398 results about "Risk indicator" patented technology

The invention discloses a method and system for predicting leakage of a water supply pipe network, and the method comprises the steps: modeling nodes and pipe sections of the pipe network into a graph topological structure, and endowing the nodes and the pipe sections with static attributes; collecting operation data of the water supply network, and constructing time-varying graph data corresponding to the graph topology; combining the time-varying graph data with the static attributes to form space-time input features; constructing a graph time sequence prediction model based on a deep learning framework, performing graph structure feature extraction on node graph features and pipe section graph features of each time step to obtain node space features and pipe section space features, and outputting a node and pipe section space-time representation set; evaluating and analyzing the leakage level of each DMA or pressure partition; generating a pipe section leakage risk space distribution set; constructing a joint loss function, and training and updating the graph time sequence prediction model; and inputting operation data acquired in real time into the trained graph time sequence prediction model, and generating a leakage rate prediction value of each partition and a leakage risk index of each pipe section on line for leakage prediction and operation and maintenance decision.

The invention relates to a grid-connected scheduling management method and device constructed in combination with a knowledge graph, equipment and a medium. According to the method, a comprehensive data set is constructed by integrating multi-source data such as new energy output, power grid topology, load, weather and historical fault records, and then a dynamic knowledge graph is formed by using entity recognition and relation extraction technologies; a probability causal graph model is constructed by extracting a causal path and adding probability parameters, and uncertainty propagation intensity is quantified in combination with a sequence diagram neural network; on the basis of a propagation model, risk index conditional probability is calculated by adopting probability causal reasoning, and a fault propagation sequence is simulated through a cascade failure theory to realize multi-level risk assessment; based on a multi-objective optimization model and deep reinforcement learning, an adaptive scheduling strategy is generated, a complete technical closed loop from data fusion and causal reasoning to intelligent decision is realized, and the technical effects of describing a new energy uncertainty propagation path, prospectively evaluating a power grid risk situation and dynamically generating an optimal grid-connected scheduling scheme are achieved.

The invention provides a fault map learning-based early abnormality diagnosis method and system for an anion exchange membrane electrolysis system, and relates to the technical field of fault diagnosis. The method comprises the following steps of: acquiring multi-working-condition operation measurement and constructing a time sequence characteristic fragment, establishing a fault map containing parts, working conditions and measurement nodes by combining a part relationship and working condition dependency, training and calibrating a map learning model on the fault map, and extracting a self-adaptive baseline; and projecting the feature fragments to a fault map, outputting multi-level risk indexes, generating a comprehensive risk score and an anomaly candidate set, and implementing multi-scale threshold judgment and stability test in combination with a baseline to form early warning and disposal suggestions. And performing further attribution analysis on the abnormal candidate set, identifying key components, and reinjecting maintenance feedback to update the atlas and the model to form a continuously optimized knowledge base. According to the method, causal correlation modeling and interpretable diagnosis are realized, and the accuracy and timeliness of early abnormality identification of the AEM electrolysis system are effectively improved.

The invention discloses a reinforcement learning unmanned ship path control method based on double-track regulation and control random network distillation. The method comprises the operation steps that an unmanned ship builds a path tracking simulation environment and a kinetic model; the unmanned ship builds a core algorithm flexible action evaluation algorithm framework; the unmanned ship deploys a priority experience playback pool based on quality and success guidance; an uncertainty perception and risk perception mechanism is introduced into the unmanned ship; the unmanned ship builds a success rate-based reward attenuation and cold start module, and the unmanned ship calculates a total reward and designs a reward softening mechanism to smooth the total reward; the unmanned ship imports hyper-parameters of all the modules, starts training circulation in a simulation environment, and dynamically adjusts exploration intensity and the like; according to the method, uncertainty and risk indexes are introduced, the exploration intensity of the intelligent agent is controlled, the intelligent agent is prevented from making dangerous actions, and the robustness is improved; a priority experience playback pool based on quality and success guidance is introduced, high-quality samples are better played back, and strategy convergence is accelerated.

A building-integrated artificial intelligence system forming a continuous ambient neural field is disclosed. The system includes a distributed multimodal sensor lattice, an on-premise symbolic cognition engine, and an adaptive environmental control kernel operating entirely at the building edge without reliance on external cloud services. Sensor data from optical, thermal, acoustic, airflow, pressure, structural, electrical, and chemical modalities are transformed into non-identifying occupancy vectors, behavioral glyphs, risk indicators, and environmental state descriptors. A privacy-governed policy graph determines sensor permissions, redaction thresholds, consent conditions, emergency overrides, and jurisdiction-specific compliance parameters. The neural field predicts occupancy loads, optimizes HVAC, ventilation, and lighting, detects accidents and structural anomalies, classifies emergent risks, and generates redacted event capsules for audit and emergency dispatch. A federated topology enables multiple buildings to exchange compressed symbolic templates to improve predictive accuracy without transmitting raw data. The system provides a universal, regulation-aligned AI nervous system for autonomous building operations.

The invention provides a risk monitoring method for a coal mine goaf, and the method comprises the steps: obtaining risk association information provided by a plurality of information collection devices in the coal mine goaf in real time according to a preset encryption communication mode; based on all the risk association information and a pre-trained risk assessment model, determining a current risk index value of the coal mine goaf, the risk assessment model being used for reflecting an association relationship between the risk index value of the coal mine goaf and all the risk association information; determining the current risk level of the coal mine goaf based on the current risk index value of the coal mine goaf and the index weight of the risk index value; and executing a risk early warning strategy corresponding to the current risk level of the coal mine goaf. According to the technical scheme, reliable technical guarantee is provided for safe production of the coal mine goaf.

The invention discloses a car following control method and a related product. According to the scheme, multi-source data are acquired, and the multi-source data are fused to obtain a fusion result; based on the fusion result, performing multi-modal trajectory prediction on the driving behavior of the preceding vehicle to obtain a prediction result of the driving behavior of the preceding vehicle; the multi-modal trajectory prediction comprises longitudinal motion prediction and transverse motion prediction; quantifying a car following risk by using a multi-dimensional risk index to obtain a risk level; constructing a multi-target cost function based on the fusion result, the front vehicle driving behavior prediction result and a control vector, and solving the multi-target cost function by using a preset constraint condition to obtain an initial control parameter of the vehicle; and on the basis of the risk level, initial control parameters of the vehicle are adjusted, and target control parameters of the vehicle are obtained. Compared with the problem of response lag in adaptive cruise control in the prior art, the adaptive cruise control method has obvious advantages.

The invention relates to the technical field of financial transactions, in particular to a financial transaction real-time anti-fraud monitoring system which comprises a data processing module, a graph construction module, a graph traversal module, a risk estimation module, a risk level determination module and a transaction management and control module. Performing feature quantization on the historical transaction sequence to obtain an intermediate state feature vector; taking the transaction subject identifier and the transaction counterparty information as nodes, taking the transaction behavior as an edge, and constructing a transaction association graph; on the basis of a preset fraud mode rule set, performing graph traversal on the transaction association graph, and identifying abnormal transaction sub-graphs; generating a comprehensive risk index according to the deviation degree of the abnormal transaction sub-graph and the intermediate state feature vector; performing grade judgment on the abnormal transaction sub-graph to obtain a risk disposal grade, and managing and controlling a corresponding transaction behavior of the transaction platform; according to the invention, the real-time performance of real-time anti-fraud monitoring of financial transactions can be improved.

The invention discloses a time sequence production simulation multi-dimensional risk analysis method and system based on deep learning, and the method comprises the steps: constructing a time sequence production simulation model, and carrying out the calculation to obtain system operation state data; constructing a multi-dimensional risk index analysis system, and calculating an initial weight of each risk index; introducing a real-time state quantity corresponding to each risk index as a driving factor to correct the initial weight of each risk index to obtain a corrected dynamic weight; carrying out weighted fusion on the risk index values based on the dynamic weight to obtain a comprehensive risk value; forming a feature vector by the product of the drive factor normalized value corresponding to each risk index and the dynamic weight, taking the feature vector as input, taking the corresponding comprehensive risk value as output, training the GRU model, and taking the trained GRU model as a risk analysis model; and processing the operation data of the target system according to the steps to obtain a corresponding feature vector, and inputting the feature vector into the analysis model to obtain a comprehensive risk value of the target system. According to the invention, the sensitivity and accuracy of risk identification are improved.

The invention relates to an intelligent driving assistance method and system based on driver state grading and a medium. The method comprises the steps that behavior characteristic data of a driver is collected through a driver monitoring system; based on the behavior characteristic data, calculating and generating a risk score and a confidence coefficient of a driver state through multi-evidence fusion; obtaining a driving scene risk factor, and carrying out coupling calculation on the risk score, the confidence coefficient and the driving scene risk factor to generate a comprehensive risk index; determining a target intervention level based on the comprehensive risk index; generating an auxiliary driving system control parameter set corresponding to the target intervention level according to the target intervention level; and issuing the control parameter set to a corresponding controller of the auxiliary driving system so as to realize self-adaptive adjustment of the auxiliary driving function. According to the method, the response strategy of the auxiliary driving function can be adaptively adjusted according to the real-time state of the driver, so that gradient safety guarantee measures matched with the driver are provided under different risk levels.

The invention discloses a risk assessment method and system for access of new energy photovoltaic power generation to a power distribution network, and belongs to the technical field of operation safety of the power distribution network, and the method comprises the steps: collecting illumination resource data and load data of the power distribution network, and generating a basic data set covering typical and extreme scenes through empirical mode decomposition, Copula function and improved Latin hypercube sampling; a coupling evaluation model fusing the graph neural network and the risk propagation theory is constructed, an improved CRITIC-entropy weight method is adopted to dynamically empowerly weight the multi-dimensional risk indexes, and a time-varying comprehensive risk level matrix is output; analyzing the risk characteristics, matching a pre-constructed measure risk knowledge graph, and generating a dynamic self-adaptive coping scheme through conflict verification and priority ranking; verifying the scheme in the digital twin power distribution network simulation model, and updating model parameters and a basic data set by incremental data feedback when the scheme does not reach the standard; according to the invention, the comprehensiveness and timeliness of risk assessment are improved.

The invention provides an electric power facility dynamic safety evaluation system based on blasting vibration propagation characteristics, which relates to the field of electric digital data processing and comprises a sensing and data acquisition module, a physical modeling and parameter library module, a real-time evaluation and risk inference module and a feedback execution and self-adaption module. The sensing and data acquisition module is used for converting on-site physical signals into high-quality data streams, the physical modeling and parameter library module is used for providing and building physical modeling and managing parameter information, and the real-time evaluation and risk inference module maps sensing data into system states and risk indexes through observation-model assimilation. The feedback execution and self-adaption module is responsible for driving decision execution, collecting feedback and continuously improving a system model; the system can monitor the influence of blasting vibration on electric power facilities in real time, accurately evaluate the structural damage degree and the system operation risk, and provide a scientific basis for safety protection and emergency decision-making of the electric power facilities.

The invention provides an industrial project cross-department collaborative approval and supervision system based on trusted computing, and relates to the technical field of cross-department processing, and the system comprises an extraction module which is used for triggering an intelligent contract deployed on an alliance chain based on an on-chain evidence storage record, generating an approval event, and sending the approval event to a server; according to the data integrity, the process time sequence and the operation behavior characteristics in the approval process event, each behavior characteristic is quantified into a risk index corresponding to each dimension through a preset risk model, and a set of the risk indexes of each dimension forms a multi-dimensional vector; through on-chain credible evidence storage, multi-dimensional risk measurement, dynamic credit evaluation and closed-loop regulation and control, precision, credibility and dynamic optimization of industrial project cross-department collaborative approval and supervision are realized, approval efficiency and supervision efficiency are improved, and scientificity and safety of collaborative management and control are guaranteed.

The invention discloses a risk assessment method for power grid cascading failures, and the method comprises the steps: firstly determining an assessment range and a target, and carrying out the data collection after the assessment range and the target are determined; establishing a system model, and constructing a digital model capable of accurately reflecting physical characteristics, operation states and control behaviors of a real power grid in a computer; constructing an initial fault scene; analyzing a cascading failure propagation path; quantifying risk indexes; and evaluating the cross-domain association influence, supplementing and evaluating the linkage influence on the association system in combination with the cross-domain association characteristics, and proposing a risk prevention and control and optimization strategy. According to the method, potential cascading failure inducements in a power grid can be actively mined through risk assessment of power grid cascading failures, a traditional passive mode of failure reprocessing is broken through, weak links are positioned, key equipment which easily triggers the cascading failures is found out through initial failure scene screening and propagation path simulation, and long-term latent hidden dangers are avoided.

The invention relates to the technical field of pressure-bearing structure welding defect identification, and discloses a pressure-bearing structure welding defect automatic identification and risk prediction method and system, and the method comprises the steps: obtaining an initial defect feature data set, carrying out the serialization analysis according to the initial defect feature data set, determining the initial defect extension trend, and carrying out the risk prediction of the initial defect extension trend; carrying out stress response analysis according to the defect initial expansion trend to obtain a defect stress response sequence, carrying out microscopic difference extraction analysis based on the defect stress response sequence to obtain a defect microcrack expansion trend, and carrying out time sequence prediction processing according to the defect microcrack expansion trend to obtain an expansion prediction vector; and carrying out characteristic evolution analysis and fatigue life calculation on the extended prediction vector, determining a potential failure time point, and carrying out risk index fusion calculation according to the potential failure time point to obtain a final failure risk prediction report. The method can solve the problem of insufficient dynamic monitoring in the prior art.

The invention relates to the technical field of cardiology nursing informatization, and discloses a cardiology nursing information monitoring system based on cloud computing. A physical sign monitoring module of the system integrates real-time data streams of a wearable device and a bedside monitor, collects multi-dimensional physiological parameters and generates a dynamic vital sign set; the health assessment module identifies abnormal fluctuation nodes of the physical sign data, matches historical medical record records, quantifies the association strength of the current physical sign and a typical pathological mode, and generates a multi-dimensional health risk index; the resource allocation module analyzes a nursing resource occupation state based on the risk index, calculates a task emergency degree weight, dynamically allocates a working path and an equipment use sequence of medical staff, and generates a hierarchical nursing scheduling scheme; and the intervention analysis module executes the scheduling scheme, compares sign changes before and after resource allocation, detects abnormal response delay time, positions execution deviation nodes and generates a nursing effect traceability report. The system assists intelligent and refined management of the nursing process of the department of cardiology.

The invention provides a vehicle track generation method and system, and relates to the technical field of data processing. Obtaining vehicle trajectory data, and screening out vehicle following fragments with a traffic oscillation phenomenon; carrying out risk grade division on the vehicle following fragment, calculating a risk index value based on a peak speed variation coefficient, an aggregation approaching risk and an aggregation collision risk, and dividing the risk index value into discrete risk tags; a conditional diffusion Transform model guided by physical information is constructed and trained; and using the trained conditional diffusion Transformer model to generate high-fidelity and physically consistent vehicle trajectory data according to a specified risk tag. When the vehicle track is generated, the data authenticity and the physical rule consistency are ensured.

The invention discloses a cerebral hemorrhage patient tracheotomy risk prediction method and system based on machine learning, and the method comprises the steps: obtaining an initial clinical data set of a target patient, calculating laboratory inspection data according to a predefined rule, and constructing a composite physiological state index to generate a feature vector for prediction; inputting the feature vector for prediction into a risk prediction model pre-trained based on an ensemble learning algorithm to obtain a risk quantitative index; the model interpretation module generates an individualized prediction contribution decomposition result based on an SHAP value calculation framework, and explains the specific influence of each feature on the risk index; and finally comprehensively generating a risk prediction report. According to the method, the feature representation and model prediction capability is enhanced by constructing the composite indexes, and meanwhile, the decision process is transparent and credible by utilizing interpretability analysis, so that clinical risk assessment and decision support are effectively assisted.

A set of one or more permissions associated with an identity is determined. One or more risk metrics and corresponding usage associated with the one or more permissions associated with the identity are determined. Access associated with at least one permission from the set of one or more permissions associated with the identity is modified based on the one or more determined risk metrics and corresponding usage associated with the one or more permissions associated with the identity.

The invention relates to the technical field of intelligent traffic monitoring, and discloses a road surface accumulated water detection and identification method based on multi-view feature fusion. The method comprises the following steps: acquiring image data, depth information and environmental parameters through sensor nodes deployed at multiple positions of a road surface to form an original monitoring data set; performing multi-source feature extraction and fusion processing on the data set to obtain an accumulated water feature image set; performing spatial domain analysis by using the image set to generate a pavement partition consistency map; time dimension data is extracted based on the map, dynamic change evaluation is carried out, and a ponding evolution report is output; detecting an abnormal mode from the report, and identifying an abnormal ponding area; and calculating a risk index according to the abnormal region, and generating a final road surface ponding risk map. According to the method, through multi-source data fusion and spatio-temporal conjoint analysis, accurate detection, dynamic evolution tracking and risk assessment of pavement ponding are realized, and the accuracy and early warning capability of urban road ponding monitoring are remarkably improved.

The invention provides a coronary heart disease recurrence risk assessment method and device, equipment and a storage medium, and relates to the technical field of medical data processing. The method comprises the following steps: acquiring dynamic behavior data, physiological data and static risk indexes of a target patient; calculating a treatment compliance index and a rehabilitation health index of the target patient based on the dynamic behavior data and the physiological data of the target patient; and inputting the treatment compliance index, the rehabilitation health index and the static risk index into a trained coronary heart disease recurrence risk scoring model to obtain a coronary heart disease recurrence risk score of the target patient. According to the method, objective and quantitative recurrence risk scores can be obtained, more accurate decision support is provided for clinicians, and early warning and personalized intervention can be realized, so that the prognosis of patients is improved, and the medical cost is reduced.

The invention discloses a risk indicator construction method and system for new energy output prediction, and belongs to the technical field of data processing and power system management, and the method comprises the steps: obtaining the preprocessing data of a plurality of new energy stations, carrying out the spatial-temporal feature analysis, and generating a spatial-temporal feature set; obtaining meteorological prediction data, and performing medium and long term output prediction through an extreme gradient boosting tree based on the space-time feature set; comparing the output prediction result with historical real output data to obtain a prediction error, and performing probability coupling modeling and quantile regression prediction in combination with a meteorological element evolution sequence to generate a dynamic quantile value; and determining a dynamic risk interval, and constructing a graded early warning index. According to the method, probabilistic coupling modeling and quantile regression prediction are adopted, prediction errors are deeply associated with dynamic evolution of meteorological elements, predicted uncertain risks can be quantized and graded, and decision support is provided for optimal scheduling and risk management of a power grid.

The invention discloses a financial risk prediction method based on multi-modal dynamic alignment and a graph neural network. The method comprises the following steps: firstly, acquiring and processing structured data and unstructured data in the financial field to obtain preliminary feature representation; then, structural features are enhanced through event attention enhancement and periodic coding, unstructured features are enhanced and deepened through domain adaptability, and dynamic alignment and fusion are performed on the two enhanced feature representations by using a bidirectional cross attention mechanism to generate unified fusion feature representations; then, dynamically updating the edge weight of the financial knowledge graph based on the fusion feature representation; and finally, simulating a nonlinear propagation process of the risk on the dynamic map by using a map neural network, predicting a risk state of each mechanism, and calculating a systematic risk index. According to the method, the accuracy, timeliness and interpretability of financial risk prediction are remarkably improved.

The invention discloses a novel power system supply and demand risk assessment method based on multiple spatial-temporal scales, and relates to the technical field of power system supply and demand risk assessment, and the method comprises the following steps: obtaining a time-varying fault rate set and a dynamic transmission limit set of power grid equipment based on extreme weather prediction data; fusing the time-varying fault rate set, the dynamic transmission limit set and power supply, load and new energy power generation data to obtain a risk scene library; screening the risk scene library to obtain a typical risk scene set; analyzing the typical risk scene set, and outputting a supply and demand risk index under each time-space unit; and performing regional division on the power system based on the supply and demand risk indexes, and generating corresponding early warning and management and control suggestions. According to the method, the problem that the evaluation result is not accurate due to the fact that the time-varying characteristics of the equipment state are ignored in extreme weather is solved, and accurate decision support is provided for disaster prevention and supply protection of a power system.

The invention discloses a tree safety risk assessment method, system and device based on a key index priority mechanism, and a storage medium. The method comprises the following steps: constructing an inference model according to a fault tree model of a street tree and a fuzzy Bayesian network; based on the inference model, through reverse inference, sensitivity analysis and most approximate cause chain analysis, screening out a key index set and a conventional index set from a preset risk candidate index set; and based on a hierarchical risk assessment principle, respectively assessing the key index set and the conventional index set to obtain an assessment result. According to the inference model, key factors having significant contributions to the overall risk of the border tree are identified from a large number of risk indexes; a conjoint analysis method of reverse reasoning, sensitivity analysis and most approximate cause chain analysis is adopted, so that false alarm and missing alarm are effectively reduced; according to the hierarchical risk assessment principle, key factors are assessed preferentially, it is ensured that the street tree with structural fatal defects can be recognized and controlled in time, and the reliability and timeliness of assessment are improved.

A smart home monitoring device is configured to plug into an electrical outlet and includes a multi-sensor array, excluding audio and video sensors, for monitoring environmental conditions, behavioral patterns, and odor-related indicators within a monitored environment. The device includes Bluetooth Low Energy communication with angle-of-arrival capability to enable sub-meter asset tracking, and supports home automation control through programmable automation frameworks such as IFTTT or Node-RED to initiate safety responses. The system further provides multi-tiered burglar detection through correlated event analysis across multiple sensor inputs, and generates health-related risk indicators based on detected patterns. Visual indicators, including multi-color LED illumination and blinking patterns, are used to communicate alert types and severity levels. The system enables non-intrusive monitoring while supporting proactive safety, security, and health management.

The invention relates to the technical field of logistics data analysis, and discloses a logistics data analysis decision-making system and method based on a cloud platform, and the method comprises the steps: obtaining a multi-source data set, carrying out the time-space standardization processing, and calculating a credibility score; constructing a logistics network graph, predicting the state of each node in the logistics network graph, and identifying bottleneck nodes and key paths; order time sequence features and external factor features are extracted, and the logistics demand quantity of each region is predicted; constructing a distribution task set, predicting road section passing time, and optimizing a distribution path and task distribution; carrying out supply and demand matching analysis, and carrying out resource scheduling optimization; identifying a key scene, simulating an execution process and calculating expected performance and risk indexes of the decision scheme; according to the method, intelligence, automation and optimization of logistics decision making are realized, logistics operation efficiency is improved, operation cost is reduced, service quality is improved, and the method has important theoretical value and practical significance.

The invention relates to a multi-factor dynamic correction investment benefit intelligent evaluation system, in particular to the field of investment evaluation, and aims to capture macroscopic event and project risk index impact in real time and dynamically generate factor impact coefficients through an event-driven factor correction module, so that the response speed and sensitivity of a model to project sudden change are remarkably improved, and the evaluation efficiency is improved. The self-adaptive weight optimization module utilizes a deep reinforcement learning framework to continuously optimize factor weight distribution based on revenue after risk adjustment in a rolling time window so as to realize self-adaptive adjustment of an investment strategy and optimization of a revenue-risk ratio, and the abnormal conduction early warning module constructs a dynamic directed graph and combines a graph neural network technology to realize early warning of the abnormal conduction. According to the method, abnormal factor conduction paths can be perspectively recognized, systematic engineering risks can be effectively warned, a strategy execution module fuses weight optimization results and risk warning signals, rebalance is executed under the condition that investment cost and risk control constraints are strictly met, and finally a project configuration scheme giving consideration to revenue, robustness and real-time performance is generated.

The invention relates to the technical field of software engineering and automatic testing, and discloses a self-adaptive multi-dimensional risk testing strategy design method and system and a storage medium. The method comprises the following steps: triggering a risk assessment service through a CI / CD assembly line, preprocessing original data obtained by a plurality of heterogeneous data sources, and generating a risk feature vector; outputting a multi-dimensional risk index by using the risk assessment model, fusing the multi-dimensional risk index and the dynamic weight thereof based on an adaptive weighted fusion model, and calculating a comprehensive risk score; dynamically updating a risk score threshold value used for dividing the risk level, automatically generating a test strategy matched with the current risk level according to a preset mapping rule of the risk level and the test strategy, and calling the test suite to automatically execute the test strategy; and collecting a test execution result and online operation data as training data, and regularly optimizing parameters of the risk assessment model. The test strategy can be dynamically adjusted according to the real-time risk situation.

The invention relates to the technical field of network security, in particular to a network security situation awareness method and system based on a large model and threat assessment. The method comprises the following steps: firstly, acquiring and standardizing multi-modal security data in a cloud service environment in real time, distributing a behavior modal cluster for security event metadata through clustering analysis, and generating a security feature vector containing business semantics and behavior dynamic features based on a cluster center relocation technology; then constructing a local situation map reflecting asset topology and an access link by using a cloud security association model; semantic reasoning is performed on the atlas through a large language model, an attack intention is recognized, and an attack path is predicted; and finally, combining the path probability, the asset value and the vulnerability feature to quantitatively calculate a risk index, and automatically generating a response strategy. Semantic compression of massive logs is realized through modal clustering, and the calculation bottleneck of processing original data by a large model is overcome; and in combination with graph correlation and large model reasoning, the crossing from passive warning to active intention prediction is realized.