Artificial Intelligence-Based Safety Risk Prediction Method and System for Hazardous Chemical Production Scenarios

Abstract

This application relates to the field of safety monitoring technology for hazardous chemical production scenarios, and provides an artificial intelligence-based method and system for predicting safety risks in hazardous chemical production scenarios. The method includes: obtaining the equipment type and status of each hazardous chemical production device in the target hazardous chemical production scenario from a hazardous chemical management system; predicting the walking path of target personnel based on point cloud data of target personnel and passageways collected by lidar; identifying the protective equipment worn by target personnel based on the camera and target detection algorithm; locating the position of hazardous chemical production equipment that is in operation and belongs to the first equipment type; and predicting the behavioral safety risk level of target personnel based on their walking path, protective equipment wearing status, the position of the hazardous chemical production equipment in operation and belonging to the first equipment type, and a preset coordination coefficient. This invention achieves refined prediction of the risks faced by personnel behavior, resulting in more accurate risk assessment and prediction.

Description

Technical Field

[0001] This application relates to the field of safety monitoring technology in hazardous chemical production scenarios, and in particular to a method and system for predicting safety risks in hazardous chemical production scenarios based on artificial intelligence. Background Technology

[0002] Currently, automated monitoring of personnel in hazardous chemical production settings includes whether they are wearing protective clothing and safety helmets. When personnel are detected not wearing protective clothing or safety helmets, they are deemed to be in a dangerous situation and a warning is issued. However, this method only judges a person as facing a high risk when the behavior of not wearing protective clothing or safety helmets is detected; it does not classify the risks involved in their behavior, which is not precise enough. Second, assessing the safety risks of personnel behavior by only considering whether personnel are wearing protective equipment is too simplistic and inaccurate. Third, the existing method cannot achieve a refined prediction of the risks faced by personnel behavior. Summary of the Invention

[0003] To address the aforementioned technical problems, the purpose of this application is to provide an artificial intelligence-based method and system for predicting safety risks in hazardous chemical production scenarios, aiming to solve at least one of the aforementioned technical problems.

[0004] In a first aspect, embodiments of this application provide a method for predicting safety risks in hazardous chemical production scenarios based on artificial intelligence, including:

[0005] Obtain the equipment type and status of each hazardous chemical production equipment in the target hazardous chemical production scenario from the hazardous chemical management system; wherein, the equipment type of the hazardous chemical production equipment includes a first equipment type and a second equipment type; wherein, the equipment status includes working status and non-working status;

[0006] The walking path of the target personnel is predicted based on the point cloud data of the target personnel collected by the lidar and the channel point cloud data in the target hazardous chemical production scene.

[0007] The camera and target detection algorithm are used to identify the protective equipment worn by the target personnel.

[0008] For hazardous chemical production equipment that is in operation and is of type 1 equipment, its location information is located using radio frequency positioning technology.

[0009] The target personnel's behavioral safety risk level is predicted based on their walking path, the protective equipment they are wearing, the location information of the hazardous chemical production equipment that is in operation and is of the first equipment type, and a preset coordination coefficient.

[0010] Furthermore, the step of predicting the behavioral safety risk level of the target personnel based on their walking path, their protective equipment wearing status, the location information of the hazardous chemical production equipment (of type 1) in operation, and a preset coordination coefficient includes:

[0011] Multiple location information is obtained from the walking path at preset distance intervals;

[0012] For each location information of the target personnel, the distance between the target personnel and the hazardous chemical production equipment of the first equipment type that is in operation is calculated based on the location information of the target personnel and the location information of the hazardous chemical production equipment of the first equipment type that is in operation.

[0013] The safety risk level of the target person's behavior at the corresponding location is determined based on the distance, the protective equipment worn by the target person, and a preset coordination coefficient.

[0014] By integrating the behavioral safety risk levels corresponding to each location information in chronological order, the behavioral safety risk levels of the target person at each location on the walking path are obtained.

[0015] Furthermore, the step of predicting the behavioral safety risk level of the target personnel based on their walking path, their protective equipment wearing status, the location information of the hazardous chemical production equipment (of type 1) in operation, and a preset coordination coefficient includes:

[0016] Multiple location information is obtained from the walking path at preset distance intervals;

[0017] For each location information of the target personnel, the distance between the target personnel and the hazardous chemical production equipment of the first equipment type that is in operation is calculated based on the location information of the target personnel and the location information of the hazardous chemical production equipment of the first equipment type that is in operation.

[0018] The safety risk level of the target person's behavior at the corresponding location is determined based on the distance, the protective equipment worn by the target person, and a preset coordination coefficient.

[0019] By integrating the behavioral safety risk levels corresponding to each location information in chronological order, the behavioral safety risk levels of the target person at each location on the walking path are obtained.

[0020] Furthermore, the step of determining the behavioral safety risk level of the target person at the corresponding location based on the distance, the protective equipment worn by the target person, and a preset coordination coefficient includes:

[0021] Based on the distance and a preset risk diffusion coefficient, the distance risk value is calculated using a Gaussian model.

[0022] The protective equipment required for the target hazardous chemical production scenario is determined from the hazardous chemical management system.

[0023] The type of protective equipment missing by the target personnel is determined based on the required protective equipment and the protective equipment worn by the target personnel.

[0024] Calculate the risk value of missing protective equipment based on the types of missing protective equipment.

[0025] The behavioral safety risk value of the target personnel at the corresponding location is calculated based on the risk value of missing protective equipment, the risk value of distance, and a preset coordination coefficient.

[0026] The behavioral safety risk level of the target personnel at the corresponding location is determined based on the target personnel's behavioral safety risk value and the preset safety risk level classification rules.

[0027] Furthermore, the step of calculating the behavioral safety risk value of the target personnel at the corresponding location based on the risk value of missing protective equipment, the distance risk value, and a preset coordination coefficient includes:

[0028] The behavioral safety risk value of the target personnel is calculated using the following formula:

[0029] ;

[0030] in, The behavioral safety risk value of the target personnel. The synergy coefficient is preset. This is the distance risk value. Risk value for lack of protective equipment.

[0031] Furthermore, after the step of integrating the behavioral safety risk levels corresponding to each location information in chronological order to obtain the behavioral safety risk levels of the target person at each location on the walking path, the method further includes:

[0032] The walking path is rendered differently at each location based on the level of behavioral safety risk and displayed on the monitoring screen.

[0033] Secondly, embodiments of this application provide an artificial intelligence-based safety risk prediction system for hazardous chemical production scenarios. The system includes a lidar and a camera, both installed at the target hazardous chemical production site. The system also includes a safety risk prediction device, which comprises:

[0034] The acquisition module is used to acquire the equipment type and equipment status of each hazardous chemical production equipment in the target hazardous chemical production scenario from the hazardous chemical management system; wherein, the equipment type of the hazardous chemical production equipment includes a first equipment type and a second equipment type; wherein, the equipment status includes working status and non-working status;

[0035] The walking path prediction module is used to predict the walking path of the target personnel based on the point cloud data of the target personnel collected by the lidar and the channel point cloud data in the target hazardous chemical production scene.

[0036] The identification module is used to identify the protective equipment worn by the target personnel based on the camera and the target detection algorithm.

[0037] The positioning module is used to locate the position information of hazardous chemical production equipment that is in operation and is of the first type of equipment, using radio frequency positioning technology.

[0038] The behavioral safety risk level prediction module is used to predict the behavioral safety risk level of the target personnel based on the target personnel's walking path, the target personnel's protective equipment wearing status, the location information of the hazardous chemical production equipment that is in operation and is of the first equipment type, and a preset coordination coefficient.

[0039] This invention, by installing LiDAR in hazardous chemical production scenarios, can accurately identify the walking position and direction of target personnel, as well as the location and layout of passageways, based on the point cloud data obtained by the LiDAR. This allows for precise prediction of the target personnel's walking path. By predicting the target personnel's walking path, the location the target personnel will pass through can be predicted. The location of the hazardous chemical production equipment in operation (Type I) can be obtained, and the distance between the target personnel and the equipment can be calculated. Artificial intelligence algorithms, such as target detection algorithms, are used to identify the protective equipment status of the target personnel. Combining distance and protective equipment wearing status, the safety risk level of the target personnel's behavior can be identified. Furthermore, considering the synergistic coefficient of the interaction strength between quantifying distance risk and the risk of missing protective equipment, the safety risk level of personnel behavior can be more accurately identified and predicted. This achieves refined prediction of the risks faced by personnel and a more accurate risk assessment. Attached Figure Description

[0040] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0041] Figure 1 This is a flowchart illustrating the safety risk prediction method for hazardous chemical production scenarios based on artificial intelligence provided in the embodiments of this application;

[0042] Figure 2 This is a schematic diagram of the structure of an artificial intelligence-based safety risk prediction system for hazardous chemical production scenarios provided in an embodiment of this application. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] Those skilled in the art will understand that, unless explicitly stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the word “comprising” as used in this application’s specification means the presence of features, integers, steps, operations, elements, modules, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The word “and / or” as used herein includes all or any modules and all combinations of one or more associated listed items.

[0045] Those skilled in the art will understand that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0046] Please see Figure 1 This application provides an artificial intelligence-based method for predicting safety risks in hazardous chemical production scenarios, wherein the hazardous chemical production scenarios are equipped with lidar and cameras, and the method includes:

[0047] S1. Obtain the equipment type and equipment status of each hazardous chemical production equipment in the target hazardous chemical production scenario from the hazardous chemical management system; wherein, the equipment type of the hazardous chemical production equipment includes a first equipment type and a second equipment type; wherein, the equipment status includes working status and non-working status;

[0048] S2. Based on the point cloud data of the target personnel collected by the lidar and the point cloud data of the passage in the target hazardous chemical production scene, predict the walking path of the target personnel.

[0049] S3. Identify the protective equipment worn by the target personnel based on the camera and target detection algorithm;

[0050] S4. For hazardous chemical production equipment that is in operation and is of the first type of equipment, its location information is located by radio frequency positioning technology.

[0051] S5. Based on the target personnel's walking path, the target personnel's protective equipment wearing status, the location information of the hazardous chemical production equipment that is in operation and is of the first equipment type, and a preset coordination coefficient, predict the target personnel's behavioral safety risk level.

[0052] In step S1 above, the hazardous chemicals management system stores information on hazardous chemical production equipment in various hazardous chemical production scenarios. For example, if a smart hazardous chemical production plant includes three hazardous chemical production scenarios, the information on the hazardous chemical production equipment in these three scenarios is entered into the hazardous chemicals management system. This equipment information includes equipment type and equipment operating status. The equipment type includes a first equipment type and a second equipment type. The first equipment type refers to the type of equipment where the closer personnel get to the machine while it is running, the greater the safety risk. The second equipment type refers to the type of equipment where there is no safety risk for personnel to approach the machine while it is running. The equipment type is entered manually. The equipment operating status includes both working and non-working states, and the equipment operating status is updated in real time by the communication module back to the hazardous chemicals management system.

[0053] In step S2 above, multiple lidars are installed in the target hazardous chemical production scene to cover the target hazardous chemical production scene. The lidars scan the target hazardous chemical production scene, and the point cloud data obtained by the lidars are clustered, segmented, and target detected to obtain the point cloud data of the target personnel and the channel point cloud data.

[0054] In step S3 above, multiple cameras are installed in the target hazardous chemical production scene to cover it. Based on the cameras and a target detection algorithm, the protective equipment worn by the target personnel is identified. Specifically, the cameras are used to acquire images of the target hazardous chemical production scene, and the target detection algorithm is used to detect the target personnel and identify whether they are wearing protective equipment, such as protective clothing, safety helmets, goggles, and gas masks. A target detection algorithm, such as YOLO, is used, combined with a specific training set, such as personnel wearing protective clothing and those not wearing protective clothing, to train a YOLO model for detecting whether personnel are wearing protective clothing. Similarly, personnel wearing safety helmets and those not wearing safety helmets are used to train a YOLO model for detecting whether personnel are wearing safety helmets.

[0055] In step S4 above, for hazardous chemical production equipment that is in operation and belongs to the first equipment type, its location information is determined using radio frequency (RFID) positioning technology. Since a hazardous chemical production scenario may include multiple or various types of hazardous chemical production equipment, to accurately obtain the location information of each type of equipment, RFID positioning technology is used to locate the location information of each piece of equipment. Specifically, RFID tags are installed on the hazardous chemical production equipment, and these tags store equipment information, including the equipment type. Active readers are installed in the target hazardous chemical production scenario. The active readers transmit radio waves, collect the signals emitted by the RFID tags, and upload them to the positioning server to pinpoint the specific location of the hazardous chemical production equipment. Hazardous chemical production equipment includes reaction vessels, tower reactors, distillation towers, centrifuges, extraction equipment, storage tanks, high-pressure equipment, etc.

[0056] In step S5 above, the synergy coefficient is used to quantify the interaction strength between distance risk and the risk of lack of protective equipment. The synergy coefficient reflects the additional amplification effect produced by the superposition of the two risks when personnel are simultaneously in close proximity and lack protection. The synergy coefficient is determined based on different hazardous chemical production equipment and their reacting chemicals. Specifically, historical accident cases can be collected, recording data such as the distance of personnel at the time of the accident, the lack of protective equipment, and the actual severity of injury. Regression analysis can then be used to solve a model to predict the synergy coefficient that best matches the risk and the actual injury. Alternatively, it can be obtained by experts based on experience.

[0057] This invention, by installing LiDAR in hazardous chemical production scenarios, can accurately identify the walking position and direction of target personnel, as well as the location and layout of passageways, based on the point cloud data obtained by the LiDAR. This allows for precise prediction of the target personnel's walking path. By predicting the target personnel's walking path, the location the target personnel will pass through can be predicted. The location of the hazardous chemical production equipment in operation (Type I) can be obtained, and the distance between the target personnel and the equipment can be calculated. Artificial intelligence algorithms, such as target detection algorithms, are used to identify the protective equipment status of the target personnel. Combining distance and protective equipment wearing status, the safety risk level of the target personnel's behavior can be identified. Furthermore, considering the synergistic coefficient of the interaction strength between quantifying distance risk and the risk of missing protective equipment, the safety risk level of personnel behavior can be more accurately identified and predicted. This achieves refined prediction of the risks faced by personnel and a more accurate risk assessment.

[0058] In one embodiment, the step of predicting the walking path of the target personnel based on the point cloud data of the target personnel collected by the lidar and the point cloud data of the passageway in the target hazardous chemical production scene includes:

[0059] The location of the target person is determined based on the point cloud data of the target person;

[0060] Extract the movement direction vector of the target person from the point cloud data of multiple consecutive frames;

[0061] The movement direction of the target personnel is determined based on the movement direction vector and the map of the target hazardous chemical production scene;

[0062] Based on the location of the target person, the movement direction and the channel layout are used to predict the target person's walking path.

[0063] In this embodiment, the lidar point cloud data contains positional information of the object's surface. Each point has three coordinate values: X, Y, and Z, representing its position in three-dimensional space. Therefore, the position of the target personnel can be determined through the point cloud data of the target personnel, and the position of the channel can be determined through the point cloud data of the channel, thus determining the channel layout. Here, the centroid of the target personnel's point cloud data can be used as the target personnel's position. The movement direction of the target personnel is extracted based on multiple consecutive frames of point cloud data. Specifically, multiple consecutive frames of the target personnel's centroid point cloud are projected onto the ground plane and connected to obtain a movement direction vector. The movement direction vector and the target hazardous chemical production scene map can determine the target personnel's movement direction. The target hazardous chemical production scene map can be dynamically updated using SLAM technology. Finally, based on the target personnel's position, movement direction, and channel layout, the walking path of the target personnel within N meters is predicted. N cannot be too small; N is generally greater than 3.6 meters to ensure a 2-3 second warning response time (system alarm + personnel reaction).

[0064] This invention, through the fusion of lidar point clouds and scene maps, achieves a closed loop of "high-precision positioning - dynamic direction analysis - environmental constraint prediction" in hazardous chemical production scenarios, thereby improving the accuracy of personnel path prediction.

[0065] In one embodiment, the step of predicting the behavioral safety risk level of the target personnel based on the target personnel's walking path, the target personnel's protective equipment wearing status, the location information of the hazardous chemical production equipment that is in operation and is of the first equipment type, and a preset coordination coefficient includes:

[0066] Multiple location information is obtained from the walking path at preset distance intervals;

[0067] For each location information of the target personnel, the distance between the target personnel and the hazardous chemical production equipment of the first equipment type that is in operation is calculated based on the location information of the target personnel and the location information of the hazardous chemical production equipment of the first equipment type that is in operation.

[0068] The safety risk level of the target person's behavior at the corresponding location is determined based on the distance, the protective equipment worn by the target person, and a preset coordination coefficient.

[0069] By integrating the behavioral safety risk levels corresponding to each location information in chronological order, the behavioral safety risk levels of the target person at each location on the walking path are obtained.

[0070] In this embodiment, the behavioral safety risk levels corresponding to each location information are integrated in chronological order to obtain the behavioral safety risk levels of the target person at each location on the walking path, i.e., a walking path with risk levels is obtained. Let the points on the extracted walking path be (in the order in which the target person arrives at the locations) ( The corresponding risk level is (). Then, the behavioral safety risk level of the target person at each location on the walking path can be expressed as ( ) By integrating the behavioral security risk levels corresponding to each location information in chronological order, the behavioral security risk levels of the target personnel at each location can be clearly identified.

[0071] In this embodiment of the application, when the predicted safety risk level of behavior at any location on the walking path is greater than the preset risk level, a warning is issued to the target personnel through a voice playback module installed in the target hazardous chemical production scene.

[0072] In one embodiment, the step of determining the behavioral safety risk level of the target person at the corresponding location based on the distance, the protective equipment worn by the target person, and a preset coordination coefficient includes:

[0073] Based on the distance and a preset risk diffusion coefficient, the distance risk value is calculated using a Gaussian model.

[0074] The protective equipment required for the target hazardous chemical production scenario is determined from the hazardous chemical management system.

[0075] The type of protective equipment missing by the target personnel is determined based on the required protective equipment and the protective equipment worn by the target personnel.

[0076] Calculate the risk value of missing protective equipment based on the types of missing protective equipment.

[0077] The behavioral safety risk value of the target personnel at the corresponding location is calculated based on the risk value of missing protective equipment, the risk value of distance, and a preset coordination coefficient.

[0078] The behavioral safety risk level of the target personnel at the corresponding location is determined based on the target personnel's behavioral safety risk value and the preset safety risk level classification rules.

[0079] In this embodiment of the application, the step of calculating the distance risk value using a Gaussian model based on the distance and the risk diffusion coefficient includes: calculating the distance risk value according to the following formula.

[0080] ;

[0081] Where d is the distance between personnel and hazardous chemical production equipment that is in operation and is of the first type of equipment; This is the risk diffusion coefficient, which is determined based on the properties of the equipment and the chemicals within it. This represents the distance risk value.

[0082] The step of calculating the risk value of missing protective equipment based on the type of unworn protective equipment includes:

[0083] Calculate the risk value of missing protective equipment using the following formula:

[0084] ;

[0085] in, This indicates the missing weight of the i-th type of protective equipment. This is an indicator function; 1 indicates missing values, and 0 indicates no missing values. This represents the risk value of missing protective equipment, where n is the total number of types of protective equipment required in the target hazardous chemical production scenario.

[0086] In one embodiment, the step of calculating the behavioral safety risk value of the target person at the corresponding location based on the risk value of missing protective equipment, the distance risk value, and a preset coordination coefficient includes:

[0087] The behavioral safety risk value of the target personnel is calculated using the following formula:

[0088] ;

[0089] in, The behavioral safety risk value of the target personnel. The synergy coefficient is preset. This is the distance risk value. Risk value for lack of protective equipment.

[0090] In one embodiment, the step of integrating the behavioral safety risk levels corresponding to each location information in chronological order to obtain the behavioral safety risk levels of the target person at each location on the walking path includes:

[0091] The walking path is rendered differently at each location based on the level of behavioral safety risk and displayed on the monitoring screen.

[0092] In this embodiment, different behavioral safety risk levels correspond to different colors. Differentiating the rendering of each location on the walking path according to the behavioral safety risk level and displaying it on the monitoring screen allows monitoring personnel to clearly understand the behavioral safety risk level that personnel are facing or will face in the hazardous chemical production scenario, so that they can quickly take corresponding measures, such as issuing warnings through broadcasts.

[0093] In one embodiment, after the step of obtaining the equipment type and equipment status of each hazardous chemical production equipment in the target hazardous chemical production scenario from the hazardous chemical management system, the method further includes:

[0094] Obtain the hazardous temperature thresholds corresponding to various hazardous chemicals in the target hazardous chemical production scenario from the hazardous chemical management system;

[0095] Extract the lowest dangerous temperature value from all the aforementioned dangerous temperature thresholds;

[0096] Temperature data from the target hazardous chemical production site is collected using temperature sensors.

[0097] Determine whether the temperature of the target hazardous chemical production scene is greater than the minimum dangerous temperature value. If so, issue a temperature environment risk warning.

[0098] In this embodiment, multiple temperature sensors are deployed in the target hazardous chemical production scenario to collect temperatures from multiple locations. The highest temperature value is selected as the temperature of the target hazardous chemical production scenario. By comparing this highest temperature with the lowest temperature hazard threshold, the perception of temperature environmental risks in the hazardous chemical production scenario is more sensitive, facilitating timely implementation of corresponding measures to reduce the safety risks posed by ambient temperature.

[0099] This application embodiment also provides a system including a lidar and a camera, both of which are installed at the target hazardous chemical production site. The system further includes a safety risk prediction device, which includes:

[0100] The acquisition module 1 is used to acquire the equipment type and equipment status of each hazardous chemical production equipment in the target hazardous chemical production scenario from the hazardous chemical management system; wherein, the equipment type of the hazardous chemical production equipment includes a first equipment type and a second equipment type; wherein, the equipment status includes a working status and a non-working status;

[0101] The walking path prediction module 2 is used to predict the walking path of the target personnel based on the point cloud data of the target personnel collected by the lidar and the channel point cloud data in the target hazardous chemical production scene.

[0102] The identification module 3 is used to identify the protective equipment worn by the target personnel based on the camera and the target detection algorithm;

[0103] Positioning module 4 is used to locate the position information of hazardous chemical production equipment that is in operation and is of the first equipment type, using radio frequency positioning technology.

[0104] The behavioral safety risk level prediction module 5 is used to predict the behavioral safety risk level of the target personnel based on the target personnel's walking path, the target personnel's protective equipment wearing status, the location information of the hazardous chemical production equipment that is in operation and is of the first equipment type, and a preset coordination coefficient.

[0105] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0106] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0107] The above description is only a preferred embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural changes made based on the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for predicting safety risks in hazardous chemical production scenarios based on artificial intelligence, characterized in that, The hazardous chemical production scenario is equipped with lidar and cameras, and the method includes: Obtain the equipment type and status of each hazardous chemical production equipment in the target hazardous chemical production scenario from the hazardous chemical management system; wherein, the equipment type of the hazardous chemical production equipment includes a first equipment type and a second equipment type; wherein, the equipment status includes working status and non-working status; the first equipment type refers to the equipment type in which the closer personnel get to the machine while it is running, the greater the safety risk. The walking path of the target personnel is predicted based on the point cloud data of the target personnel collected by the lidar and the channel point cloud data in the target hazardous chemical production scene. The camera and target detection algorithm are used to identify the protective equipment worn by the target personnel. For hazardous chemical production equipment that is in operation and is of type 1 equipment, its location information is located using radio frequency positioning technology. Based on the target personnel's walking path, their protective equipment wearing status, the location information of the hazardous chemical production equipment (of type 1) in operation, and a preset coordination coefficient, the behavioral safety risk level of the target personnel is predicted. The coordination coefficient quantifies the interaction strength between distance risk and the risk of missing protective equipment. It reflects the additional amplification effect of the two risks combined when personnel are simultaneously in close proximity and lack protection. The coordination coefficient is determined according to different hazardous chemical production equipment and their reacting chemicals. Specifically, historical accident cases are collected, and the following data are recorded: the distance of personnel at the time of the accident, the status of missing protective equipment, and the actual severity of injury. Regression analysis is used to solve the model and predict the coordination coefficient that best matches the risk and the actual injury. The step of predicting the behavioral safety risk level of the target personnel based on the target personnel's walking path, the target personnel's protective equipment wearing status, the location information of the hazardous chemical production equipment that is in operation and is of the first equipment type, and a preset coordination coefficient includes: Multiple location information is obtained from the walking path at preset distance intervals; For each location information of the target personnel, the distance between the target personnel and the hazardous chemical production equipment of the first equipment type that is in operation is calculated based on the location information of the target personnel and the location information of the hazardous chemical production equipment of the first equipment type that is in operation. The safety risk level of the target person's behavior at the corresponding location is determined based on the distance, the protective equipment worn by the target person, and a preset coordination coefficient. By integrating the behavioral safety risk levels corresponding to each location information in chronological order, the behavioral safety risk levels of the target person at each location on the walking path are obtained; The step of determining the behavioral safety risk level of the target person at the corresponding location based on the distance, the protective equipment worn by the target person, and a preset coordination coefficient includes: Based on the distance and a preset risk diffusion coefficient, the distance risk value is calculated using a Gaussian model; wherein, the distance risk value is calculated according to the following formula: ； Where d is the distance between personnel and hazardous chemical production equipment that is in operation and is of the first type of equipment; This is the risk diffusion coefficient, which is determined based on the properties of the equipment and the chemicals within it. This represents the distance risk value. The protective equipment required for the target hazardous chemical production scenario is determined from the hazardous chemical management system. The type of protective equipment missing by the target personnel is determined based on the required protective equipment and the protective equipment worn by the target personnel. Calculate the risk value of missing protective equipment based on the types of missing protective equipment. The behavioral safety risk value of the target personnel at the corresponding location is calculated based on the risk value of missing protective equipment, the risk value of distance, and a preset coordination coefficient; wherein, the behavioral safety risk value of the target personnel is calculated according to the following formula: ； in, The behavioral safety risk value of the target personnel. The synergy coefficient is preset. This is the distance risk value. Risk value for missing protective equipment; The behavioral safety risk level of the target personnel at the corresponding location is determined based on the target personnel's behavioral safety risk value and the preset safety risk level classification rules.

2. The method for predicting safety risks in hazardous chemical production scenarios based on artificial intelligence according to claim 1, characterized in that, After the step of integrating the behavioral safety risk levels corresponding to each location information in chronological order to obtain the behavioral safety risk levels of the target person at each location on the walking path, the method further includes: The walking path is rendered differently at each location based on the level of behavioral safety risk and displayed on the monitoring screen.

3. A safety risk prediction system for hazardous chemical production scenarios based on artificial intelligence, characterized in that, The system includes a lidar and a camera, both of which are installed at the target hazardous chemical production site. The system also includes a safety risk prediction device, which is used to execute the artificial intelligence-based safety risk prediction method for hazardous chemical production scenarios as described in any one of claims 1-2. The device includes: The acquisition module is used to acquire the equipment type and equipment status of each hazardous chemical production equipment in the target hazardous chemical production scenario from the hazardous chemical management system; wherein, the equipment type of the hazardous chemical production equipment includes a first equipment type and a second equipment type; wherein, the equipment status includes working status and non-working status; The walking path prediction module is used to predict the walking path of the target personnel based on the point cloud data of the target personnel collected by the lidar and the channel point cloud data in the target hazardous chemical production scene. The identification module is used to identify the protective equipment worn by the target personnel based on the camera and the target detection algorithm. The positioning module is used to locate the position information of hazardous chemical production equipment that is in operation and is of the first type of equipment, using radio frequency positioning technology. The behavioral safety risk level prediction module is used to predict the behavioral safety risk level of the target personnel based on the target personnel's walking path, the target personnel's protective equipment wearing status, the location information of the hazardous chemical production equipment that is in operation and is of the first equipment type, and a preset coordination coefficient.

Images