Method and device for evaluating safety condition of working environment

Abstract

The invention provides a working environment safety condition assessment method and device, and the method comprises the steps: obtaining to-be-processed data from a plurality of cooperation devices, the to-be-processed data carrying timestamp information; based on the timestamp information, performing time synchronization on the to-be-processed data to generate synchronous data; judging whether the synchronous data is abnormal or not based on a preset judgment rule, and generating a safety supervision result; and carrying out weighted fusion calculation on the safety supervision result, and evaluating the safety condition of the working environment according to the calculation result. In the technical scheme provided by the embodiment of the invention, the safety supervision system receives and analyzes the multi-source data from different cooperative devices in real time, and can comprehensively sense the safety state of the operation site, so that the safety state of the operation environment is more accurately evaluated, potential risk factors are more accurately identified, and the safety of the operation environment is improved. The accuracy of safety supervision is improved, and the possibility of missing report and false report is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of gas operation, and in particular to a method and device for evaluating the safety status of an operation environment. Background Art

[0002] Gas operations are often accompanied by potential risks such as gas leaks, hazardous substances, dangerous working conditions, and irregular work behaviors. Therefore, numerous safety monitoring systems are gradually being applied to the supervision of gas operation environments to ensure their safety. Existing safety monitoring systems typically manage gas concentration data, data from smart wearable devices, and video surveillance data independently, lacking a unified processing framework. This makes it impossible to combine multiple data types for comprehensive monitoring and early warning. This creates the risk of data omissions or misjudgments during safety monitoring. Furthermore, existing safety monitoring systems can monitor abnormal situations through simple threshold alarms, but lack the ability to handle complex situations. This makes it impossible to promptly identify potential safety hazards and guarantee the safety of the gas operation environment. Summary of the Invention

[0003] In view of the above problems, an embodiment of the present invention provides a method and device for assessing the safety status of an operating environment, which is used to solve the problem in the existing technology that it is impossible to combine multiple data for comprehensive supervision and early warning, and there is a risk of data omission or misjudgment in the safety supervision process.

[0004] In a first aspect, an embodiment of the present invention provides a method for assessing the safety status of a working environment, the method comprising:

[0005] respectively obtain data to be processed from a plurality of collaborative devices, wherein the data to be processed carries timestamp information;

[0006] Based on the timestamp information, time synchronization is performed on the data to be processed to generate synchronized data;

[0007] Determine whether the synchronization data is abnormal based on preset judgment rules and generate a safety supervision result;

[0008] A weighted fusion calculation is performed on the safety supervision results, and the safety status of the operating environment is evaluated based on the calculation results.

[0009] In one possible implementation, the collaborative device includes a gas sensor, a smart wearable device, and a video monitoring device, and the data to be processed includes gas concentration data, health monitoring data, and video monitoring data;

[0010] The step of respectively obtaining the data to be processed from the plurality of collaborative devices includes:

[0011] Obtain gas concentration data collected by gas sensors and health monitoring data collected by smart wearable devices through TCP, HTTP or MQTT protocols;

[0012] Obtain video surveillance data collected by video surveillance equipment through the RTSP protocol.

[0013] In a possible implementation, performing time synchronization on the data to be processed based on the timestamp information to generate synchronization data includes:

[0014] Calibrate the clocks of the gas sensor and the smart wearable device through the NTP server to obtain the timestamp information of the gas concentration data and the timestamp information of the health monitoring data;

[0015] Extracting timestamp information of the video surveillance data through a video stream processing tool;

[0016] Time synchronization is performed on the timestamp information of the gas concentration data, the timestamp information of the health monitoring data, and the timestamp information of the video monitoring data to generate synchronized data.

[0017] In one possible implementation, the synchronized data includes gas concentration synchronized data, the gas concentration synchronized data is used to indicate concentration data of various gases in the air, the safety supervision result includes an abnormal concentration supervision result or a normal concentration supervision result, and the determining whether the synchronized data is abnormal based on a preset judgment rule and generating the safety supervision result includes:

[0018] Determine whether the concentration data of various gases are within the preset normal concentration range;

[0019] If it is determined that the concentration data of at least one gas is outside the normal concentration range, generating a concentration abnormality monitoring result, and sending the concentration abnormality monitoring result to the abnormality processing center;

[0020] If it is determined that the concentration data of various gases are all within the normal concentration range, a normal concentration monitoring result is generated, and the step of determining whether the concentration data of various gases are all within the preset normal range is continued.

[0021] In one possible implementation, the synchronized data includes health monitoring synchronized data, the security monitoring result includes a data abnormality monitoring result and a data normality monitoring result, and the determining whether the synchronized data is abnormal based on a preset judgment rule and generating the security monitoring result includes:

[0022] Determining whether the health monitoring synchronization data is within a preset normal data range;

[0023] If it is determined that the health supervision synchronization data is outside the normal range of the data, a data abnormality supervision result is generated, and the data abnormality supervision result is uploaded to the abnormality processing center;

[0024] If it is determined that the health monitoring synchronization data is within the normal data range, a normal data monitoring result is generated, and the step of determining whether the health monitoring synchronization data is within the preset normal range is continued.

[0025] In one possible implementation, the synchronized data includes video surveillance synchronized data, the security monitoring result includes an abnormal monitoring monitoring result and a normal monitoring monitoring result, and the determining whether the synchronized data is abnormal based on a preset judgment rule and generating the security monitoring result includes:

[0026] Perform real-time analysis of the video surveillance data using artificial intelligence (AI) video analysis algorithms;

[0027] When an abnormal situation is detected, a monitoring abnormality supervision result is generated and uploaded to the abnormality processing center;

[0028] When no abnormal situation is detected, a normal monitoring result is generated, and the step of performing real-time analysis on the video monitoring data through the AI ​​video analysis algorithm is continued.

[0029] In a possible implementation, the abnormal monitoring result includes at least one of an equipment missing abnormal result, an open flame abnormal result, an abnormal behavior result, and a non-operating personnel intrusion abnormal result.

[0030] In one possible implementation, the real-time analysis of the video surveillance data using an AI video analysis algorithm includes:

[0031] Detecting at least one target in the video surveillance data based on a target detection algorithm and generating at least one target frame;

[0032] Tracking the at least one target based on a target tracking algorithm to generate a target motion trajectory;

[0033] The video surveillance data is analyzed in real time based on the target motion trajectory.

[0034] In a possible implementation, the method further includes:

[0035] Push abnormal supervision results and synchronized data to the front end and data sharing platform; the abnormal supervision results include at least one of abnormal concentration supervision results, abnormal data supervision results and abnormal monitoring supervision results.

[0036] In a second aspect, an embodiment of the present invention provides a device for assessing the safety status of a working environment, the device comprising:

[0037] An acquisition module, configured to acquire data to be processed from a plurality of collaborative devices, wherein the data to be processed carries timestamp information;

[0038] A synchronization module, configured to perform time synchronization on the data to be processed based on the timestamp information and generate synchronized data;

[0039] A judgment module, configured to judge whether the synchronization data has any anomalies based on preset judgment rules and generate a safety supervision result;

[0040] The evaluation module is used to perform weighted fusion calculation on the safety supervision results and evaluate the safety status of the operating environment based on the calculation results.

[0041] In the technical solution provided by the embodiment of the present invention, the safety supervision system receives and analyzes multi-source data from different collaborative devices in real time, and can fully perceive the safety status of the work site, thereby more accurately evaluating the safety status of the working environment and more accurately identifying potential hazardous factors. It avoids the problem of relying on a single data source (such as gas sensors or smart wearable devices) for monitoring in existing technologies, while ignoring the complex relationship between different data, resulting in missed detection of hazardous factors, thereby improving the accuracy of safety supervision and reducing the possibility of missed reports and false alarms.

[0042] In an embodiment of the present invention, when at least one of the gas concentration synchronization data, health monitoring synchronization data and video monitoring synchronization data is abnormal, the safety supervision system can quickly detect the abnormality and report it, with strong real-time performance, greatly improving the response speed of the system and ensuring the safety of personnel at the work site.

[0043] In this embodiment of the present invention, the security monitoring system pushes exception monitoring results and synchronized data to a data sharing platform via HTTP, enabling cross-platform data sharing and collaboration. This cross-platform data push and sharing capability enhances the flexibility and collaborative capabilities of the security monitoring system, enabling data synchronization and timely responses across different management platforms. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A flowchart of a method for assessing the safety status of an operating environment provided by an embodiment of the present invention.

[0045] Figure 2 A schematic diagram of a safety supervision strategy in a plugging operation scenario provided by an embodiment of the present invention.

[0046] Figure 3A schematic diagram of a safety supervision strategy for a confined space operation scenario provided by an embodiment of the present invention.

[0047] Figure 4 A schematic diagram of a comprehensive security supervision system provided by an embodiment of the present invention.

[0048] Figure 5 A schematic diagram of the structure of a device for assessing the safety status of an operating environment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions, and advantages of the present invention more clear and understandable, the present invention is further described below with reference to the accompanying drawings and specific embodiments. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0050] Figure 1 A flowchart of a method for evaluating the safety status of an operating environment provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method includes:

[0051] Step 101: Obtain data to be processed from multiple collaborative devices respectively, where the data to be processed carries timestamp information.

[0052] Each step in the embodiment of the present invention is performed by a safety monitoring system. Upon startup, the safety monitoring system automatically detects and establishes a data connection with the collaborating devices. The collaborating devices include gas sensors, smart wearable devices, and video surveillance equipment. The data to be processed includes gas concentration data, health monitoring data, and video surveillance data. The smart wearable device is a smart helmet.

[0053] In an embodiment of the present invention, gas concentration data collected by gas sensors and health monitoring data collected by smart wearable devices are obtained through the Transmission Control Protocol (TCP), Hypertext Transfer Protocol (HTTP), or Message Queuing Telemetry Transport (MQTT) protocol; video monitoring data collected by video monitoring equipment is obtained through the Real Time Streaming Protocol (RTSP).

[0054] As an optional solution, when the safety supervision system obtains gas concentration data and health supervision data through the TCP protocol, the safety supervision system acts as a TCP server, and the gas sensors and smart wearable devices act as TCP clients. The gas sensors and smart wearable devices actively send data to be processed to the safety supervision system. Alternatively, the safety supervision system acts as a TCP client, and the gas sensors and smart wearable devices act as TCP servers. The safety supervision system sends data query instructions to the gas sensors and smart wearable devices at a preset frequency; the gas sensors and smart wearable devices respond to the data query instructions and return the data query results to the safety supervision system; the safety supervision system parses the data to be processed from the data query results. It should be noted that in addition to timestamp information, the data to be processed also includes information such as data value, data type, and device number.

[0055] As another optional solution, when the safety monitoring system obtains gas concentration data and health monitoring data through the HTTP protocol, the gas sensors and smart wearable devices send the data to be processed to the cloud, and the safety monitoring system queries the data to be processed from the cloud through the HTTP protocol.

[0056] Specifically, when the security monitoring system obtains video surveillance data via RTSP, it connects to the video surveillance device via RTSP and uses video stream processing tools such as FFmpeg, OpenCV, and GStreamer to decode and extract video frames from the video stream. Video stream processing tools include FFmpeg, OpenCV, and GStreamer, and the video stream supports H.264 and H.265 formats.

[0057] Step 102: Based on the timestamp information, time synchronization is performed on the data to be processed to generate synchronized data.

[0058] In this step, the clocks of the gas sensor and smart wearable device are calibrated through the Network Time Protocol (NTP) server to obtain the timestamp information of the gas concentration data and the timestamp information of the health monitoring data; the timestamp information of the video monitoring data is extracted through the video stream processing tool; the timestamp information of the gas concentration data, the timestamp information of the health monitoring data, and the timestamp information of the video monitoring data are synchronized to generate synchronized data.

[0059] In an embodiment of the present invention, in order to achieve effective association of data to be processed from different collaborative devices, the security supervision system needs to align the data time of various collaborative devices, and use a combination of hardware clock synchronization and software compensation to attach precise timestamps to all data to be processed to ensure the timing consistency of data across devices.

[0060] Step 103: Determine whether there is any anomaly in the synchronization data based on the preset judgment rules and generate a security supervision result.

[0061] In this embodiment of the present invention, the synchronized data includes gas concentration synchronized data, which indicates the concentrations of various gases in the air. The safety monitoring results include abnormal concentration monitoring results or normal concentration monitoring results. Step 103 includes: determining whether the concentrations of various gases are all within a preset normal concentration range; if the concentration of at least one gas is determined to be outside the normal concentration range, generating an abnormal concentration monitoring result and transmitting the abnormal concentration monitoring result to the abnormality handling center; if the concentrations of various gases are all within the normal concentration range, generating a normal concentration monitoring result and continuing with the step of determining whether the concentrations of various gases are all within the preset normal range. For example, the gas concentration synchronized data indicates the concentrations of methane, carbon monoxide, hydrogen sulfide, and oxygen in the air. The normal concentration ranges for different gases may vary. Excessively high concentrations of carbon monoxide and hydrogen sulfide can pose a serious threat to human health. Excessively high methane concentrations can cause the oxygen concentration in the air to become too low, resulting in an oxygen-deficient environment. Furthermore, excessively high methane concentrations can cause fires or explosions when exposed to open flames. Therefore, when the gas concentration in the air meets at least one of the following conditions: methane gas concentration is too high, carbon monoxide gas concentration is too high, hydrogen sulfide gas concentration is too high, and oxygen gas concentration is too low, an abnormal gas concentration supervision result is generated, realizing gas concentration supervision of the safety supervision system.

[0062] In an embodiment of the present invention, the synchronized data includes health monitoring synchronized data, and the safety monitoring results include data abnormality monitoring results and data normal monitoring results. Step 103 includes: determining whether the health monitoring synchronized data is within a preset data normal range; if it is determined that the health monitoring synchronized data is outside the data normal range, generating a data abnormality monitoring result, and uploading the data abnormality monitoring result to the abnormality processing center; if it is determined that the health monitoring synchronized data is within the data normal range, generating a data normal monitoring result, and continuing to execute the step of determining whether the health monitoring synchronized data is within the preset normal range. For example, the health monitoring synchronized data includes blood oxygen data and heart rate data. When the blood oxygen data and / or heart rate data are outside the data normal range, a data abnormality monitoring result is generated, thereby realizing personnel health monitoring of the safety monitoring system.

[0063] In this embodiment of the present invention, the synchronized data includes video surveillance synchronized data, and the security supervision results include abnormal monitoring results and normal monitoring results. Step 103 includes: performing real-time analysis of the video surveillance data using an artificial intelligence (AI) video analysis algorithm; when an abnormality is detected, generating an abnormal monitoring supervision result and uploading the abnormal monitoring supervision result to the abnormality processing center; when no abnormality is detected, generating a normal monitoring supervision result and continuing to perform real-time analysis of the video surveillance data using the AI ​​video analysis algorithm.

[0064] Specifically, AI video analysis algorithms include target detection and tracking algorithms. The target detection algorithm detects at least one target in video surveillance data and generates at least one target frame. The target tracking algorithm tracks at least one target and generates a target motion trajectory. Video surveillance data is analyzed in real time based on the target motion trajectory. The target detection algorithm uses the YOLOv5, YOLOv8s, or Faster R-CNN algorithms; the target tracking algorithm uses the BoT-SORT or ByteTrack algorithms.

[0065] The YOLO algorithm is a deep learning-based object detection algorithm that achieves highly efficient detection performance by transforming the object detection problem into a regression problem. YOLOv8 is the next-generation, lightweight model in the YOLO family. Its core improvements lie in its dynamic detection head design and a more efficient feature fusion mechanism, achieving a better balance between accuracy and speed than previous models. YOLOv8s is a lightweight version of the YOLOv8 family, optimized for speed and suitable for devices with low computing resources or real-time monitoring scenarios.

[0066] The BoT-SORT algorithm is an efficient multi-object tracking (MOT) algorithm. By integrating motion prediction models, appearance feature matching, and camera motion compensation (CMC), it significantly improves the robustness and accuracy of tracking in complex scenarios compared to traditional SORT algorithms.

[0067] In this embodiment of the present invention, the monitoring abnormality supervision result includes at least one of an equipment missing abnormality result, an open flame abnormality result, a behavior abnormality result, and a non-operating personnel intrusion abnormality result. The equipment missing abnormality result indicates that equipment such as a tripod, fire extinguisher, blower, sign, warning tape, cone, or protective mask is missing; the open flame abnormality result indicates the presence of an open flame at the work site; the behavior abnormality result indicates that an operator has engaged in unsafe behaviors such as smoking, making phone calls, or not wearing a smart helmet; and the non-operating personnel intrusion result indicates that a non-operating personnel has entered a preset electronic fence area.

[0068] Optionally, when the safety supervision system generates an abnormal supervision result, a voice broadcast reminder is issued. Different types of abnormal supervision results correspond to different voice broadcast reminders. For example, based on the abnormal concentration supervision result, a voice reminder such as "Please note that the gas concentration in the air is abnormal. Please suspend operations" is generated.

[0069] In an embodiment of the present invention, the security monitoring system adopts a front-end and back-end separation architecture, using FastAPI, which is based on the Python programming language, as the back-end framework. On the one hand, FastAPI can provide a lightweight, high-performance application programming interface (API); on the other hand, the Python language can better integrate with the intelligent analysis functions in the security monitoring system. Modern web technologies (such as Vue.js) are used to build the front end, providing a user-friendly web interface. The front end and back end exchange data through a RESTful API to display video surveillance data and abnormal monitoring results.

[0070] FastAPI supports asynchronous programming, using the async and await keywords to handle I / O-intensive operations such as database queries, external API calls, TCP connections, and messaging. This asynchronous processing enables the system to efficiently handle large numbers of concurrent requests, improving response speed and system throughput. Furthermore, for computationally intensive intelligent analysis tasks, the security monitoring system incorporates multithreading technology. For each channel of video surveillance data collected by a video surveillance device, a separate sub-thread is enabled to perform real-time intelligent analysis of the video surveillance data, thereby improving the security monitoring system's analysis speed and processing capabilities.

[0071] Optionally, FastAPI can be integrated with a SQLite database for efficient data manipulation and API development. SQLite is a lightweight, embedded database that requires no complex configuration and is suitable for rapid iteration. FastAPI integration with SQLite offers high flexibility and can be applied to a variety of scenarios.

[0072] In an embodiment of the present invention, the safety supervision system pushes the abnormal supervision results and synchronized data to the front end and the data sharing platform. Specifically, the safety supervision system pushes the abnormal supervision results and synchronized data to the front end through the WebSocket protocol. By pushing the data to the front end, it is ensured that users can obtain the latest data, thereby improving the user experience. The safety supervision system pushes the abnormal supervision results and synchronized data to the data sharing platform through the HTTP protocol. By pushing the data to the data sharing platform, cross-platform data sharing and collaborative work are achieved. The abnormal supervision results include at least one of the abnormal concentration supervision results, the abnormal data supervision results, and the abnormal monitoring supervision results.

[0073] Step 104: Perform weighted fusion calculation on the safety supervision results, and evaluate the safety status of the working environment based on the calculation results.

[0074] In this embodiment of the present invention, normal monitoring results are marked as 0, and abnormal monitoring results are marked as 1. Each abnormal monitoring result is weighted based on actual usage requirements. By performing a weighted fusion calculation on abnormal monitoring results, a comprehensive assessment of the safety status of the operating environment is achieved. Normal monitoring results include normal concentration monitoring results, normal data monitoring results, and normal monitoring monitoring results. Abnormal monitoring results include abnormal concentration monitoring results, abnormal data monitoring results, and abnormal monitoring monitoring results.

[0075] In this embodiment of the present invention, safety supervision is described in detail for two typical gas operation scenarios. The first scenario is a plugging operation scenario, and the second scenario is a confined space operation scenario. The plugging operation scenario refers to the process of repairing or replacing a gas pipeline under pressure by drilling a hole in the pipeline and installing a plugging device. The confined space operation scenario refers to the process of operating in a closed or semi-enclosed space within a gas facility or equipment.

[0076] Figure 2 A schematic diagram of a safety supervision strategy in a blocking operation scenario provided by an embodiment of the present invention is shown as follows: Figure 2 As shown, in the plugging operation scenario, the safety supervision system's safety supervision strategies include gas concentration monitoring, personnel health monitoring, electronic fence monitoring, and operating condition monitoring. Gas concentration monitoring and personnel health monitoring are described in step 103 and are not repeated here. The safety supervision system uses AI video analysis algorithms to analyze video surveillance data in real time, implementing electronic fence monitoring and operating behavior monitoring.

[0077] Specifically, the safety supervision system uses AI video analysis algorithms to determine whether people appearing within the preset electronic fence range are wearing work clothes; if it is determined that the person is wearing work clothes that meet the standards, the person is marked as an operating personnel; if it is determined that the person is not wearing work clothes that meet the standards in N consecutive video frames, the person is marked as a non-operating personnel, and an abnormal result of non-operating personnel intrusion is generated, and the abnormal result of non-operating personnel intrusion is uploaded to the abnormality processing center, realizing the electronic fence supervision of the safety supervision system.

[0078] Specifically, the safety supervision system uses AI video analysis algorithms to determine whether there is any missing equipment at the work site, that is, to determine whether there are any missing equipment such as tripods, fire extinguishers, blowers, signs, warning tapes, cones, protective masks, etc. at the work site; if it is determined that there is no missing equipment at the work site, the video frame is marked as normal equipment; if it is determined that there is missing equipment at the work site, the video frame is marked as missing equipment; if N consecutive video frames are marked as missing equipment, an equipment missing exception result is generated, and the equipment missing exception result is uploaded to the exception handling center.

[0079] Specifically, the safety supervision system uses AI video analysis algorithms to determine whether there is an open flame at the work site. If it is determined that there is an open flame at the work site, it will immediately generate an open flame abnormality result and upload the open flame abnormality result to the abnormality handling center.

[0080] Specifically, the safety monitoring system uses AI video analysis algorithms to determine whether workers have engaged in unsafe behavior. If it determines that workers have engaged in unsafe behavior in N consecutive video frames, it generates an abnormal behavior result and uploads it to the abnormality processing center. Unsafe behaviors include smoking, making phone calls, and not wearing a smart helmet.

[0081] In an embodiment of the present invention, the operating condition supervision of the safety supervision system in the blocking operation scenario is achieved by judging whether there is equipment missing in the operating environment, whether there is open flame in the operating environment, and whether personnel have engaged in unsafe behavior.

[0082] The security monitoring system uses AI video analysis algorithms to analyze video surveillance data in real time. It then uploads any detected anomaly monitoring results, along with the video frame data where the anomaly was detected, to the anomaly handling center. For example, if the security monitoring system detects that a person is not wearing standard work clothes in N consecutive video frames, it generates a non-operating personnel intrusion anomaly result and uploads the non-operating personnel intrusion anomaly result and the N video frames to the anomaly handling center. Furthermore, the anomaly monitoring results and the N video frames are saved in a local database for subsequent query and access.

[0083] In the blocking operation scenario, after receiving the abnormal supervision results, the exception handling center generates different voice reminders based on the different types of abnormal supervision results. Each type of abnormal supervision result is set with a corresponding broadcast frequency, and the highest priority abnormal supervision result of that type is broadcast every M seconds.

[0084] Figure 3 A schematic diagram of a safety supervision strategy in a confined space operation scenario provided by an embodiment of the present invention is shown as follows: Figure 3As shown in the figure, in confined space operation scenarios, the safety supervision system's safety supervision strategy includes pre-operation supervision and in-operation supervision. Pre-operation supervision includes pre-operation gas concentration monitoring and pre-operation working condition monitoring to ensure that all preparatory work before gas operation meets standards. In-operation supervision includes gas concentration monitoring, personnel health monitoring, and working condition monitoring. Furthermore, AI video analysis algorithms are used to analyze video surveillance data in real time to monitor operational behavior.

[0085] If three different gas concentration readings are obtained, and all three are within the corresponding normal concentration range, the gas concentrations at the top, middle, and bottom of the confined space are determined to meet the standards, and pre-operation operating condition supervision is implemented. If at least one gas concentration is outside the normal concentration range, a ventilation voice reminder is generated. The safety supervision system implements pre-operation gas concentration supervision by determining whether the gas concentrations at the top, middle, and bottom of the confined space meet the standards.

[0086] The safety supervision system uses AI video analysis algorithms to determine whether there is any equipment missing at the work site and whether the workers have received construction technical briefings. If it is determined that there is equipment missing at the work site and the workers have received construction technical briefings, a voice reminder of missing equipment is generated. If it is determined that there is no equipment missing at the work site and the workers have not received construction technical briefings, a voice reminder of construction technical briefings is generated. If it is determined that there is equipment missing at the work site and the workers have not received construction technical briefings, a voice reminder of missing equipment and construction technical briefings is generated. If it is determined that there is no equipment missing at the work site and the workers have received construction technical briefings, a voice reminder of starting work is generated. Based on this, the safety supervision system realizes pre-operation working condition supervision.

[0087] In existing technologies, manual recording and verification are primarily used to determine whether key steps in the work process (such as construction technology briefings) have been executed, which carries the risk of human oversight. In embodiments of the present invention, digital detection and verification of construction technology briefings are performed to ensure the compliance of the work process. Optionally, the safety supervision system can also digitally detect and verify other key steps besides construction technology briefings based on actual usage needs.

[0088] In the embodiment of the present invention, after the operator receives the voice reminder to start the operation, the safety supervision system performs on-the-job supervision, which includes gas concentration supervision, personnel health supervision, and operation behavior supervision.

[0089] In confined space work scenarios, the safety monitoring system uses AI video analysis algorithms to determine whether workers are engaging in unsafe behavior. If it determines that a worker has engaged in unsafe behavior in N consecutive video frames, it generates a behavioral anomaly result and uploads it to the anomaly processing center. Unsafe behaviors include smoking, making phone calls, and not wearing a smart helmet. Optionally, the safety monitoring system can generate a voice reminder when a worker exhibits unsafe behavior.

[0090] In the prior art, fixed alarm logic and algorithm models are usually used for safety supervision in different operating scenarios, which cannot adapt to the diverse scenario requirements in gas operations. In the embodiments of the present invention, different safety supervision strategies are formulated for typical scenarios in gas operations (such as blocking operation scenarios and confined space operation scenarios). Different operating scenarios have corresponding video monitoring parameters and safety supervision strategies. The safety supervision system can automatically adjust the video monitoring parameters and safety supervision strategies according to different operating scenarios, monitor the operating environment in real time, and ensure that the operating environment meets the safety operation requirements to ensure the safety of the operators.

[0091] In the technical solution provided by the embodiment of the present invention, the safety supervision system receives and analyzes multi-source data from different collaborative devices in real time, and can fully perceive the safety status of the work site, thereby more accurately evaluating the safety status of the working environment and more accurately identifying potential hazardous factors. It avoids the problem of relying on a single data source (such as gas sensors or smart wearable devices) for monitoring in existing technologies, while ignoring the complex relationship between different data, resulting in missed detection of hazardous factors, thereby improving the accuracy of safety supervision and reducing the possibility of missed reports and false alarms.

[0092] In an embodiment of the present invention, when at least one of the gas concentration synchronization data, health monitoring synchronization data and video monitoring synchronization data is abnormal, the safety supervision system can quickly detect the abnormality and report it, with strong real-time performance, greatly improving the response speed of the system and ensuring the safety of personnel at the work site.

[0093] In this embodiment of the present invention, the security monitoring system pushes exception monitoring results and synchronized data to a data sharing platform via HTTP, enabling cross-platform data sharing and collaboration. This cross-platform data push and sharing capability enhances the flexibility and collaborative capabilities of the security monitoring system, enabling data synchronization and timely responses across different management platforms.

[0094] Figure 4 A schematic diagram of a comprehensive safety supervision system provided by an embodiment of the present invention is shown in FIG. Figure 4As shown in Figure 1, the integrated safety supervision system adopts a layered architecture design, consisting of a perception layer, an edge computing layer, and an application layer. The perception layer realizes data perception through collaborative devices, including gas sensors, smart wearable devices, and video surveillance equipment. The edge computing layer implements corresponding functions through the safety supervision system. The perception layer transmits the data to be processed to the edge computing layer via wireless or wired means.

[0095] The edge computing layer is used to obtain the data to be processed from multiple collaborative devices respectively. The data to be processed carries timestamp information (i.e., data acquisition); based on the timestamp information, the data to be processed is synchronized in time to generate synchronized data (i.e., data synchronization); based on the preset judgment rules, it is judged whether there are any anomalies in the synchronized data and a safety supervision result is generated (i.e., safety supervision); the safety supervision results are weighted and fused, and the safety status of the working environment is evaluated based on the calculation results (i.e., status assessment).

[0096] The application layer provides a web interface for querying real-time video surveillance data, historical anomaly monitoring results, setting judgment rules, and managing collaborative device configuration, model configuration, and scenario configuration. In practice, by distributing and configuring corresponding model files and scripts through a web interface, model and scenario configuration can be implemented for scenarios beyond blocking and confined space operations. This expands the scope of safety monitoring applications. Safety monitoring policies can be set based on these expanded application scenarios.

[0097] In the technical solution provided by the embodiment of the present invention, the safety supervision system receives and analyzes multi-source data from different collaborative devices in real time, and can fully perceive the safety status of the work site, thereby more accurately evaluating the safety status of the working environment and more accurately identifying potential hazardous factors. It avoids the problem of relying on a single data source (such as gas sensors or smart wearable devices) for monitoring in existing technologies, while ignoring the complex relationship between different data, resulting in missed detection of hazardous factors, thereby improving the accuracy of safety supervision and reducing the possibility of missed reports and false alarms.

[0098] In an embodiment of the present invention, when at least one of the gas concentration synchronization data, health monitoring synchronization data and video monitoring synchronization data is abnormal, the safety supervision system can quickly detect the abnormality and report it, with strong real-time performance, greatly improving the response speed of the system and ensuring the safety of personnel at the work site.

[0099] In this embodiment of the present invention, the security monitoring system pushes exception monitoring results and synchronized data to a data sharing platform via HTTP, enabling cross-platform data sharing and collaboration. This cross-platform data push and sharing capability enhances the flexibility and collaborative capabilities of the security monitoring system, enabling data synchronization and timely responses across different management platforms.

[0100] Figure 5 A schematic diagram of a device for evaluating the safety status of an operating environment provided by an embodiment of the present invention is shown in FIG. Figure 5 As shown, the device includes an acquisition module 11, a synchronization module 12, a judgment module 13, and an evaluation module 14. The acquisition module 11 is used to obtain the data to be processed from multiple collaborative devices, and the data to be processed carries timestamp information. The synchronization module 12 is used to synchronize the data to be processed based on the timestamp information and generate synchronized data. The judgment module 13 is used to determine whether there are any anomalies in the synchronized data based on preset judgment rules and generate a safety supervision result. The evaluation module 14 is used to perform weighted fusion calculation on the safety supervision results and evaluate the safety status of the working environment based on the calculation results.

[0101] In an embodiment of the present invention, the collaborative devices include gas sensors, smart wearable devices and video surveillance devices, and the data to be processed include gas concentration data, health monitoring data and video surveillance data; the acquisition module 11 is specifically used to obtain the gas concentration data collected by the gas sensor and the health monitoring data collected by the smart wearable device through the TCP protocol, HTTP protocol or MQTT protocol; and obtain the video surveillance data collected by the video surveillance device through the RTSP protocol.

[0102] In an embodiment of the present invention, the synchronization module 12 is specifically used to calibrate the clocks of the gas sensor and the smart wearable device through the Network Time Protocol NTP server, obtain the timestamp information of the gas concentration data and the timestamp information of the health monitoring data; extract the timestamp information of the video monitoring data through the video stream processing tool; synchronize the timestamp information of the gas concentration data, the timestamp information of the health monitoring data and the timestamp information of the video monitoring data to generate synchronized data.

[0103] In an embodiment of the present invention, the synchronized data includes gas concentration synchronized data, which is used to indicate the concentration data of various gases in the air. The safety monitoring results include abnormal concentration monitoring results or normal concentration monitoring results. The judgment module 13 is specifically configured to determine whether the concentration data of various gases are all within a preset normal concentration range. If the concentration data of at least one gas is determined to be outside the normal concentration range, an abnormal concentration monitoring result is generated and sent to the abnormality processing center. If the concentration data of various gases are all within the normal concentration range, a normal concentration monitoring result is generated, and the step of determining whether the concentration data of various gases are all within the preset normal range is continued.

[0104] In this embodiment of the present invention, the synchronized data includes health monitoring synchronized data, and the safety monitoring results include data abnormality monitoring results and data normality monitoring results. The judgment module 13 is specifically configured to determine whether the health monitoring synchronized data is within a preset normal data range. If the health monitoring synchronized data is determined to be outside the normal data range, a data abnormality monitoring result is generated and uploaded to the abnormality processing center. If the health monitoring synchronized data is determined to be within the normal data range, a data normality monitoring result is generated, and the step of determining whether the health monitoring synchronized data is within the preset normal range is continued.

[0105] In an embodiment of the present invention, the synchronization data includes video surveillance synchronization data, and the security supervision results include monitoring abnormality supervision results and monitoring normal supervision results. The judgment module 13 includes an analysis submodule, an abnormal submodule and a normal submodule. The analysis submodule is used to perform real-time analysis of video surveillance data through an artificial intelligence AI video analysis algorithm; when the analysis submodule detects an abnormal situation, it triggers the abnormal submodule to generate a monitoring abnormality supervision result, and uploads the monitoring abnormality supervision result to the abnormality processing center; when the analysis submodule does not detect an abnormal situation, it triggers the normal submodule to generate a monitoring normal supervision result, and continues to execute the step of performing real-time analysis of video surveillance data through the AI ​​video analysis algorithm.

[0106] In the embodiment of the present invention, the monitoring abnormal supervision result includes at least one of an equipment missing abnormal result, an open flame abnormal result, a behavior abnormal result, and a non-operating personnel intrusion abnormal result.

[0107] In an embodiment of the present invention, the analysis submodule is specifically used to detect at least one target in the video surveillance data based on a target detection algorithm and generate at least one target frame; track at least one target based on a target tracking algorithm and generate a target motion trajectory; and perform real-time analysis of the video surveillance data based on the target motion trajectory.

[0108] In an embodiment of the present invention, the device also includes a push module 15, which is used to push abnormal supervision results and synchronous data to the front end and the safety supervision system front end and data sharing platform; the abnormal supervision results include at least one of concentration abnormal supervision results, data abnormal supervision results and monitoring abnormal supervision results.

[0109] In the technical solution provided by the embodiment of the present invention, the safety supervision system receives and analyzes multi-source data from gas sensors, smart wearable devices and video surveillance equipment in real time, and can fully perceive the safety status of the work site, thereby more accurately assessing the safety status of the working environment and more accurately identifying potential hazardous factors. It avoids the problem of relying on a single data source (such as a gas sensor or smart wearable device) for monitoring in the existing technology, while ignoring the complex relationship between different data, resulting in missed detection of hazardous factors, thereby improving the accuracy of safety supervision and reducing the possibility of missed reports and false alarms.

[0110] In an embodiment of the present invention, when at least one of the gas concentration synchronization data, health monitoring synchronization data and video monitoring synchronization data is abnormal, the safety supervision system can quickly detect the abnormality and report it, with strong real-time performance, greatly improving the response speed of the system and ensuring the safety of personnel at the work site.

[0111] In this embodiment of the present invention, the security monitoring system pushes exception monitoring results and synchronized data to a data sharing platform via HTTP, enabling cross-platform data sharing and collaboration. This cross-platform data push and sharing capability enhances the flexibility and collaborative capabilities of the security monitoring system, enabling data synchronization and timely responses across different management platforms.

[0112] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program. When the program is running, the device where the computer-readable storage medium is located is controlled to execute the steps of the embodiment of the above-mentioned working environment safety status assessment method. For a specific description, please refer to the embodiment of the above-mentioned working environment safety status assessment method.

[0113] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for assessing the safety status of an operating environment, characterized in that: The method comprises: respectively obtain data to be processed from a plurality of collaborative devices, wherein the data to be processed carries timestamp information; Based on the timestamp information, time synchronization is performed on the data to be processed to generate synchronized data; Determine whether the synchronization data is abnormal based on preset judgment rules and generate a safety supervision result; A weighted fusion calculation is performed on the safety supervision results, and the safety status of the operating environment is evaluated based on the calculation results.

2. The method according to claim 1, characterized in that The collaborative devices include gas sensors, smart wearable devices and video monitoring devices, and the data to be processed include gas concentration data, health supervision data and video monitoring data; The step of respectively obtaining the data to be processed from the plurality of collaborative devices includes: Obtain gas concentration data collected by gas sensors and health monitoring data collected by smart wearable devices through the Transmission Control Protocol (TCP), Hypertext Transfer Protocol (HTTP), or Message Queue Telemetry Transmission (MQTT) protocol; The video surveillance data collected by the video surveillance device is obtained through the real-time streaming protocol RTSP.

3. The method according to claim 2, characterized in that Based on the timestamp information, time synchronization is performed on the data to be processed to generate synchronization data, including: Calibrate the clocks of the gas sensor and the smart wearable device through a Network Time Protocol (NTP) server to obtain timestamp information of the gas concentration data and timestamp information of the health monitoring data; Extracting timestamp information of the video surveillance data through a video stream processing tool; Time synchronization is performed on the timestamp information of the gas concentration data, the timestamp information of the health monitoring data, and the timestamp information of the video monitoring data to generate synchronized data.

4. The method according to claim 1, wherein The synchronized data includes gas concentration synchronized data, which is used to indicate concentration data of various gases in the air. The safety supervision result includes an abnormal concentration supervision result or a normal concentration supervision result. The determining whether the synchronized data is abnormal based on a preset judgment rule and generating the safety supervision result include: Determine whether the concentration data of various gases are within the preset normal concentration range; If it is determined that the concentration data of at least one gas is outside the normal concentration range, generating a concentration abnormality monitoring result, and sending the concentration abnormality monitoring result to the abnormality processing center; If it is determined that the concentration data of various gases are all within the normal concentration range, a normal concentration monitoring result is generated, and the step of determining whether the concentration data of various gases are all within the preset normal range is continued.

5. The method according to claim 1, wherein The synchronized data includes health monitoring synchronized data, the safety monitoring result includes data abnormality monitoring result and data normal monitoring result, and the determining whether the synchronized data is abnormal based on a preset judgment rule and generating the safety monitoring result includes: Determining whether the health monitoring synchronization data is within a preset normal data range; If it is determined that the health supervision synchronization data is outside the normal range of the data, a data abnormality supervision result is generated, and the data abnormality supervision result is uploaded to the abnormality processing center; If it is determined that the health monitoring synchronization data is within the normal data range, a normal data monitoring result is generated, and the step of determining whether the health monitoring synchronization data is within the preset normal range is continued.

6. The method according to claim 1, characterized in that The synchronized data includes video surveillance synchronized data, the security supervision result includes abnormal monitoring supervision results and normal monitoring supervision results, and the determining whether the synchronized data is abnormal based on a preset judgment rule and generating the security supervision result includes: Perform real-time analysis of the video surveillance data using artificial intelligence (AI) video analysis algorithms; When an abnormal situation is detected, a monitoring abnormality supervision result is generated and uploaded to the abnormality processing center; When no abnormal situation is detected, a normal monitoring result is generated, and the step of performing real-time analysis on the video monitoring data through the AI ​​video analysis algorithm is continued.

7. The method according to claim 6, characterized in that The monitoring abnormal supervision result includes at least one of an equipment missing abnormal result, an open flame abnormal result, an abnormal behavior result, and a non-operating personnel intrusion abnormal result.

8. The method according to claim 6, characterized in that The real-time analysis of the video surveillance data using the AI ​​video analysis algorithm includes: Detecting at least one target in the video surveillance data based on a target detection algorithm and generating at least one target frame; Tracking the at least one target based on a target tracking algorithm to generate a target motion trajectory; The video surveillance data is analyzed in real time based on the target motion trajectory.

9. The method according to claim 4, 5 or 6, characterized in that The method further comprises: Push abnormal supervision results and synchronized data to the front end and data sharing platform; the abnormal supervision results include at least one of abnormal concentration supervision results, abnormal data supervision results and abnormal monitoring supervision results.

10. A device for evaluating the safety status of an operating environment, characterized in that: The device comprises: An acquisition module, configured to acquire data to be processed from a plurality of collaborative devices, wherein the data to be processed carries timestamp information; A synchronization module, configured to perform time synchronization on the data to be processed based on the timestamp information and generate synchronized data; A judgment module, configured to judge whether the synchronization data has any anomalies based on preset judgment rules and generate a safety supervision result; The evaluation module is used to perform weighted fusion calculation on the safety supervision results and evaluate the safety status of the operating environment based on the calculation results.

Images