUNet combines residual mechanisms, CNNs, and multi-axial self-attention mechanisms as an encoder module, playing a role in more completely extracting lung nodule features.

A system is provided for wireless online processing using a peer to peer network. In particular, the system may comprise a plurality of wireless computing devices in nearby proximity to one another that may form the peer to peer (“P2P”) network through a shared wireless communication channel. The system may identify the network connection strengths of each of the plurality of wireless computing devices and subsequently designate one or more devices as the hubs of the P2P network. Each of the wireless computing devices may have one or more online processes to be completed through the network connections formed by the P2P network. The processes may be propagated to the devices within the P2P network, where the data used to complete the processes may be encrypted within the memory or storage devices of each of the computing devices.

The embodiment of the invention provides a communication method and device, and relates to the technical field of networks, and the method comprises the steps: first network equipment and other network equipment except the first network equipment in a local area network carry out main equipment election; if the first network device is elected as a main device, the first network device is configured with a DHCP server function, and a DHCP server backoff function is started, the first network device detects other DHCP servers; after other DHCP servers are detected, the dynamic IP addresses distributed by the other DHCP servers take effect, and the DHCP server function is closed; and if the other network devices are elected as slave devices, the other network devices are configured with the DHCP server function, and the DHCP server backoff function is started, the DHCP server function is closed, and the DHCP server backoff function is closed. According to the scheme, automatic networking during startup under the condition of multiple gateways can be realized, and zero-configuration opening of a user can be realized.

A satellite space target cooperative observation distributed planning method based on multi-agent reinforcement learning, comprising: constructing an on-board Actor network and a ground centralized Critic network to form a centralized training and distributed execution architecture; designing a structured mixed space action, training the Actor-Critic network through a large number of ground simulations and using an improved MAHPPO algorithm and a curriculum learning technique; deploying the trained Actor network on each satellite of the constellation executing the cooperative observation task to realize distributed execution; after uploading the list of space targets to be observed to each satellite through satellite-ground communication, each satellite does not need to communicate with other satellites, each satellite executes the observation target of each time slice decided by the on-board Actor network, and controls the satellite to implement the observation task, thereby realizing the space target cooperative observation task. The present application enables each satellite to complete the cooperative observation without communication during the on-orbit execution, and can be extended to large-scale constellation applications.

The invention discloses a fish gelatin jelly gel performance evaluation method based on multi-scale correlation analysis, which comprises the following steps: extracting multi-scale characteristics of molecular characteristics, network structures and macro performance of three fish gelatin jellies, performing data correlation based on three dimensions of molecular characteristics-network structures-macro performance, and combining a stoichiometric analysis method to evaluate the gel performance of the fish gelatin jellies. And a multi-level systematic evaluation system is constructed. According to the method, data of molecular characteristics (secondary structure ratio, particle size, Zeta potential and amide A wave number), network structure (aperture size) and macroscopic performance (imino acid content and gel strength) are integrated to construct the relevance of molecular characteristics-network structure-macroscopic performance, and full-dimensional analysis of the quality of the fish gelatin jelly is realized for the first time. The method breaks through the limitation that the traditional method only depends on single indexes such as protein or collagen content and the like, and clarifies the regulation and control of molecular aggregation behavior, imino acid content and electrostatic interaction on the formation of a gel network.

The invention discloses a power transmission line traveling wave fault positioning method and system based on data analysis, and relates to the technical field of big data analysis, and the method comprises the steps: firstly collecting fault original records, temperature, humidity and weather states, and building a historical library; carrying out time-frequency domain conjoint analysis on the traveling wave signal, and calculating a phase difference and a time difference; the signals are adaptively decomposed to different frequency bands to calculate fractal dimensions, and advanced features are extracted in combination with a deep convolutional neural network; associating the phase difference, the instantaneous electric field intensity and the magnetic field rotation direction to construct a three-dimensional matrix rotation graph structure, and forming a high-dimensional feature vector by using a graph attention network; and calculating a fault posterior probability according to the high-dimensional vector, dynamically adjusting a threshold value according to an environment parameter, and generating an adversarial network virtual sample to match a fault type when the probability is insufficient. The system comprises a data acquisition module, a time-frequency analysis module, a feature extraction module, a feature association module and a fault determination module, all the modules cooperate to realize data acquisition, feature analysis and fault positioning, and the positioning reliability is enhanced.

This invention discloses a multi-agent collaborative decision-making method for ramp merging based on spatiotemporal interaction capture and action priority, belonging to the field of intelligent transportation technology. The method includes: acquiring a sequence of observation information from multiple agents and obtaining a spatiotemporal interaction code through temporal encoding and self-attention interaction encoding; gating the spatiotemporal interaction code based on ramp merging features (TTMP); inputting the enhanced code into a policy priority ranking network, sampling the action generation order using Plackett-Luce, and then reordering the enhanced code, before inputting it into a decoding policy network with cross-attention to form a spatiotemporal interactive Plackett-Luce multi-agent reinforcement learning model. The model then autoregressively generates actions for each agent according to the reordered order and integrates them to form a joint action, generating vehicle control commands to control vehicles to perform collaborative driving in the ramp merging area. This invention can dynamically model the multi-vehicle interaction relationship and the action priority dependency between individual decisions in a strong interaction scenario of ramp merging, thereby improving the stability of collaborative decision-making and traffic efficiency.

The invention provides an electronic detonator traceability management early warning method and system based on big data. According to the method, a supply chain topology network is initialized in an electronic detonator production link, and a bidirectional tracing link is formed; in the transportation and storage links, a wireless energy field coupling device is used for activating a detonator built-in sensing unit, and movement track data and environment fluctuation data are collected; deploying a collaborative reasoning engine, and carrying out space-time association on the moving track and the environment fluctuation data to generate a detonator state abnormity probability value; inputting the raw material quality data and the transportation vibration intensity data into a collaborative inference engine for cross-level feature fusion, and generating safety traceability features; and when the abnormal probability value and the security traceability feature form an associated abnormal combination, triggering a grading early warning instruction and freezing the operation authority of the associated node. According to the invention, the full-life-cycle safety traceability and the real-time risk collaborative early warning precision of the electronic detonator in the links of transportation, storage and blasting are improved.

The invention provides a mooring load automatic prediction method and system based on wind wave big data, and the method comprises the steps: firstly collecting the wind wave action data of a target sea area to construct a wind wave action chain, and forming a wind wave action sequence set; then, a load conduction network is constructed according to the structural component distribution of the mooring equipment, and a load conduction model is formed; inputting the wind wave action chain into a load conduction network, and establishing a dynamic response mapping model in combination with historical data; and acquiring real-time wind wave action data, converting the data into a real-time wind wave action chain, inputting the real-time wind wave action chain into the dynamic response mapping model to obtain a real-time load conduction distribution result, and determining a mooring load prediction value. And finally, a load regulation and control strategy is generated according to the mooring load prediction value and is sent to the control module to realize dynamic load adjustment, so that the accuracy and the real-time performance of mooring load prediction are improved, and the ocean engineering safety is guaranteed.

The application discloses a kind of bare rock slope ecological restoration scheme demonstration method based on VR simulation, comprising: obtaining the multi-source data of bare rock slope, fusion processing generates three-dimensional digital model and ecological parameter distribution chart.Key nodes in slope are identified, and the connection network between key nodes is constructed, forming preliminary resource allocation scheme and network topology structure.With the aid of VR simulation technology, the water resource flow and vegetation growth process are simulated, and the resource allocation scheme and water flow distribution prediction map are optimized.Combining the vegetation coverage target and the characteristics of matrix material, the optimal solution of resource allocation is obtained.According to the distribution table of key nodes, a visual model of the slope restoration process is generated using three-dimensional rendering and dynamic simulation, and the ecological response data is analyzed to optimize the restoration scheme, and finally a complete slope ecological restoration scheme is formed.The application realizes the intelligent generation and optimization of bare rock slope ecological restoration scheme, and improves the restoration efficiency and accuracy.

The application discloses a kind of asphalt pavement technical literature mining methods based on TongYiQianWan semantic graph, comprising: obtaining road engineering-computer vision cross field academic literature data, carries out multidimensional preprocessing and forms standardized data set;Improved double-tower semantic coding network is constructed, the uniform mapping of keyword and title semantic features is realized;Optimized GraphSAGE spatial feature extraction network is constructed;CNN-LSTM hybrid architecture is used to construct timing feature extraction network, and form a multidimensional fusion feature representation system;Weighted random forest classifier with class weight balance strategy is introduced;Key word co-occurrence prediction and technology trend quantitative prediction system is established, finally, multi-module collaborative Qwen-SGCL intelligent model is obtained;The training, verification and comprehensive performance evaluation of the model are completed.The application provides an efficient intelligent tool for intelligent mining and technology trend quantitative analysis of literature in the cross field of road engineering and computer vision.

This invention belongs to the field of computer application technology. It discloses an information aggregation and retrieval system and method based on artificial intelligence, comprising: receiving user task objectives; combining associated resource sets and search preferences; calling a semantic model to generate semantic anchors; thereby generating a university-specific contextual blueprint to form a standardized information foundation within the university; parallel scheduling of various information interfaces within the university according to permissions; expanding query intents; filtering and sorting results to form a blueprint-bound retrieval result set; performing university-specific scenario analysis; generating dynamic aggregation summaries and difference comparison tables; grouping according to preset logic and performing visualization processing to form a structured presentation view; collecting user interaction data in the structured presentation view to generate a university user feedback dataset; and optimizing the university-specific contextual blueprint and university-specific knowledge unit network to form a closed-loop link.