A safety and environment multi-parameter monitoring system for limited space operation in a thermal power plant

Abstract

The application discloses a kind of thermal power plant limited space operation safety and environment multi-parameter monitoring system, it is related to operation safety monitoring technical field, the present application is based on space volume structure, gas exchange path, risk type and operation position construction structure parameter model, configure multi-parametric sensing node, and extract the abnormal fluctuation characteristics of environmental parameter by edge computing node;Combining gas flow model and heat distribution model carries out space-time inversion, locates risk source, dynamically generates space risk heat map, can obtain the position and action state of operating personnel in real time, calculates exposure probability, realizes the automatic identification and graded early warning of high-risk area;The present application has high-precision, low-delay, full-coverage monitoring capability, improves the intrinsic safety of thermal power plant limited space operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of work safety monitoring, in particular to a limited space work safety and environment multi-parameter monitoring system for thermal power plants. BACKGROUND

[0002] In the limited space of boilers, flues, coal bunkers, induced draft fan pipelines, and internal dust collectors of thermal power plants, maintenance, repair, or dust removal work is often arranged. Such limited spaces are usually accompanied by dangerous factors such as high temperature, oxygen deficiency, accumulation of toxic gases, and accumulation of combustible dust, and the working environment is extremely harsh.

[0003] In actual work, the oxygen concentration of some limited spaces decreases at a much faster rate than that of general ventilation places due to their strong structural closure and poor internal gas exchange. Once workers enter, oxygen deficiency and suffocation may occur within a short time. At the same time, the metal fume and high temperature released during electric welding may also cause local explosion of combustible gas. Due to the limited space, the existing conventional environmental monitoring devices (such as single gas detectors) are limited in layout and response, cannot realize multi-point linkage monitoring, and cannot effectively cover the entire space.

[0004] In addition, in complex limited space work of thermal power plants, there are high-altitude work, cross work, and interference of sealed electrical equipment. Without systematic monitoring means, it is not only impossible to accurately determine the source of abnormal changes, but also impossible to realize real-time state linkage response of workers. SUMMARY

[0005] The purpose of the present application is to provide a limited space work safety and environment multi-parameter monitoring system for thermal power plants to solve the problems in the background art.

[0006] In order to achieve the above purpose, the present application provides the following technical scheme: a limited space work safety and environment multi-parameter monitoring system for thermal power plants, comprising:

[0007] A structure parameter modeling module: acquires a set of structure parameters corresponding to the limited space work object of the thermal power plant, including the volume structure V of the space, the gas exchange path L, the risk type R, and the corresponding set of work positions X;

[0008] A monitoring parameter configuration module: sets a corresponding set of monitoring parameters P according to the set of work positions X and the risk type R;

[0009] A sensor node deployment module: a plurality of micro multi-parameter sensor nodes Ni are arranged in the volume structure V according to the gas exchange path L;

[0010] An edge computing analysis module: processes the monitoring data of each sensor node Ni by an edge computing node, and extracts a set of parameter fluctuation characteristics Q; wherein the edge computing node performs spatiotemporal inversion analysis on Q according to a gas flow model and a thermal distribution model;

[0011] Early warning module: according to the abnormal change trend in the feature set Q, the current position and action state of the operator are combined to construct a spatial risk dynamic heat map M, and a warning mechanism is triggered;

[0012] Linkage control execution module: according to the risk gradient distribution in the spatial risk dynamic heat map M, the sound and light alarm device, the exhaust system and the emergency communication system in the operation site are linked and responded, and the remote safety control is triggered.

[0013] Preferably, according to the operation position set X and the risk type R, the corresponding monitoring parameter set P is set, including:

[0014] Based on the spatial coordinates of each operation point in the operation position set X and the associated risk type R, a mapping relationship table of operation points and risk factors is established;

[0015] For each operation point in the mapping relationship table, a preset risk parameter matching rule library is called to determine the environmental parameter items to be monitored, including oxygen concentration, toxic gas concentration, temperature and humidity, and particulate matter concentration;

[0016] According to the position of the operation point in the volume structure V and its adjacent gas exchange path L, the update frequency and alarm threshold of the corresponding monitoring parameters are set.

[0017] Preferably, a plurality of micro multi-parameter sensing nodes Ni are arranged inside the volume structure V according to the gas exchange path L, including:

[0018] Based on the spatial coordinates of each operation position in the volume structure and the topological relationship of the gas exchange path, a three-dimensional space layout model is constructed, and a risk coverage heat map is generated;

[0019] According to the risk factor density distribution area in the risk coverage heat map, the priority layout area is determined, and the number of sensing nodes is allocated in order from high to low according to the risk level value;

[0020] In the priority layout area, the specific installation coordinates of each micro multi-parameter sensing node are determined in combination with the spatial wind direction simulation results and the gas diffusion model.

[0021] Preferably, the edge computing node performs spatio-temporal inversion analysis on Q according to the gas flow model and the heat distribution model, including:

[0022] Based on the time series data of each environmental parameter in the parameter fluctuation feature set Q, a time series change matrix is constructed, and the instantaneous gradient value is extracted according to the collection time sequence;

[0023] The time-varying matrix is input into a gas flow model, and based on the gas convection velocity, diffusion coefficient and spatial geometry structure in the model, the propagation path of the environmental parameters in the three-dimensional space is reversely calculated to obtain a source position set;

[0024] Based on the reverse calculation result, a heat distribution model is introduced to solve the coupling relationship between the temperature and humidity parameters, correct the source position deviation of the gas flow model, and generate a joint inversion result;

[0025] The joint inversion result is mapped to the three-dimensional coordinate system of the volume structure, and the risk trigger point after inversion is output.

[0026] Preferably, the spatial risk dynamic heat map M is constructed according to the abnormal change trend in the feature set Q, the current position and action state of the operation personnel, comprising:

[0027] The abnormal change trend of each monitoring parameter in the parameter fluctuation feature set is weighted and superimposed to generate a time-segmented risk change matrix;

[0028] The risk change matrix is fused with the real-time position coordinates and action state of the operation personnel to establish a human-parameter coupled risk model, and the probability of the operation personnel exposed to a high-risk factor is identified;

[0029] According to the three-dimensional coordinate system, the high-risk parameter region and the personnel-intensive activity region are spatially superimposed to generate a spatial risk dynamic heat map represented by a color gradient indicating the risk intensity;

[0030] When the risk value of a certain operation unit in the heat map exceeds the set warning threshold, a hierarchical warning signal is automatically triggered.

[0031] Preferably, the color gradient setting mode comprises:

[0032] The risk score is between 0 and 0.3, displayed in green, indicating low risk; the risk score is between 0.3 and 0.6, displayed in yellow, indicating medium risk; and the risk score is higher than 0.6, displayed in red, indicating high risk.

[0033] Preferably, the spatial risk dynamic heat map M is constructed according to the abnormal change trend in the feature set Q, the current position and action state of the operation personnel, comprising:

[0034] The risk gradient value of each operation unit in the spatial risk dynamic heat map is analyzed, and the heat map is regionally classified and marked according to the risk level division rule;

[0035] The position of the operation personnel in the high-risk region is positioned in real time, and if the exposure probability exceeds the set threshold, a start instruction is sent to the sound and light alarm device near the corresponding position;

[0036] Synchronous start of exhaust system, directional ventilation operation in high-risk area, wind speed and opening time are adjusted according to risk gradient value;

[0037] If the risk continues to rise or multiple high-risk areas appear at the same time, an emergency communication signal is automatically sent to the remote safety control center, and a linkage control instruction is generated to start the operation pause, personnel evacuation or remote forced power-off operation.

[0038] In the above technical solution, the technical effects and advantages provided by the present application are:

[0039] 1. Based on structural parameter modeling, dynamic risk perception and edge intelligent analysis technology, the present application can realize real-time monitoring of multiple environmental parameters such as oxygen, toxic and harmful gases, dust, temperature and humidity in limited space, and generate a space risk heat map dynamically by combining the position and action state of the operation personnel, so as to effectively identify high-risk operation units and risk evolution trend, greatly improving the accuracy of risk warning and the timeliness of response.

[0040] 2. By introducing the spatiotemporal inversion analysis mechanism of the fusion of gas flow model and thermal distribution model, as well as the adaptive exhaust control and multi-device linkage response strategy based on risk gradient, the present application not only realizes the reverse source positioning and predictive control of risk source, but also builds a full-chain safety control system covering perception, evaluation, response and linkage, significantly enhancing the intrinsic safety level of limited space operation in thermal power plants. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings described below are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0042] Figure 1 System module flowchart of the present application. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0044] Embodiment, please refer to Figure 1The safety and environment multi-parameter monitoring system for limited space operation of a thermal power plant according to the embodiment shown comprises:

[0045] The structural parameter modeling module: acquires a structural parameter set corresponding to the limited space operation object of the thermal power plant, including the volume structure V of the space, the gas exchange path L, the risk type R and the corresponding operation position set X.

[0046] In order to realize comprehensive perception and intelligent monitoring of the limited space operation environment of the thermal power plant, the target operation object needs to be modeled first to obtain a complete structural parameter set. The present application provides a structural parameter modeling module for extracting and constructing the structural information and risk distribution characteristics of the limited space operation object of the thermal power plant. The structural parameter set includes:

[0047] The volume structure V: the volume structure V is used to describe the three-dimensional physical size and geometric boundary information of the operation space, including the length, width, height, irregular boundary contour, opening structure (such as manhole, maintenance door) and the like of the space, which is used to determine the monitoring point layout range and the gas diffusion modeling basis subsequently. The volume structure can be obtained by the existing building BIM model, structural drawing or laser scanning modeling technology, and is converted into a three-dimensional coordinate model for data processing.

[0048] The gas exchange path L: the gas exchange path L is used to represent the gas flow channel between the inside and outside of the limited space or between the regions inside the space, including the natural ventilation path (such as the top ventilation hole, the channel gap), the forced exhaust path (such as the ventilation duct, the fan outlet) and the potential gas gathering area. The path information can be identified and mapped by field fluid simulation analysis (CFD simulation), historical maintenance records and sensor layout data as the basis for laying the sensor nodes.

[0049] The risk type R: the risk type R is used to define the types of possible danger sources in the target space, specifically including but not limited to insufficient oxygen concentration, toxic and harmful gases (such as carbon monoxide, hydrogen sulfide, ammonia), flammable and explosive gases (such as methane, hydrogen), high temperature and high humidity environment, high dust concentration, strong noise area, strong electromagnetic interference area and the like. The risk type is comprehensively determined according to the historical operation statistics, the accident type classification and the process flow analysis results, and is marked with the risk level according to the space section.

[0050] The operation position set X: the operation position set X is used to demarcate the typical operation points of the operation personnel in the limited space, including the maintenance point, the patrol path, the welding operation position, the manual cleaning position and the like. Each operation position is identified by a space coordinate and is associated with the risk type R and the space structure information thereof, which is used to determine the parameter monitoring key area and the personnel state fusion analysis subsequently.

[0051] The monitoring parameter configuration module sets the corresponding monitoring parameter set P according to the job position set X and the risk type R.

[0052] First, according to the spatial coordinate information of each job point in the job position set obtained as described above, and in combination with the type to which it belongs in the risk type classification, each job point is bound to the corresponding risk factor, and a job point-risk factor mapping relationship table is generated.

[0053] The mapping relationship table is stored in the form of a two-dimensional table, with the horizontal direction being the job point number and the vertical direction being the risk factor classification, specifically including but not limited to hypoxia risk, toxic and harmful gas risk, temperature and humidity limit risk, dust concentration exceeding standard risk, high temperature and heat radiation risk, etc.

[0054] Each mapping relationship item is constructed through field historical accident records, job working condition analysis, and job personnel feedback data, and a risk level value is assigned using a statistical weighting method. The risk level value is calculated as follows: the risk level value is equal to the historical accident occurrence frequency multiplied by the influence factor weight, plus the expert evaluation risk score, and finally normalized to a floating point number between 0 and 1. The higher the value, the more significant the risk.

[0055] After obtaining the mapping table of job points and risk factors, according to the risk factor corresponding to each job point, a preset risk parameter matching rule library is called to correspondingly output the environmental parameter item that needs to be monitored for the job point.

[0056] The matching rule library is constructed through expert experience model and scene classification analysis, and is managed in the form of a mapping matrix, with each risk factor corresponding to a set of priority monitoring parameter items. For example:

[0057] For hypoxia risk, the monitoring parameter is oxygen volume fraction;

[0058] For toxic gas risk, carbon monoxide, hydrogen sulfide, ammonia, etc. need to be monitored for concentration;

[0059] For high temperature and heat radiation risk, temperature and humidity need to be monitored;

[0060] For particulate matter risk, PM2.5 and PM10 mass concentrations need to be monitored.

[0061] The above parameter items are sorted by a defined weight function, with higher priority parameters marked as key monitoring parameters for subsequent setting of alarm thresholds and response strategies.

[0062] According to the spatial position of each job point in the volume structure and the relative distance from the gas exchange path, the collection frequency and alarm threshold of the selected monitoring parameters are dynamically set.

[0063] The setting of the update frequency uses the following calculation logic:

[0064] If the distance between the work point and the nearest gas exchange entrance is less than 3 meters, set it to high frequency collection, 1 collection per second;

[0065] If the distance is between 3 to 6 meters, set it to medium frequency collection, 1 collection every 3 seconds;

[0066] If the distance exceeds 6 meters, set it to low frequency collection, 1 collection every 5 seconds.

[0067] The setting of alarm threshold is based on the safety limit value standard of various gases and environmental parameters in industry standard, and is dynamically adjusted according to the duration of the work and the personnel exposure model. For example, the basic alarm threshold of oxygen volume fraction is set to 19.5%, if the expected work duration exceeds 30 minutes, the alarm threshold will be raised to 20.0% to trigger early warning.

[0068] In addition, in order to ensure the flexibility and environmental adaptability of parameter setting, each monitoring parameter is allowed to set double threshold interval, i.e. early warning threshold and emergency threshold, when the collected data approaches the early warning threshold, it will prompt, when it exceeds the emergency threshold, it will trigger forced linkage response, such as exhaust or personnel evacuation suggestion.

[0069] Sensor node deployment module: inside the volume structure V, a plurality of micro multi-parameter sensor nodes Ni are arranged according to the gas exchange path L.

[0070] Firstly, based on the obtained volume structure, the spatial coordinate information of each work position is extracted, and a three-dimensional coordinate system is constructed. The coordinate system takes a reference corner of the work space as the origin, sets X, Y and Z axes along the length, width and height directions respectively, and maps the positions of all work points to the three-dimensional coordinate system.

[0071] At the same time, all natural or forced ventilation inlets, outlets and channel directions in the gas exchange path are extracted, and a gas flow topology graph is established, which is superimposed into the three-dimensional coordinate model to form a unified space flow model.

[0072] Next, according to the mapping relationship between the work point and the risk factor and the risk level value distribution, all risk factors are projected as density points in space, and kernel density estimation algorithm is used for interpolation processing to generate a three-dimensional risk coverage heat map. In the heat map, the color depth represents the risk factor intensity per unit volume, the high-risk area color deepens, the low-risk area color lightens, and the spatial risk visualization is realized.

[0073] On the basis of the generated three-dimensional risk coverage heat map, the limited space is divided into a plurality of regular cubic grid units, and the length of each grid unit can be set to 0.5 to 1 meter, which is set according to the monitoring accuracy requirement.

[0074] The risk factor density value in each grid cell is counted, and the grid cell is divided into high, medium and low risk level areas according to the risk level value of the work point where it is located.

[0075] The sensor nodes are preferentially arranged in the high-risk area, and the number of nodes is allocated according to the following rules:

[0076] At least one micro multi-parameter sensor node is arranged in each high-risk grid cell;

[0077] One sensor node is arranged in every 2-3 grid cells in the medium-risk area;

[0078] The low-risk area adopts boundary distribution or auxiliary monitoring to ensure the overall coverage continuity.

[0079] The number of nodes is allocated based on the normalized weighted results of the total risk level value to optimize resource allocation and ensure that the key areas are preferentially arranged and the risk concentration points are densely arranged under the condition of limited sensing resources.

[0080] To further improve the sensing effectiveness of node arrangement, wind field simulation and gas diffusion prediction methods are used in the selected priority arrangement area to model the pollutant diffusion path and concentration evolution trend.

[0081] The computational fluid dynamics modeling method is used to construct the wind speed field and wind direction distribution map inside the limited space. The space geometry, ventilation port position, fan displacement and wind speed parameters are input during the simulation process, and the steady-state wind field distribution is obtained by finite element discrete solution.

[0082] Based on this wind field, the physical properties of the diffusion species (including density, volatility, reactivity, etc.) are superimposed to construct a gas concentration distribution model based on the convection-diffusion equation, and the diffusion path of the leakage source at different positions is simulated.

[0083] By analyzing the simulation results, the airflow convergence area, diffusion path intersection point and wind speed mutation point are selected as the preferred installation coordinates of the sensor nodes to ensure that the data collected by each node can cover the key diffusion trend area.

[0084] Edge computing analysis module: the monitoring data of each sensor node Ni is processed by the edge computing node to extract the parameter fluctuation feature set Q; the edge computing node performs spatiotemporal inversion analysis on Q based on the gas flow model and heat distribution model.

[0085] Each micro multi-parameter sensor node uploads the collected environmental parameter data to the edge computing node in real time, and the edge computing node summarizes and classifies the multi-dimensional data from different spatial positions.

[0086] According to the time stamp of each sensor node and the environmental parameter type (such as oxygen concentration, temperature, humidity, carbon monoxide concentration, etc.), the data of the same parameter at adjacent time points are sequentially arranged to construct a time series matrix.

[0087] In each time series, the instantaneous gradient value of the parameter is calculated by using the finite difference method. The instantaneous gradient value is equal to the parameter value at the current time point minus the parameter value at the previous time point, and the result is divided by the time interval between the two time points, which is the change per second.

[0088] The constructed time series change matrix is used as input data to embed the gas flow model for reverse path calculation.

[0089] The gas flow model is based on a three-dimensional convection-diffusion equation and is used to describe the propagation behavior of gas in a limited space. Its basic form is: the rate of change of gas concentration in space and time is equal to the superposition of convection term (concentration transmission in wind direction) and diffusion term (random diffusion of gas molecules). The input of the model includes: initial distribution concentration of gas; local wind speed field (obtained from wind direction sensor or simulation data); gas diffusion coefficient (determined according to physical properties table); spatial geometry structure (provided by volume structure model).

[0090] Based on the above parameters, the reverse propagation algorithm is used to perform spatial backtracking from the high fluctuation area detected by each node to the source, and a source location set composed of multiple possible source points is obtained.

[0091] First, the edge computing node receives monitoring data from multiple sensor nodes and identifies high fluctuation areas. High fluctuation area refers to a node set whose instantaneous change rate of an environmental parameter exceeds the set fluctuation threshold of the parameter within a unit time window.

[0092] The fluctuation threshold is defined as follows: if the unit time period is 10 seconds, the change rate of a parameter (i.e. the current value minus the value at the previous time, divided by the time interval) is greater than 3 times the historical stable fluctuation value of the parameter, which is marked as a high fluctuation point.

[0093] Each high fluctuation point is marked as an inversion starting point, and its spatial coordinates, time stamp, parameter type and change rate value are recorded to form a high fluctuation point set.

[0094] Before inversion, a physical model of gas propagation in a limited space needs to be established. The gas flow model uses a numerical modeling method based on a three-dimensional convection-diffusion equation, and the expression is as follows: ; wherein, represents the change rate of gas concentration per unit time; C represents the gas concentration at a certain time and a certain spatial position; represents the gas convection velocity vector, which includes the size and direction of the wind speed; Gradient vector of gas concentration, indicating the tendency of gas concentration change in space; Gas diffusion coefficient; Laplacian of gas concentration, i.e. the sum of second-order partial derivatives of gas concentration in three spatial dimensions (x, y, z).

[0095] Using the back-propagation algorithm, the gas propagation path is traced back in time from each high fluctuation point. The basic principle is: assuming the current high fluctuation point as the terminal point, the wind speed field and diffusion model are applied in reverse to simulate the possible movement path of the gas parameter in the past time period. The calculation steps are as follows:

[0096] Set the backtracking time window (e.g. 60 seconds) and take the high fluctuation point as the starting point;

[0097] At each time step, based on the reverse wind speed vector, estimate the possible last position of the pollutant;

[0098] Meanwhile, considering the diffusion probability, use the Monte Carlo random simulation method to superimpose multiple particle trajectories;

[0099] Iterate and advance until the time window starting point is reached, and output all possible path sets.

[0100] Each reverse path is projected into the spatial coordinate model, and its path point set and cumulative path probability are recorded.

[0101] Based on all the reverse path data, the overlapping area of the path trajectory in the spatial model is calculated. If the overlapping density of multiple paths in a certain area reaches a certain threshold (e.g. more than 3 paths intersect and the cumulative probability value is greater than 0.7), mark the area as a possible source point.

[0102] All source points that meet the conditions are summarized to form a set of source locations, which are output in the form of three-dimensional coordinate point sets as the basis for subsequent heat map drawing and response triggering.

[0103] Since the gas flow behavior is easily disturbed by temperature gradient and humidity change, a thermal distribution model needs to be further introduced to correct the back-propagation results.

[0104] The thermal distribution model is based on the coupled equations of thermal convection and thermal conduction, and is used to simulate the influence of temperature distribution on gas flow path. The temperature distribution can be collected by multi-node temperature sensors in real time and fitted into a spatial thermal field map.

[0105] In addition, considering the adjusting effect of humidity on gas density and diffusion speed, a temperature and humidity coupling coefficient is introduced, and a temperature and humidity joint correction factor is constructed in a multiple regression way. By multiplying the speed term in the gas diffusion model by this correction factor, the original gas flow path is adjusted.

[0106] Finally, combine the outputs of the thermal distribution model and the gas flow model to form a joint inversion path and relocate potential source points to improve inversion accuracy.

[0107] Convert the joint inversion path results into a standard three-dimensional coordinate format and map them into the spatial coordinate model of the volume structure. For multiple path intersection points or areas with dense inversion results, define them as high-confidence risk trigger points.

[0108] The criteria for determining risk trigger points are as follows: if more than three inversion paths intersect within the same volume element and the parameter instantaneous gradient exceeds its set threshold (e.g., the oxygen concentration drop rate is greater than 0.3 percentage points per second), it is marked as a trigger point.

[0109] The final output is a set of spatial points with three-dimensional coordinates, which can be used to construct dynamic risk heat maps, guide the activation of response devices, or identify the source of the accident.

[0110] Early warning module: Based on the abnormal change trend in the feature set Q, the current location and action state of the operator, construct a spatial risk dynamic heat map M, and trigger the early warning mechanism.

[0111] First, extract the abnormal change trend of each type of monitoring parameter (including oxygen concentration, hydrogen sulfide concentration, carbon monoxide concentration, particulate matter concentration, temperature, humidity, etc.) included in the parameter fluctuation feature set.

[0112] Abnormal change trend is determined by the following method:

[0113] If the instantaneous change rate of a parameter exceeds the set fluctuation threshold of that parameter within a unit time period (e.g., every 10 seconds), it is determined to be an abnormal point.

[0114] Perform weighted superposition processing on all parameter abnormal points to construct a risk change matrix. The form of the weighting function is:

[0115] The abnormal degree of each parameter is multiplied by its corresponding risk impact weight, and the weighted sum of all parameters gives the total risk score in that time period.

[0116] The weight is set according to the corresponding damage level of the parameter in the risk factor mapping table established in the early stage, such as oxygen concentration being a high-risk factor with a weight of 0.9, and temperature abnormality being a secondary factor with a weight of 0.5, with weight values ranging from 0 to 1.

[0117] Finally, a two-dimensional matrix with time as the horizontal axis and the operating area as the vertical axis is obtained, which describes the spatial risk change distribution in each time period.

[0118] Real-time spatial position coordinates (X, Y, Z) and current action state (such as static, walking, bending, waving tools, etc.) of the work personnel are obtained by wearing inertial sensors and positioning modules to achieve dynamic acquisition.

[0119] The personnel coordinate data is fused with the risk change matrix in step one to build a human-part coupling risk model. The core logic of the model is:

[0120] In any time period, if the risk score of the space grid unit where the personnel is located is higher than the set reference value, and the personnel is in a high exposure action state (such as crouching, working in place), the exposure probability P is calculated.

[0121] The calculation formula of exposure probability P is: ; In the formula, represents the risk score of the space grid unit, which is derived from the risk change matrix and is the weighted superimposed environmental risk value; represents the action exposure factor, which reflects the influence degree of personnel action state on risk exposure, is an empirical parameter, the value range is 0.1 to 1, such as crouching is 1.0, walking is 0.6, standing is 0.3. represents the theoretical maximum exposure value. The model can dynamically output the potential risk degree of each worker in different time periods.

[0122] Based on the foregoing risk change matrix and personnel exposure probability model, all risk score results are mapped to a three-dimensional coordinate system. The space grid is taken as the basic unit (such as one grid unit per cubic meter), and the risk score is converted into a color gradient value.

[0123] The color gradient setting method is as follows:

[0124] The risk score between 0 and 0.3 is displayed in green, indicating low risk;

[0125] The risk score between 0.3 and 0.6 is displayed in yellow, indicating medium risk;

[0126] The risk score higher than 0.6 is displayed in red, indicating high risk.

[0127] The dynamic heat map is updated every 10 seconds to present the aggregation, diffusion and movement trend of the risk factor in space. High-frequency refresh ensures real-time reflection of sudden risk evolution, supporting the judgment basis for subsequent intelligent linkage control.

[0128] In the space heat map, the risk score value of each grid unit is compared with the preset warning threshold value in real time. If the score value exceeds the set alarm threshold T, the corresponding level of warning signal is triggered immediately.

[0129] The warning threshold T is set according to the following principles:

[0130] Primary warning: T≥0.6 (red area), immediately trigger sound and light alarm and inform the management end through the emergency communication module;

[0131] Secondary warning: 0.3≤T<0.6 (yellow area), prompt the operating personnel to pay attention to environmental changes;

[0132] Tertiary warning: T<0.3 (green area), the system records data and no alarm is needed.

[0133] The warning trigger result will be pushed to the visual terminal interface synchronously, and the high-risk personnel position will be marked by the positioning module to assist the command personnel in making a response.

[0134] The linkage control execution module: according to the risk gradient distribution in the spatial risk dynamic heat map M, the sound and light alarm device, the exhaust system and the emergency communication system in the operating site are linked and responded to, triggering remote safety control.

[0135] First, the risk gradient value of each operating unit in the spatial risk dynamic heat map is extracted. The risk gradient value is expressed as the spatial derivative of the risk score in a unit space, that is, the difference between the risk values of adjacent grids divided by the spatial distance, which is used to measure the rate of change of risk.

[0136] The specific calculation method is:

[0137] The risk gradient value is equal to the difference in risk score between adjacent grid units divided by the Euclidean distance between the centers of the two units.

[0138] After obtaining the risk gradient values of all units, the entire heat map is partitioned according to the preset risk level division rules. The division criteria are as follows:

[0139] High-risk area: risk score≥0.6 and risk gradient value≥0.3;

[0140] Medium-risk area: risk score between 0.3 and 0.6;

[0141] Low-risk area: risk score<0.3;

[0142] Each risk level area is assigned a unique identification code, which is used as the basis for subsequent linkage response instruction triggering.

[0143] Obtain the three-dimensional coordinates uploaded by the positioning device worn by each operating personnel, and determine whether the position falls within the high-risk area marked in the heat map. If it falls within, further calculate the real-time exposure probability of the personnel.

[0144] The exposure probability is obtained by multiplying the risk score of the unit by the action exposure factor corresponding to the current action and then normalizing it, referring to the aforementioned formula:

[0145] Exposure probability = Risk score × Action exposure factor / Maximum exposure value; when the exposure probability exceeds the set threshold (recommended to be set to 0.6), the system immediately sends a start instruction to the nearest audible and visual alarm device to the personnel. The start instruction is issued in a wireless communication manner, and the alarm device issues a continuous audible and visual signal to prompt the operating personnel to evacuate or standby, thereby improving the risk alertness.

[0146] After the audible and visual alarm is started, the system automatically identifies the space number of the high-risk area and activates the exhaust equipment associated therewith.

[0147] The exhaust system start parameters are adaptively set according to the current risk gradient value. Specifically:

[0148] Exhaust air speed setting rule:

[0149] If the risk gradient value is greater than or equal to 0.5, the wind speed is set to the maximum air volume (for example, 2000 cubic meters per hour)

[0150] If the risk gradient value is between 0.3 and 0.5, the wind speed is set to the middle range (for example, 1500 cubic meters per hour)

[0151] If the risk gradient value is less than 0.3, only low-speed ventilation is started (for example, 800 cubic meters per hour)

[0152] The ventilation duration is set according to the risk duration. If the risk score does not decrease in the next 3 cycles (each cycle is 10 seconds), the exhaust state is maintained until the risk score in the heat map of the region is less than 0.3.

[0153] Through this way, directional and hierarchical ventilation of the high-risk area is realized, the local harmful substance concentration is reduced, and the on-site risk mitigation is assisted.

[0154] If any of the following conditions is met:

[0155] There are more than two high-risk areas in the same heat map;

[0156] The risk score of any high-risk area is continuously higher than 0.7 and the duration is more than 60 seconds;

[0157] The exposure probability of any operating personnel is continuously higher than 0.8 for three times, an emergency event identifier is automatically generated, and an encrypted instruction data packet is sent to the remote safety control center through the communication module. The data packet includes: abnormal area number, risk parameter list, operating personnel position and state information. After receiving the instruction, the receiving end executes the following linkage instructions:

[0158] Strong operation suspension instruction: prompt all operating personnel to stop the current operation through local broadcast;

[0159] Personnel evacuation prompt: guide relevant personnel to move to low-risk areas;

[0160] Remote power-off control: cut off the power supply of the work area through the control interface in necessary cases to prevent the expansion of fire or explosion.

[0161] This process realizes a full-chain closed-loop response from environmental perception to device control and personnel evacuation.

[0162] The above merely provides a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which shall be encompassed in the protection scope of the present application.

Claims

1. A safety and environmental multi-parameter monitoring system for limited space operation in a thermal power plant, characterized in that it comprises: Comprise: A structure parameter modeling module: obtain the structure parameter set corresponding to the limited space operation object of the thermal power plant, including the volume structure V, the gas exchange path L, the risk type R and the corresponding operation position set X; A monitoring parameter configuration module: according to the operation position set X and the risk type R, set the corresponding monitoring parameter set P; A sensor node deployment module: in the volume structure V, according to the gas exchange path L, multiple micro multi-parameter sensor nodes Ni are arranged; An edge computing analysis module: the monitoring data of each sensor node Ni is processed by the edge computing node to extract the parameter fluctuation feature set Q; wherein the edge computing node carries out space-time inversion analysis on Q according to the gas flow model and the heat distribution model, and the space-time inversion analysis specifically comprises: Based on the time series data of each environmental parameter in the parameter fluctuation feature set Q, a time series change matrix is constructed, and the instantaneous gradient value is extracted according to the collection time sequence; The time series change matrix is input into the gas flow model, and the propagation path of the environmental parameter in the three-dimensional space is inversely calculated according to the gas convection velocity, diffusion coefficient and space geometry structure in the model, and the source position set is obtained; On the basis of the inverse calculation result, the heat distribution model is introduced to solve the coupling relationship between temperature and humidity parameters, correct the source position deviation of the gas flow model, and generate the joint inversion result; Map the joint inversion result to the three-dimensional coordinate system of the volume structure, and output the risk trigger point after inversion; An early warning module: according to the abnormal change trend in the feature set Q, the current position and action state of the operator are constructed to build a space risk dynamic heat map M, and the early warning mechanism is triggered; A linkage control execution module: according to the risk gradient distribution in the space risk dynamic heat map M, the sound and light alarm device, the exhaust system and the emergency communication system in the operation site are linked to respond, and the remote safety control is triggered.

2. The safety and environmental multi-parameter monitoring system for limited space operation in a thermal power plant according to claim 1, characterized in that: According to the operation position set X and the risk type R, the corresponding monitoring parameter set P is set, including: Based on the spatial coordinates of each operation point in the operation position set X and the associated risk type R, a mapping relationship table of operation point and risk factor is established; For each operation point in the mapping relationship table, a preset risk parameter matching rule library is called to determine the environmental parameter items to be monitored, including oxygen concentration, toxic gas concentration, temperature and humidity, and particulate matter concentration; According to the position of the operation point in the volume structure V and its adjacent gas exchange path L, the update frequency and alarm threshold of the corresponding monitoring parameter are set.

3. The multi-parameter monitoring system for safety and environment in confined spaces of thermal power plants according to claim 1, characterized in that: The multiple micro multi-parameter sensor nodes Ni are arranged in the volume structure V according to the gas exchange path L, including: Based on the spatial coordinates of each operation position in the volume structure and the topological relationship of the gas exchange path, a three-dimensional space layout model is constructed, and a risk coverage heat map is generated; According to the risk factor density distribution area in the risk coverage heat map, the priority arrangement area is determined, and the number of sensor nodes is allocated from high to low according to the risk level value; In the priority arrangement area, the specific installation coordinates of each micro multi-parameter sensor node are determined by combining the spatial wind direction simulation result and the gas diffusion model.

4. The system according to claim 1, wherein the system further comprises a plurality of sensors for detecting the environmental parameters. The space risk dynamic heat map M is constructed according to the abnormal change trend in the feature set Q, the current position of the operator and the action state, and comprises: The abnormal change trend of each monitoring parameter in the parameter fluctuation feature set is weighted and superimposed to generate a time segmented risk change matrix; The risk change matrix is fused with the real-time position coordinates and action state of the operator to establish a human-parameter coupling risk model to identify the probability of the operator being exposed to a high risk factor; According to a three-dimensional coordinate system, the high risk parameter region and the personnel intensive activity region are spatially superimposed to generate a space risk dynamic heat map represented by a color gradient indicating the risk intensity; When the risk value of a work unit in the heat map exceeds a set early warning threshold, a hierarchical early warning signal is automatically triggered.

5. The safety and environmental multi-parameter monitoring system for limited space operation in a thermal power plant according to claim 4, characterized in that: The color gradient setting mode comprises: The risk score is between 0 and 0.3, and green is displayed, indicating low risk; the risk score is between 0.3 and 0.6, and yellow is displayed, indicating medium risk; the risk score is higher than 0.6, and red is displayed, indicating high risk.

6. The safety and environmental multi-parameter monitoring system for limited space operation in a thermal power plant according to claim 1, characterized in that: According to the risk gradient distribution in the space risk dynamic heat map M, the sound and light alarm device, the exhaust system and the emergency communication system in the work site are linked and responded, and comprise: The risk gradient value of each work unit in the space risk dynamic heat map is analyzed, and the heat map is regionally classified and marked according to the risk level division rule; The position of the operator in the high risk region is positioned in real time, and if the exposure probability exceeds the set threshold, the start instruction is sent to the sound and light alarm device near the corresponding position; The exhaust system is started synchronously to perform directional ventilation operation on the high risk region, wherein the wind speed and the opening time are adaptively adjusted according to the risk gradient value; If the risk continues to rise or multiple high risk regions appear at the same time, an emergency communication signal is automatically sent to the remote safety control center, and a linkage control instruction is generated to start the operation pause, personnel evacuation or remote forced power-off operation.

Images