Whole-process personal safety monitoring intelligent identification and analysis system special for sewage treatment plant

Abstract

The invention discloses a whole-process personal safety monitoring intelligent identification and analysis system special for a sewage treatment plant, and the system comprises an intelligent safety management platform, a mobile terminal, a switch, an edge server, and a plurality of detectors which are used for collecting the personnel and regional environment information of the sewage treatment plant. According to feedback information of the plurality of detectors, the intelligent safety management platform supervises equipment, personnel and environment safety conditions of the whole plant in a centralized manner, can intelligently identify abnormal conditions and give an alarm, and is linked with a mobile terminal. The device is reasonable in overall structure, high in practicability and wide in application, is suitable for municipal projects such as sewage treatment plants, pump stations and regulation and storage pond projects, can also be popularized to projects such as comprehensive pipe galleries and sludge treatment plants, and has wide application prospects; personnel operation is standardized, the accident rate and loss are reduced, and the safety of a sewage treatment plant is guaranteed; and the requirements of current technology development on intelligent factories and intelligent operation and maintenance of drainage engineering are met.

Description

Technical Field

[0001] The invention relates to sewage treatment technology, and in particular to a whole-process personal safety monitoring intelligent identification and analysis system specially used in sewage treatment plants. Background Art

[0002] The operation of sewage treatment plants involves special factors such as humidity and a series of complex processes and equipment operations, which involve certain safety risks and will have an impact on personal safety. In analyzing previous safety accident cases, the more common accidents include fainting due to poisoning, drowning in sewage, slipping and falling, and electric shock. Accidents in sewage treatment plants often occur during equipment maintenance, and the probability of production safety accidents is often higher than in daily operations. The main reason is that the staff lacks correct operating instructions or ignores operating specifications, which makes it easy to perform irregular operations, including not wearing or using improper personal protective equipment, being careless when operating equipment, and entering places that are not equal to their own rank, which leads to the risk of accidents for staff.

[0003] Since sewage treatment plants usually have slippery environments and special conditions such as working at heights, workers are prone to fall accidents. For example, slipping while walking on the operating table, not using handrails or non-slip shoes correctly, etc. can all lead to falls. Fall accidents may cause physical injuries to workers and even trigger other chain accidents.

[0004] Electrical equipment and systems in wastewater treatment plants are critical components of their operation. If the equipment fails, is improperly maintained, or lacks necessary safety protection measures, workers may be at risk of electric shock. Electric shock accidents can cause electric shock injuries, burns, or even death. Electric shock can also cause fires and equipment damage, posing a major threat to the normal operation and safety of the plant.

[0005] In addition, with the improvement of my country's urbanization level and residents' environmental requirements, underground (semi-underground, fully underground) sewage treatment plants that can coordinate with the surrounding environment, are highly closed, and have no secondary pollution are gradually becoming a new idea for engineering construction. However, the space of underground sewage treatment plants is humid and sealed, and toxic and harmful gases are easy to accumulate, which is very likely to cause poisoning, suffocation, explosions and other personal safety accidents.

[0006] The patent document with application number "CN202110803606.9" discloses an internal safety monitoring system for sewage treatment, including a central processing unit and a data acquisition module, wherein the central processing unit and the data acquisition module are unidirectionally connected via a wireless transmission module, and the input end of the data acquisition module is electrically connected to the output end of a pH detector and a liquid level sensor, a COD meter, a gas concentration meter, a water hardness meter, and a dissolved oxygen meter via wires. The present invention relates to the technical field of sewage treatment monitoring. The internal safety monitoring system for sewage treatment is electrically connected to the input end of a central processing unit via a wire through the output end of a data analysis module. When monitoring the safety of sewage treatment, after inputting a standard value in a data entry module, if the data is greater than or less than the standard value, it can be monitored, so that employees can intuitively observe the abnormal data, thereby facilitating unified management thereof.

[0007] The patent document with application number "CN201810950109.X" discloses a medical sewage treatment safety three-dimensional monitoring cloud platform, system and method. Among them, a medical sewage treatment safety three-dimensional monitoring system includes: an intelligent monitoring terminal, which obtains intelligent monitoring terminal information and sends it to the cloud platform; obtains the current image data of the medical sewage treatment equipment through a full-view camera and sends it to the cloud platform; the cloud platform receives the intelligent monitoring terminal information, integrates the real-time operation data of the medical sewage treatment equipment, the real-time water quality data of the outlet sewage and the environmental safety factor data of the medical sewage treatment equipment in the intelligent monitoring terminal information, monitors the operation status of the medical sewage treatment equipment in real time, generates alarm information when data abnormalities occur, and sends an image acquisition command to the intelligent monitoring terminal at the same time, receives the current image data of the medical sewage treatment equipment, and sends it to the user terminal; the user terminal receives the alarm information sent by the cloud platform and the current image data of the medical sewage treatment equipment for display.

[0008] The above patent document, combined with the prior art, reveals that the existing intelligent identification and analysis system for full-process personal safety monitoring in sewage treatment plants has the following defects:

[0009] At present, domestic and foreign sewage treatment plants do not have a full-process safety warning system for personal safety during the operation and maintenance stage. In recent years, domestic sewage treatment plants have begun to pay attention to the operation and maintenance management of personal safety. Some sewage treatment plants have gradually introduced a safety production management platform, which is only used to standardize operating behaviors and manage corporate safety systems, and has not yet involved real-time accident alarms and post-disaster emergency rescue. Summary of the invention

[0010] In order to overcome the deficiencies of the prior art, the present invention provides a whole-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants, which solves the problem of the whole-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants.

[0011] The first aspect of the present invention is to provide a whole-process personal safety monitoring intelligent identification and analysis system dedicated to a sewage treatment plant, comprising an intelligent safety management platform, a mobile terminal, a switch, an edge server, and a plurality of detectors, wherein the plurality of detectors establish a communication connection with the edge server, the edge server establishes a communication connection with the switch, the switch establishes a communication connection with the intelligent safety management platform, and the mobile terminal is transmitted to the intelligent safety management platform via wireless transmission;

[0012] The multiple detectors are used to collect personnel and regional environmental information of the sewage treatment plant. Based on the feedback information from the multiple detectors, the intelligent safety management platform centrally monitors the equipment, personnel, and environmental conditions of the entire plant, and can intelligently identify abnormal conditions and issue alarms, while also interacting with mobile terminals.

[0013] In the first aspect of the present invention, as a preferred embodiment, the detector includes one or more of a video probe, a millimeter wave radar sensor / smart bracelet, and an environment detection sensor;

[0014] The video probe is used to collect information about the entry and exit, operation, behavior, and status of personnel in the sewage treatment plant, and the millimeter wave radar sensor / smart bracelet is used to detect the number of people in the sewage treatment plant, movement trajectory, and respiratory heart rate information;

[0015] The environmental detection sensor is used to detect H in the dangerous area of ​​the sewage treatment plant 2 S. NH 3 , CH 4 Gas concentration and O 2 , temperature and humidity data, the edge server is used for sewage treatment plant data collection, storage, calculation analysis, and display functions, and can collect video, millimeter wave radar, smart bracelet, H 2 S. NH 3 , CH 4 , O 2 , temperature and humidity sensors and other data, and provide intelligent AI computing services for the system.

[0016] In the first aspect of the present invention, as a preferred embodiment, the intelligent security management platform includes a basic functional unit, and the basic functional unit includes a user management module, a device management module, an electronic archive management module, a data query and statistics module and an electronic map module;

[0017] The user management module can view user names, user types, user levels, and has the functions of adding and removing users and changing passwords;

[0018] The device management module can monitor the status of the system equipment in real time, including the operating status of detectors, communication devices, edge servers, etc., and has the functions of adding, deleting, and modifying;

[0019] The electronic archive management module can support the management of security electronic archive files, including storage, query, upload, delete, online access, and download functions; the data query and statistics module includes data storage, equipment query, video query, alarm type, alarm level query, etc.; it provides a chart display function for data such as temperature, humidity, gas concentration, equipment abnormality, environmental abnormality, and personnel abnormality;

[0020] The electronic map module includes a three-dimensional electronic map, which can display the location of each device, personnel, and fault point of the system on the electronic map and real-time video in real time, and has real-time recording, uploading and exporting functions.

[0021] In the first aspect of the present invention, as a preferred embodiment, the intelligent safety management platform also includes a core functional unit, and the core functional unit includes a job management module; the job management module manages operator information and identifies irregular operations.

[0022] In the first aspect of the present invention, as a preferred embodiment, the core functional unit also includes a monitoring and early warning module, which performs multi-source heterogeneous data fusion analysis to achieve the function of monitoring the status of personnel and the on-site environment throughout the entire process. The monitoring and early warning module is configured with a video probe unit, which automatically identifies the status of equipment and irregular operations of personnel.

[0023] In the first aspect of the present invention, as a preferred embodiment, the core functional unit also includes an accident handling module, which includes an alarm information push unit, an accident handling data support unit, a processing process tracking unit and an accident handling summary unit, and is continuously advanced through the alarm information push unit, the accident handling data support unit, the processing process tracking unit and the accident handling summary unit.

[0024] The second aspect of the present invention provides a method for using an intelligent identification and analysis system for full-process personal safety monitoring in a sewage treatment plant, which is used for multi-source heterogeneous data fusion analysis, comprising the following steps:

[0025] Step S10: Integrate the information of multiple detectors and upload the data to the intelligent safety management platform synchronously;

[0026] Step S20: Perform data cleaning to ensure the accuracy and consistency of multi-source data;

[0027] Step S30: Use a feature-level fusion algorithm to merge multi-source feature data and input them into a unified model for analysis: standardize the features extracted from video, millimeter-wave radar, smart bracelet, and gas detection instrument so that they are fused on the same scale to avoid unbalanced effects of different features on the model. Then perform feature fusion to concatenate feature vectors from different data sources into a long feature vector. According to the importance of different data sources, weights are assigned to different features for weighted fusion. The fused features are input into the machine learning model for training and prediction.

[0028] Step S40: Calculate a comprehensive risk score based on the output of the fusion model. The score can be expressed in a range of 0 to 100, where 0 indicates no risk and 100 indicates high risk. Set an early warning threshold based on the risk score. When the risk score exceeds the set threshold, trigger the early warning system to notify relevant personnel and take corresponding measures.

[0029] Step S50: Evaluate the accuracy, recall rate and other performance indicators of the fusion model, and adjust the model parameters according to feedback and actual conditions, optimize feature selection and fusion methods, and improve the accuracy and reliability of comprehensive risk assessment.

[0030] A third aspect of the present invention provides an electronic device, comprising: a processor;

[0031] Memory; and a program, wherein the program is stored in the memory and is configured to be executed by a processor, and the program includes a method for executing an application dedicated to the full-process personal safety monitoring intelligent identification and analysis system for sewage treatment plants.

[0032] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor and is applied to a method for using a full-process personal safety monitoring intelligent identification and analysis system specifically for a sewage treatment plant.

[0033] The fifth aspect of the present invention provides a full-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants, the whole process includes three stages: pre-prevention, in-process alarm, and post-processing;

[0034] Pre-event prevention: Define warning areas based on the time periods and places where high-frequency accidents occur in the knowledge base, and issue warning reminders to people entering;

[0035] Alarm during the event: Collect detector data and conduct multi-source heterogeneous data fusion analysis to determine personal safety status, personnel operation specifications, etc.

[0036] Post-accident processing: After an accident occurs, rescue and accident handling are carried out according to the pre-set emergency plan, and an accident handling report is formed, marking the event from the two dimensions of time and location, marking the time periods and places where high-frequency accidents occur.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] (1) The present invention has a reasonable overall structure, strong practicability, and wide application. It is suitable for municipal projects such as sewage treatment plants, pumping stations, and storage tank projects. It can also be extended to integrated pipe corridors, sludge treatment plants, garbage incineration plants, integrated pipe corridors, and other similar projects, and has broad application prospects.

[0039] (2) The present invention realizes the whole process identification and comprehensive analysis of personnel safety behavior in three stages: pre-prevention, in-process alarm, and post-processing. It uses artificial intelligence technology to combine and analyze data such as operations, environment, personnel behavior, and personnel health. After abnormal working conditions are found, the alarm is given immediately to standardize personnel operations, reduce the accident rate and losses, and ensure the safety of sewage treatment plants.

[0040] (3) Introducing millimeter-wave radar human monitoring technology in the drainage industry, so that the system can establish a comprehensive AI intelligent analysis model based on multiple data such as video, human health data, and environmental data, conduct multi-source heterogeneous data fusion analysis, and conduct comprehensive research and judgment to intelligently identify scenes such as personnel health and behavior, equipment status, and environmental anomalies, with higher accident identification accuracy;

[0041] (4) Through targeted research on edge servers, the computing load of central servers can be reduced, computing power can be provided for AI analysis, edge deployment can be strengthened, and the dependence of the entire system on central servers can be reduced, thereby achieving higher system reliability.

[0042] (5) The present invention can improve the intelligence level of sewage treatment plants, especially underground sewage treatment plants, and meet the requirements of current technological development for intelligent factories and intelligent operation and maintenance of drainage projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a structural diagram of the intelligent identification and analysis system for the whole process of personal safety monitoring in sewage treatment plants;

[0044] Figure 2 This is a schematic diagram of the functional modules of the intelligent security management platform;

[0045] Figure 3 A schematic diagram of a preferred embodiment of the structure of an intelligent security management platform;

[0046] Figure 4 It is a flowchart for the fusion analysis of multi-source heterogeneous data;

[0047] Figure 5 Provides a flow chart for intelligent identification and analysis of the entire process of personal safety monitoring;

[0048] Figure 6 The present invention is a flowchart of the method of using the intelligent identification and analysis system for full-process personal safety monitoring in sewage treatment plants.

[0049] In the figure: mobile terminal-1, intelligent security management platform-2, switch-3, edge server-4~5, dangerous area-6, video probe-8~9, millimeter wave radar sensor / smart bracelet-10~11, environmental detection sensor-12~13, transformer room and control room-7, video probe-14~15, millimeter wave radar sensor / smart bracelet-16~17;

[0050] User management module-201, equipment management module-202, electronic file management module-203, data query and statistics module-204, electronic map module-205, operation management module-206, monitoring and early warning module-207, accident handling module-208;

[0051] Security control module-401, on-duty personnel-402, operation and maintenance personnel-403. DETAILED DESCRIPTION

[0052] Below, in conjunction with the accompanying drawings and specific embodiments, the invention is further described. It should be noted that, under the premise of no conflict, the various embodiments described below or the various technical features can be arbitrarily combined to form a new embodiment. Except as otherwise specified, the materials and equipment used in this embodiment can be purchased from the market. Examples of the embodiments are shown in the accompanying drawings, in which the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0053] In the description of the present application, it should be understood that the terms "upper", "lower", "front", "back", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific position, be constructed and operated in a specific position, and therefore cannot be understood as a limitation on the present application. In the description of the present application, "plurality" means two or more, unless otherwise precisely and specifically specified.

[0054] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "connected", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, or it can be connected through an intermediary medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0055] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0056] like Figure 1-5 As shown, a whole-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants includes an intelligent safety management platform, a mobile terminal, a switch, an edge server, and multiple detectors. The multiple detectors establish a communication connection with the edge server, the edge server establishes a communication connection with the switch, the switch establishes a communication connection with the intelligent safety management platform, and the mobile terminal transmits the information to the intelligent safety management platform via wireless. The multiple detectors are used to collect personnel and regional environmental information of the sewage treatment plant. According to the feedback information of the multiple detectors, the intelligent safety management platform centrally monitors the equipment, personnel, and environmental conditions of the entire plant, and can intelligently identify abnormal conditions and alarm, and at the same time interact with the mobile terminal.

[0057] In the first aspect of the present invention, as a preferred embodiment, the detector includes one or more of a video probe, a millimeter wave radar sensor / smart bracelet, and an environmental detection sensor; the video probe is used to collect information on the entry and exit, operation, behavior, and status of personnel in the sewage treatment plant; the millimeter wave radar sensor / smart bracelet is used to detect the number of people in the sewage treatment plant, movement trajectory, and respiratory heart rate information; the environmental detection sensor is used to detect H in the dangerous area of ​​the sewage treatment plant 2 S. NH 3 , CH 4 Gas concentration and O 2 , temperature and humidity data, the edge server is used for sewage treatment plant data collection, storage, calculation analysis, and display functions, and can collect video, millimeter wave radar, smart bracelet, H 2 S. NH 3 , CH 4 , O2 , temperature and humidity sensors, and provide intelligent AI computing services for the system. Specifically, various multi-source heterogeneous sensors provide rich, complete, and mutually verified analysis data for the intelligent identification and analysis system for full-process personal safety monitoring dedicated to sewage treatment plants, giving the system a high level of identification accuracy.

[0058] In the first aspect of the present invention, as a preferred embodiment, the intelligent security management platform includes a basic functional unit, which includes a user management module, a device management module, an electronic file management module, a data query and statistics module and an electronic map module; the user management module can view user names, user types, and user levels, and has the functions of adding and reducing users and changing passwords; the device management module can monitor the status of the system equipment in real time, including the operating status of detectors, communication devices, edge servers, etc., and has the functions of adding, deleting, and modifying; the electronic file management module can support security electronic file management, including storage, query, upload, delete, online viewing, and download functions; the data query and statistics module includes data storage, device query, video query, alarm type, alarm level query and other contents; it provides a chart display function for data such as temperature, humidity, gas concentration, equipment abnormality, environmental abnormality, and personnel abnormality; the electronic map module includes a three-dimensional electronic map, which can display the location and real-time video of each device, personnel, and fault point of the system on the electronic map in real time, and has real-time recording, upload and export functions.

[0059] Specifically, this application has the following advantages:

[0060] (1) The present invention has a reasonable overall structure, strong practicability, and wide application. It is suitable for municipal projects such as sewage treatment plants, pumping stations, and storage tank projects. It can also be extended to integrated pipe corridors, sludge treatment plants, garbage incineration plants, integrated pipe corridors, and other similar projects, and has broad application prospects.

[0061] (2) The present invention realizes the whole process identification and comprehensive analysis of personnel safety behavior in three stages: pre-prevention, in-process alarm, and post-processing. It uses artificial intelligence technology to combine and analyze data such as operations, environment, personnel behavior, and personnel health. After abnormal working conditions are found, the alarm is given immediately to standardize personnel operations, reduce the accident rate and losses, and ensure the safety of sewage treatment plants.

[0062] (3) Introducing millimeter-wave radar human monitoring technology in the drainage industry, so that the system can establish a comprehensive AI intelligent analysis model based on multiple data such as video, human health data, and environmental data, conduct multi-source heterogeneous data fusion analysis, and conduct comprehensive research and judgment to intelligently identify scenes such as personnel health and behavior, equipment status, and environmental anomalies, with higher accident identification accuracy;

[0063] (4) Through targeted research on edge servers, the computing load of central servers can be reduced, computing power can be provided for AI analysis, edge deployment can be strengthened, and the dependence of the entire system on central servers can be reduced, thereby achieving higher system reliability.

[0064] (5) The present invention can improve the intelligence level of sewage treatment plants, especially underground sewage treatment plants, and meet the requirements of current technological development for intelligent factories and intelligent operation and maintenance of drainage projects.

[0065] In the first aspect of the present invention, as a preferred embodiment, the intelligent safety management platform also includes a core functional unit, and the core functional unit includes a job management module; the job management module performs operator information management and irregular job identification. Irregular job identification uses AI intelligent algorithms to identify the number of operators, personnel locations, and personnel behaviors, and conducts statistical analysis to identify irregular job behaviors, such as unauthorized entry into high-level places (face recognition, matching with login information), failure to wear safety helmets, work clothes, insulated shoes, smoking (open flames), single entry into specific areas, etc., and issues an alarm.

[0066] In the first aspect of the present invention, as a preferred embodiment, the core functional unit also includes a monitoring and early warning module, which performs multi-source heterogeneous data fusion analysis to realize the function of monitoring the status of personnel and the on-site environment throughout the whole process. Through AI intelligent algorithms, the behavior and health status of personnel in specific areas, such as whether they have fallen, received electric shocks, whether their vital signs data are abnormal, etc., are identified, and statistical analysis is performed, which is intuitively displayed on the platform and mobile terminals; the on-site environment is monitored in real time, and the on-site H 2 S. NH 3 , CH 4 , O 2 , temperature, humidity and other conditions are displayed intuitively in the form of charts, and can automatically perform data analysis and alarm according to the alarm threshold. The alarm information should provide long-term alarm prompts and support manual or automatic recovery. It also has a communication alarm function. In the event of communication interruption, abnormal content of the received message, etc., the alarm information will be sent and the alarm information will be permanently saved (can be deleted manually).

[0067] In the first aspect of the present invention, as a preferred embodiment, the monitoring and early warning module is equipped with a video probe unit, which automatically identifies the equipment status and the irregular operation of personnel. Specifically, by analyzing the data of millimeter wave radar and smart bracelet, the health status (heart rate, heartbeat, breathing) and behavior status (walking, falling, sleeping, etc.) of personnel can be automatically identified; by analyzing the H 2 S. NH 3 , CH 4 , O 2 , temperature and humidity sensors and other environmental detection sensors.2 S. NH 3 , CH 4 When any concentration reaches the alarm threshold, or 2 When the concentration is lower than the threshold, the safety monitoring intelligent identification and analysis system will immediately notify the staff through the chain broadcasting system to evacuate immediately to prevent accidents.

[0068] In the first aspect of the present invention, as a preferred embodiment, the core functional unit also includes an accident handling module, which includes an alarm information push unit, an accident handling data support unit, a processing process tracking unit and an accident handling summary unit, and is continuously advanced through the alarm information push unit, the accident handling data support unit, the processing process tracking unit and the accident handling summary unit. The alarm information push is graded and pushed to the relevant person in charge, wherein the first-level alarm is the highest level, which not only requires a page pop-up alarm, but also should be pushed to the relevant responsible personnel and supervisors in the form of a short message immediately, and should be handled immediately; the accident handling data support provides relevant data for accident handling, including accident description, accident location (marked on an electronic map), video / radar and environmental data of the accident site, relevant personnel information, etc.; the processing process tracking system tracks and records each accident handling process, saves various types of data at the accident site, processing flow data, processing process data, processing time, accident handling results, etc.; the accident handling summary summarizes and evaluates the causes of the accident, proposes improvement measures to prevent similar emergency events, and optimizes the emergency plan.

[0069] For details, please refer to Figure 1 , Figure 1 The figure shows a schematic diagram of the overall framework of a specific implementation method provided by the present invention, including a mobile terminal 1, an intelligent security management platform 2, a switch 3, edge servers 4-5, video probes 8-9, millimeter-wave radar sensors / smart bracelets 10-11 and environmental detection sensors 12-13 deployed in a dangerous area 6, and video probes 14-15 and millimeter-wave radar sensors / smart bracelets 16-17 deployed in a transformer room and a control room 7.

[0070] At the field level of the sewage treatment plant, the detection locations are mainly divided into two types: dangerous area 6 and transformer room and control room 7. Video probes 8-9, millimeter wave radar sensors / smart bracelets 10-11 and environmental detection sensors 12-13 are set in the dangerous area 6. Video probes 8-9 transmit data to edge server 4 by wire, millimeter wave radar sensors / smart bracelets 10-11 transmit data to edge server 4 by wire or wireless, and environmental detection sensors 12-13 transmit data to edge server 4 by wire; video probes 14-15 and millimeter wave radar sensors / smart bracelets 16-17 are set in the transformer room and control room 7. Video probes 14-15 transmit data to edge server 5 by wire, and millimeter wave radar sensors / smart bracelets 16-17 transmit data to edge server 5 by wire or wireless. Edge servers 4-5 process the collected video, millimeter wave radar, smart bracelet, environmental parameters and other data and perform comprehensive AI intelligent analysis. When H is detected, 2 S. NH 3 , CH 4 , O 2 , temperature, humidity and other environmental data are abnormal, alarm 2 S. NH 3 , CH 4 , O 2 The first-level alarm thresholds of the environmental detection sensor can be set to 6.6ppm, 25ppm, 10%LEL, and 19.50%VOL respectively. By intelligently analyzing the video, millimeter-wave radar, and smart bracelet data, if an abnormal situation such as equipment or personnel behavior is comprehensively judged, an alarm will be issued in time, and the processed and analyzed data will be uploaded to the intelligent security management platform 2 through the switch 3.

[0071] At the central control layer of the sewage treatment plant, the intelligent safety management platform 2 is installed on the server in the central control room, which is used to centrally supervise the safety monitoring intelligent identification analysis and early warning system of the entire plant. It can intelligently judge the behavior and health status of personnel in the sewage treatment plant based on video data, millimeter wave radar, smart bracelet data and gas detection instrument data, and combine data such as operations, environment, personnel behavior, and personnel health for comprehensive analysis. When abnormal conditions are found, the alarm will be issued as soon as possible, and alarms and reports will be automatically generated. At the same time, the intelligent safety management platform 2 can be linked with the mobile terminal 1.

[0072] For details, please refer to Figure 2 The intelligent safety management platform 2 includes two types of functions: basic and core. The basic functions include a user management module 201, a device management module 202, an electronic file management module 203, a data query and statistics module 204, and an electronic map module 205; the core functions include an operation management module 206, a monitoring and early warning module 207, and an accident handling module 208.

[0073] The user management module 201 is used to efficiently and securely manage users, including functions such as user registration, login, authority management, user information management, password management, and the creation of a log record table to record user activities and important system operation logs, thereby facilitating auditing and backtracking.

[0074] The equipment management module 202 is used to monitor the operating status of the equipment in the factory in real time. Each device in this system is equipped with a unique IP, setting area, product type and other information. Users can search for devices based on IP, area, type, and display the device status information and detection data on the platform, thereby improving equipment utilization and maintenance efficiency.

[0075] The electronic archive management module 203 is used to store, manage and track electronic documents, including functions such as document uploading, storage, retrieval, downloading, archiving and backup. On this basis, a security management module is added to ensure the security of file transmission and storage, and to implement data encryption, access control and secure authentication and authorization mechanisms.

[0076] The data query and statistics module 204, as a core function in the data management system, enables users to retrieve, analyze and generate reports on relevant data sets. Staff can use SQL statements to query, filter, aggregate and count relevant data, and perform trend analysis and prediction by analyzing the trend of data changes over time, such as the number of H in a sewage treatment plant in a certain period of time. 2 The concentration of gases such as S is prone to increase or dangerous situations are prone to occur in a certain area. Finally, the data visualization library is used to display the statistical results on the front end in the form of charts, etc., and generate data reports to help the staff in the sewage treatment plant make more informed decisions based on the data.

[0077] The electronic map module 205 utilizes a combination of GIS (Geographic Information System) technology and BIM technology to display in real time the operating status of sewage treatment plant equipment, such as on / off status, abnormal operation warnings, etc., as well as the points where personnel abnormalities occur. By setting event trigger conditions and notification mechanisms, when the system detects an abnormality or important event, it automatically sends alarms and notifications to relevant personnel, effectively improving the management efficiency and response speed of the sewage treatment plant.

[0078] The operation management module 206, as a key link in the operation of the sewage treatment plant, is used to ensure the efficient and safe operation of the operation process. It is mainly used for the management of operator information and plans, especially the monitoring of the operation process. Through AI intelligent algorithms, the number of operators, personnel positioning, and personnel behavior are identified, and statistical analysis is performed. By comparing with the data in the operation plan, it is intuitively displayed whether the operation process is standardized. When the comprehensive analysis based on video, millimeter wave radar sensor / smart bracelet and other data shows that there are non-standard operation behaviors, such as unauthorized entry into high-level places (face recognition, matching with login information), failure to wear safety helmets and work clothes, smoking (open flames), etc., an alarm is immediately issued; if the operation is identified as a special operation in the database, a supervisor should be designated to supervise the operation and wear a supervisor armband. Ventilation, detection, downgrade and other operations should be performed 30 minutes before the operation. The operation can only be approved after the record is completed in the system, otherwise an alarm will be issued.

[0079] The monitoring and early warning module 207 is used to realize functions such as real-time monitoring of personnel status and on-site environment. Through AI intelligent algorithms, real-time monitoring of personnel status, identification of personnel behavior and health status in specific areas, such as whether they have fallen, electric shock, whether their vital signs data are abnormal, etc., and statistical analysis is performed, and intuitively displayed on the platform and mobile terminals; real-time monitoring of the on-site environment, such as on-site H 2 S. NH 3 , CH 4 , O 2 , temperature, humidity and other conditions, and display them visually in the form of charts. Data analysis and alarm are automatically performed through the set alarm thresholds. The alarm information should provide long-term alarm prompts and support manual or automatic recovery. It also has a communication alarm function. In the event of communication interruption, abnormal content of the received message, etc., the alarm information will be sent up and stored permanently (can be deleted manually). General video information will be automatically overwritten in 90 days.

[0080] The accident handling module 208 includes alarm information push, accident handling data support, handling process tracking and accident handling summary functions. The alarm information push is graded and pushed to relevant persons in charge, where the first-level alarm is the highest level, which not only requires a page pop-up alarm, but also should be immediately pushed to relevant responsible persons and supervisors in the form of short messages, and should be handled immediately; the accident handling data support provides relevant data for accident handling, including accident description, accident location (marked on an electronic map), video / radar and environmental data of the accident site, relevant personnel information, etc.; the handling process tracking system tracks and records each accident handling process, and saves various types of data on the accident site, handling flow data, handling process data, handling time, accident handling results, etc.; the accident handling summary summarizes and evaluates the causes of the accident, proposes improvement measures to prevent similar emergency events, and optimizes emergency plans.

[0081] For details, please refer to Figure 3 Taking the dangerous area of ​​the sewage treatment plant as an example, a video probe 8, a millimeter-wave radar sensor / smart bracelet 10 and an environmental detection sensor 12 are set in the space. The image video data collected by the video probe 8, the human health data collected by the millimeter-wave radar sensor / smart bracelet 10 and the gas concentration data collected by the environmental detection sensor 12 are uploaded to the edge server 4. The security control module 401 deployed in the edge server 4 performs a comprehensive analysis of multi-source heterogeneous data, determines the alarm event and pushes it to the operation and maintenance personnel 403 for event processing, and displays the information screen to the on-duty personnel 402 in the central control room. After handling the event, the operation and maintenance personnel 403 uploads the processing record and experience summary to the knowledge base, which is convenient for tracing the source of the event and improving and expanding the knowledge base, thereby forming a closed-loop disposal of the entire event.

[0082] The video probe 8 has a resolution of not less than 1080P, and uses a wide-angle camera with infrared function, pan / tilt, lighting compensation and other functions, supports multiple alarm processing methods, such as sending emails, FTP uploads, relay outputs, audio alarms, etc., and uploads the captured images, videos, audio and other data to the edge server 4 for processing. Use target detection to perform intelligent analysis on the video and automatically identify the following scenes: abnormal behaviors such as falling, electric shock and other behaviors of personnel postures; irregular operations such as failure to wear safety helmets, work clothes, smoking and other behaviors in accordance with regulations; equipment abnormalities such as equipment smoking and other abnormal conditions; environmental abnormalities such as whether there is water accumulation in places such as substations and other abnormal conditions.

[0083] The millimeter wave radar sensor / smart bracelet 10 can detect the number of people, movement trajectory, breathing heart rate, etc., and upload the detected number of people, movement conditions, heart rate and other data to the edge server 4 through Wi-Fi or bus for processing. Use AI intelligent analysis to analyze people within a certain range and automatically identify abnormal scenes of people: monitor the health status of people within a range of 5m, and intelligently identify their breathing and heart rate data; locate multiple people within a range of 10m at the same time, display their positions in real time on the electronic map, and count the number of people in the room; identify people's postures such as falling, electric shock, running, standing, walking, etc.

[0084] The environment detection sensor 12 detects toxic and harmful gases in the space, and the measurement parameters include H 2 S. NH 3 , CH 4 and O 2, the error is less than ±5%, and the temperature and humidity are detected. The measurement parameters include temperature and humidity, and the error is less than ±2%. Because the humidity in the area is too high, it will damage the gas detector probe and affect the service life. Therefore, the sensor is equipped with an anti-corrosion and dehumidification function. The environmental detection sensor 12 uploads the detected toxic and harmful gases and temperature and humidity parameters to the edge server 4 for processing. By setting the threshold, the H of the environment is monitored in real time. 2 S. NH 3 , CH 4 , O 2 , temperature, humidity and other parameters are abnormal.

[0085] For details, please refer to Figure 4 , which is the step of fusion analysis of multi-source heterogeneous data.

[0086] The second aspect of the present invention provides a method for using an intelligent identification and analysis system for full-process personal safety monitoring in a sewage treatment plant, which is used for multi-source heterogeneous data fusion analysis, comprising the following steps:

[0087] Step S10 (data input step): integrating the information of multiple detectors and uploading the data to the intelligent safety management platform synchronously;

[0088] Step S20 (preprocessing step): perform data cleaning to ensure the accuracy and consistency of multi-source data; in actual operation, remove noise and outliers to ensure the accuracy and consistency of multi-source data. Specifically, synchronize the data to ensure that all data from video, millimeter-wave radar, smart bracelets, and gas detection instruments are aligned in time for effective fusion and analysis.

[0089] Specifically, after the preprocessing step, a feature extraction step is also included. The features of the video data include human posture estimation and behavior recognition, the features of the millimeter wave radar and smart bracelet data include health status and behavior analysis, and the features of the gas detection instrument data include concentration changes of toxic and harmful gases and oxygen levels.

[0090] Step S30 (model fusion step): adopt feature-level fusion algorithm to merge multi-source feature data and input them into a unified model for analysis: standardize the features extracted from video, millimeter-wave radar, smart bracelet and gas detection instrument so that they are fused on the same scale to avoid unbalanced impact of different features on the model, and then perform feature fusion to splice feature vectors from different data sources into a long feature vector, assign weights to different features according to the importance of different data sources for weighted fusion, and input the fused features into the machine learning model for training and prediction;

[0091] Step S40: Calculate the comprehensive risk score based on the output of the fusion model. The score can be expressed in the range of 0 to 100, where 0 indicates no risk and 100 indicates high risk. The warning threshold is set according to the risk score. When the risk score exceeds the set threshold, the warning system is triggered to notify relevant personnel and take corresponding measures. Assume that there are three main risks: electric shock risk (R1), fall risk (R2) and poisoning risk (R3). Each risk is predicted by the corresponding data source, such as video data, millimeter wave radar, smart bracelet data and gas detection data.

[0092] The comprehensive risk score (R) can be calculated by the following formula:

[0093] R=w1*R1+w2*R2+w3*R3

[0094] in:

[0095] R1: Electric shock risk score, output by the video analysis model, millimeter wave radar, and smart bracelet model.

[0096] R2: Fall risk score, output by the video analysis model, millimeter wave radar, and smart bracelet model.

[0097] R3: Poisoning risk score, output by the video analysis model and the gas detection model.

[0098] w1, w2, w3: The weights of each risk score, which are determined based on the relative importance of different risks or the frequency of each risk event in historical data. They can be set through expert scoring or data-driven methods. For example: w1 = 0.4, w2 = 0.3, w3 = 0.3.

[0099] Step S50: Evaluate the accuracy, recall rate and other performance indicators of the fusion model, and adjust the model parameters according to feedback and actual conditions, optimize the feature selection and fusion methods, and improve the accuracy and reliability of the comprehensive risk assessment. Evaluate the accuracy, recall rate and other performance indicators of the fusion model. The accuracy rate is the proportion of samples predicted by the model to be positive, and the recall rate is the proportion of samples predicted to be positive by the model to be positive. In the safety monitoring system, the recall rate is very important, because missing any potential danger may lead to serious consequences, so the recall rate must be above 95%, while maintaining a reasonable accuracy rate. For example, the comprehensive model's fall detection accuracy rate reaches 90%, and the recall rate reaches 95%. Adjust the model parameters according to feedback and actual conditions, optimize the feature selection and fusion methods, and improve the accuracy and reliability of the comprehensive risk assessment.

[0100] In summary, through the above steps, data from different sources are effectively integrated into a unified risk assessment system to achieve comprehensive risk assessment of electric shock, falls, and poisoning, thereby improving the comprehensiveness and accuracy of the safety early warning system.

[0101] For details, please refer to Figure 5 After the accident alarm is processed, the staff will upload the location, time, cause, process, processing results and other information of the incident to the knowledge base. The system will statistically mark the incident in terms of time and location. In terms of time, the system accumulates the types and frequencies of accidents in different time periods. In terms of location, the system accumulates the number of accidents in different places, and marks and divides the time periods and places where high-frequency accidents occur, providing a basis for advance warning. For the time periods with high number of marked accidents, advance demarcation is carried out, such as the peak water inflow time of the sewage treatment plant in the morning and evening, the water inlet pump room and other single H 2 The content of gases such as S is high, which reminds personnel to pay more attention to safety during this period to reduce the accident rate; for accident-prone areas with a high number of markings, advance demarcation is carried out, such as substations are high-risk areas for electric shock, and sludge pump rooms are prone to produce toxic and harmful gases and gather as high-risk areas for poisoning, etc., to remind personnel to pay attention to safety matters when entering the place to reduce the accident rate. During the operation of the sewage treatment plant, the intelligent safety management platform 2 and the edge server 4 conduct AI comprehensive analysis of video surveillance data, human health parameters, and environmental parameters, intelligently judge the safety status of the sewage treatment plant and issue an alarm during the event. After the alarm is completed, rescue measures are immediately taken and an accident handling report is formed, which is uploaded to the knowledge base of the intelligent safety management platform 2 to form an intelligent identification and analysis process for personal safety monitoring throughout the entire process of the entire incident. In specific implementation, the invention can be used in semi-underground sewage treatment plants and fully underground sewage treatment plants, and can also be used in sludge treatment plants, pumping stations, storage tanks and other similar projects.

[0102] The whole process includes three stages: pre-event prevention, in-event alarm, and post-event handling;

[0103] Pre-event prevention: Define warning areas based on the time periods and places where high-frequency accidents occur in the knowledge base, and issue warning reminders to people entering;

[0104] Alarm during the event: Collect detector data and conduct multi-source heterogeneous data fusion analysis to determine personal safety status, personnel operation specifications, etc.

[0105] Post-accident processing: After an accident occurs, rescue and accident handling are carried out according to the pre-set emergency plan, and an accident handling report is formed, marking the event from the two dimensions of time and location, marking the time periods and places where high-frequency accidents occur.

[0106] A third aspect of the present invention provides an electronic device, comprising: a processor;

[0107] Memory; and a program, wherein the program is stored in the memory and is configured to be executed by a processor, and the program includes a method for executing an application dedicated to the full-process personal safety monitoring intelligent identification and analysis system for sewage treatment plants.

[0108] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor and is applied to a method for using a full-process personal safety monitoring intelligent identification and analysis system specifically for a sewage treatment plant.

[0109] The fifth aspect of the present invention provides a full-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants, the whole process includes three stages: pre-prevention, in-process alarm, and post-processing;

[0110] Pre-event prevention: Define warning areas based on the time periods and places where high-frequency accidents occur in the knowledge base, and issue warning reminders to people entering;

[0111] Alarm during the event: Collect detector data and conduct multi-source heterogeneous data fusion analysis to determine personal safety status, personnel operation specifications, etc.

[0112] Post-accident processing: After an accident occurs, rescue and accident handling are carried out according to the pre-set emergency plan, and an accident handling report is formed, marking the event from the two dimensions of time and location, marking the time periods and places where high-frequency accidents occur.

[0113] The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. An intelligent identification and analysis system for monitoring personal safety in the whole process of sewage treatment plants, characterized in that: It includes an intelligent security management platform, a mobile terminal, a switch, an edge server, and multiple detectors. The multiple detectors establish a communication connection with the edge server, the edge server establishes a communication connection with the switch, the switch establishes a communication connection with the intelligent security management platform, and the mobile terminal is transmitted to the intelligent security management platform via wireless; The multiple detectors are used to collect personnel and regional environmental information of the sewage treatment plant. Based on the feedback information from the multiple detectors, the intelligent safety management platform centrally monitors the equipment, personnel, and environmental conditions of the entire plant, and can intelligently identify abnormal conditions and issue alarms, while also interacting with mobile terminals.

2. The whole-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants as claimed in claim 1 is characterized by: The detector includes one or more of a video probe, a millimeter wave radar sensor / smart bracelet, and an environmental detection sensor; The video probe is used to collect information about the entry and exit, operation, behavior, and status of personnel in the sewage treatment plant, and the millimeter wave radar sensor / smart bracelet is used to detect the number of people in the sewage treatment plant, movement trajectory, and respiratory heart rate information; The environmental detection sensor is used to detect the H2S, NH3, CH4 gas concentrations and O2, temperature and humidity data in the dangerous areas of the sewage treatment plant. The edge server is used for sewage treatment plant data collection, storage, calculation analysis, and display functions. It can collect data such as video, millimeter wave radar, smart bracelet, H2S, NH3, CH4, O2, temperature and humidity sensors, and provide intelligent AI computing services for the system.

3. The whole-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants as claimed in claim 1 is characterized by: The intelligent security management platform includes a basic functional unit, which includes a user management module, a device management module, an electronic archive management module, a data query and statistics module, and an electronic map module; The user management module can view user names, user types, user levels, and has the functions of adding and removing users and changing passwords; The device management module can monitor the status of the system equipment in real time, including the operating status of detectors, communication devices, edge servers, etc., and has the functions of adding, deleting, and modifying; The electronic archive management module can support the management of security electronic archive files, including storage, query, upload, deletion, online viewing and downloading functions; The data query and statistics module includes data storage, equipment query, video query, alarm type, alarm level query and other contents; it provides chart display function for data such as temperature, humidity, gas concentration, equipment abnormality, environmental abnormality, personnel abnormality, etc. The electronic map module includes a three-dimensional electronic map, which can display the location of each device, personnel, and fault point of the system on the electronic map and real-time video in real time, and has real-time recording, uploading and exporting functions.

4. The whole-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants as claimed in claim 1 is characterized by: The intelligent safety management platform also includes a core functional unit, which includes an operation management module; the operation management module manages the information of operators and performs intelligent identification of irregular operations.

5. The whole-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants as claimed in claim 4 is characterized by: The core functional unit also includes a monitoring and early warning module, which performs multi-source heterogeneous data fusion analysis to realize the function of monitoring the status of personnel and the on-site environment throughout the entire process. It is a full-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants. The monitoring and early warning module is equipped with a video probe unit, which automatically identifies the status of equipment and irregular operations of personnel.

6. The intelligent identification and analysis system for monitoring personal safety during the whole process of a sewage treatment plant as claimed in claim 5 is characterized by: The core functional unit also includes an accident handling module, which includes an alarm information push unit, an accident handling data support unit, a processing process tracking unit and an accident handling summary unit, and is continuously advanced through the alarm information push unit, the accident handling data support unit, the processing process tracking unit and the accident handling summary unit.

7. A method for using the intelligent identification and analysis system for full-process personal safety monitoring in sewage treatment plants according to any one of claims 1 to 6, for multi-source heterogeneous data fusion analysis, characterized in that: The following steps are involved: Step S10: Integrate the information of multiple detectors and upload the data to the intelligent safety management platform synchronously; Step S20: Perform data cleaning to ensure the accuracy and consistency of multi-source data; Step S30: Use a feature-level fusion algorithm to merge multi-source feature data and input them into a unified model for analysis: standardize the features extracted from video, millimeter-wave radar, smart bracelet, and gas detection instrument so that they are fused on the same scale to avoid unbalanced effects of different features on the model. Then perform feature fusion to concatenate feature vectors from different data sources into a long feature vector. According to the importance of different data sources, weights are assigned to different features for weighted fusion. The fused features are input into the machine learning model for training and prediction. Step S40: Calculate a comprehensive risk score based on the output of the fusion model. The score can be expressed in a range of 0 to 100, where 0 indicates no risk and 100 indicates high risk. Set an early warning threshold based on the risk score. When the risk score exceeds the set threshold, trigger the early warning system to notify relevant personnel and take corresponding measures. Step S50: Evaluate the accuracy, recall rate and other performance indicators of the fusion model, and adjust the model parameters according to feedback and actual conditions, optimize feature selection and fusion methods, and improve the accuracy and reliability of comprehensive risk assessment.

8. An electronic device, characterized in that include: processor; Memory; and a program, wherein the program is stored in the memory and configured to be executed by a processor, the program including instructions for executing the method of claim 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to perform the method according to claim 7.

10. A full-process personal safety monitoring intelligent identification and analysis system dedicated to sewage treatment plants, characterized by The whole process includes three stages: pre-event prevention, in-event alarm, and post-event handling; Pre-event prevention: Define warning areas based on the time periods and places where high-frequency accidents occur in the knowledge base, and issue warning reminders to people entering; Alarm during the event: Collect detector data and conduct multi-source heterogeneous data fusion analysis to determine personal safety status, personnel operation specifications, etc. Post-accident processing: After an accident occurs, rescue and accident handling are carried out according to the pre-set emergency plan, and an accident handling report is formed, marking the event from the two dimensions of time and location, marking the time periods and places where high-frequency accidents occur.

Images