Miner lung health monitoring system

Abstract

The invention relates to the technical field of health monitoring, and discloses a miner lung health monitoring system, comprising: a data acquisition module for acquiring miner health state data and mine environment data in real time, the mine environment data including dust concentration, toxic gas concentration and airflow velocity; the health state modeling module is used for constructing a random model for describing the dynamic change of the health state along with time based on the miner health state data and the mine environment data; and the environment dynamic modeling module is used for constructing a model for describing the dust concentration and the toxic gas concentration in the mine through the environment data. Through the technical scheme of combining a closed-loop feedback mechanism and real-time health risk assessment, the technical effect of monitoring and optimizing the miner health state and environment dynamic change in real time is achieved, and compared with the scheme of conducting intervention only depending on static analysis in the prior art, the defect that the intervention effect is poor due to dynamic adjustment hysteresis is overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of health monitoring, and in particular to a miner-based lung health monitoring system. Background Art

[0002] At present, the health monitoring and working environment management of miners are gradually receiving attention, especially the prevention and control of lung health risks. However, most existing technologies adopt a combination of traditional environmental monitoring equipment and periodic physical examinations. These technical solutions are usually based on environmental indicators such as dust concentration and toxic gas concentration. Environmental data is collected through fixed-point monitoring equipment and analyzed in combination with miners' health examination records. Some technologies have attempted to introduce health status modeling methods or automate ventilation system control through preset rules. These measures have a certain positive effect on reducing health risks and improving the working environment, and have played an important role in mine safety management.

[0003] However, when faced with the dynamic and changing conditions in the mine working environment, the existing technology still has some difficult-to-overcome problems. For example, the health monitoring method of miners mainly relies on regular physical examinations or static model analysis, which makes it difficult to capture the real-time changes in the health status of miners; environmental monitoring is usually limited to the monitoring of a single variable, and fails to effectively combine the health status for comprehensive evaluation. In addition, traditional intervention measures such as ventilation adjustment methods with fixed parameters lack the ability to respond to real-time risk changes, resulting in a lag or low efficiency in the effect of health intervention. In practical applications, these problems may lead to the failure to identify and control miners' health risks in a timely manner, thereby affecting work safety and the prevention and control of occupational diseases. Summary of the invention

[0004] In view of the shortcomings of the existing technology, the present invention provides a miner lung health monitoring system, which solves the problem of untimely identification of miners' health risks and lack of accuracy of intervention strategies due to the lack of real-time and dynamic health status monitoring and environmental risk assessment mechanism in mine operations.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A miner's lung health monitoring system, comprising: The data acquisition module is used to collect the miners' health status data and mine environment data in real time. The health status data includes blood oxygen saturation, respiratory rate and heart rate, and the mine environment data includes dust concentration, toxic gas concentration and airflow velocity. The health status modeling module is used to build a stochastic model to describe the dynamic changes of health status over time based on the miners' health status data and mine environment data; Environmental dynamic modeling module, which is used to build a model to describe the spatial distribution and dynamic changes of dust concentration and toxic gas concentration in the mine through environmental data; The health risk assessment module is used to calculate the miners' lung health risk value and perform risk classification based on the outputs of the health status modeling module and the environmental dynamic modeling module; The health intervention optimization module is used to generate the optimal health intervention strategy based on the output of the health risk assessment module; The closed-loop feedback module is used to implement health intervention strategies and collect and update data in real time to correct the health status modeling module and the environmental dynamics modeling module.

[0006] Preferably, the health status modeling module uses stochastic differential equations to establish a health status model, wherein changes in the health status include deterministic changes and stochastic changes in health data. The deterministic changes are determined by the relationship between the miner's blood oxygen saturation, respiratory rate and heart rate, as well as the dust concentration and toxic gas concentration. The stochastic changes are described by Brownian motion terms.

[0007] Preferably, the environmental dynamic modeling module uses partial differential equations to establish a dynamic model of mine dust concentration and toxic gas concentration. The model considers the superposition of diffusion effect, airflow convection effect and pollutant source term, and performs numerical solution through spatial discretization and time discretization.

[0008] Preferably, the health risk assessment module calculates the miner's lung health risk value based on the following risk function: The risk function consists of three parts: the risk of health status deviating from the normal reference value, the risk of dust concentration, and the risk of toxic gas concentration; The risk of health status deviating from the normal reference value is obtained by weighting the square of the deviation between the health status variable and the normal reference value. The risk of dust concentration and the risk of toxic gas concentration are obtained by weighting the square of dust concentration and the square of toxic gas concentration, respectively.

[0009] Preferably, the health risk assessment module performs risk classification according to the value of the risk function, specifically including: When the risk value is less than the first threshold, it is determined that the health status is safe; When the risk value is between the first threshold and the second threshold, it is determined as a health status warning; When the risk value is greater than the second threshold, it is determined that the health status is dangerous.

[0010] Preferably, the health intervention optimization module uses a Lagrangian optimization algorithm to generate an optimal intervention strategy, wherein: Health intervention strategies include adjusting mine ventilation intensity, miners’ job rotation frequency, and work-rest cycles; The goal of the Lagrangian optimization algorithm is to minimize the cumulative value of health risk value within a certain time range. The optimization process is subject to the constraints of the health status model and the environmental dynamic model.

[0011] Preferably, the health intervention optimization module generates the optimal intervention strategy through the following steps: Construct a Lagrangian function consisting of a health risk function, a health status model constraint, and an environmental dynamic model constraint; Solve the extreme value of Lagrangian function based on variational method; According to the solution results, the optimal ventilation intensity, job rotation time and rest cycle are obtained.

[0012] Preferably, the data acquisition module includes: Wearable health monitoring devices are used to collect miners' blood oxygen saturation, respiratory rate and heart rate in real time, and upload the data to the central server via wireless network; Environmental monitoring equipment, including distributed dust concentration sensors, toxic gas concentration sensors and air flow velocity monitoring equipment, is used to collect mine environmental data in real time.

[0013] Preferably, the closed-loop feedback module includes: Real-time intervention feedback module, used to record the changes in miners' health status data and mine environment data after the implementation of the intervention strategy; The model updating module is used to dynamically update the parameters of the health status model and the environmental dynamic model according to the intervention feedback results.

[0014] Preferably, the health data and environmental data collected in real time by the data acquisition module are pre-processed, model calculation and intervention optimization are performed through a cloud server, and the intervention strategy is sent to the mine control center or the miner terminal through wireless communication technology.

[0015] The present invention provides a miner's lung health monitoring system. It has the following beneficial effects: 1. The present invention achieves the technical effect of real-time monitoring and optimization of miners' health status and dynamic changes in the environment through a technical solution that combines a closed-loop feedback mechanism with real-time health risk assessment. Compared with the existing solution that only relies on static analysis for intervention, it solves the problem of poor intervention effect caused by dynamic adjustment lag.

[0016] 2. The present invention introduces a multi-level modeling method based on stochastic differential equations and partial differential equations, which achieves the technical effect of accurately quantifying the interactive impact of health status and environmental factors. Compared with the single modeling method in the prior art that ignores complex environmental variables, it solves the problem of insufficient accuracy in risk assessment and further improves the adaptability of the model to the dynamic conditions of the mining area.

[0017] 3. The present invention realizes the real-time generation of personalized health intervention strategies by optimizing the dynamic adjustment function of the module and combining the feedback mechanism, achieving the technical effect of significantly reducing health risks. Compared with the traditional strategy generation method based on fixed parameters, it solves the shortcomings of the lack of flexibility of intervention measures and difficulty in adapting to the changing mining environment.

[0018] 4. The present invention seamlessly connects the closed-loop feedback module with the health intervention optimization module, and significantly improves the response speed and accuracy of the intervention strategy through error correction and self-learning mechanisms. Compared with the existing solutions in which the intervention effect relies on manual judgment, it solves the problems of insufficient automation and inability to dynamically correct. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the modules of the present invention. DETAILED DESCRIPTION

[0020] 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.

[0021] Please refer to the attached Figure 1 The embodiment of the present invention provides a miner's lung health monitoring system, comprising: The data acquisition module is used to collect the miners' health status data and mine environment data in real time. The health status data includes blood oxygen saturation, respiratory rate and heart rate, and the mine environment data includes dust concentration, toxic gas concentration and airflow velocity. The health status modeling module is used to build a stochastic model to describe the dynamic changes of health status over time based on the miners' health status data and mine environment data; Environmental dynamic modeling module, which is used to build a model to describe the spatial distribution and dynamic changes of dust concentration and toxic gas concentration in the mine through environmental data; The health risk assessment module is used to calculate the miners' lung health risk value and perform risk classification based on the outputs of the health status modeling module and the environmental dynamic modeling module; The health intervention optimization module is used to generate the optimal health intervention strategy based on the output of the health risk assessment module; A closed-loop feedback module is used to implement health intervention strategies and collect and update data in real time to calibrate the health status modeling module and the environmental dynamics modeling module; The health status modeling module uses stochastic differential equations to establish a health status model, in which the changes in health status include deterministic changes and random changes in health data. The deterministic changes are determined by the relationship between the miners' blood oxygen saturation, respiratory rate and heart rate, as well as the dust concentration and toxic gas concentration. The random changes are described by Brownian motion terms. The environmental dynamic modeling module uses partial differential equations to establish a dynamic model of mine dust concentration and toxic gas concentration. The model considers the superposition of diffusion effect, airflow convection effect and pollutant source term, and numerically solves it through spatial discretization and time discretization. The health risk assessment module calculates the miners' lung health risk value based on the following risk function: The risk function consists of three parts: the risk of health status deviating from the normal reference value, the risk of dust concentration, and the risk of toxic gas concentration; The risk of health status deviating from the normal reference value is obtained by weighting the square of the deviation between the health status variable and the normal reference value. The risk of dust concentration and the risk of toxic gas concentration are obtained by weighting the square of dust concentration and the square of toxic gas concentration, respectively. The health risk assessment module classifies risks according to the value of the risk function, including: When the risk value is less than the first threshold, it is determined that the health status is safe; When the risk value is between the first threshold and the second threshold, it is determined as a health status warning; When the risk value is greater than the second threshold, it is determined that the health status is dangerous; The health intervention optimization module uses the Lagrangian optimization algorithm to generate the optimal intervention strategy, where: Health intervention strategies include adjusting mine ventilation intensity, miners’ job rotation frequency, and work-rest cycles; The goal of the Lagrangian optimization algorithm is to minimize the cumulative value of health risk value within a certain time range. The optimization process is subject to the constraints of the health state model and the environmental dynamic model. The health intervention optimization module generates the optimal intervention strategy through the following steps: Construct a Lagrangian function consisting of a health risk function, a health status model constraint, and an environmental dynamic model constraint; Solve the extreme value of Lagrangian function based on variational method; According to the solution results, the optimal ventilation intensity, job rotation time and rest cycle are obtained; The data acquisition module includes: Wearable health monitoring devices are used to collect miners' blood oxygen saturation, respiratory rate and heart rate in real time, and upload the data to the central server via wireless network; Environmental monitoring equipment, including distributed dust concentration sensors, toxic gas concentration sensors and air flow velocity monitoring equipment, for real-time collection of mine environmental data; The closed-loop feedback module includes: Real-time intervention feedback module, used to record the changes in miners' health status data and mine environment data after the implementation of the intervention strategy; A model updating module is used to dynamically update the parameters of the health status model and the environmental dynamic model according to the intervention feedback results; The data acquisition module pre-processes the health data and environmental data collected in real time, performs model calculation and intervention optimization through the cloud server, and sends the intervention strategy to the mine control center or miner terminal through wireless communication technology.

[0022] Specifically, the data acquisition module is used to achieve real-time monitoring and efficient collection of miners' health data and mine environment data. This module plays a role in basic data support in the entire system, and is closely connected with the health status modeling and environmental dynamic analysis modules. Specifically, this module is responsible for providing data input for the variables used by stochastic differential equations (SDE) and partial differential equations (PDE), and works together through various types of sensors to ensure the integrity and accuracy of the data. The design of the data acquisition module fully considers the complexity of the mine environment, including high dust, high humidity, and possible interference factors in data transmission, aiming to provide reliable support for subsequent health risk assessment and intervention optimization.

[0023] In this embodiment, the data acquisition module includes two main parts: a wearable health monitoring device and an environmental monitoring sensor network. The functions and implementation methods of each are described in detail below.

[0024] Wearable health monitoring devices are mainly used to collect miners' physiological parameter data in real time, including but not limited to blood oxygen saturation, respiratory rate and heart rate. Generally, these data are directly reflected in the miners' lung health status and input into the stochastic differential equation (SDE) model of health status.

[0025] In some embodiments, blood oxygen saturation The measurement is achieved through an optical blood oxygen sensor, which uses the difference in the absorption of red light and infrared light in tissue to calculate the blood oxygen value. Its mathematical expression is:

[0026] in, represents the concentration of oxygenated hemoglobin, represents the concentration of deoxyhemoglobin. This data is input into the health state vector , , .

[0027] Respiratory rate The measurement uses acceleration sensor technology based on chest expansion / contraction detection. Specifically, the sensor records the number of chest expansions per unit time to obtain the respiratory frequency time series.

[0028] Heart rate The measurement is done using photoplethysmography (PPG) technology. As an option, an optical sensor emits green light and detects fluctuations in light absorption caused by changes in blood flow in the vessels, thereby inferring heart rate.

[0029] In some embodiments, the frequency of health data collection is set to 5 times per second (i.e., the sampling frequency is 5Hz) to capture subtle fluctuations in the miners' physiological state. During the collection process, in order to reduce the impact of mine dust and high humidity environment on the sensors, all wearable device shells are dust-proof and waterproof, and anti-static treatment is performed.

[0030] Environmental monitoring sensor network is used to collect dust concentration in the mining area , toxic gas concentration and air velocity These environmental data are sent to the data analysis platform in real time through the wireless transmission module, providing key input for the environmental diffusion model (PDE).

[0031] Generally speaking, the dust concentration The measurement uses light scattering particle sensor technology. Specifically, the sensor detects the intensity of light scattered by dust particles by emitting a light beam and calculates the concentration of dust particles per unit volume. The formula can be expressed as:

[0032] in, represents the scattered light intensity of a single particle, is the sampling volume.

[0033] Toxic gas concentration The measurement uses an electrochemical sensor. As a possible implementation, the sensor selectively responds to a specific gas (such as carbon monoxide or hydrogen sulfide) and outputs an electrical signal proportional to the gas concentration. The measurement formula is:

[0034] in, is the calibration factor, is the output voltage of the sensor.

[0035] In environmental parameter collection, air flow velocity The measurement relies on ultrasonic anemometers, which use the changes in the speed of sound waves in the airflow to calculate wind speed and direction. The layout of ultrasonic sensors follows the key points of the mine ventilation path to ensure the comprehensiveness of wind speed measurement.

[0036] To improve the accuracy and robustness of the data, the data acquisition module also includes edge computing functions. The collected health and environmental data are first pre-processed in the local device, including denoising, filtering, and missing data completion. Specifically, the Kalman filter is used for noise processing, and its state update formula is:

[0037] in: : The filtered state vector.

[0038] : Predicted state vector.

[0039] : Kalman gain.

[0040] : Observed value.

[0041] : Observation matrix The denoised data is transmitted to the central data processing platform via LoRa wireless communication or 5G network, and is further used for health status modeling and risk assessment.

[0042] In some embodiments, in order to meet the special needs of different mining areas, the data acquisition module can also be expanded as follows: Add temperature and humidity sensors to monitor climate conditions in the mining area.

[0043] Integrated positioning module, through GPS or mine-specific positioning technology (such as RFID) to determine the location of miners, to facilitate the matching of regional environmental monitoring data.

[0044] Adopt multi-sensor redundant design. For example, single-point dust concentration measurement can be collected by two light scattering sensors in parallel, and the data averaging method is used to improve accuracy.

[0045] The health status modeling module dynamically analyzes and predicts the miners' lung health status in real time through mathematical models. This module combines the health parameter data and environmental variable data provided by the data acquisition module and uses stochastic differential equations (SDE) to describe the dynamic evolution of health status. This module not only provides basic support for health risk assessment, but also provides accurate input for subsequent intervention optimization modules. In order to adapt to the complex working environment and data fluctuations in the mining area, this module specially designs a technical solution combining dynamic modeling and uncertainty analysis to achieve efficient description of health status.

[0046] In this embodiment, the health status modeling module uses a stochastic differential equation (SDE) for modeling to establish a dynamic change model of the health status over time. Specifically, the health status vector includes key parameters such as the blood oxygen saturation of the miner, respiratory rate, heart rate, etc.

[0047] Generally, the health status of the miner is affected by time changes and is simultaneously affected by environmental variables, such as dust concentration and toxic gas concentration . The present invention uses the following stochastic differential equation to describe the dynamic evolution of the health status:

[0048] In a possible implementation, the deterministic function represents the main change trend of the health status, and its definition is:

[0049] Where: : The autocorrelation matrix of the health status variables.

[0050] , : The influence coefficients of dust concentration and toxic gas concentration on the health status.

[0051] As an option, the stochastic perturbation function is used to quantify the impact of uncertainty on the health status. Its definition is:

[0052] Where: : The stochastic noise intensity matrix.

[0053] The stochastic perturbation part represents a dimensional Brownian motion process, which is used to simulate the uncertainty impact in the mining area environment In a specific implementation, the numerical solution of the SDE uses the Euler-Maruyama method. This method discretizes time and converts the continuous-time SDE into a discrete format, expressed as follows:

[0054] Where: : The health status at the th moment.

[0055] , : Dust concentration and toxic gas concentration at the corresponding moment.

[0056] : time step.

[0057] : Brownian motion increment, obeys normal distribution.

[0058] Specifically, the health status modeling module needs to obtain the time series data of blood oxygen saturation, respiratory rate and heart rate, as well as the dynamic change values ​​of dust and gas concentration from the data acquisition module. All data are standardized and input into the modeling formula.

[0059] In some embodiments, the module also supports dynamic adjustment of modeling parameters , , The value of is adjusted to adapt to the environmental conditions of different mining areas and the individual differences of miners.

[0060] The health status vector output by this module It can be directly used in the health risk assessment module. Generally, risk assessment is calculated by the degree of deviation between the health status and the reference value. For example, the risk function can be defined as:

[0061] in: : Reference normal value for health status.

[0062] : Weight parameter, used to adjust the risk contribution.

[0063] Through the above modeling, this module provides reliable health status input for the subsequent intervention optimization module.

[0064] The environmental dynamic modeling module aims to model and predict the spatiotemporal changes of dust concentration and toxic gas concentration in the mining area. This module works closely with the data acquisition module to construct a partial differential equation (PDE) model that describes the dynamic changes of mining environmental parameters using the collected environmental data, such as dust concentration, toxic gas concentration, airflow velocity, etc. By dynamically modeling the mining environment, this module can provide key inputs for the health status modeling module, while supporting the optimal design of risk assessment and intervention strategies. Specifically, the modeling process of this module fully considers the diffusion, convection and source term change characteristics of dust and gas in the mining area to ensure that the model has high accuracy and adaptability.

[0065] In this embodiment, the environment dynamic modeling module uses the convection-diffusion equation as a mathematical basis to establish a partial differential equation model of dust concentration and gas concentration changing with time and space. The specific implementation method is described in detail below.

[0066] Generally speaking, the dust concentration in the mining area is Will change over time and spatial location The dynamic evolution of dust concentration is mainly affected by the diffusion, convection and dust source terms. In the present invention, the dynamic evolution of dust concentration is described by the following convection-diffusion equation:

[0067] in: : Indicates time Lower position dust concentration.

[0068] : Dust diffusion coefficient, used to describe the diffusion rate of dust in the mining area.

[0069] : Laplace operator of dust concentration, representing the diffusion effect.

[0070] : Airflow velocity vector, used to describe the convective effect of airflow on dust concentration.

[0071] : The gradient of dust concentration, indicating the concentration change in the convection direction.

[0072] : Dust source term, used to describe the generation intensity of dust.

[0073] Specifically, the diffusion coefficient Usually determined by the dust particle size and the temperature and humidity conditions in the mining area, the air flow velocity Measured by ultrasonic anemometer, the dust source term It is generated by mining equipment and human activities.

[0074] In one possible implementation, the air flow velocity vector The value can be expressed by a three-dimensional wind speed distribution function, and the wind speed at each point in the mining area can be dynamically predicted by an interpolation algorithm.

[0075] For toxic gas concentration In this embodiment, the convection-diffusion equation similar to the dust concentration is used for dynamic modeling. In general, the change of toxic gas concentration is also affected by diffusion, convection and gas generation source terms, and its dynamic evolution can be expressed as:

[0076] in: : Indicates time Lower position concentration of toxic gases.

[0077] : Diffusion coefficient of gas, usually higher than the diffusion coefficient of dust.

[0078] : Laplace operator of gas concentration.

[0079] : Gas source term (unit: ppm / s\text{ppm / s}ppm / s), indicating the generation intensity of the gas.

[0080] As an option, the gas source term It is expressed as a nonlinear function related to the operating equipment and geological activities, for example:

[0081] in: : Indicates location Distance from mine ventilation openings.

[0082] : Indicates the operating intensity of the equipment.

[0083] , , : Empirical parameters for gas generation.

[0084] Generally speaking, environmental dynamic modeling of mining areas requires setting reasonable boundary conditions and initial conditions to ensure that the model solution process is stable and realistic.

[0085] In some embodiments, the boundary conditions of dust concentration and gas concentration are set as: At the mine ventilation outlet: the concentration is zero, indicating the input of clean air.

[0086]

[0087] At the closed wall of the mine area: no outflow boundary condition.

[0088]

[0089] in, Represents the concentration gradient along the wall normal.

[0090] The initial conditions were set to the background concentration at the start of the mine operation:

[0091] Specifically, the present invention uses the finite difference method to discretize and solve the partial differential equation. In general, the time derivative uses the backward difference format:

[0092] The spatial derivatives use the central difference format:

[0093] After discretization, the convection-diffusion equation can be expressed as:

[0094] In the specific implementation, the discretized linear equations are solved by an iterative algorithm (such as the Gauss-Seidel method) to obtain the spatiotemporal distribution of dust and gas concentrations.

[0095] The core function of the health risk assessment module is to calculate the miners' health risk values ​​and classify the health risk levels based on the output data of the health status modeling module and the environmental dynamic modeling module. Through the health risk assessment module, the miners' health status can be quantitatively analyzed, potential health threats can be identified in a timely manner, and a direct basis can be provided for the formulation of intervention strategies. The design of this module comprehensively considers the degree to which the health status deviates from the normal range and the dynamic impact of environmental variables, and has high flexibility and adaptability.

[0096] In this embodiment, the health risk assessment module mainly constructs a health risk function to quantify the abnormality of the miner's health status and dynamically classifies the risks in combination with real-time input data. The specific implementation method is described in detail below.

[0097] In general, the health risk value Used to measure the health status of miners The degree of deviation from the normal state, while considering the direct impact of environmental variables on health status. In the present invention, the health risk function is defined as the following integral form:

[0098] in: :time The global health risk value, unitless.

[0099] : Risk density function, used to describe a single location point local health risks.

[0100] : The spatial extent of the mining area.

[0101] As an alternative, the risk density function It can be specifically defined as:

[0102] in: : The square of the deviation of the health status from the normal value, used to measure the abnormality of physiological parameters.

[0103] , , : Weight parameter, used to adjust the contribution ratio of health status, dust concentration and gas concentration to risk.

[0104] , : Dust concentration and gas concentration are derived from the output of the environmental dynamic modeling module.

[0105] In general, the health risk value It is divided into three levels to guide subsequent intervention decisions.

[0106] In one possible implementation, the risk assessment module combines the risk classification algorithm to dynamically classify real-time health data. Specifically, the module uses the following conditional logic to perform risk assessment:

[0107] In this embodiment, the health risk assessment module needs to perform numerical calculations on the integral formula. is discretized into finite grid points, and the integral formula is discretized into a summation formula:

[0108] in: : The number of grid points after the mining area is discretized.

[0109] : Grid points The location coordinates of .

[0110] : The volume of the grid cell.

[0111] In order to detect potential health risks in advance, this module supports dynamic risk prediction based on time series. In the specific implementation, the module uses the stochastic differential equation (SDE) model of the health status modeling module to predict the health status in the future. Make predictions, and combine the environmental dynamic modeling module to predict the spatiotemporal distribution of dust and gas concentrations, and calculate the health risk value at future times.

[0112] In one implementation, the predicted health risk value It can be estimated by the following formula:

[0113] in: : Predicted health status at a future time.

[0114] , : Predicted dust and gas concentrations at future times.

[0115] Through the above-mentioned prediction functions, this module can provide longer-term support for the intervention optimization module.

[0116] The health intervention optimization module analyzes the miners’ health status and environmental data based on the real-time output of the health risk assessment module and formulates personalized intervention measures. Its purpose is to reduce the health risk value by dynamically optimizing the working conditions and working environment. , and trigger the emergency response mechanism when necessary. The design of this module fully considers the complexity of the mining environment and the individual differences of miners, and uses the variational method and control theory to realize the calculation and implementation of the optimal intervention strategy under limited resources.

[0117] In this embodiment, the health intervention optimization module takes the health risk function as the target, establishes an optimization problem and generates the optimal intervention strategy by solving it. The implementation method is described in detail below.

[0118] Generally speaking, the main purpose of health intervention is to To this end, the module takes the health risk function as the optimization objective and defines the following optimization problem:

[0119] in: : Control variables of intervention measures include ventilation intensity, job rotation frequency, rest time, etc.

[0120] : Health risk value, depends on the control variables changes.

[0121] : Optimize time interval.

[0122] As a possible implementation method, the module decomposes the optimization goal into two sub-goals: one is to reduce the degree to which the health status deviates from the normal value; the other is to control the dust and gas concentration within a reasonable range.

[0123] In this example, the intervention measures Includes the following: Adjust the operating status of the ventilation equipment, such as increasing the ventilation intensity or changing the wind direction.

[0124] Optimize job rotation frequency and reduce the working time of miners in high-risk areas.

[0125] Dynamically schedule rest periods to allow miners to temporarily exit operations when risks are higher.

[0126] As an option, the module can also combine real-time location data to implement regional interventions based on the environmental and health status of specific areas. For example, when the dust concentration in a certain location exceeds the standard, the operating power of the ventilation equipment in that area can be increased first.

[0127] In this embodiment, the health intervention optimization module adopts an optimization algorithm based on the calculus of variations. Generally, the calculus of variations solves the optimization problem with constraints by constructing a Lagrangian function. Specifically, the Lagrangian function is defined as:

[0128] in: , : Lagrange multiplier, used to introduce constraints.

[0129] : An optimization function that contains the objective function and constraints.

[0130] By solving the Euler-Lagrange equation, the optimal intervention strategy can be obtained:

[0131] As an implementation method, the module uses a discretized gradient descent method in actual calculations to iteratively solve the optimization problem to obtain a numerical solution.

[0132] This module receives the output data of the health risk assessment module, including real-time health risk values At the same time, the module also needs to obtain the dynamic change information of dust concentration, toxic gas concentration and health status from the environmental dynamic modeling module and the health status modeling module.

[0133] The module output is the optimal intervention strategy, including ventilation adjustment plans at each time point, job rotation suggestions and rest arrangements. The strategy is directly transmitted to the execution equipment, such as the ventilation system or the on-site management terminal, through the system interface.

[0134] In this embodiment, the health intervention optimization module supports a real-time feedback mechanism. Generally, the module will re-evaluate the health risk value after each intervention measure is implemented. If the risk value is not significantly reduced, the module will automatically adjust the optimization target or increase the intervention intensity.

[0135] As a possible implementation method, the feedback mechanism achieves dynamic adjustment through the following formula:

[0136] in: : The adjustment amount of the control variable.

[0137] : Learning rate, controls the adjustment speed.

[0138] Through the above feedback mechanism, the module can maintain the effectiveness of the intervention strategy in complex and changing environments.

[0139] The closed-loop feedback module aims to achieve adaptive adjustment and optimization of the entire system through real-time monitoring and evaluation of the effects of health intervention measures. This module is closely integrated with the health risk assessment module and the health intervention optimization module. Through the dynamic feedback mechanism, it ensures that the system can continuously adjust the intervention strategy according to the changes in the miners' health status and environmental parameters, thereby maintaining the optimal operating effect. The design focus of the closed-loop feedback module is to establish a real-time closed loop of data flow, reduce the lag and uncertainty of intervention measures, and significantly improve the response efficiency and accuracy of the system.

[0140] In this embodiment, the closed-loop feedback module builds a feedback loop based on real-time data streams, monitors changes in health risk values, and guides improvements in intervention measures through an error correction mechanism. The specific implementation of this module is described in detail below.

[0141] In general, the closed-loop feedback module calculates the error between the actual intervention effect and the expected effect by receiving the real-time evaluation data from the health risk assessment module. The size and direction of the error are directly used to adjust the control parameters in the health intervention optimization module, thereby achieving adaptive optimization of the intervention strategy.

[0142] In this embodiment, the module defines the error as the difference between the target change and the actual change of the health risk value. The target change is usually generated by the optimization algorithm, while the actual change comes from the output data of the health state modeling module and the environmental dynamic modeling module. Through continuous adjustment, the closed-loop feedback module can dynamically track the evolution of the health risk value to ensure that the intervention strategy always matches the actual needs.

[0143] Specifically, in this embodiment, the closed-loop feedback module performs dynamic error correction based on the time step. Generally, the system calculates the change in the health risk value after each time step and compares it with the target change.

[0144] As a possible implementation, the error is calculated using a simple difference form: The actual risk change is the health risk value at the current moment minus the health risk value at the previous moment.

[0145] The target risk change is calculated through the optimal change path generated by the optimization algorithm.

[0146] Under this mechanism, if the actual risk change is less than the target change, it means that the current intervention intensity is insufficient; if the actual change is too large, it may lead to waste of resources or excessive intervention. Based on this result, the closed-loop feedback module will make incremental adjustments to the control variables in the health intervention optimization module. For example, the ventilation intensity can be appropriately increased, the frequency of rest for miners can be increased, or the job rotation time can be further shortened.

[0147] The feedback adjustment process of the closed-loop feedback module can be described as the following dynamic iterative formula: At each time step, the control variable is adjusted proportionally to the error, taking into account the impact of historical adjustments on the current one. Specifically, the adjustment is determined by the proportional factor, the integral factor, and the differential factor.

[0148] Generally speaking, the proportional factor is used to directly reflect the impact of the current error, the integral factor is used to accumulate historical errors, and the differential factor is used to predict the trend of future error changes. Through the combination of the three, the closed-loop feedback module can achieve accurate and stable control adjustments.

[0149] The input data received by the closed-loop feedback module mainly includes: The health risk assessment module provides real-time risk values ​​and classification results.

[0150] The health status modeling module provides miners’ health status change data.

[0151] The environmental dynamic modeling module provides dynamic change data of dust concentration, toxic gas concentration and air flow velocity.

[0152] The output data includes: The adjusted intervention controlled for variables such as operating parameters of ventilation equipment, frequency of job rotation, and rest arrangements.

[0153] To ensure the real-time nature of data flow, the closed-loop feedback module automates the entire process of data collection, processing, and output through a high-performance computing framework. The module also supports anomaly detection, which can trigger an alarm or interrupt intervention operations when significant anomalies occur in the data.

[0154] In this embodiment, the closed-loop feedback module pays special attention to the dynamic adjustment and adaptability of the intervention strategy. In some embodiments, the module dynamically adjusts the parameters of the feedback mechanism itself according to changes in environmental parameters and health status. For example, in the case of a sharp increase in dust concentration, the module can temporarily increase the feedback gain to speed up the response to the intervention measures.

[0155] In addition, the closed-loop feedback module also supports self-optimization during long-term operation. By analyzing historical data, the module can automatically adjust feedback parameters to adapt to long-term changes in the mining area. For example, when the ventilation equipment is frequently overloaded over a period of time, the module will actively reduce the adjustment range of ventilation-related parameters to avoid excessive equipment wear.

[0156] In one possible implementation, the closed-loop feedback module combines machine learning algorithms to achieve self-learning. Specifically, the module records the results of each intervention adjustment and trains a prediction model based on historical data. The model is used to predict the effects of different intervention strategies and guide subsequent intervention optimization.

[0157] In general, the self-learning mechanism can significantly improve the intelligence level of the module. For example, when the environmental conditions in a mining area change drastically, the module can quickly adjust the feedback parameters based on the learned experience to avoid intervention lag caused by insufficient parameter initialization.

[0158] 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 miner's lung health monitoring system, characterized in that: include: The data acquisition module is used to collect the miners' health status data and mine environment data in real time. The health status data includes blood oxygen saturation, respiratory rate and heart rate, and the mine environment data includes dust concentration, toxic gas concentration and airflow velocity. The health status modeling module is used to build a stochastic model to describe the dynamic changes of health status over time based on the miners' health status data and mine environment data; Environmental dynamic modeling module, which is used to build a model to describe the spatial distribution and dynamic changes of dust concentration and toxic gas concentration in the mine through environmental data; The health risk assessment module is used to calculate the miners' lung health risk value and perform risk classification based on the outputs of the health status modeling module and the environmental dynamic modeling module; The health intervention optimization module is used to generate the optimal health intervention strategy based on the output of the health risk assessment module; The closed-loop feedback module is used to implement health intervention strategies and collect and update data in real time to correct the health status modeling module and the environmental dynamics modeling module.

2. A miner's lung health monitoring system according to claim 1, characterized in that: The health status modeling module uses stochastic differential equations to establish a health status model, in which the changes in health status include deterministic changes and stochastic changes in health data. The deterministic changes are determined by the relationship between the miners' blood oxygen saturation, respiratory rate and heart rate, as well as the dust concentration and toxic gas concentration. The stochastic changes are described by Brownian motion terms.

3. The miner's lung health monitoring system according to claim 1 is characterized in that: The environmental dynamic modeling module uses partial differential equations to establish a dynamic model of mine dust concentration and toxic gas concentration. The model considers the superposition of diffusion effect, airflow convection effect and pollutant source term, and performs numerical solution through spatial discretization and temporal discretization.

4. The miner lung health monitoring system according to claim 1 is characterized in that: The health risk assessment module calculates the miner's lung health risk value based on the following risk function: The risk function consists of three parts: the risk of health status deviating from the normal reference value, the risk of dust concentration, and the risk of toxic gas concentration; The risk of health status deviating from the normal reference value is obtained by weighting the square of the deviation between the health status variable and the normal reference value. The risk of dust concentration and the risk of toxic gas concentration are obtained by weighting the square of dust concentration and the square of toxic gas concentration, respectively.

5. The miner lung health monitoring system according to claim 1 is characterized in that: The health risk assessment module performs risk classification according to the value of the risk function, specifically including: When the risk value is less than the first threshold, it is determined that the health status is safe; When the risk value is between the first threshold and the second threshold, it is determined as a health status warning; When the risk value is greater than the second threshold, it is determined that the health status is dangerous.

6. The miner lung health monitoring system according to claim 1 is characterized in that: The health intervention optimization module uses the Lagrangian optimization algorithm to generate the optimal intervention strategy, where: Health intervention strategies include adjusting mine ventilation intensity, miners’ job rotation frequency, and work-rest cycles; The goal of the Lagrangian optimization algorithm is to minimize the cumulative value of health risk value within a certain time range. The optimization process is subject to the constraints of the health status model and the environmental dynamic model.

7. The miner lung health monitoring system according to claim 1 is characterized in that: The health intervention optimization module generates the optimal intervention strategy through the following steps: Construct a Lagrangian function consisting of a health risk function, a health status model constraint, and an environmental dynamic model constraint; Solve the extreme value of Lagrangian function based on variational method; According to the solution results, the optimal ventilation intensity, job rotation time and rest cycle are obtained.

8. The miner lung health monitoring system according to claim 1 is characterized in that: The data acquisition module comprises: Wearable health monitoring devices are used to collect miners' blood oxygen saturation, respiratory rate and heart rate in real time, and upload the data to the central server via wireless network; Environmental monitoring equipment, including distributed dust concentration sensors, toxic gas concentration sensors and air flow velocity monitoring equipment, is used to collect mine environmental data in real time.

9. The miner lung health monitoring system according to claim 1 is characterized in that: The closed-loop feedback module comprises: Real-time intervention feedback module, used to record the changes in miners' health status data and mine environment data after the implementation of the intervention strategy; The model updating module is used to dynamically update the parameters of the health status model and the environmental dynamic model according to the intervention feedback results.

10. The miner lung health monitoring system according to claim 1, characterized in that: The data acquisition module pre-processes the health data and environmental data collected in real time, performs model calculation and intervention optimization through the cloud server, and sends the intervention strategy to the mine control center or miner terminal through wireless communication technology.

Images