A safe production intelligent monitoring method and device

Abstract

The application provides a safe production intelligent monitoring method and device, which is suitable for the technical field of data processing, and comprises the following steps: generating current factory production state information according to equipment operation area information, current equipment running state information and current factory operation behavior information; generating historical factory production state information according to the equipment operation area information, historical equipment running state information and historical factory operation behavior information; and generating safe production monitoring instruction information according to the current factory production state information, the historical factory production state information, a safe production monitoring instruction generation model and a preset safe production risk information library, so as to perform safe production intelligent monitoring processing through the safe production monitoring instruction information. The application adaptively generates differentiated monitoring and control instructions, improves the real-time performance, accuracy and stability of the industrial factory safe production intelligent monitoring, and realizes the whole-process intelligent closed-loop control of production safety.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and in particular relates to intelligent monitoring methods and devices for safe production. Background Technology

[0002] Currently, the field of intelligent monitoring for industrial safety production has widely integrated IoT sensing, video surveillance, artificial intelligence analysis, and information system technologies. These technologies are widely used in industrial scenarios such as machining, chemical processing, and intelligent manufacturing, gradually replacing the traditional safety management model of manual inspection and post-event traceability.

[0003] In existing technologies, a cloud-based centralized processing architecture is adopted. Independent sensing and video surveillance equipment deployed in industrial plants collect equipment operation data and on-site video data, respectively. The massive amount of data is then uploaded to the cloud server via the network to achieve hazard identification, anomaly judgment, and preliminary early warning push.

[0004] However, in existing technologies, the compatibility of device data acquisition is poor, data processing relies on centralized computing in the cloud, and the transmission of massive video data can easily cause network congestion, resulting in delays in risk identification and abnormal warning response. Summary of the Invention

[0005] In view of this, the present application provides a method and device for intelligent monitoring of safe production, which aims to solve the problems of incomplete equipment data collection, slow cloud processing response, and poor adaptability of on-site control in the prior art.

[0006] The first aspect of this application provides a method for intelligent monitoring of safe production, including:

[0007] Acquire equipment operation area information, current equipment operating status information, current plant area operation behavior information, historical equipment operating status information, historical plant area operation behavior information, and generate a safety production monitoring instruction model;

[0008] Based on the equipment operation area information, current equipment operating status information, and current plant operation behavior information, multiple current plant production status information are generated.

[0009] Based on the equipment operation area information, historical equipment operation status information, and historical plant operation behavior information, multiple historical plant production status information are generated.

[0010] Based on the multiple current plant area production status information, multiple historical plant area production status information, safety production monitoring instruction generation model, and preset safety production risk information database, multiple safety production monitoring instruction information is generated to perform intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

[0011] A second aspect of this application provides a smart monitoring device for safe production, comprising:

[0012] The information acquisition module is used to acquire equipment operation area information, current equipment operating status information, current plant area operation behavior information, historical equipment operating status information, historical plant area operation behavior information, and a safety production monitoring instruction generation model.

[0013] The current plant area production status information generation module is used to generate multiple current plant area production status information based on the equipment operation area information, current equipment operation status information, and current plant area operation behavior information;

[0014] The historical plant area production status information generation module is used to generate multiple historical plant area production status information based on the equipment operation area information, historical equipment operation status information, and historical plant area operation behavior information.

[0015] The safety production monitoring instruction information generation module is used to generate multiple safety production monitoring instruction information based on the multiple current plant area production status information, multiple historical plant area production status information, safety production monitoring instruction generation model, and a preset safety production risk information database, so as to carry out intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

[0016] A third aspect of this application provides a terminal device, which includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the intelligent monitoring method for safe production described in the first aspect above.

[0017] A fourth aspect of this application provides a computer-readable storage medium, comprising: storing a computer program, wherein when executed by a processor, the computer program implements the steps of the intelligent monitoring method for safe production described in the first aspect above.

[0018] Compared with the prior art, the beneficial effects of this application are: this application realizes dynamic perception and accurate judgment of the operating conditions of equipment, personnel operations, and safety risks in the plant area, and adaptively generates differentiated monitoring and control instructions, effectively avoiding the problems of single data, delayed judgment, and poor adaptability of existing monitoring methods, thereby greatly improving the real-time performance, accuracy, and stability of intelligent monitoring of safety production in industrial plants, and realizing intelligent closed-loop management of the entire production safety process. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram illustrating the implementation process of the intelligent monitoring method for safe production provided in Embodiment 1 of this application;

[0021] Figure 2 This is a schematic diagram illustrating the implementation process of the intelligent monitoring method for safe production provided in Embodiment 2 of this application;

[0022] Figure 3 This is a schematic diagram illustrating the implementation process of the intelligent monitoring method for safe production provided in Embodiment 3 of this application;

[0023] Figure 4 This is a schematic diagram illustrating the implementation process of the intelligent monitoring method for safe production provided in Embodiment 4 of this application;

[0024] Figure 5 This is a schematic diagram illustrating the implementation process of the intelligent monitoring method for safe production provided in Embodiment 5 of this application;

[0025] Figure 6 This is a schematic diagram illustrating the implementation process of the intelligent monitoring method for safe production provided in Embodiment Six of this application;

[0026] Figure 7 This is a schematic diagram of the structure of the intelligent monitoring device for safe production provided in the embodiments of this application;

[0027] Figure 8 This is a schematic diagram of the terminal device provided in the embodiments of this application. Detailed Implementation

[0028] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0029] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0030] Figure 1 A flowchart illustrating the implementation of the intelligent monitoring method for safe production provided in Embodiment 1 of this application is shown, and is described in detail below:

[0031] Step S101: Obtain equipment operation area information, current equipment operation status information, current plant operation behavior information, historical equipment operation status information, historical plant operation behavior information, and safety production monitoring instruction generation model.

[0032] In this embodiment, equipment operation area information refers to the distribution of operation points, operation coverage, safety protection zone boundaries, and environmental conditions of various CNC machine tools, machining centers, and automated production lines in the factory area. This information is used to define the monitoring coverage area of ​​different production equipment and can be manually entered using on-site laser positioning equipment, high-definition camera acquisition equipment, and factory electronic ledgers. The safety production monitoring instruction generation model can be pre-set by the user. This model can employ a Transformer multi-head attention deep learning model, capable of extracting risk features, performing correlation analysis, and adaptively generating monitoring instructions based on multi-dimensional factory production data. It can be pre-trained using massive factory safety production condition datasets and risk anomaly datasets, and its parameters can be fine-tuned by combining industrial production scenario characteristics to adapt to various intelligent safety production monitoring scenarios for machining production lines.

[0033] In this embodiment, the current equipment operating status information includes current equipment load energy consumption information, current equipment processing operation time information, and current equipment operation anomaly information. Specifically, the current equipment load energy consumption information includes the equipment's real-time operating power, hourly power consumption, no-load energy consumption, and load energy consumption data. Under normal production conditions, the real-time operating power ranges from 20KW to 80KW, and the hourly power consumption ranges from 5 kWh to 20 kWh, which can be collected in real-time using an edge calculator combined with a power sensing module. The current equipment processing operation time information includes the equipment's single-time startup operation time, daily cumulative processing time, and effective operating time. The standard threshold for the equipment's effective daily operating time is preset to 8 hours, which can be statistically analyzed in real-time through the edge calculator's equipment operating condition acquisition channel. The current equipment operation anomaly information includes abnormal operating conditions such as equipment stalling, parameter deviation, overload errors, abnormal noises, and overheating. The abnormal equipment body temperature threshold is preset to 60 degrees Celsius, which can be identified and collected in real-time through equipment sensing components combined with an edge calculator, thereby completing the accurate recording of the equipment's real-time operating status.

[0034] In this embodiment, the current factory area operation information includes current production line standardized operation information, current production line quality inspection operation information, and current factory area equipment operation and maintenance operation information. The current production line standardized operation information refers to the standardized operational behavior of workers in completing product processing, parameter debugging, and material loading and unloading according to process specifications, which can be captured in real time through the factory area video acquisition channel combined with a multimodal AI visual recognition module. The current production line quality inspection operation information refers to the operational behavior of workers in detecting appearance defects, verifying dimensions, and sorting batches of finished and semi-finished products; the standards for judging compliant quality inspection operations can be preset manually. The current factory area equipment operation and maintenance operation information refers to the on-site operations of workers in completing equipment inspection, cleaning and maintenance, troubleshooting, and parts replacement; the threshold for identifying illegal operation and maintenance operations can be preset manually, thereby achieving comprehensive collection and recording of real-time factory area operation behavior.

[0035] In this embodiment, the historical equipment operation status information includes historical equipment load energy consumption information, historical equipment processing operation time information, and historical equipment operation anomaly information. Among them, the historical equipment load energy consumption information is the equipment's average daily operating power, average daily power consumption, and no-load and load energy consumption fluctuation data over the past 30 days, which is used to compare and analyze the equipment's energy consumption operation patterns. The system can automatically overlay and update the energy consumption data collected by the edge calculator daily to generate historical ledgers. The historical equipment processing operation time information is the equipment's cumulative daily operation time and monthly effective operation time over the past 30 days, which is used to assess the equipment's operating load status. The historical equipment operation anomaly information includes the equipment's past fault error records, anomaly occurrence frequency, and anomaly duration. The monthly anomaly warning threshold can be preset to 5 times, which is then used to construct a template for equipment operation anomaly patterns to assist in real-time risk assessment.

[0036] In this embodiment, historical factory operation information includes historical production line standardized operation information, historical production line quality inspection operation information, and historical factory equipment operation and maintenance information. The historical production line standardized operation information is a historical record of daily standardized and non-standard operations by factory workers over the past 30 days, used to analyze personnel work habits and operational risk patterns. The historical production line quality inspection operation information includes past compliance rates, defect sorting accuracy rates, and records of missed and incorrect inspections; the quality inspection pass / fail criteria can be preset. The historical factory equipment operation and maintenance information includes past records of regular equipment maintenance, fault repair, and daily inspections; the maintenance cycle standard can be preset to once a day for inspection and once a week for maintenance, thus forming a historical database of factory operation behavior, providing data support for dynamic monitoring and analysis.

[0037] Step S102: Generate multiple current factory production status information based on the equipment operation area information, current equipment operating status information, and current factory operation behavior information.

[0038] In this embodiment, the equipment operation area information is first divided into regions, coded, and its operating condition features are extracted to accurately distinguish the operation scenarios and monitoring ranges of equipment at different locations. Then, the energy consumption data, operating time data, and abnormal operating condition data in the current equipment operation status information are normalized. At the same time, the standardized operation, quality inspection operation, and maintenance operation data in the current factory operation behavior information are feature-filtered and classified and integrated. Then, the random forest algorithm is used to perform correlation and fusion analysis on the multi-dimensional real-time data to accurately match the correlation between the equipment operation area and the corresponding equipment operating conditions and personnel operation status. Finally, multiple current factory production status information, including equipment real-time energy consumption status, equipment operation load status, production line operation compliance status, and on-site operation and maintenance safety status, are generated.

[0039] Step S103: Generate multiple historical plant area production status information based on the equipment operation area information, historical equipment operation status information, and historical plant area operation behavior information.

[0040] In this embodiment, the historical monitoring range of equipment in each area is locked by the equipment operation area information. Then, the system retrieves historical equipment operation status energy consumption, duration, abnormal log data, and historical plant operation behavior compliance, quality inspection, and maintenance records stored in the system. Subsequently, the system uses a long short-term memory network deep learning model to mine patterns and fit features of the time-series historical data. The system compares and analyzes the historical operation fluctuation patterns of equipment in different operation areas and the patterns of personnel operation behavior habits, thereby generating multiple historical plant production status information including the historical energy consumption fluctuation status of equipment, the historical failure frequency status of equipment, the historical operation compliance status of the production line, and the historical maintenance safety status of the plant.

[0041] Step S104: Based on the multiple current plant area production status information, multiple historical plant area production status information, safety production monitoring instruction generation model, and preset safety production risk information database, generate multiple safety production monitoring instruction information to perform intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

[0042] In this embodiment, the preset safety production risk information database can be manually preset. This database includes various safety risk characteristic data such as equipment overload risk, parameter deviation risk, personnel violation operation risk, quality inspection oversight risk, and maintenance absence risk in the organic processing plant area. The thresholds for judging various risk levels can be manually preset, with the first-level warning risk threshold being equipment abnormality lasting for 1 minute, the second-level warning risk threshold being equipment abnormality lasting for 3 minutes, and the third-level warning risk threshold being equipment abnormality lasting for 5 minutes. First, multiple current plant area production status information and multiple historical plant area production status information are jointly encoded to generate a comprehensive feature vector of plant area production operation. Then, the comprehensive feature vector is input into the trained safety production monitoring instruction generation model. The model completes the linkage comparison between real-time status and historical patterns, risk correlation calculation, and risk level judgment. Then, it matches the preset safety production risk information database to complete risk tracing and scenario classification. Finally, it generates multiple safety production monitoring instruction information including equipment parameter adjustment instructions, on-site risk warning instructions, personnel operation rectification instructions, equipment maintenance scheduling instructions, and production line shutdown control instructions. Then, through various monitoring instructions, intelligent monitoring and closed-loop management of safety production in the entire plant area are completed.

[0043] The intelligent monitoring method for safe production provided in this application enables dynamic perception and accurate assessment of equipment operating conditions, personnel operations, and safety risks in the plant area. It adaptively generates differentiated monitoring and control instructions, effectively avoiding the problems of single data, delayed assessment, and poor adaptability of existing monitoring methods. This significantly improves the real-time performance, accuracy, and stability of intelligent monitoring for safe production in industrial plants, and realizes intelligent closed-loop management of the entire production safety process.

[0044] Figure 2 The flowchart illustrating the implementation of the intelligent monitoring method for safe production provided in Embodiment 2 of this application is shown. The difference between this method and Embodiment 1 is that step S104 specifically includes:

[0045] Step S201: Based on the preset safety production risk feature extraction model, feature extraction processing is performed according to the preset safety production risk information database to generate multiple safety production risk feature information.

[0046] In this embodiment, the preset safety production risk feature extraction model can be manually preset. This model can employ a convolutional neural network deep learning model, capable of extracting high-dimensional features from structured risk data and scenario risk data. The preset safety production risk information database can also be manually preset. This database uniformly collects texture features, operating condition threshold features, and behavioral violation features of various safety risks in the machine processing production line. Specifically, the critical threshold for equipment parameter deviation is manually preset to ±5%, the weight for judging personnel violation risks is manually preset to 0.8, and the weight for judging equipment overload risks is manually preset to 0.9. First, the convolutional neural network deep learning model reads all risk sample data from the preset safety production risk information database. Then, it extracts the dimensional features corresponding to various equipment risks, operational risks, maintenance risks, and quality inspection risks in a hierarchical manner. Finally, the extracted scattered risk features are organized and integrated to generate multiple safety production risk feature information, including equipment energy consumption risk features, equipment failure risk features, personnel operation violation features, quality inspection oversight risk features, and maintenance absence risk features.

[0047] Step S202: Perform time-series encoding processing on the multiple historical factory area production status information to generate multiple factory area production status time-series information.

[0048] In this embodiment, the historical equipment energy consumption data, historical equipment operating time data, historical equipment anomaly data, and historical work behavior data contained in the production status information of multiple historical plant areas are first sorted and organized in time sequence. The data is then uniformly standardized according to the time dimension of the past 30 days. Then, the continuous time sequence production data is structured and encoded using a time sequence coding algorithm. Finally, the discrete historical production status data is converted into a time sequence feature vector with time correlation attributes, thereby generating multiple plant area production status time sequence information that can reflect the periodic fluctuation pattern of plant area production status, the time sequence pattern of fault occurrence, and the time sequence pattern of personnel work behavior.

[0049] Step S203: Based on the multiple safety production risk characteristic information and the multiple plant area production status time sequence information, the safety production monitoring instruction generation model is iteratively trained to generate a trained safety production monitoring instruction generation model.

[0050] In this embodiment, multiple safety production risk characteristics are first used as the baseline features for model risk identification, and multiple production status time-series information of factory areas are used as the production status fitting features of the model. Then, the two types of feature information are spliced ​​and fused. The fused comprehensive feature data is then input into the safety production monitoring command generation model. The model continuously learns the correlation mapping relationship between production time-series status and safety risks, and iteratively updates the internal parameters of the model. At the same time, the training process is controlled by a pre-set iterative training threshold. The iterative convergence accuracy threshold is pre-set to 0.001, and the maximum number of iterations is pre-set to 1000, thereby completing the iterative training process of the model and generating the trained safety production monitoring command generation model.

[0051] Step S204: Based on the multiple current plant area production status information and the trained safety production monitoring instruction generation model, generate multiple safety production monitoring instruction information to perform intelligent safety production monitoring processing through the multiple safety production monitoring instruction information.

[0052] In this embodiment, multiple current plant production status information obtained from real-time collection and analysis are first standardized and preprocessed to remove abnormal interference data. Then, the preprocessed current plant production status information is input into the trained safety production monitoring instruction generation model. The model then completes risk matching, risk level determination, and control strategy matching for real-time production status, thereby adaptively outputting multiple safety production monitoring instruction information adapted to the current plant production scenario. Furthermore, through various monitoring instructions, the model completes full-dimensional intelligent safety production monitoring and processing for plant equipment operation, personnel operation, quality inspection and maintenance, and risk warning.

[0053] The intelligent safety production monitoring method provided in this application deeply mines the temporal risk patterns of historical production data, enabling the model to fully learn the dynamic correlation between production status and safety risks. This significantly improves the model's ability to identify hidden safety risks and periodic safety hazards, thereby making safety production monitoring instructions more accurate and aligned with the actual production conditions of the plant, effectively enhancing the comprehensiveness and foresight of the overall intelligent safety production monitoring.

[0054] Figure 3 The flowchart illustrating the implementation of the intelligent monitoring method for safe production provided in Embodiment 3 of this application is shown. The difference between this method and Embodiment 2 is that step S203 specifically includes:

[0055] Step S301: Based on the multiple safety production risk characteristic information and the multiple factory area production status time sequence information, cross-sponge processing is performed to generate multiple safety production risk and factory area production status splicing information.

[0056] In this embodiment, multiple safety production risk feature information is first dimensionally supplemented and standardized to ensure the uniformity of various risk feature dimensions. Then, the time-series information of multiple factory production status is time-series aligned to achieve one-to-one matching of risk features and production time-series features. Then, the two types of feature information are deeply fused and spliced ​​through a feature cross-splitting algorithm to break the independent limitations of a single feature, thereby generating multiple safety production risk and factory production status spliced ​​information that combines risk dimension features and time-series production status features.

[0057] Step S302: Based on the spliced ​​information of multiple safety production risks and factory production status, the safety production monitoring instruction generation model is iteratively trained to generate a trained safety production monitoring instruction generation model.

[0058] In this embodiment, the model training learning rate can be preset manually, and the preset value is 0.01. First, all the spliced ​​information of multiple safety production risks and factory production status is input into the safety production monitoring instruction generation model. Then, the spliced ​​features are weighted through the model's multi-head attention mechanism, focusing on fitting the feature correlation relationship of high-risk time series scenarios. Then, the model is continuously iterated and trained according to the preset number of 1000 iterations and convergence accuracy of 0.001, continuously optimizing the model's risk identification and instruction generation capabilities, thereby generating the trained safety production monitoring instruction generation model.

[0059] The intelligent monitoring method for safe production provided in this application realizes a deep correlation between risk characteristics and production time-series data, solves the technical problems of traditional models having single characteristics and a disconnect between spatiotemporal correlation, improves the model's ability to perceive periodic and latent safety production risks, and thus enhances the accuracy and predictive ability of intelligent monitoring for safe production.

[0060] Figure 4 The flowchart illustrating the implementation of the intelligent monitoring method for safe production provided in Embodiment 4 of this application is shown. The difference between this method and Embodiment 1 is that step S104 specifically includes:

[0061] Step S401: Match the multiple current production status information of the factory area with the preset safety production risk information database to generate multiple matched current safety production risk information.

[0062] In this embodiment, the preset safety production risk information database can be manually preset. The database contains standard features and judgment thresholds for safety risks in all categories of machining production. First, the real-time equipment operating conditions, real-time operation behaviors, and real-time maintenance status data contained in multiple current factory production status information are compared and matched with the preset safety production risk information database one by one. Then, risk points such as abnormal equipment energy consumption, potential equipment failures, personnel violations, quality inspection omissions, and maintenance deficiencies in the current production status are screened out. Then, the various real-time risk points are organized and classified to generate multiple matched current safety production risk information.

[0063] Step S402: Match the multiple historical factory area production status information with the preset safety production risk information database to generate multiple matched historical safety production risk information.

[0064] In this embodiment, multiple historical production status information of the factory area stored in the system are retrieved first. The fluctuation patterns of equipment operation, personnel work habits, failure cycles, and frequent quality inspection problems in the historical production process are extracted. Then, various historical production characteristics are matched and compared with periodic risk characteristics in the preset safety production risk information database. Then, high-frequency risks, periodic hidden dangers, and routine violations in the historical production process of the factory area are screened out, thereby generating multiple matched historical safety production risk information.

[0065] Step S403: Based on the multiple matched current safety production risk information, multiple matched historical safety production risk information, multiple current plant area production status information, multiple historical plant area production status information, and the safety production monitoring instruction generation model, generate multiple safety production monitoring instruction information to perform intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

[0066] In this embodiment, the four types of data—current safety production risk information, historical safety production risk information, current plant production status information, and historical plant production status information—are first aligned and unified in terms of dimensions. Then, the aligned comprehensive data is input into the safety production monitoring instruction generation model. The model then combines the four dimensions of real-time risk, historical risk, real-time production conditions, and historical production conditions to complete a comprehensive assessment, classify early warning levels and control measures, and generate multiple safety production monitoring instruction information that are adapted to the real-time risk situation on site. Finally, the intelligent monitoring and closed-loop handling of the entire process of safety production in the plant is completed through various monitoring instructions.

[0067] The intelligent safety production monitoring method provided in this application takes into account both instantaneous sudden risks and long-term periodic hidden dangers, avoiding the problems of risk misjudgment and omission caused by relying solely on real-time data, thereby improving the rationality of hierarchical control of monitoring instructions and enhancing the long-term stable control effect of safety production monitoring in the plant area.

[0068] Figure 5 The flowchart illustrating the implementation of the intelligent monitoring method for safe production provided in Embodiment 5 of this application is shown. The difference between this method and Embodiment 4 above is that step S403 specifically includes:

[0069] Step S501: Based on the multiple matched historical safety production risk information and multiple historical plant area production status information, optimize the safety production monitoring instruction generation model to generate an optimized safety production monitoring instruction generation model.

[0070] In this embodiment, time-series correlation features are first extracted from multiple matched historical safety production risk information and multiple historical plant area production status information to form a historical risk training dataset. The total number of samples in the dataset is preset to 50,000. Then, the historical risk training dataset is input into the safety production monitoring instruction generation model to complete basic optimization. The feature weights of the model for periodic historical hazards are adjusted, and then the interference weights of occasional minor risks on the model output results are reduced, thereby generating an optimized safety production monitoring instruction generation model.

[0071] Step S502: Based on the multiple matched current safety production risk information, perform parameter fine-tuning on the optimized safety production monitoring instruction generation model to generate a fine-tuned safety production monitoring instruction generation model.

[0072] In this embodiment, the fine-tuning learning rate can be preset manually, with a preset value of 0.0005. First, multiple matched current safety production risk information are extracted to form a real-time risk fine-tuning dataset. Then, the real-time risk fine-tuning dataset is input into the optimized safety production monitoring instruction generation model to perform parameter fine-tuning, adapting it to the real-time production risk scenario of the factory on that day. Then, the fluctuation range of parameters in a single fine-tuning is limited to prevent the model from overfitting, thereby generating the fine-tuned safety production monitoring instruction generation model.

[0073] Step S503: Based on the multiple current plant area production status information and the fine-tuned safety production monitoring instruction generation model, generate multiple safety production monitoring instruction information to perform intelligent safety production monitoring processing through the multiple safety production monitoring instruction information.

[0074] In this embodiment, multiple current production status information of the plant area are first standardized and normalized, and the data value range is unified to 0 to 1. Then, the normalized current production status information of the plant area is input into the fine-tuned safety production monitoring instruction generation model. The model then combines historical optimization parameters and real-time fine-tuning parameters to comprehensively calculate the risk level and push corresponding control instructions, thereby generating multiple hierarchical safety production monitoring instruction information. Finally, the plant area safety production intelligent monitoring and handling of the whole scenario is completed by relying on the monitoring instructions.

[0075] The intelligent monitoring method for safe production provided in this application adopts a two-level model optimization logic: first, the model is globally optimized based on historical risk data, and then the model is locally fine-tuned based on real-time risk data. This balances the model's long-term risk identification capability with its ability to adapt to real-time operating conditions, thereby significantly reducing the probability of missed detections and false detections, and improving the overall operational stability and control accuracy of intelligent monitoring for safe production.

[0076] Figure 6 The flowchart illustrating the implementation of the intelligent monitoring method for safe production provided in Embodiment Six of this application is shown. Its difference from Embodiment One described above lies in:

[0077] The safety production monitoring instruction information includes safety level early warning instruction information, equipment parameter adjustment instruction information, personnel dispatch instruction information, and production line control type instruction information;

[0078] Following step S104, the method further includes:

[0079] Step S601: Based on the preset mapping relationship between safety production monitoring instructions and the start / stop parameters of audible and visual alarm devices, generate multiple start / stop parameter information for audible and visual alarm devices according to the multiple safety level early warning instruction information and multiple device parameter control instruction information.

[0080] In this embodiment, the mapping relationship between the preset safety production monitoring instructions and the start / stop parameters of the audible and visual alarm equipment can be preset manually. This mapping relationship is based on a standardized rule base constructed from a large amount of historical safety early warning event data in the plant area, which can achieve one-to-one matching of early warning level, equipment control requirements, and audible and visual alarm operating parameters. If the safety level early warning instruction information is determined to be a Level 1 early warning and the equipment parameter control instruction information includes equipment shutdown requirements, then high-frequency continuous start / stop parameter information for audible and visual alarms is generated; if the safety level early warning instruction information is determined to be a Level 2 early warning and the equipment parameter control instruction information includes equipment load reduction requirements, then medium-frequency intermittent start / stop parameter information is generated; if the safety level early warning instruction information is determined to be a Level 3 early warning and the equipment parameter control instruction information only includes parameter fine-tuning requirements, then low-frequency single-time prompt start / stop parameter information is generated. The start / stop parameter information of the audible and visual alarm equipment is transmitted to the plant area safety light call hardware module in the form of digital control electrical signals. The control module completes the start / stop scheduling of the alarm horn and warning light according to the internal values ​​of the parameter information, realizing the coordinated linkage of alarm audible and visual intensity, duration, and on-site risk level. The preset audio-visual gradient parameters include: Level 1 warning lights have a brightness gradient range of 800 to 1000 lumens, a horn sounding frequency of 3 times per second, and a continuous working duration of 300 seconds; Level 2 warning lights have a brightness gradient range of 400 to 700 lumens, a horn sounding frequency of 1 time per second, a single sound duration of 10 seconds, and an interval duration of 60 seconds; Level 3 warning lights have a brightness gradient range of 100 to 300 lumens, a single short horn sound duration of 2 seconds, and only one warning is triggered. All gradient values ​​are uniformly stored in the mapping rule base as the parameter matching judgment benchmark.

[0081] Step S602: Based on the preset mapping relationship between safety production monitoring instructions and visual monitoring equipment operation control parameters, multiple visual monitoring equipment operation control parameter information is generated according to the multiple personnel scheduling instruction information and multiple production line control type instruction information.

[0082] In this embodiment, the mapping relationship between preset safety production monitoring instructions and visual monitoring equipment operation control parameters can be manually preset. This mapping relationship integrates experimental calibration data from multimodal AI quality inspection and on-site personnel behavior recognition, covering the complete set of operating parameters of camera equipment corresponding to personnel scheduling scenarios and production line control scenarios. The manually preset visual monitoring gradient parameters include a video capture frame rate range of 5 to 30 frames per second, a zoom observation distance range of 1 to 50 meters, an AI defect recognition accuracy threshold range of 0.1 to 1 millimeter, and an image exposure brightness range of 100 to 1000 nits. This mapping relationship covers four types of corresponding matching rules: personnel emergency evacuation scheduling instructions correspond to high frame rate camera capture and wide-range zoom observation parameters; personnel fixed-point rectification scheduling instructions correspond to medium frame rate and local close-up zoom parameters; production line full-line shutdown control instructions correspond to full-area high-definition continuous recording and the highest precision AI recognition parameters; and production line local inspection control instructions correspond to low-power intermittent capture and conventional precision recognition parameters. The complete mapping relationship construction process is as follows: First, all types of personnel dispatch instructions and production line control instructions are collected as sample data. Then, the various operating parameters of the visual monitoring equipment are divided into gradient combinations, including 5 levels of capture frame rate, 10 levels of zoom distance, and 8 levels of recognition accuracy. Next, the risk recognition accuracy under different instructions and different camera parameters is collected, with a target of at least 98% accuracy in hazard identification. The optimal matching combination is then statistically selected using a random forest algorithm, thus solidifying the complete mapping rule matrix. For example, when an emergency evacuation dispatch instruction is combined with a full production line shutdown control instruction, the visual monitoring equipment operation control parameters are generated: a capture frame rate of 30 frames per second, a zoom observation distance of 50 meters, an AI defect recognition accuracy of 0.1 millimeters, and an exposure brightness of 1000 nits. When a personnel fixed-point rectification dispatch instruction is combined with a production line partial inspection control instruction, the visual monitoring equipment operation control parameters are generated: a capture frame rate of 5 frames per second, a zoom observation distance of 10 meters, an AI defect recognition accuracy of 1 millimeter, and an exposure brightness of 300 nits. The generated operating control parameter information of multiple visual monitoring devices is transmitted to the internal video acquisition channel of the super edge calculator. The operating status of all AI visual camera devices in the factory is adjusted synchronously through the video acquisition channel to adapt to the current production line management and personnel scheduling needs.

[0083] The intelligent safety production monitoring method provided in this application embodiment realizes the coordinated control of safety production monitoring instructions, audible and visual alarm devices, and visual monitoring devices. It automatically converts monitoring instructions of different risk levels and control types into operating control parameters adapted to the on-site hardware devices, realizing adaptive dynamic adjustment of factory area safety audible and visual warnings and multimodal AI visual recognition. This is used to eliminate the problems of missed judgments and mismatched warning intensity caused by the fixed parameter operation of alarm and camera equipment in traditional monitoring solutions, thereby improving the intuitiveness of factory area safety production risk warning and the accuracy of visual recognition, and realizing closed-loop collaborative control of all hardware devices for risk management in the production site.

[0084] Corresponding to the method in the above embodiments, Figure 7 The diagram shows a structural block diagram of the intelligent monitoring device for safe production provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown. Figure 7 The example of a smart safety monitoring device can be the subject of the smart safety monitoring method provided in the aforementioned embodiment 1.

[0085] Reference Figure 7 The intelligent safety monitoring device includes:

[0086] The information acquisition module 710 is used to acquire equipment operation area information, current equipment operating status information, current plant area operation behavior information, historical equipment operating status information, historical plant area operation behavior information, and a safety production monitoring instruction generation model.

[0087] The current factory area production status information generation module 720 is used to generate multiple current factory area production status information based on the equipment operation area information, current equipment operation status information, and current factory area operation behavior information;

[0088] The historical plant area production status information generation module 730 is used to generate multiple historical plant area production status information based on the equipment operation area information, historical equipment operation status information, and historical plant area operation behavior information.

[0089] The safety production monitoring instruction information generation module 740 is used to generate multiple safety production monitoring instruction information based on the multiple current plant area production status information, multiple historical plant area production status information, safety production monitoring instruction generation model, and a preset safety production risk information database, so as to carry out intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

[0090] The process by which each module in the intelligent safety monitoring device provided in this application implements its respective function can be found in the foregoing. Figure 1 The description of Embodiment 1 shown will not be repeated here.

[0091] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0092] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0093] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0094] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0095] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first," "second," etc., are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, a first table may be named a second table, and similarly, a second table may be named a first table, without departing from the scope of the various described embodiments. Both the first table and the second table are tables, but they are not the same table.

[0096] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0097] The intelligent monitoring method for safe production provided in this application can be applied to terminal devices such as mobile phones, tablets, wearable devices, vehicle-mounted devices, augmented reality / virtual reality devices, laptops, super mobile personal computers, netbooks, and personal digital assistants. This application does not impose any restrictions on the specific type of terminal device.

[0098] For example, the terminal device may be a station in a WLAN, a cellular phone, a cordless phone, a session initiation protocol phone, a wireless local loop station, a personal digital processing device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, a set-top box, a user premises equipment, and / or other devices for communication over a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks or mobile terminals in future evolved public terrestrial mobile networks, etc.

[0099] Figure 8 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For example... Figure 8 As shown, the terminal device 8 of this embodiment includes: at least one processor 80 ( Figure 8 Only one is shown in the image), and a memory 81 is stored in which a computer program 82 that can run on the processor 80 is stored. When the processor 80 executes the computer program 82, it implements the steps in the above-described embodiments of the intelligent monitoring method for safe production, for example... Figure 1 Steps S101 to S104 are shown. Alternatively, when the processor 80 executes the computer program 82, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 7 The functions of modules 710 to 740 are shown.

[0100] The terminal device 8 can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 80 and a memory 81. Those skilled in the art will understand that... Figure 8 This is merely an example of terminal device 8 and does not constitute a limitation on terminal device 8. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input transmission devices, network access devices, buses, etc.

[0101] The processor 80 may be a central processing unit, or it may be other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0102] In some embodiments, the memory 81 may be an internal storage unit of the terminal device 8, such as a hard disk or memory of the terminal device 8. The memory 81 may also be an external storage device of the terminal device 8, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc., equipped on the terminal device 8. Furthermore, the memory 81 may include both internal and external storage units of the terminal device 8. The memory 81 is used to store operating systems, applications, bootloaders, data, and other programs, such as the program code of computer programs. The memory 81 can also be used to temporarily store data that has been sent or will be sent.

[0103] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0104] This application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it causes the terminal device to implement the steps in any of the above method embodiments.

[0105] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0106] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0107] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0108] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0109] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0110] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0111] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for intelligent monitoring of safe production, characterized in that, include: Acquire equipment operation area information, current equipment operating status information, current plant area operation behavior information, historical equipment operating status information, historical plant area operation behavior information, and generate a safety production monitoring instruction model; Based on the equipment operation area information, current equipment operating status information, and current plant operation behavior information, multiple current plant production status information are generated. Based on the equipment operation area information, historical equipment operation status information, and historical plant operation behavior information, multiple historical plant production status information are generated. Based on the multiple current plant area production status information, multiple historical plant area production status information, safety production monitoring instruction generation model, and preset safety production risk information database, multiple safety production monitoring instruction information is generated to perform intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

2. The intelligent monitoring method for safe production as described in claim 1, characterized in that, The current equipment operating status information includes current equipment load energy consumption information, current equipment processing operation time information, and current equipment operating abnormality information; The current plant area operation information includes current production line standardized operation information, current production line quality inspection operation information, and current plant area equipment operation and maintenance information. The historical equipment operating status information includes historical equipment load energy consumption information, historical equipment processing and operation time information, and historical equipment abnormal operation information. The historical plant operation information includes historical production line standardized operation information, historical production line quality inspection operation information, and historical plant equipment operation and maintenance information.

3. The intelligent monitoring method for safe production as described in claim 1, characterized in that, The step of generating multiple safety production monitoring instructions based on the multiple current plant area production status information, multiple historical plant area production status information, a safety production monitoring instruction generation model, and a preset safety production risk information database, and then performing intelligent safety production monitoring processing through the multiple safety production monitoring instructions, specifically includes: Based on a preset safety production risk feature extraction model, feature extraction processing is performed according to a preset safety production risk information database to generate multiple safety production risk feature information; Based on the multiple historical plant area production status information, time-series encoding processing is performed to generate multiple plant area production status time-series information. Based on the multiple safety production risk characteristics and the time sequence information of multiple factory production status, the safety production monitoring instruction generation model is iteratively trained to generate a trained safety production monitoring instruction generation model. Based on the multiple current plant production status information and the trained safety production monitoring instruction generation model, multiple safety production monitoring instruction information is generated to perform intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

4. The intelligent monitoring method for safe production as described in claim 3, characterized in that, The step of iteratively training the safety production monitoring instruction generation model based on the multiple safety production risk characteristic information and the time sequence information of multiple plant production statuses to generate a trained safety production monitoring instruction generation model specifically includes: Based on the aforementioned multiple safety production risk characteristic information and multiple plant area production status time sequence information, cross-splitting processing is performed to generate multiple safety production risk and plant area production status splicing information. Based on the spliced ​​information of multiple safety production risks and plant production status, the safety production monitoring instruction generation model is iteratively trained to generate a trained safety production monitoring instruction generation model.

5. The intelligent monitoring method for safe production as described in claim 1, characterized in that, The step of generating multiple safety production monitoring instructions based on the multiple current plant area production status information, multiple historical plant area production status information, a safety production monitoring instruction generation model, and a preset safety production risk information database, and then performing intelligent safety production monitoring processing through the multiple safety production monitoring instructions, specifically includes: The multiple current production status information of the plant area and the preset safety production risk information database are matched to generate multiple matched current safety production risk information. Based on the multiple historical plant area production status information and the preset safety production risk information database, a matching process is performed to generate multiple matched historical safety production risk information; Based on the multiple matched current safety production risk information, multiple matched historical safety production risk information, multiple current plant area production status information, multiple historical plant area production status information, and the safety production monitoring instruction generation model, multiple safety production monitoring instruction information is generated to perform intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

6. The intelligent monitoring method for safe production as described in claim 5, characterized in that, The step of generating multiple safety production monitoring instruction information based on the multiple matched current safety production risk information, multiple matched historical safety production risk information, multiple current plant area production status information, multiple historical plant area production status information, and a safety production monitoring instruction generation model, and then performing intelligent safety production monitoring processing through the multiple safety production monitoring instruction information, specifically includes: Based on the multiple matched historical safety production risk information and multiple historical plant area production status information, the safety production monitoring instruction generation model is optimized to generate an optimized safety production monitoring instruction generation model. Based on the multiple matched current safety production risk information, the optimized safety production monitoring instruction generation model is fine-tuned to generate a fine-tuned safety production monitoring instruction generation model. Based on the multiple current plant production status information and the fine-tuned safety production monitoring instruction generation model, multiple safety production monitoring instruction information is generated to perform intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

7. The intelligent monitoring method for safe production as described in claim 1, characterized in that, The safety production monitoring instruction information includes safety level early warning instruction information, equipment parameter adjustment instruction information, personnel dispatch instruction information, and production line control type instruction information; After the step of generating multiple safety production monitoring instruction information based on the multiple current plant area production status information, multiple historical plant area production status information, a safety production monitoring instruction generation model, and a preset safety production risk information database, and then performing intelligent safety production monitoring processing through the multiple safety production monitoring instruction information, the method further includes: Based on the preset mapping relationship between safety production monitoring instructions and the start / stop parameters of audible and visual alarm devices, multiple start / stop parameter information for audible and visual alarm devices is generated according to the multiple safety level early warning instruction information and multiple device parameter control instruction information. Based on the preset mapping relationship between safety production monitoring instructions and visual monitoring equipment operation control parameters, multiple visual monitoring equipment operation control parameter information is generated according to the multiple personnel scheduling instruction information and multiple production line control type instruction information.

8. A smart monitoring device for safe production, characterized in that, include: The information acquisition module is used to acquire equipment operation area information, current equipment operating status information, current plant area operation behavior information, historical equipment operating status information, historical plant area operation behavior information, and a safety production monitoring instruction generation model. The current plant area production status information generation module is used to generate multiple current plant area production status information based on the equipment operation area information, current equipment operation status information, and current plant area operation behavior information; The historical plant area production status information generation module is used to generate multiple historical plant area production status information based on the equipment operation area information, historical equipment operation status information, and historical plant area operation behavior information. The safety production monitoring instruction information generation module is used to generate multiple safety production monitoring instruction information based on the multiple current plant area production status information, multiple historical plant area production status information, safety production monitoring instruction generation model, and a preset safety production risk information database, so as to carry out intelligent safety production monitoring and processing through the multiple safety production monitoring instruction information.

9. A terminal device, characterized in that, The terminal device includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images