Open-cast mining operator safety monitoring method and system based on Internet of Things

Abstract

The invention relates to the technical field of safety monitoring, and discloses an open-cast mining operator safety monitoring method and system based on the Internet of Things, and the method comprises the following steps: S1, designing and integrating a five-point safety belt which comprises a chest belt, two shoulder belts and two leg belts, and integrating a sensor on the safety belt; s2, integrating various sensors, including a heart rate sensor, a gravity sensor (accelerometer), a position sensor, a gas detection sensor, a temperature sensor and an acoustic sensor; s3, transmitting sensor data to a central monitoring system in real time by using a wireless communication technology; and S4, carrying out data processing and analysis in the central monitoring system. By using the five-point type safety belt and various sensors, the environment and individual health conditions can be monitored in real time in an omnibearing manner, various risk factors are recognized and responded in time, the accident rate and potential health problems are remarkably reduced, and the working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety monitoring, and in particular to a method and system for safety monitoring of open-pit mining workers based on the Internet of Things. Background Art

[0002] Open-pit mining operations have long faced various safety risks and health issues due to their open and dynamic working environment. These risks include toxic gas leakage, extreme temperatures, equipment failures, and abnormal health of personnel, which directly threaten the life safety and health of operators. Therefore, how to effectively monitor these risk factors and take timely countermeasures is a major challenge in the current open-pit mining safety management.

[0003] In recent years, with the development of Internet of Things technology, the application of intelligent sensors and big data analysis technology in various industries has become increasingly widespread, providing new technical means for the upgrade of open-pit mining safety monitoring systems. By using gas detection sensors and position sensors to monitor the mining environment and operators in real time, and providing accurate risk warnings and emergency responses based on intelligent algorithms.

[0004] Although the existing mining safety monitoring system can realize real-time monitoring of the mining environment and workers, and provide accurate risk warnings and emergency responses based on intelligent algorithms, there are still some shortcomings. The existing mining safety monitoring system is difficult to achieve comprehensive coverage of the entire mining area, especially the dynamic environment and mobile workers cannot be monitored in real time. At the same time, in open-pit mining, accidents involving workers are usually related to the abnormal health of the workers themselves. The existing mining safety monitoring system has limited capabilities in data collection and processing, and cannot make full use of big data and intelligent algorithms for in-depth analysis and prediction, resulting in inadequate risk identification and prevention measures. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a method and system for safety monitoring of open-pit mining workers based on the Internet of Things, which solves the problem that the existing mining safety monitoring system cannot fully cover the entire mining area, especially the real-time monitoring of dynamic environments and mobile workers. In addition, the existing system is limited in data collection and processing, and it is difficult to use big data and intelligent algorithms for in-depth analysis and prediction, resulting in the inability to effectively identify and prevent risks.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method and system for monitoring safety of open-pit mining workers based on the Internet of Things, comprising the following steps:

[0007] S1. Design and integrate a five-point safety belt, the safety belt comprising a chest belt, two shoulder straps and two leg straps, and the sensor is integrated in the safety belt;

[0008] S2, integrates multiple sensors, including heart rate sensor, gravity sensor, position sensor, gas detection sensor, temperature sensor and acoustic sensor;

[0009] S3, using wireless communication technology to transmit sensor data to the central monitoring system in real time;

[0010] S4. Perform data processing and analysis in the central monitoring system, including data cleaning, standardization, anomaly detection, risk prediction, and dynamic route planning;

[0011] S5. Automatically trigger alarms and send evacuation instructions when environmental risks, individual health risks, physical and safety accidents, and location risk events occur.

[0012] Preferably, the heart rate sensor uses photoelectric volumetric pulse wave sensing technology to measure heart rate through changes in light intensity;

[0013] The formula is:

[0014] ΔI=I0-I(t), where ΔI is the change in light intensity, I0 is the initial light intensity, and I(t) is the light intensity during heart beat.

[0015] Preferably, the gravity sensor (accelerometer) adopts micro-electromechanical system technology, and the formula for calculating the acceleration vector amplitude is: where a x ,a y ,a z are the components of acceleration on the three spatial axes.

[0016] Preferably, the anomaly detection uses the isolation forest algorithm, and the calculation formula for the anomaly score is: Where E(h(x)) is the average path length of sample point x, and c(n) is the normalization constant;

[0017] The risk prediction adopts the Cox proportional risk model, and the formula of the risk function is h(t|x)=h0(t)exp(β1x1+β2x2+…+β n x n ), where h0(t) is the baseline hazard function, x i is the risk factor, β i is the corresponding coefficient;

[0018] Equipment failure prediction uses a random forest model by averaging the prediction results of multiple decision trees.

[0019] Preferably, the model parameters of the isolation forest algorithm include the number of trees n trees = 100 and the number of samples per tree n samples =256, the formula of the normalization constant c(n) is:

[0020]

[0021] Among them, H(i) is the i-th harmonic number, that is,

[0022] Preferably, the risk factor x of the Cox proportional hazard model is i Including gas concentration, temperature and heart rate, the baseline risk function h0(t) is obtained by maximum likelihood estimation, and the model parameter β i Calculated by the following formula:

[0023]

[0024] Where R(t i ) is at time t i The collection of individuals who were previously still at risk.

[0025] Preferably, the parameters of the random forest model include the number of decision trees n trecs = 100 and the maximum depth of each tree max_depth = None, the generation process of each tree includes the following steps:

[0026] Randomly sample with replacement from the training data;

[0027] At each node, a subset of features is randomly selected, and the split point is chosen to maximize the information gain;

[0028] The segmentation process is repeated recursively until the sample purity of all leaf nodes reaches a preset threshold or the maximum depth is reached.

[0029] Preferably, the dynamic route planning adopts Dijkstra algorithm, and the shortest path calculation formula is d(v)=min u∈S {d(u)+l(u,v)}, where S is the set of vertices with known shortest paths, and l(u,v) is the length of the edge from vertex u to vertex v.

[0030] Preferably, automatically triggering an alarm and sending an evacuation instruction when an environmental risk, an individual health risk, a physical and safety accident, or a location risk event occurs comprises the following steps:

[0031] When gas concentration, temperature, heart rate, acceleration or noise data exceeds preset thresholds, the central monitoring system automatically triggers an alarm;

[0032] The alarm signal activates the sound and light signal sirens in the mine area;

[0033] Broadcast emergency announcements over the public address system;

[0034] Send emergency notifications to all operators via personal communication devices;

[0035] Uses the Dijkstra algorithm to calculate the safest evacuation route and provides detailed evacuation information and visual maps through display screens and personal communication devices;

[0036] Update evacuation instructions and routes in real time based on new sensor data.

[0037] The present invention also provides an open-pit mining operator safety monitoring system based on the Internet of Things, comprising:

[0038] Five-point safety belt with integrated heart rate sensor and gravity sensor;

[0039] A variety of sensors, including position sensors, gas detection sensors, temperature sensors, and acoustic sensors;

[0040] Wireless communication module for transmitting sensor data to the central monitoring system in real time;

[0041] A central monitoring system including a data processing and analysis unit for data cleaning, standardization, anomaly detection, risk prediction, and dynamic route planning;

[0042] The alarm system and evacuation command system automatically triggers alarms and sends evacuation commands when environmental risks, individual health risks, physical and safety accidents, and location risk events occur.

[0043] The present invention provides a method and system for monitoring the safety of open-pit mining workers based on the Internet of Things. It has the following beneficial effects:

[0044] 1. The present invention can monitor the environment and individual health status in all directions and in real time by using a five-point safety belt and a variety of sensors. By timely identifying and responding to various risk factors, the accident rate and potential health problems can be significantly reduced, and work efficiency can be improved.

[0045] 2. By utilizing the isolation forest algorithm, the Cox proportional risk model and the random forest model, the present invention can quickly extract useful information from a large amount of data, conduct in-depth risk analysis and fault prediction, improve the effect of data analysis and processing, ensure rapid response to environmental hazards and individual health impacts, automatically trigger alarms and promptly notify operators, greatly improving the timeliness and accuracy of emergency response.

[0046] 3. The present invention adopts the Dijkstra algorithm for dynamic route planning, which not only calculates the shortest path from the current position to the safe exit, but also adjusts the evacuation route according to the real-time environmental data. In an emergency, it can provide the optimal evacuation route for the operating personnel, ensuring that the personnel can evacuate to a safe area safely and quickly.

[0047] 4. The present invention provides personalized safety information and evacuation instructions according to the specific location and health status of each worker, and the use of personal communication devices makes emergency notifications more accurate and timely. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flow chart of the method of the present invention;

[0049] Figure 2 It is a system structure diagram of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the specification 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.

[0051] Please see attached Figure 1 The embodiment of the present invention provides a method for monitoring the safety of open-pit mining workers based on the Internet of Things, comprising the following steps:

[0052] S1. Design and integrate a five-point safety belt. The alarm safety belt includes a chest belt, two shoulder straps and two leg straps. The sensor is integrated in the safety belt. The five-point safety belt provides comprehensive body protection to avoid additional injuries caused by improper seat belt fixation in an accident and provides a stable platform for sensor integration. The sensors are integrated in different positions of the safety belt to ensure the accuracy and effectiveness of data collection.

[0053] S2. Integrate multiple sensors, including heart rate sensor, gravity sensor (accelerometer), position sensor, gas detection sensor, temperature sensor and acoustic sensor. The heart rate sensor is installed on the chest strap, close to the heart, to monitor heart activity and provide instant feedback on abnormal heart rate. The gravity sensor (accelerometer) is used to measure the acceleration change caused by movement and can also sense the impact caused by gravity. It is used to monitor the movement state, posture changes, and potential falls and collisions of operators. The position sensor uses GPS positioning technology to ensure that the location of each operator can be accurately tracked. The gas detection sensor is used to monitor the concentration of toxic gases such as carbon monoxide and methane to prevent health problems caused by toxic gases. The temperature sensor is used to monitor the ambient temperature and body surface temperature to prevent health problems caused by extreme temperatures. The acoustic sensor is used to detect environmental noise, especially mechanical failures and explosions. Through all-round monitoring, it ensures that all aspects from physiology, environment to position are supported by real-time data. When any abnormal situation is detected, the operator can respond quickly, thereby reducing the probability of accidents and reducing potential injuries.

[0054] S3. Use wireless communication technology to transmit sensor data to the central monitoring system in real time. Use LoRa and Wi-Fi wireless technologies to ensure stable data transmission even underground or in remote areas. Wireless technology enables data to be transmitted to the monitoring center in real time, allowing instant data analysis and decision making.

[0055] S4. Data processing and analysis in the central monitoring system, including data cleaning, standardization, anomaly detection, risk prediction and dynamic route planning. The received data is formatted and cleaned to remove noise or invalid data. The provided data can be analyzed to identify abnormal behavior or potential dangers. The risk level of operators is analyzed through risk prediction, and accidents such as chemical exposure and heat-related diseases are predicted. Finally, the safe evacuation route is calculated and updated through dynamic route planning. Automated data processing reduces human errors, improves processing speed and accuracy, and improves the safety of operations and reduces accidents.

[0056] S5. Automatically trigger alarms and send evacuation instructions when environmental risks, individual health risks, physical and safety accidents, and location risk events occur. When the system detects environmental risks, individual health risks, physical and safety accidents, and location risk events, it sends specific alarm information and emergency instructions to each operator through the wireless network. At the same time, it uses a fixed or mobile broadcasting system to play pre-recorded or real-time emergency evacuation instructions and safety information. The automated alarm and evacuation system reduces human delays, improves the response speed in emergencies, ensures that operators are quickly and effectively guided to evacuate safely in dangerous situations, and maximizes the protection of personnel safety.

[0057] Please see attached Figure 1 ,In a preferred embodiment of the present invention, the alarm heart rate sensor uses photoplethysmography sensing technology to measure heart rate through changes in light intensity;

[0058] The formula is:

[0059] ΔI=I0-I(t), where ΔI is the change in light intensity, I0 is the initial light intensity, and I(t) is the light intensity when the heart beats. By placing the heart rate sensor close to the heart, the blood flow changes caused by the heart beat can be captured more sensitively. The sensor includes an LED light source and a photodetector. The light emitted by the LED penetrates the skin and is absorbed by the blood. The unabsorbed light is received by the photodetector. When the heart pumps blood, the amount of blood in the blood vessels increases, absorbing more light, and the light intensity detected by the photodetector decreases accordingly. By continuously monitoring this process, heart rate changes can be accurately tracked, and real-time, continuous heart rate data can be provided, which helps monitor the health of workers in high-intensity labor and extreme environments. Abnormal heart rates can be detected and warned in time, allowing workers to suspend work and undergo medical examinations, etc., thereby preventing possible health risks.

[0060] Please see attached Figure 1 In a preferred embodiment of the present invention, the alarm gravity sensor (accelerometer) adopts micro-electromechanical system technology, and the formula for calculating the acceleration vector amplitude is: where a x ,a y ,a z Gravity sensors are installed on the chest strap and shoulder strap of the five-point safety belt to maximize the capture of the dynamic movement of the body. The gravity sensor uses the physical displacement of the internal microstructure to measure acceleration. Movement or vibration in any direction will cause the relative displacement of the internal mass block, which in turn affects the capacitance value. The change is converted into an acceleration reading through the circuit. By monitoring the operator's body movement in real time, the system can instantly detect falls and rapid movements, and can initiate emergency response procedures to provide timely rescue.

[0061] Please see attached Figure 1 In a preferred embodiment of the present invention, the above anomaly detection uses the isolation forest algorithm, which isolates sample points by constructing multiple trees until each data point is isolated on a leaf node of the tree. The calculation formula for the anomaly score is: Where E(h(x)) is the average path length of sample point x, c(n) is the normalization constant, and the isolation forest algorithm does not need to calculate the distance between data points, so it can quickly identify abnormal points and detect abnormal changes in data in a timely manner, such as sudden health problems and dangerous behaviors, so that measures can be taken quickly to improve safety;

[0062] The risk prediction adopts the Cox proportional risk model, and the formula of the risk function is h(t|x)=h0(t)exp(β1x1+β2x2+…+β n x n ), where h0(t) is the baseline hazard function, x i is the risk factor, β i As the corresponding coefficient, the Cox model is fitted using historical data to determine the coefficient β of each risk factor i The Cox model considers the combined impact of multiple risk factors, provides personalized risk assessment for each operator, and can update model parameters as new data is acquired to ensure the accuracy and real-time nature of risk assessment, helping decision makers understand the current scenario and formulate response strategies in advance;

[0063] Equipment failure prediction uses a random forest model. By averaging the prediction results of multiple decision trees, random forests reduce the risk of overfitting by constructing multiple decision trees, improve the accuracy and robustness of predictions, and predict potential problems before failures occur, reducing unexpected downtime and repair costs. By predicting maintenance needs, it helps develop more effective equipment maintenance plans, thereby extending the service life of equipment.

[0064] Please see attached Figure 1 In a preferred embodiment of the present invention, the model parameters of the alarm isolation forest algorithm include the number of trees n trees = 100 and the number of samples per tree n samples =256, the formula for the alarm normalization constant c(n) is:

[0065]

[0066] Among them, H(i) is the i-th harmonic number, that is, Isolation forest assumes that outliers are more easily isolated due to different attribute values. Therefore, outliers have shorter path lengths. The normalization constant c(n) is used to adjust the path length so that the results of sample sets of different sizes are comparable. It adjusts the path length based on the sample size n in order to more accurately evaluate the degree of abnormality of data points independent of the influence of the sample size.

[0067] Please see attached Figure 1 In a preferred embodiment of the present invention, the risk factor x of the alarm Cox proportional hazard model i Including gas concentration, temperature and heart rate, the baseline risk function h0(t) is obtained by maximum likelihood estimation, and the model parameter β i Calculated by the following formula:

[0068]

[0069] Where R(t i ) is at time t i For the set of individuals who were still at risk, the proportional effect of risk is modeled based on the linear combination of risk factors. The model assumes that the instantaneous risk of an individual is the product of the baseline risk function h0(t) and the exponential function of the risk factor. Gas concentration, temperature and heart rate are selected as risk factors because they are the most direct factors affecting the health status of miners. By accurately calculating the risk level, the safety management team can design more accurate preventive measures targeting specific risk factors, such as strengthening ventilation to reduce the concentration of harmful gases, and adjusting work plans to avoid high temperature periods. The Cox model can update risk assessments in real time and provide immediate safety warnings so that measures can be taken before dangerous situations actually occur.

[0070] Please see attached Figure 1 In a preferred embodiment of the present invention, the parameters of the alarm random forest model include the number of decision trees n trecs = 100 and the maximum depth of each tree max_depth = None. The generation process of each tree includes the following steps: input the collected data into the data processing system, perform necessary cleaning and formatting for subsequent risk analysis, and select gas concentration, temperature and heart rate as risk factors of the model:

[0071] Randomly extract samples with replacement from the training data, and randomly extract samples with replacement from the entire training data set to form multiple training subsets. Each subset is used to build a decision tree. In the process of building each decision tree, a feature subset is randomly selected for each node, and the best split point is selected according to the information gain criterion. Information gain is the degree to which the feature improves the classification effect of the sample set;

[0072] At each node, a subset of features is randomly selected, and the split point is chosen to maximize the information gain;

[0073] The segmentation process is repeated recursively until the sample purity of all leaf nodes reaches the preset threshold or the maximum depth is reached. Random forest is an integrated learning technology that improves the accuracy and stability of the overall prediction by constructing multiple decision trees and aggregating their prediction results. At the same time, the construction of each tree is based on randomly selected features and samples. This randomness helps reduce the risk of overfitting of the model and enhances the model's generalization ability to new data. By integrating multiple decision trees, random forest can effectively reduce the overfitting that may occur in the model on a single decision tree and improve the accuracy of the prediction. Even if some data are noisy, the integrated characteristics of random forest make the overall output of the model more stable and have strong resistance to outliers and noise. Random forest can provide an assessment of the importance of each feature in the prediction process, helping researchers and engineers understand which factors most affect the health of the equipment, so as to carry out targeted maintenance and optimization.

[0074] Please see attached Figure 1 In a preferred embodiment of the present invention, the alarm dynamic route planning adopts the Dijkstra algorithm, and the shortest path calculation formula is d(v)=min u∈S {d(u)+l(u,v)}, where S is the set of vertices with known shortest paths, l(u,v) is the length of the edge from vertex u to vertex v. Dijkstra's algorithm is a greedy algorithm that searches for unprocessed minimum path vertices at each step and then tries to improve the path to other vertices through it. The algorithm gradually reduces the shortest path estimate to each vertex through relaxation techniques until these estimates show the true shortest path length, thereby being able to efficiently process large graph data and is suitable for complex mining terrain and changing environmental conditions, ensuring real-time route planning in dynamic environments. Accurately calculated shortest paths can prevent the use of longer routes in dangerous situations, thereby reducing the risk of accidents and injuries.

[0075] Please see attached Figure 1 In a preferred embodiment of the present invention, automatically triggering an alarm and sending an evacuation instruction when an environmental risk, an individual health risk, a physical and safety accident, or a location risk event occurs includes the following steps:

[0076] When the gas concentration, temperature, heart rate, acceleration or noise data exceeds the preset threshold, the central monitoring system automatically triggers an alarm. By setting the gas concentration warning threshold to 25ppm, the alarm threshold to 35ppm, the temperature alarm threshold to -5℃~35℃, the body surface temperature alarm threshold to 50bpm~120bpm, the gravity (acceleration) alarm threshold to 3g, and the noise alarm threshold to 85dB, the safety monitoring system can effectively warn of potential dangers in advance and ensure the safety of open-pit mining operations and the health of operators.

[0077] The alarm signal activates the sound and light signal alarms in the mine area, and activates the sound and light signal alarms in the mine area through the system to warn all personnel on site;

[0078] Broadcast emergency announcements over the public address system. Broadcast pre-recorded emergency announcements over the mine site’s public address system to ensure everyone is warned and notified of necessary evacuations;

[0079] Emergency notifications are sent to all operators via personal communication devices and to personnel's personal communication devices via radio, providing personalized safety information;

[0080] Use the Dijkstra algorithm to calculate the safest evacuation route and provide detailed evacuation information and visual maps through display screens and personal communication devices, and use the Dijkstra algorithm to calculate the shortest path from the current location to the nearest safe exit;

[0081] Update evacuation instructions and routes in real time based on new sensor data, display detailed evacuation route maps through display screens and personal communication devices in the mine area, and finally continuously monitor environmental changes and new data inputs, update evacuation routes and instructions in real time, and adapt to avoid new risks and obstacles.

[0082] The open-pit mining worker safety monitoring system based on the Internet of Things described below and the open-pit mining worker safety monitoring method based on the Internet of Things described above can be referred to in correspondence with each other.

[0083] Please see attached Figure 2 The present invention also provides an open-pit mining operator safety monitoring system based on the Internet of Things, including:

[0084] Five-point safety belt with integrated heart rate sensor and gravity sensor (accelerometer);

[0085] A variety of sensors, including position sensors, gas detection sensors, temperature sensors, and acoustic sensors;

[0086] Wireless communication module for transmitting sensor data to the central monitoring system in real time;

[0087] Central monitoring system, including data processing and analysis unit, alarm data processing and analysis unit for data cleaning, standardization, anomaly detection, risk prediction and dynamic route planning;

[0088] The alarm system and evacuation command system automatically triggers alarms and sends evacuation commands when environmental risks, individual health risks, physical and safety accidents, and location risk events occur.

[0089] The system of this embodiment can be used to execute the above method embodiment, and its principle and technical effect are similar, which will not be repeated here.

[0090] 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. Safety monitoring method for open-pit mining workers based on the Internet of Things, It is characterized in that The following steps are involved: S1. Design and integrate a five-point safety belt, the safety belt comprising a chest belt, two shoulder straps and two leg straps, and the sensor is integrated in the safety belt; S2, integrates multiple sensors, including heart rate sensor, gravity sensor, position sensor, gas detection sensor, temperature sensor and acoustic sensor; S3, using wireless communication technology to transmit sensor data to the central monitoring system in real time; S4. Perform data processing and analysis in the central monitoring system, including data cleaning, standardization, anomaly detection, risk prediction, and dynamic route planning; S5. Automatically trigger alarms and send evacuation instructions when environmental risks, individual health risks, physical and safety accidents, and location risk events occur.

2. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 1, It is characterized in that The heart rate sensor uses photoelectric capacitance pulse wave sensing technology to measure heart rate through changes in light intensity; The formula is: ΔI=I 0 -I(t), where ΔI is the change in light intensity, I 0 is the initial light intensity, and I(t) is the light intensity during heart beat.

3. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 1, It is characterized in that The gravity sensor (accelerometer) adopts micro-electromechanical system technology, and the formula for calculating the acceleration vector amplitude is: where a x ,a y ,a z are the components of acceleration on the three spatial axes.

4. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 1, It is characterized in that The anomaly detection uses the isolation forest algorithm, and the calculation formula for the anomaly score is: Where E(h(x)) is the average path length of sample point x, and c(n) is the normalization constant; The risk prediction adopts the Cox proportional risk model, and the formula of the risk function is h(t|x)=h 0 (t)exp(β 1 x 1 +β 2 x 2 +…+β n x n ), where h 0 (t) is the baseline risk function, x i is the risk factor, β i is the corresponding coefficient; Equipment failure prediction uses a random forest model by averaging the prediction results of multiple decision trees.

5. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 4, It is characterized in that The model parameters of the isolation forest algorithm include the number of trees n trees = 100 and the number of samples per tree n samples =256, the formula of the normalization constant c(n) is: Among them, H(i) is the i-th harmonic number, that is, 6. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 1, It is characterized in that The risk factor x of the Cox proportional hazards model i Including gas concentration, temperature and heart rate, the baseline risk function h 0 (t) is obtained by maximum likelihood estimation, and the model parameter β i Calculated by the following formula: Where R(t i ) is at time t i The collection of individuals who were previously still at risk.

7. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 4, It is characterized in that The parameters of the random forest model include the number of decision trees n trecs = 100 and the maximum depth of each tree max_depth = None, the generation process of each tree includes the following steps: Randomly sample with replacement from the training data; At each node, a subset of features is randomly selected, and the split point is chosen to maximize the information gain; The segmentation process is repeated recursively until the sample purity of all leaf nodes reaches a preset threshold or the maximum depth is reached.

8. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 1, It is characterized in that The dynamic route planning adopts Dijkstra algorithm, and the shortest path calculation formula is d(v)=min u∈S {d(u)+l(u,v)}, where S is the set of vertices with known shortest paths, and l(u,v) is the length of the edge from vertex u to vertex v.

9. The method for safety monitoring of open-pit mining workers based on the Internet of Things according to claim 1, It is characterized in that Automatically triggering alarms and sending evacuation instructions when environmental risks, personal health risks, physical and safety accidents, and location risk events occur includes the following steps: When gas concentration, temperature, heart rate, acceleration or noise data exceeds preset thresholds, the central monitoring system automatically triggers an alarm; The alarm signal activates the sound and light signal sirens in the mine area; Broadcast emergency announcements over the public address system; Send emergency notifications to all operators via personal communication devices; Uses the Dijkstra algorithm to calculate the safest evacuation route and provides detailed evacuation information and visual maps through display screens and personal communication devices; Update evacuation instructions and routes in real time based on new sensor data.

10. An open-pit mining worker safety monitoring system based on the Internet of Things, used to implement the method according to any one of claims 1 to 9, It is characterized in that include: Five-point safety belt with integrated heart rate sensor and gravity sensor; A variety of sensors, including position sensors, gas detection sensors, temperature sensors, and acoustic sensors; Wireless communication module for transmitting sensor data to the central monitoring system in real time; A central monitoring system including a data processing and analysis unit for data cleaning, standardization, anomaly detection, risk prediction, and dynamic route planning; The alarm system and evacuation command system automatically triggers alarms and sends evacuation commands when environmental risks, individual health risks, physical and safety accidents, and location risk events occur.

Images