61 results about "Network mining" patented technology

The invention relates to the technical field of geological exploration, and discloses a water-rich sandy stratum tunnel base settlement prediction method and system, and the method comprises the steps: constructing a multi-source cooperative exploration system; deploying a sanding-seepage-stress coupling model at an edge computing node, and mining the relevance of data of different dimensions through a space-time attention fusion network; based on the prediction matrix and mineral exploitation working condition parameters; when the predicted settlement exceeds a safety threshold value or the sanding expansion rate is abnormal, a high-precision remeasurement mechanism is triggered, and a cross-hole radar and acoustic logging combined verification module is activated; and iteratively optimizing coupling model parameters by adopting a transfer learning algorithm through deviation analysis of measured data and a prediction result. According to the method, a scientific basis is provided for stratum stability evaluation and prevention and control measure formulation in the mineral exploitation process, and the probability of engineering risks caused by base settlement of the water-rich sandy stratum is effectively reduced.

The invention discloses a knowledge base knowledge association fusion method based on a knowledge graph, and the method comprises the steps: integrating structured, semi-structured and non-structured data through a cross-modal alignment technology, and constructing a multi-source heterogeneous data association network of a unified semantic space; newly added external data are fused to a multi-source heterogeneous data association network in real time by using a dynamic attention mechanism, and entity conflicts are eliminated by combining rule reasoning and a machine learning model; hidden association among entities in the multi-source heterogeneous data association network is mined based on the graph neural network, and a knowledge graph logic chain is complemented; based on the knowledge graph, intelligent question and answer and risk assessment decision scenes are supported through a hybrid retrieval architecture and an inference engine; and automatically updating and associating the knowledge base of the network extension knowledge graph by adopting an incremental learning technology. Natural language questions and answers are supported, accurate answers are generated through knowledge reasoning, and user experience is remarkably enhanced.

The invention relates to the technical field of fund information management, in particular to a fund exchange information management system based on big data analysis. The system comprises a multi-source heterogeneous data acquisition and fusion module, an enterprise dynamic behavior portrait construction module, a hidden association network mining module and a trade background authenticity dynamic evaluation module. The method comprises the following steps: firstly, collecting and processing data from a plurality of data sources to generate a fused fund exchange record; constructing an enterprise dynamic behavior portrait based on the record; furthermore, a hidden association network graph is constructed based on the data, the data are received at the same time, the transaction behavior deviation degree is identified through behavior consistency check, fund closed-loop analysis is carried out on abnormal transactions, and finally a trade background authenticity index and a risk decision signal are generated. According to the invention, dynamic evaluation and risk early warning of supply chain financing trade background authenticity are realized, and the risk control capability of a financial institution is effectively improved.

The invention discloses an atmospheric and vacuum distillation unit intelligent optimization method and system based on deep reinforcement learning, and relates to the field of oil refining and chemical engineering, and the method comprises the steps: collecting historical operation data from a distributed control unit and a real-time database of an atmospheric and vacuum distillation unit, and carrying out the preprocessing, and obtaining a preprocessed historical data set; constructing a process mechanism feature network through a graph neural network by using the preprocessed historical data set, mining an association relationship among equipment parts, extracting abstract features representing a process principle, and constructing a digital twin environment; in a digital twin environment, a meta-reinforcement learning algorithm is adopted to train a deep reinforcement learning agent, so that the deep reinforcement learning agent obtains meta-strategies adapted to different crude oil properties. According to the method, safe, stable and efficient intelligent optimization control is realized through a closed-loop operation mechanism of safe mirror image drilling and credibility driving, and the problem of optimization response lag caused by manual model adjustment due to crude oil property change in a traditional optimization technology is solved.

The invention relates to a multivariable system control method based on causal reasoning and reinforcement learning, and the method comprises the steps: collecting the power, the flywheel rotating speed and the temperature of a flywheel energy storage system in the operation process through a multi-sensor group, and carrying out the preprocessing of the power, the flywheel rotating speed and the temperature, and obtaining a standardized feature vector; mining a causal relationship among the power, the flywheel rotating speed and the temperature through a three-layer residual network for the standardized feature vector, and constructing a causal graph; a flywheel kinetic equation is injected into the third-layer residual network; optimizing a control strategy of the flywheel energy storage system by adopting a reinforcement learning algorithm based on the causal diagram; and a cascade fault suppression strategy is adopted to correct the control strategy based on the stability of the flywheel energy storage system, the corrected control strategy is converted into an instruction, the instruction is transmitted to an actuator, and the control strategy is executed. According to the method, the causal missetup rate is reduced, strategy convergence acceleration is accelerated, and the service life of equipment is prolonged.

The invention discloses a multi-source feature collaborative knowledge mining and semantic association method, and belongs to the technical field of full-life-cycle knowledge processing of railway track engineering. According to the method, field-related documents are screened, an element range and a processing sequence are limited in combination with a knowledge graph, the documents are customized to generate a term candidate set, term statistical weights are calculated based on a full-life-cycle corpus, dynamic semantic vectors are generated by utilizing a field self-adaptive pre-training model, and composite concept representation is obtained through fusion of an attention mechanism. And linking the knowledge graph to construct a heterogeneous graph, mining association through a graph attention network, analyzing, querying and verifying sequential logic, and then outputting structured knowledge. According to the method, the problems of railway track engineering data islands and cross-stage semantic segmentation can be solved, low-frequency key terms are accurately recognized, the result interpretability is enhanced, applications such as design optimization and construction management and control are directly supported, and the method is adaptive to multiple engineering professions and high in practicability.

The invention relates to the field of network public opinion management and control, in particular to a cross-modal network emergency command system based on a network space model. The method comprises the following steps: acquiring a network public opinion multi-modal data stream, positioning a malicious public opinion propagation source account based on the network public opinion multi-modal data stream, and performing network space cross-layer holographic portraying based on the malicious public opinion propagation source account to generate a source account holographic portraying information set; on the basis, multi-modal feature extraction and network space behavior model construction are carried out on hidden negative guidance behaviors, and an implicit negative information propagation mode model is generated; on the basis, potential malicious account active tracking and risk propagation network mining are carried out, and a potential risk propagation network information set is generated; on the basis, multi-level differential intervention strategies are generated and executed, and a network public opinion intervention strategy instruction set is generated. In the network public opinion emergency command process, pollution of false information to network space is reduced, and construction of healthy and sustainable network ecology is assisted.

The invention discloses a main shaft cutter wear monitoring method based on physical guidance and semi-supervised learning. The method comprises the following steps: collecting cutter acceleration signals, extracting multi-domain statistical features, carrying out cross-cutter robustness screening, and constructing accumulated features reflecting a historical trend to form a mixed input set; mining an explicit physical equation conforming to a degradation mechanism by using a parallel symbol regression network with monotonicity penalty introduced; performing prediction and smooth monotonization processing on unlabeled data by using the equation, generating a physical weak label, and constructing a mixed training set; designing a physical guidance cross attention mechanism, and performing weighted fusion on statistical and cumulative features by taking a physical equation as prior; and constructing a lightweight neural network processing fusion feature, and designing a differential loss function for joint training based on the mixed training set. The method can break through the dependence on a large amount of labeled data, guarantees the physical consistency of prediction results, and meets the high-precision monitoring requirements of an industrial site.

The invention discloses a spatio-temporal trajectory prediction method and device based on a multi-modal condition potential diffusion generation model, electronic equipment and a storage medium, and the method comprises the steps: cleaning and aligning multi-source data such as trajectory data, environment semantics and a road network, and constructing standardized input; extracting spatio-temporal context representation through multi-modal attention, mining an environment topological structure in combination with an iterative graph network, and generating structured node embedding; adopting dual-channel cross-modal attention fusion trajectory dynamic features and graph structure information to form unified semantic representation; the conditional variation auto-encoder fuses feature codes to a low-dimensional potential space, conditional reverse denoising generation is executed by using a potential diffusion model, and a future trajectory sequence is recovered step by step; and light weight of the model is realized through diffusion consistency distillation, and reverse sampling is compressed. Through the multimode topology-diffusion distillation integrated architecture, the precision, continuity and reasoning efficiency of trajectory prediction under sparse noise data are improved, and the method is suitable for scenes such as ecological monitoring, intelligent traffic and navigation.

The invention relates to the technical field of enterprise development evaluation, and particularly discloses an enterprise development potential evaluation method based on big data, and the method comprises the steps: collecting agricultural enterprise multi-source heterogeneous data through distributed nodes, importing a deep learning driven multi-modal fusion model, and generating a high-dimensional feature vector through cross-modal feature alignment and weighting; inputting the high-dimensional feature vector into a risk and pressure resistance evaluation model, constructing a dynamic risk conduction network mining risk association, generating a risk evaluation vector in combination with a time sequence feature, cascading the risk evaluation vector with the high-dimensional feature vector, and then generating a fusion feature vector through nonlinear transformation; and on the basis, a quantitative index is generated through multi-dimensional feature matching, and a comprehensive evaluation result is output through multi-scale decision analysis. According to the method, deep fusion of multi-source data is realized, risk changes are dynamically captured, evaluation comprehensiveness and accuracy are improved, and a reliable basis is provided for decision making of agricultural enterprises.

This invention discloses a method and system for cross-domain fault diagnosis of rolling bearings with unknown inter-domain data label relationships. It relates to the field of rolling bearing fault diagnosis technology and aims to solve the problem of low fault diagnosis accuracy in existing transfer learning methods due to unknown data label relationships between the source and target domains. The key technical points of this invention include: constructing a comparative general domain adaptation model; introducing a BYOL network to mine the unique structure of the target domain data and proposing an entropy separation strategy to reject unknown class samples; simultaneously, designing a source class weighting mechanism to improve the transfer semantic enhancement method, assigning different class-level weights to the source classes, so that the feature distributions of the two domains are better aligned in the shared label space, further constructing a fault diagnosis model; and performing fault diagnosis through the trained fault diagnosis model. This invention exhibits superior fault diagnosis accuracy under different operating conditions.

The present disclosure relates to a system and a method for determining relevant entities and products using an LLM model. A patent information extraction unit extracts patent related information. A large language model (LLM) unit analyzes the extracted claims for identifying top companies, startups, and products, forming a first list. A background collection module employs the LLM units along with advanced searching unit to generate a second list of relevant entities and products. A result combiner unit generates a list of relevant entities and products. A web mining unit search for relevant hyperlinks disclosing features of the identified products. A RAG module embeds background text extracted from identified hyperlinks. A claim chart module generates a claim chart table for each of the identified products. A ranking module ranks the patent via a weightage-based score and a report generation unit prepares a summarized report comprising an image-report and a textual-report.

The application discloses a kind of based on deep reinforcement learning's atmospheric and vacuum distillation unit intelligent optimization method and system, it is related to oil refining chemical industry, including, from the distributed control unit of atmospheric and vacuum distillation unit and real-time database collection historical operation data, and pre-processing, obtain the historical data set after pre-processing;Using the historical data set after pre-processing, process mechanism characteristic network is constructed by graph neural network, the correlation between equipment components is mined and abstract feature representing process principle is extracted, and digital twin environment is constructed;In digital twin environment, meta-reinforcement learning algorithm is used to train deep reinforcement learning agent, so that deep reinforcement learning agent obtains meta-strategy suitable for different crude oil properties.The application realizes safe, smooth, efficient intelligent optimization control through safety mirror image drill and trust degree driven closed-loop operation mechanism, solves the optimization response lag problem caused by manual adjustment model due to the change of traditional optimization technology crude oil property.

The invention belongs to the technical field of underwater robot fault diagnosis, and particularly relates to an underwater robot fault diagnosis method based on multi-sensor sequence chain graph composite feature extraction and fusion, a program, equipment and a storage medium. According to the invention, a sequence chain graph construction method is designed, time sequence data of a plurality of sensors are integrated into a sequence chain graph structure adaptive to AUV characteristics, reliable support is provided for subsequent network mining fault characteristics, the method is more adaptive to AUV large-inertia dynamic characteristics, and a more accurate AUV graph data structure can be established. The method introduces a composite feature extraction and feature fusion method, can extract the features of a single sensor and the correlation features among multiple sensors at the same time, achieves the effective fusion of the two types of features, improves the fault diagnosis precision, and makes up the limitation of the prior art in the aspect of AUV fault feature extraction.

This invention relates to the field of bioinformatics processing and discloses a method and system for mining the structure-activity relationship (SMR) of porcine neurotrophic peptides based on sequence features. The method includes constructing an original sample index table and fusing multi-source production data, extracting peptide sequence features to generate a sequence feature matrix, constructing a sequence-process joint graph containing peptide nodes and process state nodes, training a structure-activity relationship graph neural network to mine SMR relationships, and deriving a process control decision table based on a process response sample set generated by the network, thereby achieving online optimization of the porcine neurotrophic peptide preparation process. This invention solves the problem of SMR mining caused by the separation of process parameters, sequence information, and activity data, achieving accurate characterization of the synergistic effect of sequence and process and reverse optimization of process parameters, thus improving the targeted enrichment efficiency and bioactivity retention level of target neurotrophic peptides.

This invention belongs to the field of highway management technology, specifically relating to a method and system for temporal prediction and analysis of highway traffic flow based on LSTM. The method includes collecting temporal parameters of traffic flow from a target highway segment. These parameters include traffic flow parameters, environmental disturbance data, traffic event data, and temporal feature data. By introducing a multi-dimensional feature fusion mechanism (temporal, spatiotemporal, and disturbance-related), particularly employing an attention mechanism to dynamically weight disturbance features, and constructing a deep LSTM network to mine long-term temporal dependencies, this invention can more comprehensively and adaptively characterize the complex nonlinear dynamics of traffic flow. Combined with post-processing optimizations such as residual correction and physical rationality verification, the system significantly enhances the accuracy and stability of prediction results when facing sudden disturbances such as rainfall and accidents, effectively overcoming the problem of large prediction biases in complex scenarios using traditional methods and single LSTM models.

The application relates to a code summary extraction method and device and a computer device. The method comprises the following steps: inputting a detected code into a pre-trained function feature extraction model to obtain feature information of an unknown library function which is not pre-defined in the detected code; inputting the feature information of the unknown library function into a pre-trained function summary generation model to generate a first summary of the unknown library function according to a pre-set summary representation format; the summary representation format is used for representing a plurality of attributes of the unknown library function; supplementary information related to the unknown library function is obtained by network mining on the unknown library function, and a second summary of the unknown library function is generated according to the pre-set summary representation format according to the supplementary information; and the first summary and the second summary are merged to obtain a target summary of the unknown library function. According to the method, the summary of the unknown library function in the software code can be accurately extracted.

The invention discloses a multi-mode photovoltaic power prediction method, system and device based on improved ViT and a medium, and belongs to the technical field of photovoltaic power generation system monitoring, and the method comprises the steps: processing a sky image sequence and historical meteorological data: extracting the spatial-temporal dynamic characteristics of a cloud layer through Transform layering, mining the high-dimensional law of meteorological data through a deep network, and obtaining the spatial-temporal dynamic characteristics of the cloud layer; a single factor prediction is generated through LSTM. And fusing the image and the meteorological features by adopting a cross-modal attention mechanism, inputting the fused image and meteorological features into a MambaTS network, and carrying out adaptive fusion on a single-factor prediction result and a multi-modal prediction result. According to the invention, through spatial-temporal feature extraction of a sky image sequence, high-dimensional feature mining of historical meteorological data, cross-modal attention fusion and adaptive result integration, a multi-level and collaborative optimization photovoltaic power prediction framework is constructed. According to the framework, the comprehensive sensing ability of local details, global trends, short-term fluctuation and multi-mode association is unified.

The invention provides a time-space spectrum strong correlation optimization method based on water ecological characteristic elements, and relates to the technical field of water ecological environment monitoring and simulation. Comprising five steps of synchronous acquisition and processing of multi-source water ecological data, spatial-temporal spectrum three-dimensional feature primitive structured construction, spatial-temporal spectrum strong association rule analysis based on a graph model, parameter optimization of a coupling physical mechanism and an association rule, and model closed-loop optimization under dual verification. According to the method, a standardized data set is generated by fusing multi-source data, space-time spectrum feature primitives are constructed, then space-time spectrum strong association rules among elements are mined by utilizing a graph attention network, then the rules are embedded into a mechanism model in a constraint form, parameter optimization is performed by adopting a hybrid optimization algorithm, dual driving of a physical mechanism and a data rule is realized, and the method is suitable for the spatial-time spectrum strong association rules. Finally, the optimization model is output through closed-loop verification iteration, and therefore the precision, rationality and generalization ability of water ecological simulation are effectively improved.

The invention discloses a numerical weather forecast cold start error suppression method and device, and relates to the technical field of offshore wind power, and the suppression method comprises the steps: S1, collecting initial data through a marine meteorological sensor array, including sea wave height, period, wave direction and corresponding wind field initial observation data, and constructing a multi-dimensional data set after preprocessing; s2, on the basis of the data set and historical sea wave-wind field time series data, a physical mechanism and data driving combination method is adopted to construct a sea wave-wind field coupling model, and the nonlinear coupling relation between sea waves and the wind speed and the wind direction is quantified. According to the method, a physical mechanism and data driving fusion modeling is adopted, so that the meteorological rule conformity is guaranteed, the nonlinear coupling relationship is accurately captured, the limitation of a pure physical and pure data driving model is avoided, and the prediction core support is tamped; based on neural network mining time sequence and mode characteristics, a lattice point relaxation algorithm parameter strategy is dynamically adjusted, and the problem that traditional fixed parameters are poor in adaptability is solved.

The invention discloses a relation network mining and visualization system oriented to an international behavior body, which relates to the technical field of international relation network mining and comprises a data acquisition module, an economic dependency relation analysis module, an external exchange relation analysis module, a collaborative relation analysis module, a visual relation construction module and a display terminal. Through a complete technical link constructed by data acquisition, economic dependency analysis, external cross relationship analysis, organization collaborative analysis and a visual map, automatic mining and structure visual presentation of the international behavior multi-dimensional relationship are realized, and the objectivity, coherence and constitutive property of an analysis result are improved; scientific, systematic and extensible technical support is provided for international relationship research under a global background; the complex international relation structure can be presented in a transparent and visual mode, and the technical level and visual experience of international relation data analysis are remarkably improved.

The invention discloses a federated learning client trust degree evaluation method, system and device, a medium and a program product, and belongs to the field of cyberspace security, and the method comprises the steps: firstly, mining structural features of nodes by using a heterogeneous graph attention network, and then obtaining initial embedding of the nodes; secondly, respectively taking initial embedding of a server node and a newly added client node as initial features of an evaluator and an evaluated person, mining transmissibility and combinatorial features of the nodes by using a graph convolutional neural network, obtaining final embedding of the nodes by combining the two embedding, finally serially connecting the final embedding, and fitting the final embedding to a standard full-connection layer, so as to obtain an evaluation result of the evaluator and the evaluated person; using a softmax function to predict a trust relationship between the two nodes; the system, the equipment and the medium are used for implementing the method. The program product comprises a computer program for implementing the method; according to the method, the cold start processing capability in the federated learning environment is improved, and the method has relatively high trust evaluation accuracy and efficient calculation performance.

The invention discloses an organization implicit cooperation relation identification and optimization method based on multi-modal social network mining, which comprises the following steps of: multi-source cooperation data acquisition and preprocessing: acquiring mail exchange records, conference participation logs, instant messaging interaction data and project management system data in an organization, and performing anonymization processing on the data; reserving department and role labels; a node set comprises department-level nodes and employee-level nodes, and the two layers of nodes are connected through a membership relationship; according to the method, multi-source heterogeneous data (mail communication, conference logs, instant messaging records, project management data and the like) are collected and subjected to anonymization and cleaning processing, the department data dispersion barrier is broken, a complete data basis is provided for collaborative analysis, and the collaborative interaction behavior between the nodes is represented by the edge set and comprises communication frequency, task association and an information transmission path. And the analysis one-sidedness caused by the fact that the prior art only depends on single structured data is avoided.

The application discloses an intelligent exercise recommendation system and method based on multi-dimensional target optimization and belongs to the technical field of artificial intelligence. The method first extracts six-dimensional difficulty features from a question text and completes automatic labeling, and then, in combination with historical answering records of students, uses a graph attention network to mine causal coupling relationships among various ability dimensions, and identifies the root cause of the short board that truly restricts the score. On the basis, the six-dimensional recommendation weights are dynamically adjusted by using deep reinforcement learning, and personalized questions with high training value are continuously output to different students. The method has completed engineering deployment verification in learning machines, education APPs, AI intelligent learning platforms and other scenes.

The invention relates to the technical field of hotel guest room demand prediction, and discloses a big data-driven hotel guest room demand prediction model, which comprises a data acquisition module, a preprocessing and fusion module, a feature extraction module, a dynamic adaptive prediction module and a monitoring and optimization module. Through deep fusion of multi-source heterogeneous data and accurate extraction of spatio-temporal features, the model can comprehensively capture internal and external factors and dynamic rules affecting guest room demands, and internal historical orders, customer portraits, external tourism flow, social media popularity and other data are associated through a knowledge graph. The demand fluctuation of the time dimension and the geographic association of the space dimension are mined in combination with the space-time Transform network, so that the prediction result is more suitable for the actual market change, and an accurate basis is provided for hotel resource allocation.

The application discloses a hyperspectral image classification method based on an improved Swin-Transformer network, and comprises the following steps: S1, according to the characteristics of hyperspectral data, a spatial spectrum recombination module is proposed to preprocess the data; S2, the Swin-Transformer network is improved, and a cross-layer fusion module is added to the network to avoid information loss in the interlayer feedforward process; the output of the current layer in the network is fused with the output of the previous layer to realize the transmission of information from the shallow layer to the deep layer, thereby avoiding the loss of effective information in the feedforward process; and S3, the spatial spectrum recombination module is inserted into the improved Swin-Transformer network, and public hyperspectral image data is used as training data; the application has the advantages of higher accuracy, improved efficiency of network mining of spectral information, and effectively reduced loss of effective information in the feedforward process of the network.

The invention discloses a park integrated energy system reliability evaluation method and system based on physical data fusion and a storage medium, and relates to the technical field of integrated energy systems. Forming a natural gas network pipeline average flow rate data set including a system operation state and a corresponding scheduling process; mining a mapping relation in the data set by using a dual-path neural network, and constructing a natural gas pipeline average flow velocity prediction model based on the comprehensive energy operation state; embedding the predicted average airflow velocity into a linearized dynamic optimal energy flow model of the integrated energy system, and constructing a linearized dynamic optimal energy flow model of physical data fusion; a Monte Carlo simulation method is utilized to form a system operation state, and a load shedding condition of the system state is analyzed by adopting an optimal energy flow model of physical data fusion, so that a reliability index is calculated. The method solves the problems that a traditional data driving method is low in interpretability and insufficient in precision.

The invention relates to an intelligent referral decision-making method for cooperative matching of a patient and a hospital and a storage medium, and the method comprises the steps: firstly constructing a patient-hospital-adaptation degree integrated triple sample set, then constructing a medical conjunct heterogeneous network graph containing the patient, the hospital and two types of association edges, mining deep association features between nodes in combination with a graph attention network, and obtaining a medical conjunct heterogeneous network graph containing the patient, the hospital and the two types of association edges; compared with a traditional manual or single feature matching mode, the accuracy of evaluation of the patient-hospital fitness degree is remarkably improved, and mismatching and mismatching conditions can be reduced. Then feature aggregation and fitness calculation are automatically completed based on a graph attention network, fitness scores of patients and candidate hospitals are rapidly output, the tedious processes of manual screening and evaluation are replaced, referral decision-making time can be greatly shortened, the patients are guided to hospitals with matched diagnosis and treatment ability and resource conditions, high-quality medical resources are prevented from being excessively crowded, and medical treatment efficiency is improved. Meanwhile, the resource utilization rate of primary hospitals is increased, and medical conjoined grading diagnosis and treatment are assisted.

The invention relates to the technical field of voice signal processing and industrial communication, and discloses a mine scene voice enhancement method and device based on multi-modal noise feature learning, and the method comprises the following steps: synchronously collecting acoustics, vibration and acceleration signals of a mine scene, and extracting multi-dimensional features after trend term removal and standardization preprocessing; dynamically fusing the vibration and acceleration characteristics representing the physical working condition of the pure noise source with the acoustic characteristics by using a self-attention mechanism; inputting the fusion features into a CNN-LSTM deep network, and reasoning to obtain an ideal ratio masking vector; and using the masking vector to correct the amplitude spectrum, and combining original phase reconstruction to generate enhanced speech. According to the method, the non-acoustic physical mode is introduced to serve as auxiliary information, causal association between deep network mining mechanical vibration and acoustic noise is utilized, the problem that the non-stationary impact noise is difficult to restrain under the low signal-to-noise ratio is effectively solved, and the definition and intelligibility of voice communication in the complex mine environment are improved.

The invention discloses a power system visualization display method and system based on a knowledge graph, and particularly relates to the technical field of power system intelligent visualization, and the method comprises the steps: constructing a power system knowledge graph, and carrying out the initialization configuration; user interaction behaviors are recorded and quantified, and the dynamic interestingness and the common-view relation are calculated; mining a user operation time sequence mode by using an LSTM network, and constructing and updating a user interest model; according to the interest model, dynamically adjusting a force-oriented layout parameter and a visual channel priority, and realizing adaptive visual rendering; and a reinforcement learning closed loop is established, and a visualization strategy is continuously optimized through a DQN algorithm. According to the method, user behaviors are perceived and the visualization effect is adaptively adjusted, so that the expression definition and the interaction efficiency of complex power grid data are effectively improved, and the cognitive load of the user is reduced.