Enterprise safety production standardization evaluation management method and system

Abstract

The invention relates to the technical field of enterprise safety production standardization assessment, in particular to an enterprise safety production standardization assessment management method and system. The method comprises the following steps: firstly, providing an environmental safety production monitoring model to evaluate gas safety, fire safety, flood safety and collapse safety, and outputting safety production evaluation values in various aspects through a safety evaluation layer for safety evaluation; according to the model, comprehensive evaluation of safety production in the aspect of mine environment is realized; secondly, providing a safety production assessment method for the worker, which is used for assessing the position, dressing and action of the worker; the health and the safety of workers are guaranteed through the evaluation in all the aspects; finally, the invention provides a safety production assessment method of the working equipment, and the method monitors the health degree of the health function equipment through the equipment, and assesses the comprehensive health state of the working equipment. According to the invention, comprehensive and accurate safety production evaluation in a mine environment is realized from multiple aspects.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise safety production standardization assessment, and in particular to an enterprise safety production standardization assessment management method and system. Background Art

[0002] Enterprise safety production refers to the process of production and operation of enterprises, through effective management measures and system arrangements, to ensure the safety of employees, facilities and environment, prevent accidents and losses, and ensure that production activities are carried out safely, stably and continuously. This is part of corporate social responsibility and an important guarantee for protecting employee health and improving production efficiency.

[0003] Nowadays, all walks of life pay more and more attention to safety in production, especially in high-risk industries such as the coal industry. In the process of mining, the risk of safety accidents is increased due to the complex and harsh environment. Therefore, real-time monitoring and evaluation during mining operations are very necessary for the safety of mines. A good safety production assessment method can not only help identify and manage safety risks, but also improve management strategies, prevent accidents and enhance corporate reputation. It is an indispensable and important tool in mine safety management.

[0004] The standardized assessment of mine safety production is a very complex task; it includes many aspects, such as safety assessment of the working environment, safety assessment of staff and safety assessment of working equipment; these aspects together constitute the important content of the standardized assessment of mine safety production. Each aspect requires detailed data, analysis and comprehensive judgment to ensure that the mine's safety production reaches a high level.

[0005] Many methods for evaluating mine safety production standards have been proposed in existing literature; for example, Chinese patent publication number CN111894677A discloses a mine monitoring and early warning system based on information fusion theory; the invention uses the STM32 embedded system as a platform, collects environmental parameters such as gas concentration, carbon monoxide concentration, wind speed, temperature, humidity, water level of the reservoir, etc. in mine tunnels, fully mechanized mining working faces, and fully mechanized excavation working faces, and uses information fusion algorithms to evaluate the current working status. According to the output results, PWM speed control of fans and control of water pumps can be realized. Control, early warning and other functions, thereby realizing the mine monitoring system to ensure the safe production of coal mines; for example, the Chinese patent with publication number CN105117857A discloses a coal mine safety dynamic assessment and diagnosis system and method; the invention is based on a comprehensive integrated coal mine information system and monitoring system to establish a comprehensive coal mine safety database and expert knowledge base; combined with the daily transactional data of the coal mine and the real-time monitoring data, based on the three-dimensional virtual mine model, a coal mine safety dynamic diagnosis system is established, which ultimately reduces the occurrence of coal mine safety accidents and realizes the informatization and modernization of coal mine safety management. However, the existing literature mentioned above only evaluates a single aspect of mine safety production, is not comprehensive, and cannot achieve accurate evaluation of safety production in a mine environment.

[0006] To this end, the present invention proposes an enterprise safety production standardization assessment management method and system. Summary of the invention

[0007] The purpose of the present invention is to propose a method and system for enterprise safety production standardization assessment management to realize safety production assessment in the mine operation process; the present invention respectively realizes the safety production assessment of the mine working environment, the relevant safety production assessment of the staff and the safety production assessment of the working equipment; among them, in terms of the safety production assessment of the working environment, the present invention monitors and assesses the common safety accidents in the mine, including: gas safety, fire safety, flood safety and landslide safety. The comprehensive use of these monitoring and assessment methods can effectively improve the safety of the mine working environment, prevent and reduce the occurrence of common safety accidents such as gas hazards, fire, flood and landslide, and ensure the safety of workers and equipment; in terms of the relevant safety production assessment of the staff, the present invention The evaluation of the staff includes: whether they are in the prescribed working position, whether their attire complies with the regulations, and whether there are any dangerous behaviors in the operation process. Through the above evaluation, the working position, attire and working behavior of the staff can be effectively monitored and evaluated, and safety accidents caused by human operation can be prevented and reduced, so as to ensure the health and safety of the staff. In terms of the safe production evaluation of working equipment, the present invention obtains the working parameters of the current working equipment and performs characterization operations; then, the corresponding health status indicators are constructed according to the type of working equipment; finally, the health status of the working equipment is calculated through the equipment monitoring health function; in this way, comprehensive health status evaluation and management of the working equipment can be realized, equipment failures and accidents can be effectively prevented, and work safety can be improved.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] The present invention proposes an enterprise safety production standardization assessment management method, comprising:

[0010] A mine environment safety production monitoring model is used to perform safety production monitoring; wherein the mine environment safety production monitoring model includes: a data acquisition layer, a data processing layer, a safety monitoring layer, a safety assessment layer and an output layer;

[0011] The data acquisition layer acquires multi-source data inside the mine through various instruments and equipment; wherein the multi-source data includes: gas concentration data, air pressure data, temperature data, humidity data, smoke concentration data, hydrological data, geological data, mine tunnel image data and mine structure data;

[0012] Further, the data collected by the data collection layer is input into the data processing layer;

[0013] Furthermore, the data processing layer is used to pre-process the multi-source data to obtain standard multi-source data;

[0014] Furthermore, the standard multi-source data is input into the safety monitoring layer to obtain multi-type safety monitoring results; wherein the safety monitoring layer includes: a gas safety monitoring module, a fire safety monitoring module, a flood safety monitoring module and a landslide safety monitoring module;

[0015] The gas safety monitoring module is used to monitor oxygen and harmful gases inside the mine. The specific gas safety monitoring process includes: obtaining standard gas data of each monitoring point from the standard multi-source data; comparing the standard gas data with the safety threshold of each gas, and recording the gas safety error; and performing gas safety assessment according to the gas safety error;

[0016] The fire safety monitoring module is used to monitor the occurrence of fire inside the mine. The specific fire safety monitoring process includes: obtaining standard temperature data from the standard multi-source data; judging whether the standard temperature data exceeds the fire temperature threshold; when exceeding, obtaining the standard mine tunnel image and standard carbon monoxide data of the corresponding position according to the acquisition position of the standard temperature data; performing image segmentation on the standard mine tunnel image to obtain a suspected fire source image; extracting the features of the suspected fire source image to obtain a suspected fire source feature map; wherein the suspected fire source feature map includes: color features, texture features and shape features; converting the suspected fire source feature map into numerical values ​​to obtain multiple suspected fire source values; and calculating the fire degree in combination with the standard carbon monoxide data and the multiple suspected fire source values.

[0017] The flood safety monitoring module is used to monitor the occurrence of floods inside the mine. The specific flood safety monitoring process includes: obtaining real-time standard hydrological data of each water level monitoring point; extracting features from the standard hydrological data using feature engineering to obtain flood feature vectors; inputting the flood feature vectors into a neural network to obtain the probability of mine floods;

[0018] The collapse safety monitoring module is used to monitor the collapse probability inside the mine. Specifically, the collapse safety monitoring module includes: obtaining standard hydrological data, standard geological data, standard mine tunnel image data and standard mine structure data through the data processing layer; extracting features from the standard hydrological data, the standard geological data, the standard mine tunnel image data and the standard mine structure data; respectively constructing a hydrological feature set, a geological feature set, a mine tunnel feature set and a structure feature set; analyzing the features in each feature set to obtain feature weights; weighting the original standard data with the feature weights to obtain weighted standard data; fusing the weighted standard data with each feature set to obtain collapse prediction fusion feature data; analyzing the collapse prediction fusion feature data using a collapse regression analysis equation to obtain the collapse probability; wherein the collapse regression analysis equation represents: Wherein, P() represents the landslide regression analysis equation; X represents the set of landslide prediction fusion feature data; x i It is represented as the fusion feature data of the i-th landslide prediction; β0 is represented as the regression coefficient; β i It is represented as the contribution coefficient of the i-th landslide prediction fusion feature data; n is represented as the total number of landslide prediction fusion feature data.

[0019] Further, the multi-type safety monitoring results are input into the safety assessment layer to obtain multi-type safety production assessment values ​​of the mine environment;

[0020] Further, when any one of the multiple types of production safety assessment values ​​does not reach the mine production safety standard, the mine operation is stopped; otherwise, the mine operation is continued;

[0021] Furthermore, the safety production monitoring of the staff is carried out, and the specific process includes:

[0022] Determine whether the worker is at the specified task point based on the positioning information; issue a warning when the distance between the positioning point and the specified task point exceeds the maximum allowable distance;

[0023] When the maximum allowable distance is not exceeded, obtaining a surveillance video of the staff member;

[0024] Process the surveillance video to obtain N surveillance images; perform clothing recognition and behavior recognition on the N surveillance images; wherein the clothing recognition process includes: segmenting the staff images in the N surveillance images; marking the staff images with clothing; wherein the content of the clothing mark includes: head, hands, upper body and lower body; according to the clothing mark, identify: safety helmet, safety glasses, dust mask, safety gloves, work clothes and protective shoes; when any item in the clothing mark is not recognized, issue a safety warning;

[0025] Furthermore, various data of the staff in the monitoring time period are extracted, including: posture, motion trajectory and gait; the monitoring time is evenly divided into multiple time points; the features of the data corresponding to the time points are extracted to construct a time feature sequence; the time feature sequence is input into the behavior safety recognition model to obtain the behavior safety recognition result corresponding to the time point; wherein the behavior safety recognition model includes: an input layer for receiving the time feature sequence and performing feature correlation processing on the time feature sequence; a local limb feature attention layer for paying attention to some features in the time feature sequence; a weighted layer for performing weighted operations on the features paid attention to by the local limb feature attention layer; and an output layer for outputting the behavior danger probability at each time point. When the behavior safety recognition result shows illegal behavior, a safety warning is issued.

[0026] Furthermore, the working equipment during the mine operation is monitored for production safety, and the specific process includes: collecting the current working parameters of the working equipment;

[0027] Further, the current working parameters are subjected to data preprocessing to obtain current equipment working characteristics;

[0028] Further, constructing a device operating feature set according to the working device type and the current device operating feature;

[0029] Further, obtaining a health status indicator of a corresponding device according to the device type;

[0030] Further, the health status of the corresponding working equipment in the equipment working feature set is analyzed according to the health status indicator; wherein the health status is analyzed using an equipment monitoring health function; the equipment monitoring health function is expressed as: Among them, EHV i represents the monitored health value of working device i; f() represents the device monitoring health function; HSI represents the health status index; Represented as the current equipment working characteristics of working equipment i;

[0031] Furthermore, whether the working equipment is safe is determined based on the monitored health value.

[0032] An enterprise safety production standardization assessment and management system, comprising: a monitoring unit, a positioning unit, a data acquisition unit, a mine environment safety production assessment unit, a staff safety production assessment unit, a work equipment safety production assessment unit, a display unit and a feedback unit;

[0033] Wherein, the monitoring unit is used to monitor the interior of the mine;

[0034] The positioning unit is used to locate the staff and the working equipment; the data collection unit is used to use the data related to the safety assessment;

[0035] The mine environment production safety assessment unit is used to monitor and assess the internal environment of the mine for production safety; it includes: a gas safety monitoring module, a fire safety monitoring module, a flood safety monitoring module and a landslide safety monitoring module;

[0036] The worker production safety assessment unit is used to monitor and assess the worker's production safety; it includes: a work position safety monitoring module, a clothing safety monitoring module and a behavior safety monitoring module;

[0037] Wherein, the work position safety monitoring module determines whether the worker is at the specified task point according to the positioning unit; when the distance between the positioning point and the specified task point exceeds the maximum allowable distance, the feedback unit provides feedback to the management personnel;

[0038] When the maximum allowed distance is not exceeded, the monitoring video of the staff is obtained; the clothing safety monitoring module processes the monitoring video to obtain N monitoring images; the clothing recognition and behavior recognition are performed on the N monitoring images; wherein the clothing recognition process includes: segmenting the staff images in the N monitoring images; marking the staff images with clothing; wherein the content of the clothing mark includes: head, hands, upper body and lower body; recognition is performed according to the clothing mark; wherein, safety helmets, safety glasses, dust masks, safety gloves, work clothes and protective shoes; when any item of the clothing mark is not recognized, the feedback unit provides feedback to the management personnel;

[0039] The behavior safety monitoring module extracts various data of the staff in the monitoring time period, including: posture, motion trajectory and gait; divides the monitoring time evenly into multiple time points; extracts the features of the data corresponding to the time points, and constructs a time feature sequence; inputs the time feature sequence into the behavior safety recognition model to obtain the behavior safety recognition result corresponding to the time point; wherein the behavior safety recognition model includes: an input layer, which is used to receive the time feature sequence and perform feature correlation processing on the time feature sequence; a local limb feature attention layer, which is used to pay attention to some features in the time feature sequence; a weighted layer, which is used to perform weighted operations on the features paid attention to by the local limb feature attention layer; an output layer, which is used to output the behavior danger probability at each time point;

[0040] The feedback unit provides feedback to the management personnel according to the safety identification result;

[0041] The working equipment production safety assessment unit is used to assess the production safety of working equipment; wherein, the working equipment production safety assessment unit comprises: collecting current working parameters of working equipment; performing data preprocessing on the current working parameters to obtain current equipment working characteristics; constructing a set of equipment working characteristics according to the type of working equipment combined with the current equipment working characteristics; obtaining a health status indicator of the corresponding equipment according to the equipment type; analyzing the health status of the corresponding working equipment in the equipment working characteristic set according to the health status indicator; wherein, the health status is analyzed using an equipment monitoring health function; the equipment monitoring health function is expressed as: Among them, EHV i represents the monitored health value of working device i; f() represents the device monitoring health function; HSI represents the health status index; Represented as the current equipment working characteristics of the working equipment i; judging whether the working equipment is safe according to the monitored health value;

[0042] The display unit is used to display the scene inside the mine;

[0043] The feedback unit is used to feed back various production safety assessment results to management personnel.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. The present invention proposes a mine environment production safety monitoring model for monitoring and evaluating gas safety, flood safety, fire safety and landslide safety; the model collects multi-source data in the mine to evaluate the production safety of the above-mentioned aspects; in the implementation process, the gas safety is ensured by comparing the oxygen content and the concentration of harmful gases with the safety threshold; in terms of flood safety, the groundwater level changes at each monitoring point are mainly analyzed by hydrological data to evaluate flood safety; in terms of fire safety, the suspected fire source is judged by temperature; and the fire safety of the suspected fire source is evaluated based on the image and carbon monoxide concentration; in terms of landslide safety, the multi-source data is combined for analysis, the original multi-source data is updated in a weighted manner, and the landslide probability is calculated by the landslide regression analysis equation; using these monitoring and evaluation methods, a comprehensive evaluation of production safety in the mine environment can be achieved, and the safety of the mine working environment can be effectively improved.

[0046] 2. The present invention proposes a method for evaluating the safe production of staff; the method evaluates the position, clothing and movements of the staff; in terms of the position of the staff, the distance between the staff and the specified task point is calculated by positioning means to evaluate whether the staff performs their duties as required; in terms of clothing, the equipment worn by the staff is identified to determine whether the staff wears the equipment as required, so as to provide timely feedback; in terms of movement, a behavioral safety recognition model is used to identify abnormalities in the behavioral movements of the staff; wherein, in the model, the hands, feet and upper body of the staff are paid attention to by adding a local limb feature attention layer to improve the accuracy of recognition; the above evaluation method ensures the health and safety of the staff, and realizes a comprehensive evaluation of the staff in the safe production process.

[0047] 3. The present invention proposes a method for evaluating the production safety of working equipment. The method obtains the working parameters of the current working equipment and performs characterization operations; then, the corresponding health status indicators are constructed according to the type of working equipment; finally, the health status of the working equipment is calculated through the equipment monitoring health function; in this way, comprehensive health status evaluation and management of the working equipment can be achieved, and the occurrence of equipment failures and accidents can be effectively prevented. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flowchart of a method for enterprise safety production standardization assessment management provided by an embodiment of the present invention;

[0049] Figure 2 A structural diagram of an enterprise safety production standardization assessment management system provided by an embodiment of the present invention;

[0050] Figure 3 A structural diagram of a mine environment safety production monitoring model provided by an embodiment of the present invention;

[0051] Figure 4 A structural diagram of a mine environment safety production assessment unit provided by an embodiment of the present invention;

[0052] Figure 5 A structural diagram of a worker production safety assessment unit provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Enterprise safety production refers to the process of production and operation of enterprises, through effective management measures and system arrangements, to ensure the safety of employees, facilities and environment, prevent accidents and losses, and ensure that production activities are carried out safely, stably and continuously. This is part of corporate social responsibility and an important guarantee for ensuring employee safety and improving production efficiency.

[0055] Nowadays, enterprises attach great importance to production safety, especially the coal industry. For high-risk industries such as the coal industry, enterprise production safety is particularly important. There are many safety factors involved in the mining process, such as the complex environment inside the mine, which hides huge risks; the non-standard operation methods of the staff will also bring safety hazards to production. It is necessary to carry out multi-faceted monitoring and evaluation during mine operations for production safety. It is crucial to adopt effective standardized evaluation methods during the monitoring process to ensure production safety; through standardized evaluation of the production safety process, potential safety hazards and problems in the mine can be identified in time, accidents can be prevented, and the overall production safety level and work efficiency can be improved.

[0056] In the prior art, most methods for evaluating mine safety production only focus on one aspect, which results in an incomplete evaluation of mine safety production and the inability to well guarantee the safety of workers and equipment; therefore, factors affecting safety production in a complex mine environment require a comprehensive evaluation of multiple aspects. To this end, the present invention proposes a standardized evaluation management method and system for safety production; the method and system achieve a comprehensive and accurate safety evaluation of the mine environment, workers, and working equipment. The following will be explained from two embodiments.

[0057] Embodiment 1:

[0058] In the embodiment of the present application, the present solution will be described for Mine A. It is known that Mine A has 4 mine tunnels, so the safety production assessment of the 4 mine tunnels is performed; for the method flow and system structure proposed in the present invention, refer to Figure 1 and Figure 2 Content. Figure 1 The specific steps of the method of the present invention include: S10. Using the mine environment safety production monitoring model to perform safety production monitoring; S20. Determining whether the mine safety production monitoring results meet the mine safety production standards; S30. Performing safety production monitoring on the staff; S40. Performing safety production monitoring on the working equipment. Figure 2 In the embodiment, the structure of the system is: a monitoring unit, a positioning unit, a data acquisition unit, a mine environment safety production assessment unit, a staff safety production assessment unit, a work equipment safety production assessment unit, a display unit and a feedback unit.

[0059] Further, according to Figure 1 and Figure 2 The contents of the embodiment are described in detail;

[0060] The four mine tunnels of the mine are monitored in real time according to the monitoring unit of the system; wherein the monitoring unit monitors through a high-pixel camera installed in the mine tunnel, and the monitoring scope includes: the internal environment of the mine tunnel, the staff and the working equipment;

[0061] Furthermore, data collection is performed by a data collection unit of the system; wherein the data collection unit is used to obtain multi-source data related to production safety assessment;

[0062] The data acquisition unit corresponds to the data acquisition layer of the mine environment safety production monitoring model in step S10; the unit not only uses the monitoring unit to obtain video information, but also facilitates data acquisition using various instruments such as temperature sensors, humidity sensors, gas sensors and smoke sensors placed inside the mine tunnel; the structure of the mine environment safety production monitoring model refers to Figure 3 ;

[0063] In the embodiment of the present application, the data acquired by the data acquisition unit includes: gas concentration data, air pressure data, temperature data, humidity data, smoke concentration data, hydrological data, geological data, mine tunnel image data and mine structure data.

[0064] In the embodiment of the present application, a series of data is obtained through the data acquisition unit for mine environment safety production monitoring; multiple types of data are collected for subsequent multi-faceted safety production monitoring; comprehensive data acquisition and analysis can effectively improve the comprehensiveness and timeliness of mine environment safety monitoring, enhance the ability to respond to sudden safety incidents, and ensure the life safety of miners and the operational stability of mine facilities. This comprehensive monitoring method not only helps to improve the level of safety production, but also effectively reduces the occurrence of accidents and losses.

[0065] Further, the collected multi-source data is input into the data processing layer of the mine environment safety production monitoring model; the data processing layer is used to pre-process the multi-source data to obtain standard multi-source data; the data processing methods for different data types in the data processing layer are mainly divided into two categories, numerical data pre-processing and image data pre-processing; wherein, numerical data pre-processing includes: data clarity, standardization processing, feature selection and data integration, etc.; image data pre-processing includes: image correction, image denoising, image enhancement and image cropping, etc.;

[0066] Further, the standard multi-source data is input into the safety monitoring layer of the mine environment safety production monitoring model, corresponding to the mine environment safety production assessment unit of the system, referring to Figure 4; The safety monitoring layer includes: a gas safety monitoring module, a fire safety monitoring module, a flood safety monitoring module and a landslide safety monitoring module;

[0067] The gas safety monitoring module is used to monitor oxygen and harmful gases inside the mine. The specific gas safety monitoring process includes: obtaining standard gas data of each monitoring point from the standard multi-source data; comparing the standard gas data with the safety threshold of each gas, and recording the gas safety error; and performing gas safety assessment according to the gas safety error.

[0068] In the embodiment of the present application, a gas safety monitoring module is used for gas safety monitoring in the mine, which is mainly used to obtain oxygen and harmful gases in the mine; among them, oxygen is an essential gas for miners to breathe and is used to maintain normal physiological functions. The oxygen content in the mine directly affects the health and work efficiency of the miners. If the oxygen content is insufficient or abnormal, the miners may suffer from hypoxia, leading to dizziness, fatigue or even coma, and even life-threatening in severe cases; the harmful gases in the mine mainly include: methane, carbon dioxide, carbon monoxide and ammonia, etc.; when the harmful gases exceed a certain amount, it will endanger the life safety of the workers. Table 1 shows some of the gases obtained in mine tunnel 2;

[0069] Table 1. Gas values ​​for some mine tunnels 2

[0070] gas Gas concentration value oxygen 21.3％ Methane 1.25％ carbon dioxide 0.54％ Carbon monoxide 50pmm Ammonia 25ppm

[0071] The safety of harmful gases in mine tunnel 2 is judged based on the set minimum threshold; the oxygen content is judged based on the actual staff in the mine tunnel, and it is considered a dangerous state when the oxygen content is lower than the minimum safety standard;

[0072] A fire safety monitoring module is used to monitor the fire in each mine tunnel of Mine A. The specific monitoring process includes: the fire safety monitoring module is used to monitor the occurrence of fire inside the mine. The specific fire safety monitoring process includes: obtaining standard temperature data from the standard multi-source data; judging whether the standard temperature data exceeds the fire temperature threshold; when exceeding, obtaining the standard mine tunnel image and standard carbon monoxide data of the corresponding position according to the acquisition position of the standard temperature data; performing image segmentation on the standard mine tunnel image to obtain a suspected fire source image; extracting the features of the suspected fire source image to obtain a suspected fire source feature map; wherein the suspected fire source feature map includes: color features, texture features and shape features; converting the suspected fire source feature map into numerical values ​​to obtain multiple suspected fire source values; calculating the fire degree in combination with the standard carbon monoxide data and the multiple suspected fire source values;

[0073] In the embodiment of the present application, the fire in the mine is monitored; first, the temperature inside the mine tunnel is obtained according to the temperature sensor; a threshold is set to judge that the obtained temperature data has a high temperature and is suspected of a fire; in order to further identify the authenticity of the fire, the probability of the fire is calculated by obtaining an image at a high temperature and the concentration of carbon monoxide; in a mine environment, the temperature may rise due to machine operation and other conditions; the reliability of fire identification can be increased by using image recognition technology and carbon monoxide concentration;

[0074] The flood safety monitoring module is used to monitor the occurrence of floods inside the mine. The specific flood safety monitoring process includes: obtaining real-time standard hydrological data of each water level monitoring point; using feature engineering to extract features from the standard hydrological data to obtain a flood feature vector; inputting the flood feature vector into a neural network to obtain the probability of mine floods; wherein, a long short-term memory neural network is used to obtain the probability of mine floods, and the network can accurately analyze the probability of floods based on the changes in hydrological data at different times.

[0075] In the embodiments of the present application, the collected hydrological data is used for flood monitoring. The application of this data in mine flood monitoring not only provides key real-time information and trend analysis, but also supports scientific decision-making and effective emergency response, thereby improving the safety and reliability of mines.

[0076] The collapse safety monitoring module is used to monitor the collapse probability inside the mine. Specifically, the collapse safety monitoring module includes: obtaining standard hydrological data, standard geological data, standard mine tunnel image data and standard mine structure data through the data processing layer; extracting features from the standard hydrological data, the standard geological data, the standard mine tunnel image data and the standard mine structure data; respectively constructing a hydrological feature set, a geological feature set, a mine tunnel feature set and a structure feature set;

[0077] In the embodiment of the present application, the hydrological feature set includes: groundwater level change trend, absolute height of groundwater level, groundwater flow velocity and direction, groundwater permeability and water permeability, and groundwater erosion on rock formations, etc.; the geological feature set includes: rock type and lithology, formation dip and stratigraphic relationship, and cracks and fragmentation of rock formations, etc.; the mine tunnel feature set includes: fissures and joints and the integrity of mine support structures; the structural feature set includes: the geometric shape and size of the mine tunnel and the geological structure around the mine tunnel, etc.;

[0078] The features in each feature set are analyzed to obtain feature weights; the feature weights are weighted on the original standard data to obtain weighted standard data; the weighted standard data are fused with each feature set to obtain landslide prediction fusion feature data; the landslide prediction fusion feature data are analyzed using a landslide regression analysis equation to obtain the landslide probability; wherein the landslide regression analysis equation is expressed as: Wherein, P() represents the landslide regression analysis equation; X represents the set of landslide prediction fusion feature data; x i It is represented as the fusion feature data of the i-th landslide prediction; β0 is represented as the regression coefficient; β i It is represented as the contribution coefficient of the i-th landslide prediction fusion feature data; n is represented as the total number of landslide prediction fusion feature data.

[0079] Furthermore, the monitoring results of each module are input into the safety assessment layer to obtain multiple types of safety production assessment values ​​of the mine environment; including: gas safety assessment, fire safety assessment, flood safety assessment and landslide safety assessment.

[0080] In the embodiment of the present application, the mine environmental production safety monitoring model performs environmental production safety monitoring; wherein, a safety monitoring layer is added to the model to realize comprehensive monitoring of the mine environment from multiple aspects such as gas safety monitoring, fire safety monitoring, flood safety monitoring and landslide safety monitoring; the comprehensive use of these monitoring and evaluation methods can effectively improve the safety of the mine working environment, prevent and reduce the occurrence of common safety accidents such as gas hazards, fires, floods and landslides, and ensure the safety of workers and equipment.

[0081] Further, according to Figure 1 In step S20, it is determined whether each safety assessment result meets the mine safety production assessment standard; if any one of them does not meet the standard, the mine operation is stopped; otherwise, the mine operation is continued;

[0082] Further, the staff safety production assessment unit of the system performs a safety production assessment, corresponding to the above-mentioned step S30; wherein the staff safety production assessment unit is as follows: Figure 5 As shown, it includes: work position safety monitoring module, clothing safety monitoring module and behavior safety monitoring module;

[0083] Among them, the work position safety monitoring module uses the positioning unit of the system to obtain the position of the staff; the staff in the mine are equipped with a GPS positioning device to track and record the position of each staff member in real time to ensure their safety; in emergency situations, such as collapse or accidents, the position of trapped people can be quickly determined to improve rescue efficiency.

[0084] Furthermore, the work position safety monitoring module calculates the distance between the worker and the specified task point according to the acquired positioning information; when the distance error is greater than the maximum allowable distance, the feedback unit of the system provides feedback to the management personnel and warns the worker;

[0085] When the maximum allowable distance is not exceeded, the monitoring unit of the system obtains the monitoring video of the staff according to the positioning position, and the staff's clothing is safely monitored through the clothing safety monitoring module; safe clothing is crucial to preventing accidents and protecting workers' safety during actual mine operations; among them, protective equipment includes: mine work clothes, safety helmets, safety shoes, and dust masks, etc.

[0086] The specific monitoring process of the clothing safety monitoring module includes: performing video processing on the monitoring video to obtain N monitoring images; performing clothing recognition and behavior recognition on the N monitoring images; wherein the clothing recognition process includes: segmenting the staff images in the N monitoring images; performing clothing marking on the staff images; wherein the content of the clothing marking includes: head, hands, upper body and lower body; identifying according to the clothing marking, including: safety helmets, safety glasses, dust masks, safety gloves, work clothes and protective shoes; and issuing a safety warning when any item of the clothing marking is not recognized.

[0087] Furthermore, the behavior safety monitoring module of the worker production safety assessment unit monitors the worker's movements; the specific process includes: extracting various data of the worker during the monitoring time period, including: posture, motion trajectory and gait; evenly dividing the monitoring time into multiple time points; extracting the features of the data corresponding to the time points, and constructing a time feature sequence; inputting the time feature sequence into the behavior safety recognition model to obtain the behavior safety recognition results corresponding to the time points; wherein the behavior safety recognition model includes: an input layer, for receiving the time feature sequence and performing feature correlation processing on the time feature sequence; a local limb feature attention layer, for paying attention to some features in the time feature sequence; a weighted layer, for performing weighted operations on the features paid attention to by the local limb feature attention layer; an output layer, for outputting the behavior hazard probability at each time point;

[0088] In the embodiment of the present application, a behavioral safety recognition model is used to monitor the underground behavior of the workers; in the model, the worker's behavior is feature extracted according to the time point. A local limb feature attention layer is added to the model to pay attention to the hands, feet, legs and upper body of the workers, and then the weighted layer is used to weight the attention features; this can achieve effective monitoring and safety assessment of the workers' actions, effectively prevent accidents and improve the overall safety of the workplace.

[0089] When the behavior safety identification results show any violation, a safety warning will be issued.

[0090] In the embodiments of the present application, the present invention performs safety monitoring on the position, clothing and movements of the staff, which can not only improve the overall safety level of the workplace, but also improve work efficiency, reduce the risk of accidents, and provide a scientific basis for management decisions, thereby ensuring the health and safety of the staff.

[0091] Furthermore, the working equipment safety production assessment unit of the system performs safety monitoring on the currently working equipment, corresponding to the above-mentioned step S40; the specific process includes: collecting the current working parameters of the working equipment; performing data preprocessing on the current working parameters to obtain the current equipment working characteristics; constructing a set of equipment working characteristics according to the working equipment type and the current equipment working characteristics; obtaining the health status indicator of the corresponding equipment according to the equipment type; analyzing the health status of the corresponding working equipment in the equipment working characteristic set according to the health status indicator; wherein the health status is analyzed using the equipment monitoring health function; the equipment monitoring health function is expressed as: Among them, EHV i represents the monitored health value of working device i; f() represents the device monitoring health function; HSI represents the health status index; Represented as the current equipment working characteristics of working equipment i; judging whether the working equipment is safe according to the monitored health value.

[0092] In the embodiment of the present application, a working equipment safety assessment method is adopted to perform a safety assessment on the currently running working equipment, by obtaining the working parameters of the current working equipment and performing characterization operations; then, a corresponding health status indicator is constructed according to the type of working equipment; finally, the health status of the working equipment is calculated through the equipment monitoring health function; in this way, comprehensive health status assessment and management of the working equipment can be achieved, effectively preventing the occurrence of equipment failures and accidents, and improving work safety.

[0093] Furthermore, the system display unit displays the working image of mine A.

[0094] In the application embodiment, the method of the present invention is combined with the system to realize the safety production assessment of mine A; the present invention mainly starts from the following aspects; first, the present invention proposes a mine environment safety production monitoring model for monitoring and evaluating the gas safety, fire safety, flood safety and landslide safety of mine A; the model includes: a data acquisition layer, a data processing layer, a safety monitoring layer, a safety assessment layer and an output layer; the safety monitoring layer realizes comprehensive monitoring of the mine environment from multiple aspects of gas safety monitoring, fire safety monitoring, flood safety monitoring and landslide safety monitoring; the comprehensive use of these monitoring and evaluation methods can effectively improve the safety of the mine working environment, prevent and reduce the occurrence of common safety accidents such as gas hazards, fires, floods and landslides, and ensure the safety of workers and equipment. Secondly, the present invention proposes a method for monitoring and evaluating the safe production of staff. This method conducts safety monitoring on the position, clothing and movements of staff, which can not only improve the overall safety level of the workplace, but also improve work efficiency, reduce accident risks, and provide a scientific basis for management decisions, thereby ensuring the health and safety of staff; Finally, the present invention proposes a method for monitoring and evaluating the safe production of work equipment. This method obtains the working parameters of the current work equipment and performs characterization operations; then, the corresponding health status indicators are constructed according to the type of work equipment; finally, the health status of the work equipment is calculated through the equipment monitoring health function; in this way, comprehensive health status evaluation and management of the work equipment can be achieved, equipment failures and accidents can be effectively prevented, and work safety can be improved.

[0095] Embodiment 2:

[0096] In Example 1, the method and system of the present invention are combined to realize the safety production standardization evaluation of 4 mine tunnels in Mine A. In order to further illustrate the effectiveness of the present invention, in the embodiment of the present application, the safety production standardization evaluation of Mine B is performed. It is known that there are 2 mine tunnels in Mine B that are currently undergoing mining operations. The specific evaluation process is basically the same as the implementation process of Example 1, which mainly includes:

[0097] The mine environment safety production monitoring model is used to conduct environmental safety production monitoring and assessment on the mine tunnels in Mine B where mining operations are ongoing. The assessment aspects include: gas safety, fire safety, flood safety, and landslide safety.

[0098] The specific process includes:

[0099] The multi-source data of the mine B is collected by using the data collection layer of the mine environment safety production monitoring model; wherein the multi-source data mainly refers to gas concentration data, air pressure data, temperature data, humidity data, smoke concentration data, hydrological data, geological data, mine tunnel image data and mine structure data;

[0100] Further, the multi-source data about the mine B is input into the data processing layer to obtain standard multi-source data;

[0101] Furthermore, the standard multi-source data is input into the safety monitoring layer to obtain multi-type safety monitoring results; wherein the safety monitoring layer includes: a gas safety monitoring module, a fire safety monitoring module, a flood safety monitoring module and a landslide safety monitoring module;

[0102] Further, the multi-type safety monitoring results are input into the safety assessment layer to obtain multi-type safety production assessment values ​​of the mine environment;

[0103] Further, when any one of the multiple types of production safety assessment values ​​does not reach the mine production safety standard, the mine operation is stopped; otherwise, the mine operation is continued;

[0104] Furthermore, the safety production monitoring of the staff is carried out, and the specific process includes:

[0105] Determine whether the worker is at the specified task point based on the positioning information; issue a warning when the distance between the positioning point and the specified task point exceeds the maximum allowable distance;

[0106] When the maximum allowable distance is not exceeded, obtaining a surveillance video of the staff member;

[0107] Process the surveillance video to obtain N surveillance images; perform clothing recognition and behavior recognition on the N surveillance images; wherein the clothing recognition process includes: segmenting the staff images in the N surveillance images; marking the staff images with clothing; wherein the content of the clothing mark includes: head, hands, upper body and lower body; identifying according to the clothing mark; wherein, safety helmets, safety glasses, dust masks, safety gloves, work clothes and protective shoes; when any item of the clothing mark is not recognized, issuing a safety warning;

[0108] Furthermore, various data of the staff in the monitoring time period are extracted, including: posture, motion trajectory and gait; the monitoring time is evenly divided into multiple time points; the features of the data corresponding to the time points are extracted to construct a time feature sequence; the time feature sequence is input into the behavior safety recognition model to obtain the behavior safety recognition result corresponding to the time point; wherein the behavior safety recognition model includes: an input layer for receiving the time feature sequence and performing feature correlation processing on the time feature sequence; a local limb feature attention layer for paying attention to some features in the time feature sequence; a weighted layer for performing weighted operations on the features paid attention to by the local limb feature attention layer; and an output layer for outputting the behavior danger probability at each time point. When the behavior safety recognition result shows illegal behavior, a safety warning is issued.

[0109] Furthermore, the working equipment during the mine operation is monitored for production safety, and the specific process includes: collecting the current working parameters of the working equipment;

[0110] Further, the current working parameters are subjected to data preprocessing to obtain current equipment working characteristics;

[0111] Further, constructing a device operating feature set according to the working device type and the current device operating feature;

[0112] Further, obtaining a health status indicator of a corresponding device according to the device type;

[0113] Further, the health status of the corresponding working equipment in the equipment working feature set is analyzed according to the health status indicator; wherein the health status is analyzed using an equipment monitoring health function; the equipment monitoring health function is expressed as: Among them, EHV i represents the monitored health value of working device i; f() represents the device monitoring health function; HSI represents the health status index; Represented as the current equipment working characteristics of working equipment i;

[0114] Furthermore, whether the working equipment is safe is determined based on the monitored health value.

[0115] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for enterprise safety production standardization assessment management, characterized in that: include: Use the mine environment safety production monitoring model to carry out safety production monitoring; The mine environment safety production monitoring model includes: a data acquisition layer, a data processing layer, a safety monitoring layer, a safety assessment layer and an output layer; The data acquisition layer acquires multi-source data inside the mine through various instruments and equipment; the data collected by the data acquisition layer is input into the data processing layer; the data processing layer is used to pre-process the multi-source data to obtain standard multi-source data; the standard multi-source data is input into the safety monitoring layer to obtain multi-type safety monitoring results; the safety monitoring layer includes: a gas safety monitoring module, a fire safety monitoring module, a flood safety monitoring module and a landslide safety monitoring module; the multi-type safety monitoring results are input into the safety assessment layer to obtain multi-type safety production assessment values ​​of the mine environment; When any of the multiple types of safety production assessment values ​​does not meet the mine safety production standard, the mine operation is stopped; otherwise, the mine operation is continued; Conducting safety production monitoring on the staff and obtaining safety production assessment results of the staff; and handling the staff according to the safety production assessment results of the staff; Performing safety production monitoring on working equipment during mine operation to obtain safety production assessment results of the working equipment; and processing the working equipment according to the safety production assessment results of the working equipment.

2. According to claim 1, a method for enterprise safety production standardization assessment management is characterized in that: The multi-source data includes: gas concentration data, air pressure data, temperature data, humidity data, smoke concentration data, hydrological data, geological data, mine tunnel image data and mine structure data.

3. The enterprise safety production standardization assessment management method according to claim 1 is characterized in that: The gas safety monitoring module is used to monitor oxygen and harmful gases inside the mine. The specific gas safety monitoring process includes: obtaining standard gas data of each monitoring point from the standard multi-source data; comparing the standard gas data with the safety threshold of each gas, recording the gas safety error; and performing gas safety assessment according to the gas safety error; The fire safety monitoring module is used to monitor the occurrence of fire inside the mine. The specific fire safety monitoring process includes: obtaining standard temperature data from the standard multi-source data; judging whether the standard temperature data exceeds the fire temperature threshold; when exceeding, obtaining the standard mine tunnel image and standard carbon monoxide data of the corresponding position according to the acquisition position of the standard temperature data; performing image segmentation on the standard mine tunnel image to obtain a suspected fire source image; extracting the features of the suspected fire source image to obtain a suspected fire source feature map; wherein the suspected fire source feature map includes: color features, texture features and shape features; converting the suspected fire source feature map into numerical values ​​to obtain multiple suspected fire source values; and calculating the fire degree in combination with the standard carbon monoxide data and the multiple suspected fire source values. The flood safety monitoring module is used to monitor the occurrence of floods inside the mine. The specific flood safety monitoring process includes: obtaining real-time standard hydrological data of each water level monitoring point; extracting features from the standard hydrological data using feature engineering to obtain flood feature vectors; inputting the flood feature vectors into a neural network to obtain the probability of mine floods; The collapse safety monitoring module is used to monitor the collapse probability inside the mine. Specifically, the collapse safety monitoring module includes: obtaining standard hydrological data, standard geological data, standard mine tunnel image data and standard mine structure data through the data processing layer; extracting features from the standard hydrological data, the standard geological data, the standard mine tunnel image data and the standard mine structure data; respectively constructing a hydrological feature set, a geological feature set, a mine tunnel feature set and a structure feature set; analyzing the features in each feature set to obtain feature weights; weighting the original standard data with the feature weights to obtain weighted standard data; fusing the weighted standard data with each feature set to obtain collapse prediction fusion feature data; analyzing the collapse prediction fusion feature data using a collapse regression analysis equation to obtain the collapse probability; wherein, the collapse regression analysis equation represents: Wherein, P() represents the landslide regression analysis equation; X represents the set of landslide prediction fusion feature data; x i It is represented as the fusion feature data of the i-th landslide prediction; β0 is represented as the regression coefficient; β i It is represented as the contribution coefficient of the i-th landslide prediction fusion feature data; n is represented as the total number of landslide prediction fusion feature data.

4. The enterprise safety production standardization assessment management method according to claim 1 is characterized in that: The process of monitoring the safety of production of the staff includes: judging whether the staff is at a specified task point according to the positioning information; issuing a warning when the distance between the positioning point and the specified task point exceeds the maximum allowable distance; When the maximum allowed distance is not exceeded, obtain the surveillance video of the staff; process the surveillance video to obtain N surveillance images; perform clothing recognition and behavior recognition on the N surveillance images; wherein the clothing recognition process includes: segmenting the staff images in the N surveillance images; marking the staff images with clothing; wherein the content of the clothing mark includes: head, hands, upper body and lower body; identify according to the clothing mark, including: safety helmet, safety glasses, dust mask, safety gloves, work clothes and protective shoes; when any item of the clothing mark is not recognized, issue a safety warning; The behavior recognition includes: extracting various data of the staff during the monitoring time period, including: posture, motion trajectory and gait; evenly dividing the monitoring time into multiple time points; extracting the features of the data corresponding to the time points and constructing a time feature sequence; inputting the time feature sequence into a behavior safety recognition model to obtain a behavior safety recognition result corresponding to the time point; and issuing a safety warning when the behavior safety recognition result shows any violation of regulations.

5. The enterprise safety production standardization assessment management method according to claim 4 is characterized in that: The behavior safety identification model includes: an input layer, which is used to receive the time feature sequence and perform feature correlation processing on the time feature sequence; a local limb feature attention layer, which is used to focus on some features in the time feature sequence; a weighted layer, which is used to perform weighted operations on the features focused on by the local limb feature attention layer; and an output layer, which is used to output the behavior danger probability at each time point.

6. The enterprise safety production standardization assessment management method according to claim 1 is characterized in that: The process of monitoring the production safety of the working equipment includes: collecting the current working parameters of the working equipment; performing data preprocessing on the current working parameters to obtain the current equipment working characteristics; constructing a set of equipment working characteristics according to the working equipment type and the current equipment working characteristics; obtaining the health status indicator of the corresponding equipment according to the equipment type; analyzing the health status of the corresponding working equipment in the equipment working characteristic set according to the health status indicator; wherein the health status is analyzed using the equipment monitoring health function; the equipment monitoring health function is expressed as: Among them, EHV i represents the monitored health value of working device i; f() represents the device monitoring health function; HSI represents the health status index; Represented as the current equipment working characteristics of working equipment i; judging whether the working equipment is safe based on the monitored health value.

7. An enterprise safety production standardization assessment and management system, characterized in that: The system includes: a monitoring unit, a positioning unit, a data acquisition unit, a mine environment production safety assessment unit, a staff production safety assessment unit, a work equipment production safety assessment unit, a display unit and a feedback unit; wherein the monitoring unit is used to monitor the interior of the mine; the positioning unit is used to locate the staff and the work equipment; the data acquisition unit is used to adopt data related to safety assessment; the mine environment production safety assessment unit is used to monitor and assess the internal environment of the mine for production safety; wherein, it includes: a gas safety monitoring module, a fire safety monitoring module, a flood safety monitoring module and a landslide safety monitoring module; the staff production safety assessment unit is used to monitor and assess the staff for production safety; wherein, it includes: a work position safety monitoring module, a clothing safety monitoring module and a behavior safety monitoring module; the work equipment production safety assessment unit is used to assess the production safety of the work equipment; the display unit is used to display the scene inside the mine; the feedback unit is used to feedback various production safety assessment results to management personnel.

8. The enterprise safety production standardization evaluation and management system according to claim 7 is characterized in that: The staff safety production assessment unit includes: The work position safety monitoring module determines whether the worker is at the specified task point according to the positioning unit; when the distance between the positioning point and the specified task point exceeds the maximum allowable distance, the feedback unit provides feedback to the management personnel; When the maximum allowed distance is not exceeded, the surveillance video of the staff is obtained; the clothing safety monitoring module processes the surveillance video to obtain N surveillance images; clothing recognition and behavior recognition are performed on the N surveillance images; wherein the clothing recognition process includes: segmenting the staff images in the N surveillance images; clothing marking the staff images; wherein the content of the clothing marking includes: head, hands, upper body and lower body; recognition is performed according to the clothing marking, including: safety helmets, safety glasses, dust masks, safety gloves, work clothes and protective shoes; when any item of the clothing marking is not recognized, the feedback unit provides feedback to the management personnel; The behavioral safety monitoring module extracts various data of the staff during the monitoring time period, including: posture, motion trajectory and gait; divides the monitoring time evenly into multiple time points; extracts the features of the data corresponding to the time points and constructs a time feature sequence; inputs the time feature sequence into the behavioral safety recognition model to obtain the behavioral safety recognition results corresponding to the time points; and the feedback unit provides feedback to the management personnel based on the safety recognition results.

9. The enterprise safety production standardization evaluation and management system according to claim 8, characterized in that: The behavior safety identification model includes: an input layer, which is used to receive the time feature sequence and perform feature correlation processing on the time feature sequence; a local limb feature attention layer, which is used to focus on some features in the time feature sequence; a weighted layer, which is used to perform weighted operations on the features focused on by the local limb feature attention layer; and an output layer, which is used to output the behavior danger probability at each time point.

10. The enterprise safety production standardization evaluation and management system according to claim 7, characterized in that: The working equipment production safety assessment unit includes: collecting current working parameters of working equipment; performing data preprocessing on the current working parameters to obtain current equipment working characteristics; constructing a set of equipment working characteristics according to the type of working equipment combined with the current equipment working characteristics; obtaining a health status indicator of the corresponding equipment according to the equipment type; analyzing the health status of the corresponding working equipment in the equipment working characteristic set according to the health status indicator; wherein the health status is analyzed using an equipment monitoring health function; the equipment monitoring health function is expressed as: Among them, EHV i represents the monitored health value of working device i; f() represents the device monitoring health function; HSI represents the health status index; Represented as the current equipment working characteristics of working equipment i; judging whether the working equipment is safe based on the monitored health value.

Images