Early-warning, flame-retardant and explosion-suppression method and system for production workshop

Abstract

The invention provides an early-warning flame-retardant explosion suppression method and system for a production workshop, and relates to the technical field of safety production, and the method comprises the steps: collecting equipment parameters, material attributes and environment parameters of the production workshop, and carrying out the preprocessing based on a Kalman filtering technology; based on a # imgabs0 # technology, constructing a workshop equipment geometric model, and performing dynamic calibration on the workshop equipment geometric model to obtain a dynamic workshop equipment geometric model; based on a multi-dimensional model technology, constructing a multi-dimensional physical field coupling model, performing simulation risk evolution on the dynamic workshop equipment geometric model according to the multi-dimensional physical field coupling model and the preprocessed material attributes and environmental parameters to obtain corresponding workshop risk factors, and then executing a corresponding solution; based on a neural convolutional network, according to the workshop risk factors, constructing a production workshop early warning model, further generating a production workshop early warning score, and according to the production workshop early warning score, generating graded response early warning; and the early warning accuracy is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of safe production, and in particular to an early warning flame retardant and explosion suppression method and system for a production workshop. Background Art

[0002] With the in-depth development of industry and intelligent manufacturing, the automation and complexity of production workshops are constantly increasing, and safety risks such as fire and explosion are characterized by multi-parameter coupling and dynamic evolution. Traditional early warning systems rely on single-point sensor threshold alarms, which have the following defects:

[0003] Data isolation and risk omissions: lack of analysis of multi-sensor data; static model and poor adaptability: workshop equipment displacement and material property changes are not dynamically calibrated, and fixed thresholds are difficult to cope with complex working conditions; single physical field analysis and linkage are insufficient: lack of coupled simulation of flow field, thermal field, and electromagnetic field, unable to simulate the risk evolution path, and the fire protection and process system linkage strategy is extensive, with a high risk of accident expansion.

[0004] In this context, there is an urgent need for a new generation of early warning technology that integrates dynamic modeling, multi-physics field coupling and intelligent algorithms. Therefore, a method and system for early warning flame retardant and explosion suppression in a production workshop is provided. Summary of the Invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide an early warning flame retardant and explosion suppression method and system for a production workshop.

[0006] In order to achieve the above object, the present invention provides the following technical solutions: A method for early warning flame retardancy and explosion suppression in a production workshop, the method comprising: Collect equipment parameters, material properties, and environmental parameters in the production workshop, and pre-process them based on Kalman filtering technology; based on Technology, and according to the pre-processed equipment parameters, construct a workshop equipment geometric model, dynamically calibrate the workshop equipment geometric model, and obtain a dynamic workshop equipment geometric model; Based on multidimensional modeling technology, a multidimensional physical field coupling model is constructed. According to the multidimensional physical field coupling model and the pre-processed material properties and environmental parameters, the risk evolution of the dynamic workshop equipment geometry model is simulated to obtain the corresponding workshop risk factors, and then the corresponding solutions are implemented; Based on the neural convolutional network and according to the workshop risk factors, a production workshop early warning model is constructed to generate a production workshop early warning score, and a graded response early warning is generated according to the production workshop early warning score.

[0007] According to one preferred embodiment of the present invention, the process of collecting equipment parameters, material properties, and environmental parameters of a production workshop includes: A data acquisition device is preset; the data acquisition device includes: an equipment parameter acquisition unit, a material attribute acquisition unit, and an environmental parameter acquisition unit; and a corresponding acquisition cycle is set, the acquisition cycle includes several acquisition moments; The equipment parameter acquisition unit is used to acquire equipment parameters in the production workshop; the material attribute acquisition unit is used to acquire material attributes in the production workshop; and the environmental parameter acquisition unit is used to acquire environmental parameters in the production workshop.

[0008] According to one preferred embodiment of the present invention, the process of pre-processing equipment parameters, material properties, and environmental parameters includes: Set up the preprocessing state equation: ;in, Collection time The state vector of is the state transfer matrix; is the input vector; is the input matrix; is the process noise vector, and , is the covariance matrix of process noise; Set up the preprocessing measurement equation: ;in, is the measurement vector; is the measurement matrix; The noise vector is measured, and , is the covariance matrix of the measurement noise; Based on Kalman filtering, the steps for predicting data are: ;in, is the state prediction value; is the forecast error covariance matrix; The steps to update the data are: ;in, is the Kalman gain; is the preprocessed data; is the estimated error covariance matrix; According to the prediction step and the update step, the corresponding pre-processing equipment parameters, material properties and environmental parameters are obtained.

[0009] According to one preferred embodiment of the present invention, the process of constructing a geometric model of workshop equipment includes: Obtaining preprocessed equipment parameters, and obtaining a distribution map of the equipment in the production workshop based on the preprocessed equipment parameters; Import the plan, elevation, and section drawings of the production workshop into the BIM modeling software to construct a geometric model of the workshop; The distribution map of the equipment in the production workshop is imported into the workshop geometric model, and the workshop equipment geometric model is constructed according to the actual size of the production workshop and the corresponding equipment.

[0010] According to one preferred embodiment of the present invention, the process of dynamically calibrating the geometric model of workshop equipment to obtain the dynamic geometric model of workshop equipment includes: At the time of collection , get the location coordinates of the device ; Compared with the last collection time Location coordinates Calculate and obtain the device offset distance ; Preset device offset distance threshold ; If the device is offset , no dynamic calibration is performed; If the device is offset , then dynamically update the corresponding device; Offset the device by a distance The position coordinates of the equipment at that time are mapped to the workshop equipment geometric model to generate a dynamic workshop equipment geometric model.

[0011] According to one preferred embodiment of the present invention, based on multidimensional model technology, the process of constructing a multidimensional physical field coupling model includes: The multi-dimensional model technology includes computational fluid dynamics algorithm, Fourier equations and Maxwell equations; the multi-dimensional physical field coupling model includes: dust diffusion model, heat conduction model and electric spark generation model; Based on the computational fluid dynamics algorithm, the process of building a dust diffusion model includes: Use unstructured grids to divide the production workshop into grids, and set the grid density based on the actual production workshop conditions. Based on the actual dust concentration requirements of the production workshop, select the Euler-Lagrange model, Euler-Euler model, or turbulence model as the basic model. Set dust diffusion boundary conditions, including inlet boundary conditions, outlet boundary conditions, and wall boundary conditions. Constructing a dust diffusion model according to the unstructured grid, grid density, basic model, inlet boundary condition, outlet boundary condition, and wall boundary condition; Based on the Fourier equation, the process of building a heat conduction model includes: Select the unstructured grid type to mesh the production workshop and equipment, and set the grid density according to the actual situation of the production workshop and equipment; Set heat conduction boundary conditions, including: first-class boundary conditions, second-class boundary conditions, and third-class boundary conditions; Construct a heat conduction model based on unstructured grids, grid density, first-type boundary conditions, second-type boundary conditions, and third-type boundary conditions; Based on Maxwell's equations, the process of building an EDM generation model includes: Select the unstructured grid type to mesh the production workshop and equipment, and set the grid density according to the actual situation of the production workshop and equipment; Setting the boundary conditions for spark generation, including electric field boundary conditions and magnetic field boundary conditions; An EDM generation model is constructed based on unstructured grids, grid density, electric field boundary conditions, and magnetic field boundary conditions.

[0012] According to one preferred embodiment of the present invention, the process of simulating risk evolution of a dynamic workshop equipment geometric model, obtaining corresponding workshop risk factors, and then executing corresponding solutions includes: The pre-treated material properties and environmental parameters are used as the initial state conditions of the physical field, and the dust diffusion model, heat conduction model, and spark generation model are used as boundary conditions; Importing the initial state conditions and boundary conditions into the dynamic workshop equipment geometric model, and setting risk indicator thresholds, including: a high concentration dust risk indicator threshold, an equipment thermal anomaly risk indicator threshold, and an electric spark risk indicator threshold; According to different boundary conditions, the corresponding risk indicators are output, including: high concentration dust risk indicator, equipment thermal anomaly risk indicator and electric spark risk indicator; If the high-concentration dust risk index is less than or equal to the high-concentration dust risk index threshold, the production workshop has a lower probability of dust explosion risk. If the high-concentration dust risk index is greater than the high-concentration dust risk index threshold, the production workshop has a higher probability of dust explosion risk. The simulation outputs the corresponding workshop risk factor, reminds relevant personnel to take corresponding solutions, and uploads the records to the backend for storage via wireless transmission technology. If the equipment thermal anomaly risk index is less than or equal to the equipment thermal anomaly risk index threshold, the probability of the equipment in the production workshop generating thermal anomaly risk is smaller. If the equipment thermal anomaly risk index is greater than the equipment thermal anomaly risk index threshold, the probability of the equipment in the production workshop generating thermal anomaly risk is greater. The simulation outputs the corresponding workshop risk factor, reminds relevant personnel to take corresponding solutions, and uploads the records to the background for storage through wireless transmission technology. If the spark risk index is less than or equal to the spark risk index threshold, the probability of the production workshop's lines generating spark risk is smaller. If the spark risk index is greater than the spark risk index threshold, the probability of the production workshop's lines generating spark risk is greater. The simulation outputs the corresponding workshop risk factor, reminds relevant personnel to take corresponding solutions, and uploads the records to the background for storage through wireless transmission technology.

[0013] According to one preferred embodiment of the present invention, a production workshop early warning model is constructed to generate a production workshop early warning score. The process of generating a graded response early warning based on the production workshop early warning score includes: Obtain several groups of historical workshop risk factors; group and label several groups of historical workshop risk factors, and record them as is a natural number; The historical workshop risk factors of the group are used as sample data, and is less than and using the sample data to obtain the mean of the sample data, which is recorded as a sample set; using the remaining groups of historical workshop risk factors as test sets; and forming a training sample set based on the sample set and the test set; Based on convolutional network neural networks, a standard risk warning model is constructed; The training sample set is input into the standard risk warning model to train the standard risk warning model, and the trained standard risk warning model is recorded as the production workshop warning model; According to the production workshop early warning model, the production workshop early warning score under the current environmental factors is generated , the production workshop early warning score for: ;in, The convolutional network neural network output feature map is the first The weight corresponding to the convolution kernel; is the workshop risk factor corresponding to the collection time; It is The bias value corresponding to the convolution kernel; Preset production workshop warning threshold range ; If the production workshop early warning score If the production workshop has no risk, it is deemed to be in a safe operating state and the corresponding data is recorded in real time; if the production workshop early warning score is If the production workshop has an accident risk, it will be identified as a medium-level warning, triggering local alarms and remote notifications, and reminding relevant maintenance personnel to perform corresponding maintenance on the equipment; if the production workshop warning score is When the alarm is triggered, there is an accident risk in the production workshop, which is identified as a high-level alarm and triggers a full alarm. The siren will sound an on-site warning, the on-site video will be saved, the power supply equipment will be cut off in time, and the maintenance personnel will be reminded to implement the corresponding emergency treatment measures.

[0014] The present invention further provides an early warning flame retardant and explosion suppression system for a production workshop, which implements the above-mentioned early warning flame retardant and explosion suppression method for a production workshop, comprising: Data acquisition module, used to collect equipment parameters, material properties and environmental parameters in the production workshop; The data preprocessing module preprocesses equipment parameters, material properties, and environmental parameters based on Kalman filtering technology; Workshop geometry model building module, based on Technology, and according to the pre-processed equipment parameters, construct a workshop equipment geometric model, dynamically calibrate the workshop equipment geometric model, and obtain a dynamic workshop equipment geometric model; The simulation analysis engine module, based on multidimensional model technology, constructs a multidimensional physical field coupling model. Based on the multidimensional physical field coupling model and pre-processed material properties and environmental parameters, it simulates the risk evolution of the dynamic workshop equipment geometric model, obtains the corresponding workshop risk factors, and implements the corresponding solutions. The intelligent early warning module is based on a neural convolutional network and constructs a production workshop early warning model according to workshop risk factors, thereby generating a production workshop early warning score, and generating a graded response early warning based on the production workshop early warning score.

[0015] The present invention further provides a computer-readable storage medium, which stores a computer program. The computer program can be executed by a processor to implement the above-mentioned early warning flame retardant and explosion suppression method for a production workshop.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. Kalman filtering technology pre-processes equipment parameters, material properties, and environmental parameters, effectively filtering out noise and repairing abnormal data to ensure the reliability of input data. Combining a dynamically calibrated workshop equipment geometry model with a multi-dimensional physical field coupling model, it can simulate complex physical processes such as dust diffusion, heat conduction, and spark generation in real time, accurately capturing the hidden risks caused by multi-parameter coupling.

[0017] 2. The early warning model, built using a convolutional neural network, automatically learns the characteristic associations in historical risk data, generates dynamic early warning scores, and supports graded responses. Through a closed loop of "data collection - simulation - strategy execution - effect feedback," it self-optimizes early warning thresholds and control strategies based on actual operating conditions, significantly reducing false alarm rates and significantly improving production shop safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0019] Figure 1 The figure is a schematic diagram of the steps of an early warning flame retardant and explosion suppression method in a production workshop.

[0020] Figure 2 This is a process judgment diagram of an early warning flame retardant and explosion suppression method and system for a production workshop.

[0021] Figure 3 This is a module diagram of an early warning flame retardant and explosion suppression system for a production workshop. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other implementations obtained by those of ordinary skill in the art without inventive effort are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, a method for early warning flame retardant and explosion suppression in a production workshop includes the following steps: Collect equipment parameters, material properties, and environmental parameters in the production workshop, and pre-process them based on Kalman filtering technology; based on Technology, and according to the pre-processed equipment parameters, construct a workshop equipment geometric model, dynamically calibrate the workshop equipment geometric model, and obtain a dynamic workshop equipment geometric model; Based on multidimensional modeling technology, a multidimensional physical field coupling model is constructed. According to the multidimensional physical field coupling model and the pre-processed material properties and environmental parameters, the risk evolution of the dynamic workshop equipment geometry model is simulated to obtain the corresponding workshop risk factors, and then the corresponding solutions are implemented; Based on the neural convolutional network and according to the workshop risk factors, a production workshop early warning model is constructed to generate a production workshop early warning score, and a graded response early warning is generated according to the production workshop early warning score.

[0024] It should be further explained that, during the specific implementation process, the specific process of collecting equipment parameters, material properties and environmental parameters in the production workshop includes: A data acquisition device is preset; the data acquisition device includes: an equipment parameter acquisition unit, a material attribute acquisition unit, and an environmental parameter acquisition unit; and a corresponding acquisition cycle is set, the acquisition cycle includes several acquisition moments; The equipment parameter acquisition unit is used to acquire equipment parameters in the production workshop; the material attribute acquisition unit is used to acquire material attributes in the production workshop; the environmental parameter acquisition unit is used to acquire environmental parameters in the production workshop; It should be further explained that equipment parameters include the operating parameters of the equipment, such as temperature, pressure, speed, current, voltage, etc., as well as the location information of the equipment, such as the coordinate position of the equipment in the production workshop; material properties include the physical properties of the materials in the production workshop, such as density, specific heat capacity, thermal conductivity, etc., as well as the chemical properties of the materials, such as chemical composition, reactivity, etc.; environmental parameters include the temperature, humidity, air pressure, light intensity, noise level, etc. in the production workshop.

[0025] For example, in a production workshop, each device requiring location monitoring is equipped with a location sensor, such as a GPS positioning module (suitable for larger workshops where precision is not a priority), a Bluetooth-based beacon and receiver combination (suitable for indoor positioning with relatively high precision), or a UWB (ultra-wideband) positioning system (which offers high centimeter-level accuracy and is suitable for workshop equipment with high precision requirements). These sensors collect the device's location coordinates in real time.

[0026] It should be further explained that, in the specific implementation process, the specific process of pre-processing equipment parameters, material properties and environmental parameters based on Kalman filtering technology includes: Set up the preprocessing state equation: ;in, Collection time The state vector of is the state transfer matrix; is the input vector; is the input matrix; is the process noise vector, and , is the covariance matrix of the process noise.

[0027] Set up the preprocessing measurement equation: ;in, is the measurement vector; is the measurement matrix; The noise vector is measured, and , is the covariance matrix of the measurement noise.

[0028] Based on Kalman filtering, the steps for predicting data are: ;in, is the state prediction value; is the prediction error covariance matrix.

[0029] The steps to update the data are: ;in, is the Kalman gain; is the preprocessed data; is the estimated error covariance matrix; According to the prediction step and the update step, the corresponding pre-processing equipment parameters, material properties and environmental parameters are obtained; and the influence of noise on subsequent analysis is reduced.

[0030] For example, set the device at time The temperature is recorded as ; If the temperature of the device changes slowly at adjacent moments, the state transfer matrix ; Measurement matrix ; Input vector ; Process noise refers to the temperature variation of the equipment at adjacent moments due to some unpredictable factors (such as small thermal fluctuations inside the equipment). ,in, , which represents the variance of the process noise, is set based on experience or preliminary understanding of the equipment.

[0031] Therefore, the pre-processing state equation of the equipment corresponding to the temperature is: ; The pre-processing measurement equation of the device's corresponding temperature: ;in, Collection time The measured temperature, is the temperature measurement noise, which reflects the measurement error of the corresponding sensor; , according to the accuracy index of the sensor, set , represents the variance of the measurement noise.

[0032] According to the normal operating temperature range of the device, it is assumed that the estimated value of the device temperature at the initial moment is ; Set the initial estimate error covariance matrix according to the confidence level of the initial estimate ; It should be further explained that the covariance matrix The size of represents the uncertainty of the initial estimate, with larger values indicating higher uncertainty.

[0033] Assume that at the first moment, the temperature collected by the corresponding sensor is ; Based on Kalman filtering, the steps for predicting temperature are: ; The steps to update the temperature are: ; Assume that at the second moment, the temperature collected by the corresponding sensor is ; Based on Kalman filtering, the steps for predicting temperature are: ; The steps to update the temperature are: .

[0034] It should be further explained that, in the specific implementation process, based on The specific process of building the geometric model of workshop equipment based on the pre-processed equipment parameters includes: Obtaining preprocessed equipment parameters, and obtaining a distribution map of the equipment in the production workshop based on the preprocessed equipment parameters; Import the plan, elevation, and section drawings of the production workshop into the BIM modeling software to construct a geometric model of the workshop; The distribution map of the equipment in the production workshop is imported into the workshop geometric model, and the workshop equipment geometric model is constructed according to the actual size of the production workshop and the corresponding equipment.

[0035] It should be further explained that, in the specific implementation process, the specific process of dynamically calibrating the geometric model of the workshop equipment and obtaining the dynamic geometric model of the workshop equipment includes: At the time of collection , get the location coordinates of the device ; Compared with the last collection time Location coordinates Calculate and obtain the device offset distance ; Preset device offset distance threshold ; If the device is offset , no dynamic calibration is performed; If the device is offset , then dynamically update the corresponding equipment to reduce accidents caused by equipment deviation; Offset the device by a distance The position coordinates of the equipment at that time are mapped to the workshop equipment geometric model to generate a dynamic workshop equipment geometric model.

[0036] It should be further explained that, in the specific implementation process, based on the multidimensional model technology, the specific process of building a multidimensional physical field coupling model includes: It should be further explained that the multi-dimensional model technology includes computational fluid dynamics algorithm, Fourier equations and Maxwell equations; the multi-dimensional physical field coupling model includes: dust diffusion model, heat conduction model and electric spark generation model; Based on the computational fluid dynamics algorithm, the specific process of building a dust diffusion model includes: Use unstructured grids to divide the production workshop into grids, and set the grid density according to the actual situation of the production workshop; For example, near dust sources and around ventilation holes, the grid is encrypted to improve calculation accuracy; in areas with relatively gentle changes, the grid density is appropriately reduced to reduce the amount of calculation.

[0037] According to the actual dust concentration requirements of the production workshop, the turbulence model is selected as the basic model; it should be further explained that the basic model is not limited to the turbulence model, and the Euler-Lagrange model and the Euler-Euler model can be selected according to actual conditions.

[0038] It should be further explained that the Euler-Lagrangian model is suitable for situations with low dust concentration, the Euler-Euler model is suitable for situations with high dust concentration and strong interactions between particles, and the turbulence model is used for scenarios with high accuracy requirements; Set dust diffusion boundary conditions, including inlet boundary conditions, outlet boundary conditions and wall boundary conditions; The inlet boundary conditions are used to further define the parameters at the inlets of ventilation openings, dust sources, etc., including: 1. Inlet velocity boundary conditions, fluid velocity ;in, is the rated air volume of the fan; is the cross-sectional area of the ventilation duct; for adjustable ventilation systems, consider the velocity changes under different working conditions; 2. Temperature boundary conditions: When ventilation air passes through heating or cooling equipment, the outlet temperature is determined based on the equipment's operating parameters. If ventilation air comes from the outside environment, the temperature can be obtained based on local meteorological data or real-time monitoring. 3. Dust concentration boundary conditions: Based on the production process and equipment operating parameters, estimate the generation rate and spatial distribution of dust sources to determine the dust concentration; for intermittent dust generation processes, it is necessary to consider the changes in dust concentration in different time periods.

[0039] The outlet boundary conditions include: 1. Pressure boundary conditions: For outlets connected to the atmosphere, the pressure is usually set to the local atmospheric pressure. For outlets connected to other equipment or piping systems, the outlet pressure needs to match the pressure of the connected system. 2. Outlet flow rate boundary condition. When pressure is used as the outlet boundary condition, the flow rate is obtained by solving the flow field. When setting the outlet boundary condition, it is generally not necessary to directly specify the flow rate. It is necessary to ensure the rationality of the pressure value to ensure that the calculated flow rate conforms to the actual situation.

[0040] The wall boundary conditions include: 1. No-slip boundary condition, that is, for solid walls such as workshop walls and equipment surfaces, the velocity of the fluid at the wall is zero; 2. Adsorption boundary conditions: Different wall materials (such as metal, plastic, concrete, etc.) have different adsorption capacities for dust. The adsorption coefficient of the wall material is obtained based on the actual situation. 3. Deposition boundary conditions: A deposition model is used to calculate the deposition rate of dust on the wall. Based on parameters such as dust particle size distribution, flow velocity, and wall characteristics, the deposition rate of dust of different particle sizes is calculated, thereby obtaining the change of dust deposition on the wall over time. 4. Resuspension boundary conditions: When dust deposited on the wall is disturbed by the airflow or other external forces, it may be resuspended and re-enter the airflow. The shear force of the airflow and the adhesion force between the dust and the wall are used to determine whether the dust will be resuspended, and the amount of resuspended dust is calculated.

[0041] A dust diffusion model is constructed according to the unstructured grid, grid density, basic model and dust diffusion boundary conditions.

[0042] Based on the Fourier equation, the specific process of constructing the heat conduction model includes: Select the unstructured grid type to mesh the production workshop and equipment, and set the grid density according to the actual situation of the production workshop and equipment; It should be further explained that for areas with drastic temperature changes, such as near heat sources and heat dissipation surfaces, the grid should be denser; for areas with gentle temperature changes, the grid density should be appropriately reduced.

[0043] Set the heat conduction boundary conditions, including: 1. The first type of boundary condition: the temperature value on the given boundary For example, if the temperature of the part of the equipment surface that contacts the coolant is known to be constant, the boundary of that part can be set as the first type of boundary condition.

[0044] 2. The second type of boundary condition: heat flux density on a given boundary ,in is the normal direction of the boundary. For example, when a device exchanges heat with its surroundings, if the heat dissipation power of the device surface is known, this part of the boundary can be set as a second-class boundary condition.

[0045] 3. The third type of boundary condition: convection heat transfer coefficient on a given boundary and the temperature of the surrounding fluid , the boundary conditions are: For example, when conducting heat in pipes in a production workshop, the third type of boundary condition can be used to consider the convection heat transfer between the pipes and the surrounding air.

[0046] A heat conduction model is constructed based on unstructured grids, grid density, and heat conduction boundary conditions.

[0047] Based on Maxwell's equations, the specific process of building an EDM generation model includes: Select the unstructured grid type to mesh the production workshop and equipment, and set the grid density according to the actual situation of the production workshop and equipment; It should be further explained that due to the dramatic changes in the electromagnetic field during spark generation, the mesh needs to be denser in key areas, such as near the electrodes and in the plasma region, to accurately capture these changes. Choose an appropriate mesh type, such as tetrahedral or hexahedral.

[0048] Set the boundary conditions for spark generation, including: 1. Electric field boundary condition: the potential on the conductor surface is constant , the normal electric field strength of the insulating surface is zero, that is, ; 2. Magnetic field boundary conditions: continuity of magnetic flux at the boundary , the magnetic field is zero at infinity .

[0049] Based on the finite-difference time-domain method, the time-domain form of Maxwell's equations is: ; ; ;in, is the electric field strength, is the magnetic field strength, is the electric displacement vector, is the magnetic induction intensity, is the current density, is the charge density; Get the electric and magnetic fields in The update formula in the direction is: ;in, Represents the index of the spatial grid node; It represents the dielectric constant of the medium at that position in space; 、 Respectively The spatial step length in the direction; Represents the index of the time step; represents the time step; 、 Respectively represent time steps and time steps, the spatial position ( The electric field intensity component at ; 、 、 Represents a specific time step and spatial position ( The magnetic field strength component at .

[0050] Similarly, the electric and magnetic fields are obtained The updated formula in the direction and the obtained electric and magnetic fields in Update formula in direction.

[0051] An EDM model is constructed based on unstructured grid, grid density, EDM boundary conditions and update formulas.

[0052] It should be further explained that during the specific implementation process, the risk evolution of the dynamic workshop equipment geometry model is simulated based on the multi-dimensional physical field coupling model and the pre-processed material properties and environmental parameters to obtain the corresponding workshop risk factors, and then the specific process of implementing the corresponding solution includes: Obtain dust diffusion model, heat conduction model, spark generation model, and pre-processed material properties and environmental parameters; Using the pretreated material properties and environmental parameters as initial state conditions of the physical field, and using the dust diffusion model, heat conduction model, and spark generation model as boundary conditions; Importing the initial state conditions and boundary conditions into the dynamic workshop equipment geometric model, and setting risk indicator thresholds, including: a high concentration dust risk indicator threshold, an equipment thermal anomaly risk indicator threshold, and an electric spark risk indicator threshold; According to different boundary conditions, the corresponding risk indicators are output, including: high concentration dust risk indicator, equipment thermal anomaly risk indicator and electric spark risk indicator; If the high-concentration dust risk index is less than or equal to the high-concentration dust risk index threshold, the production workshop has a lower probability of dust explosion risk. If the high-concentration dust risk index is greater than the high-concentration dust risk index threshold, the production workshop has a higher probability of dust explosion risk. The simulation outputs the corresponding workshop risk factor and reminds relevant personnel to take corresponding solutions, such as turning on the exhaust fan for ventilation, and uploads the records to the background for storage through wireless transmission technology.

[0053] If the equipment thermal anomaly risk index is less than or equal to the equipment thermal anomaly risk index threshold, the probability of the equipment in the production workshop generating thermal anomaly risk is smaller. If the equipment thermal anomaly risk index is greater than the equipment thermal anomaly risk index threshold, the probability of the equipment in the production workshop generating thermal anomaly risk is greater. The simulation outputs the corresponding workshop risk factor and reminds relevant personnel to take corresponding solutions, such as checking whether the equipment is overloaded, etc., and uploads the records to the background for storage through wireless transmission technology.

[0054] If the spark risk index is less than or equal to the spark risk index threshold, the probability of the production workshop's lines generating spark risk is smaller. If the spark risk index is greater than the spark risk index threshold, the probability of the production workshop's lines generating spark risk is greater. The simulation outputs the corresponding workshop risk factor and reminds relevant personnel to take corresponding solutions, such as checking whether the equipment's lines are damaged, etc., and uploads the records to the background for storage through wireless transmission technology.

[0055] It should be further explained that, in the specific implementation process, based on the neural convolutional network and according to the workshop risk factors, a production workshop early warning model is constructed, and then a production workshop early warning score is generated. Based on the production workshop early warning score, the specific process of generating a graded response early warning includes:

[0056] Obtain historical workshop risk factors for several groups; Group and label several groups of historical workshop risk factors, denoted as is a natural number; Will The historical workshop risk factors of the group are used as sample data, and is less than The sample data is used to obtain the mean of the sample data, which is recorded as a sample set; the remaining groups of historical workshop risk factors are used as test sets; A training sample set is formed according to the sample set and the test set; Based on convolutional network neural networks, a standard risk warning model is constructed; The training sample set is input into the standard risk warning model to train the standard risk warning model, and the standard risk warning model after training is recorded as the production workshop warning model.

[0057] It should be further explained that, in the specific implementation process, the specific process of generating graded response warnings based on the production workshop warning scores includes: According to the production workshop early warning model, the production workshop early warning score under the current environmental factors is generated , the production workshop early warning score for: ;in, The convolutional network neural network output feature map is the first The weight corresponding to the convolution kernel; is the workshop risk factor corresponding to the collection time; It is The bias value corresponding to the convolution kernel.

[0058] Preset production workshop warning threshold range ; If the production workshop early warning score When the production workshop is safe, there is no risk and it is considered to be in a safe operating state, and the corresponding data is recorded in real time; If the production workshop early warning score When the alarm is raised, there is an accident risk in the production workshop, which is identified as a medium-level warning, triggering local alarms and remote notifications, and reminding relevant maintenance personnel to perform corresponding maintenance on the equipment; If the production workshop early warning score When the alarm is triggered, there is an accident risk in the production workshop, which is identified as a high-level alarm and triggers a full alarm. The siren will sound an on-site warning, the on-site video will be saved, the power supply equipment will be cut off in time, and the maintenance personnel will be reminded to implement the corresponding emergency treatment measures.

[0059] like Figure 3 As shown, a flame retardant and explosion suppression early warning system for a production workshop includes: a data acquisition module, a data preprocessing module, a workshop geometry model building module, a simulation analysis engine module, and an intelligent early warning module; Data acquisition module, used to collect equipment parameters, material properties and environmental parameters in the production workshop; The data preprocessing module preprocesses equipment parameters, material properties, and environmental parameters based on Kalman filtering technology; Workshop geometry model building module, based on Technology, and according to the pre-processed equipment parameters, construct a workshop equipment geometric model, dynamically calibrate the workshop equipment geometric model, and obtain a dynamic workshop equipment geometric model; The simulation analysis engine module, based on multidimensional model technology, constructs a multidimensional physical field coupling model. Based on the multidimensional physical field coupling model and pre-processed material properties and environmental parameters, it simulates the risk evolution of the dynamic workshop equipment geometric model, obtains the corresponding workshop risk factors, and implements the corresponding solutions. The intelligent early warning module is based on a neural convolutional network and constructs a production workshop early warning model according to workshop risk factors, thereby generating a production workshop early warning score, and generating a graded response early warning based on the production workshop early warning score.

[0060] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for early warning flame retardant and explosion suppression in a production workshop, characterized in that: The method comprises: Collect equipment parameters, material properties, and environmental parameters in the production workshop, and pre-process them based on Kalman filtering technology; based on Technology, and according to the pre-processed equipment parameters, construct a workshop equipment geometric model, dynamically calibrate the workshop equipment geometric model, and obtain a dynamic workshop equipment geometric model; Based on the computational fluid dynamics algorithm, the dust diffusion model is constructed in the following steps: using an unstructured grid to divide the production workshop into grids and setting the grid density according to the actual production workshop conditions; selecting a turbulence model as the base model based on the actual dust concentration requirements of the production workshop; setting dust diffusion boundary conditions; and constructing the dust diffusion model based on the unstructured grid, grid density, base model, and dust diffusion boundary conditions. Based on the Fourier equation, the process of constructing a heat conduction model includes: selecting an unstructured grid as the grid type to mesh the production workshop and equipment, and setting the grid density according to the actual situation of the production workshop and equipment; setting heat conduction boundary conditions; and constructing a heat conduction model based on the unstructured grid, grid density, and heat conduction boundary conditions. Based on Maxwell's equations, the process of constructing an EDM generation model includes: selecting an unstructured grid as the grid type to mesh the production workshop and equipment, and setting the grid density according to the actual conditions of the production workshop and equipment; setting the EDM generation boundary conditions; and constructing the EDM generation model based on the unstructured grid, grid density, and EDM generation boundary conditions. Simulating the risk evolution of the dynamic workshop equipment geometry model based on the dust diffusion model, heat conduction model, and spark generation model to obtain the corresponding workshop risk factor and then implement the corresponding solution; Based on the neural convolutional network and according to the workshop risk factors, a production workshop early warning model is constructed to generate a production workshop early warning score, and a graded response early warning is generated according to the production workshop early warning score.

2. The early warning flame retardant and explosion suppression method for a production workshop according to claim 1, characterized in that: The process of collecting equipment parameters, material properties, and environmental parameters in the production workshop includes: A data acquisition device is preset; the data acquisition device includes: an equipment parameter acquisition unit, a material attribute acquisition unit, and an environmental parameter acquisition unit; and a corresponding acquisition cycle is set, the acquisition cycle includes several acquisition moments; The equipment parameter acquisition unit is used to acquire equipment parameters in the production workshop; the material attribute acquisition unit is used to acquire material attributes in the production workshop; and the environmental parameter acquisition unit is used to acquire environmental parameters in the production workshop.

3. The early warning flame retardant and explosion suppression method for a production workshop according to claim 2, characterized in that: The process of preprocessing equipment parameters, material properties, and environmental parameters includes: Set up the preprocessing state equation: ;in, Collection time The state vector of is the state transfer matrix; is the input vector; is the input matrix; is the process noise vector, and , is the covariance matrix of process noise; Set up the preprocessing measurement equation: ;in, is the measurement vector; is the measurement matrix; The noise vector is measured, and , is the covariance matrix of the measurement noise; Based on Kalman filtering, the steps for predicting data are: ;in, is the state prediction value; is the forecast error covariance matrix; The steps to update the data are: ;in, is the Kalman gain; is the preprocessed data; is the estimated error covariance matrix; According to the prediction step and the update step, the corresponding pre-processing equipment parameters, material properties and environmental parameters are obtained.

4. The early warning flame retardant and explosion suppression method for a production workshop according to claim 3, characterized in that: The process of building a geometric model of workshop equipment includes: Obtaining preprocessed equipment parameters, and obtaining a distribution map of the equipment in the production workshop based on the preprocessed equipment parameters; Import the plan, elevation, and section drawings of the production workshop into the BIM modeling software to construct a geometric model of the workshop; The distribution map of the equipment in the production workshop is imported into the workshop geometric model, and the workshop equipment geometric model is constructed according to the actual size of the production workshop and the corresponding equipment.

5. The early warning flame retardant and explosion suppression method for a production workshop according to claim 4, characterized in that: The process of dynamically calibrating the geometric model of workshop equipment to obtain the dynamic geometric model of workshop equipment includes: At the time of collection , get the location coordinates of the device ; Compared with the last collection time Location coordinates Calculate and obtain the device offset distance ; Preset device offset distance threshold ; If the device offset distance , no dynamic calibration is performed; If the device offset distance , then dynamically update the corresponding device; Offset the device by a distance The position coordinates of the equipment at that time are mapped to the workshop equipment geometric model to generate a dynamic workshop equipment geometric model.

6. The early warning flame retardant and explosion suppression method for a production workshop according to claim 5, characterized in that: Based on multidimensional model technology, the process of building a multidimensional physical field coupling model includes: The multi-dimensional model technology includes computational fluid dynamics algorithm, Fourier equations and Maxwell equations; the multi-dimensional physical field coupling model includes: dust diffusion model, heat conduction model and electric spark generation model; Based on the computational fluid dynamics algorithm, the process of building a dust diffusion model includes: The production workshop is meshed using an unstructured grid, and the grid density is set according to the actual situation of the production workshop. The turbulence model is selected as the basic model based on the actual dust concentration requirements of the production workshop. The dust diffusion boundary conditions are set, including inlet boundary conditions, outlet boundary conditions, and wall boundary conditions. Constructing a dust diffusion model according to the unstructured grid, grid density, basic model, inlet boundary condition, outlet boundary condition and wall boundary condition; Based on the Fourier equation, the process of building a heat conduction model includes: Select the unstructured grid type to mesh the production workshop and equipment, and set the grid density according to the actual situation of the production workshop and equipment; Set heat conduction boundary conditions, including: first-class boundary conditions, second-class boundary conditions, and third-class boundary conditions; Construct a heat conduction model based on unstructured grid, grid density, first-class boundary conditions, second-class boundary conditions, and third-class boundary conditions; Based on Maxwell's equations, the process of building an EDM generation model includes: Select the unstructured grid type to mesh the production workshop and equipment, and set the grid density according to the actual situation of the production workshop and equipment; Setting the boundary conditions for spark generation, including electric field boundary conditions and magnetic field boundary conditions; An EDM generation model is constructed based on unstructured grids, grid density, electric field boundary conditions, and magnetic field boundary conditions.

7. The early warning flame retardant and explosion suppression method for a production workshop according to claim 6, characterized in that: The process of simulating the risk evolution of a dynamic workshop equipment geometry model, obtaining the corresponding workshop risk factors, and then implementing the corresponding solutions includes: The pre-treated material properties and environmental parameters are used as the initial state conditions of the physical field, and the dust diffusion model, heat conduction model, and spark generation model are used as boundary conditions; Importing the initial state conditions and boundary conditions into the dynamic workshop equipment geometric model, and setting risk indicator thresholds, including: a high concentration dust risk indicator threshold, an equipment thermal anomaly risk indicator threshold, and an electric spark risk indicator threshold; According to different boundary conditions, the corresponding risk indicators are output, including: high concentration dust risk indicator, equipment thermal anomaly risk indicator and electric spark risk indicator; If the high-concentration dust risk index is less than or equal to the high-concentration dust risk index threshold, the production workshop has a lower probability of dust explosion risk. If the high-concentration dust risk index is greater than the high-concentration dust risk index threshold, the production workshop has a higher probability of dust explosion risk. The simulation outputs the corresponding workshop risk factor, reminds relevant personnel to take corresponding solutions, and uploads the records to the backend for storage via wireless transmission technology. If the equipment thermal anomaly risk index is less than or equal to the equipment thermal anomaly risk index threshold, the probability of the equipment in the production workshop generating thermal anomaly risk is smaller. If the equipment thermal anomaly risk index is greater than the equipment thermal anomaly risk index threshold, the probability of the equipment in the production workshop generating thermal anomaly risk is greater. The simulation outputs the corresponding workshop risk factor, reminds relevant personnel to take corresponding solutions, and uploads the records to the background for storage through wireless transmission technology. If the spark risk index is less than or equal to the spark risk index threshold, the probability of the production workshop's lines generating spark risk is smaller. If the spark risk index is greater than the spark risk index threshold, the probability of the production workshop's lines generating spark risk is greater. The simulation outputs the corresponding workshop risk factor, reminds relevant personnel to take corresponding solutions, and uploads the records to the background for storage through wireless transmission technology.

8. The early warning flame retardant and explosion suppression method for a production workshop according to claim 7, characterized in that: The process of constructing a production workshop early warning model, generating a production workshop early warning score, and generating a graded response early warning based on the production workshop early warning score includes: Obtain several groups of historical workshop risk factors; group and label several groups of historical workshop risk factors, and record them as is a natural number; The historical workshop risk factors of the group are used as sample data, and is less than and using the sample data to obtain the mean of the sample data, which is recorded as a sample set; using the remaining groups of historical workshop risk factors as test sets; and forming a training sample set based on the sample set and the test set; Based on convolutional network neural networks, a standard risk warning model is constructed; The training sample set is input into the standard risk warning model to train the standard risk warning model, and the trained standard risk warning model is recorded as the production workshop warning model; According to the production workshop early warning model, the production workshop early warning score under the current environmental factors is generated , the production workshop early warning score for: ;in, The convolutional network neural network output feature map is the first The weight corresponding to the convolution kernel; is the workshop risk factor corresponding to the collection time; It is The bias value corresponding to the convolution kernel; Preset production workshop warning threshold range If the production workshop early warning score If the production workshop has no risk, it is deemed to be in a safe operating state and the corresponding data is recorded in real time; if the production workshop early warning score is If the production workshop has an accident risk, it will be identified as a medium-level warning, triggering local alarms and remote notifications, and reminding relevant maintenance personnel to perform corresponding maintenance on the equipment; if the production workshop warning score is When the alarm is triggered, there is an accident risk in the production workshop, which is identified as a high-level alarm and triggers a full alarm. The siren will sound an on-site warning, the on-site video will be saved, the power supply equipment will be cut off in time, and the maintenance personnel will be reminded to implement the corresponding emergency treatment measures.

9. An early warning flame retardant and explosion suppression system for a production workshop, implementing the early warning flame retardant and explosion suppression method for a production workshop as described in any one of claims 1 to 8, comprising: Data acquisition module, used to collect equipment parameters, material properties and environmental parameters in the production workshop; The data preprocessing module preprocesses equipment parameters, material properties, and environmental parameters based on Kalman filtering technology; Workshop geometry model building module, based on Technology, and according to the pre-processed equipment parameters, construct a workshop equipment geometric model, dynamically calibrate the workshop equipment geometric model, and obtain a dynamic workshop equipment geometric model; The simulation analysis engine module, based on multidimensional model technology, constructs a multidimensional physical field coupling model. Based on the multidimensional physical field coupling model and pre-processed material properties and environmental parameters, it simulates the risk evolution of the dynamic workshop equipment geometric model, obtains the corresponding workshop risk factors, and implements the corresponding solutions. The intelligent early warning module is based on a neural convolutional network and constructs a production workshop early warning model according to workshop risk factors, thereby generating a production workshop early warning score, and generating a graded response early warning based on the production workshop early warning score.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement the early warning flame retardant and explosion suppression method for a production workshop as described in any one of claims 1 to 8.

Images