43 results about "Safety knowledge" patented technology

The invention discloses a sensitive data identification method, system and device based on rule matching and LLM deep semantic analysis, a medium and a product, and relates to the field of artificial intelligence, the method comprises the following steps: preprocessing data in an information data set; the information data set comprises multi-industry corpora, personal information and safety knowledge; the multi-industry corpora cover structured data format features and unstructured text semantic scenes; according to the preprocessed data, a LoRA technology is adopted to carry out field adaptation fine tuning on the large language model; the large language model comprises a rule matching layer and an LLM depth recognition layer; the rule matching layer screens and identifies the structured data; the LLM depth recognition layer recognizes unstructured data and implicit sensitive data; performing semantic recognition on the to-be-tested data according to the fine-tuned large language model, and outputting sensitive data; the sensitive data comprises information such as an identity card number, a home address and a mobile phone number, and the sensitive data can be accurately recognized.

The invention discloses a personalized safety training content generation method, system and device and a medium. The personalized safety training content generation method comprises the steps that answer data, historical performance data and future operation tasks of operators in safety training are collected; safety knowledge weak points and learning behavior modes of the operating personnel are identified, and related risk association points are determined; based on the safety knowledge weak point, the learning behavior mode and the risk association point, constructing a structured user portrait, based on a deep neural network model fused with an attention mechanism, screening basic training content matched with the user portrait from a dynamic training material library, and adjusting the basic training content to obtain a basic training result; according to the method, the personalized target safety training content is generated, and the safety training is performed on the operator according to the target safety training content, so that the generation efficiency of the personalized safety training content can be improved.

The invention discloses an electric power safety knowledge testing method based on a large language model and enhanced retrieval generation. The method mainly comprises the steps of constructing a vector knowledge base, constructing a dynamic interaction generation workflow framework driven by a multi-role large language model, performing safety knowledge testing and quantifying the electric power operation safety awareness of a user. According to the invention, the problem with high scene adaptation degree can be dynamically generated in combination with the scene of the operation work order and the user interaction result, so that the test problem accuracy is improved; the quality of test content is improved through cooperative work of a multi-role large language model and in combination with an enhanced retrieval generation module; a dynamic adjustment strategy is implemented according to a user feedback mechanism, so that knowledge blind spots of the user can be effectively identified; the safety awareness of the user is quantified, targeted consolidation is carried out, and the electric power safety knowledge level of the user is improved.

The invention provides a construction site safety management and control method and system based on enhanced retrieval generation, and belongs to the technical field of data processing. Comprising the following steps: constructing a construction safety knowledge base, and vectorizing text data in the knowledge base; field self-adaptive fine tuning is carried out on the visual language large model through a low-rank self-adaptive method, and perception, cognition and decision-making processes of construction site safety management and control are executed: in a perception stage, construction safety hidden dangers are dynamically identified and positioned, and violation descriptions containing hidden danger types are generated; in the cognition stage, an enhanced retrieval generation process is started, and construction specification information associated with hidden dangers is obtained through multi-level semantic matching and retrieval; in a decision-making stage, based on violation description and construction specification information, a visual language large model is driven to automatically generate a structured security log through a preset structured cue word template. According to the invention, unsafe behaviors of the construction site can be timely and accurately warned, and effective data support is provided for the safety of the construction site.

The application provides a full-chain meat safety knowledge graph construction method and retrieval device, which has the following characteristics: steps S1-S2, entity and relationship extraction is performed on all meat safety standard files; step S3, meat product nodes and corresponding subordinate nodes are constructed and relationships are set; steps S4-S6, corresponding multistage nodes are set for each subordinate node; step S7, the extracted entities and relationships are assigned to the corresponding nodes; step S8, CCP selection nodes including specific hazards and corresponding control measures are set according to the HACCP principle; steps S9-S10, non-directly connected nodes are directly connected as needed through artificial relationships. In summary, the method can construct a more comprehensive meat safety knowledge graph, and the retrieval device can more quickly, conveniently and accurately retrieve and display the data of the meat safety knowledge graph.

The invention discloses a method for improving an industrial safety education training effect based on an AI large model. The method comprises the following steps: analyzing, vectorizing and storing latest industrial safety related documents to obtain a safety knowledge base; the related documents comprise industrial standard specifications, laws and regulations and enterprise specifications related to industrial safety; the knowledge base comprises a plurality of sub knowledge bases with different themes and different types; matching an associated target sub-knowledge base from the safety knowledge base based on the training plan, so as to generate an employee training task based on the target sub-knowledge base; based on the training task, utilizing an AI large model to generate a training problem in the training process, and using the training problem for employee training to obtain a training result of the employee; and based on the training result, using an AI large model to generate test questions corresponding to different employees, and examining the employees based on the test questions to obtain an examination result. The training effect of employee safety education can be improved, and potential safety hazards are reduced.

This invention discloses a big data-based safety design management method for chemical projects, relating to the field of big data technology in chemical safety. The method includes: collecting multi-source heterogeneous data and preprocessing it to form a project baseline dataset; collecting design change data and performing differential comparison with a safety knowledge graph to identify differential objects; starting from the differential objects, performing dependency propagation analysis along the safety dependencies in the safety knowledge graph to output affected objects and risk propagation chains; determining the review order of affected objects based on the risk propagation chains, sequentially extracting the specifications to be reviewed that match the affected objects from the specification knowledge base, determining whether the affected objects meet the specifications to be reviewed, and outputting the specification review conclusion. This invention realizes dynamic and precise management and automated compliance review of chemical project safety design, shifting from experience-driven to big data-driven management.

This invention discloses a method and system for power operation safety monitoring and question answering based on a multimodal large model. The method includes the following steps: a data acquisition step, acquiring visual data and safety regulation text data from the power operation site; a multimodal semantic fusion step, generating a unified fused semantic vector through feature extraction and cross-modal alignment fusion technology; a knowledge reasoning step, matching the fused semantic vector with a pre-constructed power safety knowledge graph and performing compliance reasoning based on a graph neural network; and an intelligent question answering generation step, inputting the fused semantic vector and the reasoning results into a large language model to generate natural language question answering information. This invention, through the synergistic innovation of multimodal semantic fusion, knowledge graph reasoning, and a large language model, solves the technical problems of insufficient semantic understanding, disconnect between the question answering system and the field, and unstructured knowledge representation in existing technologies, achieving fully automated and interpretable intelligent safety supervision from perception to cognition.

The present application relates to the technical field of knowledge graph construction, in particular to a knowledge graph construction method based on semantic network. Production data set of knowledge graph to be constructed is acquired, safety knowledge recognizer is used to perform safety knowledge recognition, and safety knowledge word vector set is obtained; basic safety knowledge graph is constructed; scale coefficient set and proportion coefficient set are obtained; according to training features of the safety knowledge recognizer, accident coefficient set and reason coefficient set are obtained, the proportion coefficient set is corrected, accident labeling coefficient set and reason labeling coefficient set are processed, edge scale of each accident type word vector and reason type word vector is labeled and processed, and safety knowledge graph is obtained. The present application increases the visualization degree of safety knowledge graph, realizes clear presentation of logical association of “accident type-reason type”, and quantifies accident severity, reason influence weight and the association strength of both.

This invention provides a construction AI-based risk analysis and decision-making method for safety management. Through a knowledge base construction mechanism involving intelligent parsing, vectorized storage, and incremental updates, it transforms scattered safety knowledge from projects into digital assets, enabling the continuous accumulation of enterprise safety knowledge assets. A RAG-enhanced large language model question-answering system equips each on-site personnel with a personal AI expert, providing near-instantaneous responses to on-site issues and improving efficiency. By employing a performance quantitative analysis model, an Attention-LSTM time-series risk prediction model, and a hazard clustering analysis algorithm, it uncovers management shortcomings and predicts risk trends from massive amounts of data, making safety management decisions more data-driven and moving away from experience-based, extensive management. AI time-series prediction identifies high-risk areas and hazard trends seven days in advance, generating actionable control recommendations, shifting safety management from post-event remediation to pre-event prevention, significantly reducing the probability of accidents.

The invention relates to the technical field of knowledge graph construction, in particular to a knowledge graph construction method based on a semantic network. Obtaining a production data set of a to-be-constructed knowledge graph, and performing safety knowledge recognition by adopting a safety knowledge recognizer to obtain a safety knowledge word vector set; constructing a basic security knowledge graph; obtaining a scale coefficient set and a proportion coefficient set; according to training features of the safety knowledge recognizer, analyzing to obtain an accident coefficient set and a reason coefficient set, correcting and processing the proportion coefficient set to obtain an accident labeling coefficient set and a reason labeling coefficient set, and labeling edge scales of each accident type word vector and each reason type word vector to obtain a safety knowledge graph. According to the method, the visualization degree of the security knowledge graph is increased, the logic association of'accident type-reason type 'is clearly presented, and the accident severity, the reason influence weight and the association strength of the accident severity and the reason influence weight are quantified.

The invention relates to the technical field of artificial intelligence, in particular to an intelligent evaluation method for a harbor potential safety hazard treatment effect, and the method comprises the steps: constructing a harbor safety knowledge graph based on harbor static data and harbor potential safety hazard treatment dynamic data; performing text analysis based on semantic understanding of a port potential safety hazard treatment scheme and treatment records, and converting the text into a structured action sequence; creating a copy of the port safety knowledge graph, and mapping the structured action sequence to the copy of the port safety knowledge graph to obtain a simulation evaluation initial state graph; performing simulation deduction on the simulation evaluation initial state diagram by using Monte Carlo tree search to obtain probability distribution of a deduction result; and determining an evaluation score of the harbor potential safety hazard treatment effect based on each deduction result included in the probability distribution of the deduction results and the probability score thereof. Therefore, the accuracy of the evaluation score is improved.

The invention relates to the field of petroleum and petrochemical safety, and discloses a JSA operation safety analysis method and system based on an AI agent, and the method comprises the following steps: receiving and complementing operation description through a user layer, and obtaining complete context information; processing the context information to extract key information and converting the key information into structured data; generating an ordered tool calling sequence based on the key information; calling tools in sequence and carrying out verification, evaluation and conflict resolution on a plurality of returned results so as to integrate and form a unified analysis conclusion; and filling the analysis conclusion into a JSA document template to generate a structured report. According to the method, a system architecture of a user layer, an agent layer, a memory storage layer and a tool layer is constructed, the agent layer is utilized to autonomously plan tasks, various tools such as historical experience retrieval and operation safety analysis are called, specific risks in historical cases and general safety knowledge are integrated, and a JSA report with accurate content is generated; and the efficiency and the quality of operation safety analysis are improved.

The invention relates to the technical field of data security training and evaluation, and discloses a holographic perception evaluation method and system for data security, and the method comprises the steps: constructing a security mapping model of multiple sensory channels, and mapping data security risk features to visual, auditory and tactile perception features; personalized perception intensity adjustment and semantic correlation optimization are carried out based on user cognition characteristics; dynamic simulation and interactive experience of known and unknown security threats are realized through a potential safety hazard simulation technology; establishing a unified cross-scene perception evaluation framework; according to the method, the limitation of a traditional single sensory channel is broken through, multi-dimensional cooperative transmission of safety knowledge is realized, the visual understanding and deep cognitive ability of a user on data safety risks are improved, and a systematic solution is provided for improving the overall safety awareness of an organization.

The invention relates to the technical field of Internet of Things and machine learning, and particularly discloses a laboratory safety whole-process dynamic management system based on the Internet of Things. The system comprises an Internet of Things sensing layer, an edge calculation layer, a data fusion and knowledge construction layer, a dynamic risk assessment and decision-making layer and an execution and feedback layer. A dynamic security knowledge graph is constructed by fusing multi-source heterogeneous data, risk assessment is performed by integrating mode recognition and probability graph reasoning, a hierarchical management and control instruction is generated, and finally closed-loop management is formed through an execution layer, so that real-time and comprehensive perception, accurate risk assessment and adaptive dynamic management and control of a laboratory security state are realized.

The embodiment of the invention provides an intelligent auxiliary compilation method and system based on an electric power safety knowledge large model, and belongs to the technical field of electric power auxiliary compilation. The intelligent auxiliary compilation method comprises the following steps: acquiring electric power safety knowledge data; constructing an electric power safety knowledge base according to the electric power safety knowledge data; selecting a basic large model, and finely adjusting the basic large model by adopting the electric power safety knowledge base; acquiring real-time content input by a current user; inputting the real-time content into the basic large model to obtain a compiled first draft; checking the compiled first draft, and obtaining a compiled final draft; and inputting the programmed final draft into the electric power safety knowledge base. According to the method, data updating can be effectively and continuously carried out on the electric power safety knowledge base, so that the first draft compiling accuracy is improved.

This invention discloses an artificial intelligence-based campus safety hazard monitoring system, belonging to the field of campus security technology. Specifically, it includes: collecting campus environmental data through a multimodal sensor network and converting it into a low-level physical event stream; using causal rules in a pre-set campus safety knowledge base for matching and reasoning to generate safety hazard hypotheses containing expected consequence event sequences and initial confidence levels; identifying discrepancy nodes and selecting key verification evidence types through comparative analysis; calculating the temporal matching degree based on the enhanced collection of verification event sequences to generate risk assessment values ​​corresponding to each safety hazard hypothesis; and automatically triggering a standardized handling protocol when the risk assessment value exceeds a threshold. This invention, through a multi-level causal reasoning and verification mechanism, effectively solves the gap between physical perception signals and behavioral semantics in existing technologies, achieving accurate identification and rapid handling of campus safety hazards.

This invention discloses an intelligent assessment and evaluation system for power safety training based on machine learning, comprising a data acquisition module, a behavior analysis module, a knowledge assessment module, a comprehensive evaluation module, and a feedback generation module. Through the collaborative work of multiple modules, this invention achieves intelligent management of the entire power safety training and assessment process. The system utilizes machine learning technology to conduct multi-dimensional analysis of trainees' theoretical knowledge and practical behavior, significantly improving the objectivity and accuracy of the assessment. By constructing a power safety knowledge graph and an operational behavior standard library, the system can comprehensively cover various safety scenarios and achieve accurate assessment of trainees' safety awareness and operational skills based on deep learning-based multimodal data fusion analysis. The personalized feedback module can generate targeted training suggestions based on trainees' weaknesses, effectively improving training outcomes. Simultaneously, the system supports remote online assessment, greatly reducing training costs and time investment.

The application provides an AI fusion management method and system of a multi-dimensional safety production knowledge base, a storage medium and a program product, relates to the technical field of safety production management, and comprises the following steps: in the case that multi-source heterogeneous safety production source data is acquired, a multi-level safety knowledge graph and a bidirectional index mapping channel are constructed by using the safety production source data, the multi-dimensional safety production knowledge base comprises the multi-level safety knowledge graph and the bidirectional index mapping channel; in response to a context awareness request triggered by a terminal, real-time running parameters of a target device and behavior intention characteristics of a user are collected; the real-time running parameters and the behavior intention characteristics are mapped by using the bidirectional index mapping channel, so as to lock a target risk source node in the multi-level safety knowledge graph; the target risk source node is subjected to multi-hop graph traversal deduction according to the topological association characteristics of the multi-level safety knowledge graph, and a target safety protection strategy is generated; and the target safety protection strategy is used to perform a contextualized knowledge pushing operation on the terminal.

This invention relates to the field of network and information security technology, specifically to a risk intelligent decision optimization method based on security knowledge graph reasoning. First, multi-source security data is collected to construct a hierarchical security knowledge graph integrating behavioral and tactical relationships. Then, real-time security event sequences are mapped to initial behavioral relationships, and horizontal association expansion and vertical tactical abstraction are performed within the graph to generate causal attack paths with influence weights. Finally, the causal attack paths are input into a path simulator, and under defense resource constraints, the optimal adversarial decision sequence is solved based on Markov decision processes and reinforcement learning, driving automated response execution. This invention achieves visualized and quantitative evaluation of attack chains and global optimization configuration of defense strategies, improving proactive defense capabilities under limited security resources.

The invention provides a menu review method and a computer program product, and relates to the technical field of large language models. The method comprises the following steps: firstly, carrying out semantic analysis on an obtained original menu to obtain a structured analysis result; converting the structured analysis result into a standard quantification form to obtain a cooking instruction set; based on the cooking equipment information specified by the user, the original menu and a pre-created multi-source knowledge base, multi-dimensional safety examination is conducted on the cooking instruction set, an examination result is obtained, and the multi-source knowledge base comprises multi-dimensional safety knowledge; and finally, performing comprehensive assessment based on the review result, and outputting a risk assessment report. According to the method, after the cooking instruction set is obtained through conversion, the safety of the menu is ensured through multi-dimensional safety examination and final comprehensive risk assessment.

The application discloses a BAS system attack load automatic generation method and system based on a large language model, and belongs to the technical field of network security. In order to solve the problems that attack load generation depends on manual work, has poor compliance and is difficult to break through dynamic defense, the BAS attack instruction or environment fingerprint is received and modeled as a decision process; search enhancement technology is used to fuse safety knowledge to form prompt information; an initial load is generated by a strategy network under the constraint of a predefined syntax rule; a reward model is trained based on target environment feedback to evaluate bypassing capability; and the strategy network is updated by using a reinforcement learning mechanism and knowledge dynamic backfilling is realized. The application can significantly improve the simulation degree and bypass success rate of the BAS system, realize automatic iteration of attack strategies and mining of new vulnerability variants.

The invention provides a training and risk assessment method, device and system for an electricity utilization safety assessment model, and relates to the technical field of electricity utilization safety. The method comprises the following steps: constructing different types of small models based on different data types to form small model thinking bodies; performing primary training on the small model thinking body based on obtained historical operation data of each power device and a received power utilization safety knowledge base from a cloud side large language model to obtain a small model fusion body; and carrying out secondary training on the small model fusion body based on historical operation data of each power device so as to improve the small sample processing capability of the small model fusion body, and taking the trained small model fusion body as the power utilization risk assessment model. According to the invention, the problems of low inspection accuracy and low power utilization inspection service efficiency of the existing power utilization safety inspection mode are solved.

The invention relates to the technical field of industrial safety production training, and discloses a full-chain safety training method, system and device based on VR practice and a medium, and the method comprises the steps: collecting the multi-source data of an employee, and processing the original multi-source data through a feature extraction algorithm; constructing a security knowledge-risk heterogeneous graph, fusing structured and unstructured data by using a graph neural network, and updating embedded vectors of employee nodes; generating a dynamic employee portrait based on the embedded vector of the employee node, generating a personalized training path according to the matching degree of the portrait and the post risk gap, and dynamically adjusting the path; dynamically arranging a VR training scene through a reinforcement learning engine, adjusting scene complexity in real time, and deploying a sensor in a VR environment to collect employee operation data; and analyzing employee interaction behaviors according to the multi-modal data, updating employee portraits and optimizing training paths. According to the method, the safety awareness and the emergency disposal capability of employees are improved based on a multi-dimensional perception interaction learning environment of the Internet of Things.

This invention belongs to the field of information retrieval technology and discloses a method and system for automatically generating subway construction safety reports. The method includes: using an intelligent agent to parallelly fuse daily inspection data, historical safety knowledge, and construction specification knowledge; transforming the daily safety inspection record into three knowledge representation forms: vector, natural language description, and knowledge graph; and calculating similarity between each representation and historical knowledge to obtain a preliminary safety report; inputting the preliminary safety report and natural language data into a specification retrieval intelligent agent, which outputs the core questions for each part of the preliminary safety report; the specification retrieval intelligent agent performs multi-dimensional decomposition and knowledge completion on the core questions to obtain the corresponding question set, and calls the MCP service for subway construction specification retrieval to obtain the clause specifications for each part of the preliminary safety report; and generating the daily subway construction safety report based on these clause specifications and the preliminary safety report. This invention enables the automated and intelligent generation of subway construction safety reports.

The invention provides a railway operation safety assessment method and device, and relates to the technical field of data processing, and the method comprises the steps: collecting operation data of a target railway; generating a railway safety knowledge graph based on the operation data; performing feature extraction based on the railway safety knowledge graph to generate a text feature vector; performing risk calculation based on the railway safety knowledge graph and the text feature vector to generate a risk list; the risk list is disassembled, and a corresponding check item list is generated; and based on the inspection item list and a preset inspection standard, performing security assessment to obtain an assessment result. According to the railway operation safety assessment method provided by the invention, credible assessment of railway operation safety is realized, the traceability and the real-time performance of an assessment result are ensured, and the efficiency and the precision of railway operation safety assessment of different regions are improved.

The application discloses a safety risk assessment system based on railway station yard information, relates to the technical field of railway transportation safety, and solves the technical problem that the existing technology is time-consuming and low in efficiency in constructing a risk prediction model for a newly-built station yard and a low-risk station yard, and a small amount of training data is often used, so that the deployed safety risk assessment system is time-consuming and low in efficiency; a railway safety knowledge graph is constructed according to station yard data, and a risk prediction model is generated based on the railway safety knowledge graph; after a prediction risk event and a prediction risk level are generated according to the risk prediction model, an alarm signal is generated, the railway safety knowledge graph and the risk prediction model are fused to enhance the risk identification precision and realize early warning; a multi-source heterogeneous data time sequence coding technology is adopted to realize comprehensive analysis of unified cross-frequency data; a target station yard prediction model is trained by using multi-source station yard risk data, the problem of insufficient training data of the target station yard is solved, and the precision and application efficiency of the safety risk assessment system are improved.

The application discloses a kind of based on the safety inspection method and system of multi-model cooperation, which comprises: the multi-dimensional detection data of target area is preprocessed, and detection image and accompanying metadata are obtained;Detection image and accompanying metadata are input to lightweight target detection model, and target area image and structured metadata are obtained;Detection image and accompanying metadata, target area image and structured metadata are respectively input to multi-modal large model, and global semantic information and local semantic information are obtained;Global semantic information is searched in external professional knowledge base, vector matching and cross-modal adapter retrieval are carried out, and safety knowledge item is obtained;Local semantic information, global semantic information and safety knowledge item are input into multi-modal large model, and safety hazard area, safety hazard content, improvement suggestion and safety regulation are obtained.The application has high positioning accuracy, high output reliability, and has professional field knowledge guarantee.

The application provides a quadruple extraction and knowledge graph construction method for campus safety emergency plan, and belongs to the technical field of campus safety emergency management. The method comprises the following steps: adopting a set of structured extraction method taking database mode constraint quadruple modeling as the core; facing the long text, strong structure and strong role chain scene of the campus safety emergency plan, converting the unstructured plan text into a standard quadruple composed of place, safety event, personnel and action, and completing the instance layer construction of the campus safety knowledge graph on the basis; the main process comprises five steps of clause perception cutting, quadruple direct extraction, structure legality scoring filtering, normalization correction and three-field composite key fusion, and graph warehousing and application verification. The application solves the problems of recovery and graph entry of campus emergency response facts.

1. The name of the design product: safety education equipment (oxyacetylene operation B). 2. The use of the design product: for learning the relevant safety knowledge of oxyacetylene, including composition introduction, knowledge popularization, knowledge examination, and introduction of each component, matters needing attention, use of popular video of complete equipment, selection of answer scoring, etc. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.