200 results about "Industrial security" patented technology

The invention relates to the field of data processing, in particular to a security agent construction system and method based on a large model. Constructing a low-temperature liquid leakage knowledge base; collecting normal operation condition data, and training the leakage sensing model in an unsupervised mode; configuring a large model with a tool calling interface as a security agent; multi-modal data are synchronously collected in real time, a preliminary abnormal alarm is generated through a perception model, and associated information is retrieved; fusing the multi-modal information by the intelligent agent to carry out leakage prediction analysis and comprehensive research and judgment; and for the confirmed leakage event, a professional tool is automatically called to carry out quantitative consequence simulation, and a scenario disposal scheme is generated. According to the invention, the dependence on scarce leakage samples is effectively overcome, and the detection reliability is improved; through knowledge enhancement and multi-modal fusion, the research and judgment accuracy is greatly improved; full-process automation from early perception, intelligent analysis to consequence severity division is realized, and the intelligent level and emergency response efficiency of industrial safety monitoring are remarkably improved.

The invention provides a multi-modal fusion causal analysis method and platform for industrial safety production, and the method comprises the steps: carrying out the preprocessing of multi-modal key data from an industrial site, and generating a standardized time series data matrix; based on the standardized time sequence data matrix, combining industrial physical rules and operation process semantics to construct an industrial safety causal atlas; based on a self-supervised learning mechanism, optimizing the structure and edge weight of the industrial safety causal atlas by using the standardized time sequence data matrix to obtain an optimized causal reasoning model; and performing path-level risk activation judgment and accident causal traceability by using a real-time standardized time sequence data matrix constructed by the causal reasoning model and real-time inflow multi-modal key data to generate a risk alarm and traceability report. According to the method, the perspectiveness, the accuracy and the interpretability of industrial safety monitoring are remarkably improved.

The invention discloses an industrial internet risk analysis system and method based on a knowledge graph, and the method comprises the steps: collecting data, and defining a multi-source original data flow and the knowledge graph; performing data preprocessing, unifying sample data, and performing mapping splicing on each mode in the multi-source original data stream; constructing, storing and updating a knowledge graph; carrying out risk modeling based on attention, and introducing an attention mechanism and a causal relationship model; carrying out risk scoring and inference; and performing risk early warning and visualization, outputting an early warning list, a propagation path diagram and key hub nodes, and performing butt joint with an existing industrial safety platform for display through an analysis report interface. The invention provides an industrial internet risk analysis scheme which not only can fuse the structural modeling capability of the knowledge graph, but also can utilize a self-attention mechanism to improve the key risk feature recognition precision.

The invention discloses an agent-based security risk assessment and operation management integrated method and system, and relates to the technical field of industrial security production and artificial intelligence, and the method comprises the steps: obtaining enterprise security production operation data, carrying out the preprocessing of the data, obtaining a standardized data set, and carrying out the deep transfer learning through an agent, and extracting scene risk features; calculating a propagation entropy value of the risk mapping data and constructing a propagation path, and performing dynamic time sequence fusion to generate a risk assessment result; generating a scene control instruction based on group game operation and performing optimization to obtain a baseline check instruction; operating data are collected, the conduction probability is calculated, a safety response rule is generated, linkage optimization is carried out, and a cooperative defense strategy is formed. According to the invention, risk feature extraction, propagation path analysis and cooperative prevention and control are realized through the intelligent agent, and the intelligent management level of enterprise safety production is improved.

The invention discloses an industrial safety detection method combining time alignment and motion consistency, and belongs to the technical field of industrial safety monitoring. According to the method, through seven steps of multi-camera data acquisition and preprocessing, target feature extraction, coordinate projection, instantaneous speed estimation, multi-dimensional similarity fusion, soft matching optimization and model training and target trajectory generation, the problem of space-time inconsistency during multi-camera cooperative detection in an industrial scene is solved. A camera clock offset parameter is innovatively introduced to correct a cross-camera time difference, appearance, geometry and time prior multi-dimensional similarity are fused, a multi-loss function optimization model is constructed, and cross-camera target accurate matching and long-time trajectory tracking are realized. High-precision technical support is provided for personnel safety monitoring and equipment abnormal behavior detection in an industrial scene, and the industrial production safety protection capability is improved.

The invention relates to a gas cylinder filling method, an early warning method, an assessment method, a system and a medium in the technical field of industrial safety production. The gas cylinder filling safety assessment method based on voice analysis comprises the following steps: acquiring acoustic data of an operator; and analyzing the acoustic data and calculating to obtain a knowledge proficiency score Spro. And acoustic data are extracted, and a psychological quality score Semo is obtained through calculation. Collecting motion posture data of the gas cylinder and calculating to obtain an operation stability score Ssta; and carrying out weighted fusion on the Stro, the Semo and the Ssta to obtain an assessment score Stotal. And setting a score line based on the assessment requirement, and if the Stotal is greater than or equal to the score line, determining that the operator passes the assessment. According to the invention, through innovative multi-mode assessment, the ability of the operator is changed from feeling-based ability to data-based ability, the final result can be known, and the specific weak link can be known according to the specific condition of each link, so that the psychological quality and emergency response ability of the operator can be effectively assessed.

The invention provides a wide dynamic range methane detection system based on synchronous accumulation and system reconstruction. The device is characterized by comprising a modulated laser (1), a gas absorption cell (2), a photoelectric detector (APD) (3), a variable gain trans-impedance amplifier (TIA) (4) based on a digital potentiometer, a dual-threshold voltage comparator circuit (5) and a microcontroller processing unit (6). The device can be used for large-dynamic-range real-time monitoring of the methane gas concentration, and the problem that a traditional TDLAS system is prone to saturation or limited in detection limit when the concentration span is large is effectively solved. The method can be widely applied to the fields of gas concentration detection, industrial safety control and the like.

The invention relates to the technical field of industrial safety control, in particular to a dual-system redundancy safety circuit and a self-diagnosis method, the dual-system redundancy safety circuit is composed of a first safety system, a second safety system, a test load and a switching unit, and the first safety system and the second safety system realize information interaction through an optical coupling isolation communication interface. According to the self-diagnosis method, an encrypted diagnosis trigger pulse sequence is generated through the second safety system, a test contactor of the first safety system is driven to be connected to a test load, and a main safety loop generates simulated real safety action current; the second safety system performs high-speed sampling, filtering, correlation analysis and response delay extraction on the current waveform to obtain a diagnosis result of the contact performance and the load capacity; and when an abnormity is detected, the second safety system controls the switching unit to execute safety switching through the solid-state switching output point, so that control right taking-over is realized, and the reliability and the online diagnosis capability of the safety loop are improved.

A cloud-based industrial safety system executes safety applications and interfaces with industrial assets on the plant floor using software encoded processing (SEP). The use of SEP ensures safety reliable execution of safety applications and data communication software by redundantly executing native code, thereby implementing a level of software-based fault detection that is independent of the hardware on which the safety application operates. This allows the cloud-based industrial safety applications to achieve at least SIL3 safety ratings for safety services even though the safety applications are executed using standard COTS hardware that is commonplace in the server market. The use of SEP to reliably execute safety software on the cloud can make possible a wide variety of safety applications that would be difficult to implement using purely localized industrial safety systems.

The invention relates to the technical field of information technology and industrial safety management, and discloses a hazardous waste whole-process risk prevention and control method and system. The method comprises the following steps: acquiring a real-time monitoring data set of each link of the hazardous waste management chain; calculating the fluctuation amplitude of the environment change indexes and analyzing the distribution characteristics of the operation deviation marks; when the two meet a preset condition, judging the initial signal intensity of risk upgrading by utilizing a classification model; screening high-intensity signals and inputting the high-intensity signals into the regression model to obtain a risk transformation trend; matching the risk conversion trend with a historical risk mode to generate a dynamic monitoring adjustment parameter; if the parameter indicates that the monitoring sensitivity is insufficient, fusing additional sensor data to generate an enhanced monitoring data set; and tracking a risk propagation path based on the enhanced data set, and predicting an occurrence position and a time point of a risk peak value. According to the method, the real-time performance, the accuracy and the response speed of hazardous waste management are improved, and the effect of risk prediction and management is remarkably enhanced.

This invention relates to the field of industrial safety technology, and in particular to a safety protection device, comprising a cabinet with an installation port. Baffles are installed on the inner walls of both sides of the installation port near the inner side of the cabinet, and an openable door is provided on the inner wall of the installation port near the outer side of the cabinet. A first type of sensor is disposed between the door and the baffles to detect the opening and closing state of the door and output a first state signal. A second type of sensor is disposed at the installation port to detect intrusion into the area of ​​the installation port and output a second state signal. A safety controller is mounted on the baffles. This invention integrates physical access control sensing and open area photoelectric sensing through the first and second types of sensors, achieving comprehensive protection with both contact and non-contact warnings, thus achieving a dual warning effect to prevent personnel from entering dangerous areas.

The invention discloses an intelligent electrostatic monitoring system and method based on state self-sensing, and belongs to the technical field of industrial safety monitoring, and the system comprises a pipeline, a Faraday cup sampling mechanism driven by a cylinder, a multi-sensor module and an intelligent control unit. The core of the method is that air pressure and displacement signals of an air cylinder are collected in real time and compared with a pre-stored reference model, the motion state of the air cylinder is intelligently diagnosed, a hierarchical fault-tolerant control strategy is triggered according to a diagnosis result, and self-recovery or shutdown protection is achieved. And the sampling frequency can be adaptively adjusted. The Faraday cup adopts a double-layer shielding structure with height difference and a connector with a round top and a square bottom so as to resist interference; and a top-speed opening and closing mechanism is combined with a multiple flexible anti-blocking sealing structure, so that the problems of mechanical blocking and sealing are effectively solved. According to the invention, the conversion from passive maintenance to predictive maintenance is realized, and the reliability, availability and intelligent level of monitoring equipment in a severe industrial environment are greatly improved.

The application relates to an intelligent alarm method for industrial safety monitoring and belongs to the technical field of industrial safety production monitoring, which solves the problems of insufficient scene understanding ability and high false alarm rate of a traditional AI monitoring system. The method comprises the following steps: acquiring an industrial production scene video stream in real time and extracting a key frame image; determining a scene description structured text based on the key frame image and a scene classification prompt word by using a multi-modal visual language model; acquiring a determined high-risk scene type based on the scene description structured text and a scene reasoning prompt word by using a large language model; determining a violation feature of the scene based on the determined scene type and a violation feature prompt word by using a model calling platform; determining a judgment conclusion about the violation feature based on the violation feature of the scene and the key frame image by using a multi-modal visual language model; and generating an alarm event according to the judgment conclusion. The application realizes an intelligent alarm method for industrial safety monitoring.

This invention discloses a method and apparatus for detecting safety violations based on a visual language model, relating to the field of industrial safety monitoring technology. The method includes: collecting video data containing safety violations from work sites in multiple industrial sectors and preprocessing it; enhancing the preprocessed data using complementary data augmentation strategies; automatically generating safety analysis annotations using an open-source model after obtaining the enhanced data; constructing a standardized training dataset; fine-tuning the visual language model using LoRA fine-tuning technology; deploying the model online and processing the images input to the model using a behavior amplification module; inputting the processed images into the fine-tuned visual language model for safety violation detection inference; and generating a structured violation report and issuing real-time alarms after inference is completed. This invention can improve the efficiency and accuracy of detecting and identifying violations by workers in industrial sectors.

The invention discloses a circuit board crushing monitoring system and method based on image analysis, and belongs to the technical field of industrial safety monitoring. Image conditions of a circuit board are acquired, conditions of electronic elements on the circuit board are identified, and crushing equipment data are acquired at the same time; and evaluating the real-time state of the electronic component based on the identified electronic component condition on the circuit board, evaluating the explosion risk of the electronic component based on the real-time state of the electronic component and the cutting process, and evaluating the damage caused by explosion based on the explosion risk of the electronic component and the fragility condition of the crushing equipment. And corresponding damage early warning information is generated based on the damage caused by explosion, so that the safety and reliability of the circuit board crushing process are improved.

1. The name of the design product: industrial inspection dog charging house. 2. The use of the design product: for industrial security inspection dog charging station. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

The application provides a steam delivery pipeline leakage detection system, which comprises a metal conducting rod fixedly connected to the outer wall of a steam delivery pipeline, a sensor connected to the metal conducting rod, a signal analyzer electrically connected to the sensor, and a DCS control center electrically connected to the signal analyzer, and the sensor is used for collecting the vibration sound signal of the metal conducting rod. The steam delivery pipeline leakage detection system of the application can accurately and timely determine the leakage and the position of the leakage point by using the metal conducting rod connected to the steam delivery pipeline to transmit the vibration sound generated by the intense friction between the leakage medium and the pipeline wall, and the sound signal is not obviously attenuated in the transmission process due to the solid medium transmission, the transmission distance is long, and the sound signal is not easily interfered by other signals, the false positive rate of the leakage detection is effectively reduced, the operation and maintenance cost of the steam delivery pipeline is significantly reduced, and the industrial safety is improved.

The invention relates to the technical field of industrial safety, in particular to a coal mine operation scene anomaly detection method and system based on a causal wavelet resonance network. The method comprises the following steps: acquiring multi-modal sensing data and constructing a high-dimensional spatial-temporal characteristic field; performing transformation processing on the high-dimensional spatial-temporal characteristic field by adopting a causal wavelet resonance network, and extracting a real-time causal resonance spectrum; calling a baseline, a spectrum and a covariance matrix thereof which are learned in advance, calculating mahalanobis distances between the baseline and the real-time causal resonance spectrum and between the spectrum and the real-time causal resonance spectrum, and generating a space-time detuning diagram; and judging whether the detuning value of each time-space position in the time-space detuning diagram exceeds a preset threshold, if so, determining that the time-space position is abnormal, analyzing the causal wavelet type and parameter which contribute to detuning to the greatest extent, and generating alarm information capable of explaining abnormity, thereby improving the abnormity detection precision and improving the detection accuracy. And high-sensitivity sensing and interpretability judgment can be carried out on weak and unknown abnormal sprouts containing causal logic.

An industrial security system leverages generative artificial intelligence (AI) to automate the process of identifying software or hardware insecurities on industrial assets, generate recommendations for mitigating these vulnerabilities, and, where appropriate, deploy countermeasures to the vulnerable assets. By leveraging automated asset discovery, real-time asset and network monitoring, and generative AI-assisted vulnerability detection and remediation, the system can reduce the amount of time spent by security administrators in identifying and closing security vulnerabilities within their plant environments, and can alert administrators of potential security issues before those issues become critical.

The invention discloses an industrial safety production risk intelligent early warning system based on image recognition, and relates to the technical field of risk early warning, and the system comprises an acquisition module for deploying a multispectral visual sensor network, synchronously acquiring visible light, thermal infrared and depth image data, carrying out the real-time preprocessing, and generating a multi-modal image data stream, an extraction module, and a processing module. The multi-modal image data flow is input into a dynamic feature extraction network, risk features are extracted and optimized through a shared encoder and a task specific decoder, a risk feature map is output, an analysis module inputs the risk feature map into a causal reasoning mechanism, and the risk feature map is analyzed. Constructing a risk factor causal graph model, analyzing a risk propagation path through intervention calculation, and outputting a risk assessment report; according to the method, by constructing the dynamic feature extraction network based on the MAML framework, adaptive feature optimization of a multi-task scene is realized, and the limitation of lack of task specific optimization is effectively overcome.

This invention provides an intelligent safety management method and system for boiler walkway gratings, relating to the field of industrial safety technology. The method includes: S1, acquiring in real-time the obstruction signal sequence generated by an infrared beam sensor deployed at the grating inlet and the weight signal sequence generated by a pressure sensor deployed below the grating, and synchronizing the obstruction and weight signal sequences with timestamps; S2, inputting the synchronized obstruction and weight signal sequences into a pre-constructed lightweight spatiotemporal feature extraction network to extract multi-dimensional temporal features representing personnel flow, object presence, and weight distribution; and inputting these multi-dimensional temporal features into a risk fusion inference engine built based on Dempster-Shafer evidence theory. This system can eliminate perceptual ambiguity, overcome decision-making rigidity, and mine data value in the harsh edge environment of boiler walkways, achieving progressive intelligent safety management from precise risk identification to dynamic and flexible assessment and early hazard insight.

The application discloses a real-time leakage positioning monitoring system and method based on electrical resistance tomography, relates to the technical field of process tomography and industrial safety monitoring, and solves the technical problem that it is difficult to intelligently identify, three-dimensionally position and quantitatively evaluate leakage events; the system comprises a sensor array, a multi-channel data acquisition and excitation module, a real-time data processing and image reconstruction module, a leakage intelligent identification and positioning module, a man-machine interaction and data management module; the core lies in that a multi-layer nested annular array and a sparse measurement strategy are adopted to improve data acquisition efficiency, a physical information neural network is introduced to reconstruct an image, a measurement topological coding matrix and a physical residual correction module are used to reconstruct the image, and finally, a time-space dual-domain leakage probability field model is constructed, a virtual sensor network and an improved D-S evidence reasoning framework are combined to intelligently identify and position a leakage diffusion mode.

The invention relates to a method for positioning at least one real sensor, in particular a safety sensor or a camera for industrial safety applications, in a real environment of an industrial plant by means of an augmented reality view based on the real environment of the industrial plant, wherein at least one simulated parameter of a virtual sensor is visualized for a user in the augmented reality view by means of a display device, wherein the virtual sensor is based on the real sensor to be positioned, wherein the simulated parameter comprises at least one position parameter, wherein the at least one position parameter specifies a position and / or an orientation of the virtual sensor in the augmented reality view, and wherein a change in the position parameter is obtained.wherein the updated position parameter based on the received change is visualized for the user in the augmented reality view by means of the display device, and wherein at least one output parameter is output which is based on the updated position parameter of the virtual sensor and which enables the positioning of the real sensor in the real environment.

The invention relates to the technical field of industrial safety intelligent monitoring, in particular to an operation safety state intelligent monitoring and alarming system based on a limited space, which comprises the steps of constructing a unified time sequence signal flow by collecting various sensor data, and dividing operation stages. Multi-dimensional features are extracted for each stage, a dynamic risk evolution model is input for forward deduction, and a risk evolution path is generated; meanwhile, similar scenes are retrieved from a historical case library, cross validation and confidence correction are carried out on deduction paths by taking actual results as references, and comprehensive risk indexes of multiple time points in the future are calculated. The index is compared with a dynamically adjusted alarm threshold curve, an alarm is triggered when a threshold is broken through, and a security instruction sequence is generated by matching a preliminary strategy based on a coping strategy map and combining real-time monitoring data and personnel distribution information. According to the scheme, through fusion of dynamic prediction and historical verification, advanced identification, dynamic evaluation and accurate response of limited space operation risks are realized.

The invention discloses a gallium oxide film double-array device group for flame detection and identification and a preparation method and application thereof, and belongs to the technical field of semiconductor photoelectric devices and flame monitoring. The device group comprises two arrays which are complementary in function and independent in structure: a Ga2O3 ultraviolet detector array which is grown based on a high-temperature annealing sapphire substrate and is consistent in orientation and is used for performing high-signal-to-noise-ratio rapid detection on flame solar-blind ultraviolet radiation; and the non-oriented Ga2O3 photoelectric synapse array is deposited on the unannealed substrate, and learning and classification identification of flame spatio-temporal characteristics are realized by utilizing plasticity caused by defect capture of the non-oriented Ga2O3 photoelectric synapse array. In the preparation process, the orientation-controlled Ga2O3 active layer and the non-orientation Ga2O3 active layer are respectively obtained by controlling whether the sapphire substrate is subjected to high-temperature annealing treatment or not. The two arrays can form a solution integrating detection and identification. The scheme is simple in structure, low in power consumption, easy to expand and suitable for industrial security and protection, combustion monitoring, forest fire early warning and other scenes.

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.

The invention belongs to the technical field of industrial safety monitoring, and particularly relates to a risk situation monitoring model, method and system. The risk situation monitoring method comprises the steps that the occurrence state of each initial event and the operation states of preventive protection measures and remission protection measures are collected; determining the access probability of a single event evolution link according to the occurrence state of the single initial event and the running state of the preventive protection measure, and determining the access probability of a single accident evolution link according to the access probability of the single event evolution link and the running state of the remission protection measure; determining the occurrence probability of the top event according to the access probability of the plurality of event evolution links; and determining the occurrence probability of a single accident consequence according to the occurrence probability of the top event and the operation state of the remission protection measure. According to the method, the occurrence probability of a certain accident consequence under the condition is calculated by considering the condition that a plurality of initial events exist, so that the method is closer to the actual industrial operation condition.

The invention discloses an electronic fence system and method for a dewatering belt conveyor, and relates to the technical field of industrial safety, and the system comprises a laser fence which is arranged along the boundary of a dangerous area of the dewatering belt conveyor, adopts the design of emitting visible light to form a visual barrier, and is used for monitoring the light shielding degree when a person enters the dangerous area in real time; and the alarm host is electrically connected with the laser fence and is used for receiving the shielding signal of the laser fence and performing grading processing, and the alarm host judges the shielding degree based on a preset shielding area threshold value. According to the electronic fence system and method for the dewatering belt conveyor, real-time monitoring and visual protection of a dangerous area of the dewatering belt conveyor are achieved by deploying the visible light laser fence, the safety of an operation area is effectively improved, the risk level can be accurately judged according to the personnel shielding degree through a graded early warning and intelligent linkage mechanism, and the safety of the operation area is improved. And sound-light alarm or equipment shutdown is automatically triggered, so that the timeliness of safety protection is enhanced.

The invention relates to the technical field of industrial safety, in particular to a safety instrument system SIL evaluation method considering human factor failure. The method comprises the steps that firstly, human factor influence factors of personnel, equipment, environment and management are objectively quantified through a triangular fuzzy entropy weight method; then human factor failure probability models of three stages of observation, judgment and execution are constructed; and finally, the quantized human factor failure probability is used as a series link to be coupled into a system PFD calculation model, so that an SIL level evaluation result closer to an actual operation condition is obtained. According to the method, quantitative evaluation of human factor failure is realized, a quantitative basis containing human factor improvement measures can be provided for system redundancy configuration and test cycle optimization, and the accuracy and engineering guidance value of SIL evaluation are remarkably improved.