Explosive safety prediction method and system based on digital twin model

Abstract

The present application relates to the technical field of safety prediction, and particularly relates to an explosive safety prediction method and system based on a digital twin model. The method comprises the following steps: obtaining original explosive data and extracting particle size distribution data, constructing a simulation explosive model with equal proportions based on the data, recording the explosive forming process, extracting and labeling working condition characteristics to obtain explosive working condition data at each stage, and performing real-time stability analysis to generate explosive real-time stability data and build a stability environment field, simulating the stacking of the simulation explosive model, obtaining simulated stacking data and inferring stacking pressure bearing data, performing real-time stacking safety evaluation according to the stacking pressure bearing and working condition data, calculating the stacking cumulative bearing pressure, using the stacking cumulative bearing pressure to predict continuous pressure deformation, and combining real-time safety to comprehensively evaluate safety. The present application realizes a more efficient and safer explosive safety prediction method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety prediction, in particular to an explosive safety prediction method and system based on a digital twin model. BACKGROUND

[0002] Traditional explosive safety prediction methods based on digital twin models often rely on experience and static models for safety evaluation, lacking dynamic monitoring of explosive performance and environmental changes, leading to the inability to identify potential safety hazards in time in actual applications, especially in high-risk blasting operations, where the stability and stacking safety of explosives are often overlooked, significantly increasing the risk of accidents. Existing technologies often fail to conduct comprehensive data analysis when dealing with explosive particle size distribution and forming process characteristics, resulting in insufficient accuracy and reliability of the model, which in turn affects the effectiveness of safety evaluation. Traditional explosive stacking simulation methods often lack integration of real-time working condition data and fail to establish a dynamic stability environment field, making it impossible to reflect the actual situation of explosive stacking pressure evaluation, especially under complex storage and transportation conditions, where the deformation and stability changes of explosives are difficult to predict. The lack of scientific evaluation tools makes it difficult to implement safety management measures, and the evaluation of stacking safety in the research often relies on single static data, failing to consider the impact of multiple working conditions on explosive performance. SUMMARY

[0003] Therefore, it is necessary to provide an explosive safety prediction method and system based on a digital twin model to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, the explosive safety prediction method based on the digital twin model comprises the following steps:

[0005] Step S1: obtaining original explosive data; extracting particle size distribution data from the original explosive data; and constructing the original explosive data into a simulation explosive model with equal proportions based on the particle size distribution data;

[0006] Step S2: recording the explosive forming process; extracting working condition characteristics in the explosive forming process, and labeling explosive working condition data of each stage of the explosive forming process based on the working condition characteristics;

[0007] Step S3: performing real-time stability analysis of explosives through explosive working condition data to obtain real-time stability data of explosives; and building an explosive stability environment field according to the real-time stability data of explosives;

[0008] Step S4: simulating explosive stacking storage of the simulation explosive model based on the explosive stability environment field to obtain simulation explosive stacking data; and deducing explosive stacking pressure data according to the simulation explosive stacking data;

[0009] Step S5: Real-time stack safety evaluation is performed according to the explosive stack pressure bearing and explosive working condition data to generate real-time explosive stack safety; the stack cumulative bearing pressure is calculated through the explosive stack pressure bearing data;

[0010] Step S6: Continuous pressure deformation prediction is performed on the simulation explosive model based on the stack cumulative bearing pressure to obtain predicted explosive deformation data; continuous stack safety prediction is performed according to the predicted explosive deformation data, and safety evaluation is performed in combination with the real-time explosive stack safety.

[0011] The present application ensures the accuracy and comprehensiveness of the data by obtaining the original explosive data, and the extracted particle size distribution data provides the necessary basis for the construction of the simulation model. The equal-scale simulation explosive model generated based on these data realizes the accurate simulation of the explosive characteristics, significantly improving the reliability and practicality of the model. Recording the explosive forming process ensures comprehensive analysis of the process characteristics at each stage, and the extracted working condition characteristics provide key support for the stability evaluation of the explosive. Real-time stability analysis based on dynamic monitoring of working condition data makes the evaluation of explosive stability more scientific and accurate. The stability environment field built provides environmental basis for subsequent stack storage simulation. The explosive stack storage simulation combined with the stability environment field effectively reflects the stack bearing pressure under different working conditions. The inferred stack bearing pressure data provides an important parameter for safety evaluation. Real-time stack safety evaluation combined with working condition data improves the scientificity and effectiveness of management decisions. The calculation of the stack cumulative bearing pressure provides data support for long-term safety. Continuous pressure deformation prediction combined with real-time data provides an early warning mechanism for long-term storage safety of explosives. The comprehensive safety evaluation system formed promotes the intelligentization and modernization of explosive management, significantly improves the safety level of the industry, ensures the stability and safety of explosives under various use conditions, and provides strong technical support for the safety management of future explosive applications. It promotes the technological progress and standardized development of the explosive industry, and ensures the efficiency and safety of explosives in the production, storage and use process.

[0012] The present application also provides an explosive safety prediction system based on a digital twin model, which is used to perform the explosive safety prediction method based on a digital twin model as described above. The explosive safety prediction system based on a digital twin model comprises:

[0013] The model simulation module is configured to obtain original explosive data; extract particle size distribution data from the original explosive data; and construct the original explosive data into an equal-scale simulation explosive model based on the particle size distribution data.

[0014] The working condition recording module is configured to record the explosive forming process; extract working condition characteristics in the explosive forming process; and label explosive working condition data of each stage of the explosive forming process based on the working condition characteristics.

[0015] A stability analysis module is configured to perform real-time stability analysis on the explosive based on explosive working condition data, to obtain real-time stability data of the explosive, and to build an explosive stability environment field based on the real-time stability data of the explosive.

[0016] A stacking simulation module is configured to perform explosive stacking storage simulation on a simulation explosive model based on the explosive stability environment field, to obtain simulation explosive stacking data, and to deduce explosive stacking bearing pressure data based on the simulation explosive stacking data.

[0017] A real-time safety evaluation module is configured to perform real-time stacking safety evaluation based on the explosive stacking bearing pressure and the explosive working condition data, to generate real-time explosive stacking safety, and to calculate stacking cumulative bearing pressure based on the explosive stacking bearing pressure data.

[0018] A safety joint evaluation module is configured to perform continuous pressure deformation prediction on the simulation explosive model based on the stacking cumulative bearing pressure, to obtain predicted explosive deformation data, to perform continuous stacking safety prediction based on the predicted explosive deformation data, and to perform safety evaluation in combination with the real-time explosive stacking safety.

[0019] The model simulation module realizes comprehensive acquisition of original explosive data, ensures the integrity and accuracy of the data, and the extracted particle size distribution data provides a scientific basis for the construction of the simulation explosive model, the generated simulation explosive model of the same proportion realizes real reproduction of explosive characteristics, and the reliability and practicality of the simulation are significantly improved, the working condition recording module can comprehensively record the explosive forming process, and the extracted working condition characteristics provide important support for subsequent stability analysis, the real-time stability analysis module enhances dynamic evaluation of the safety of the explosive through real-time monitoring of the explosive working condition data, the built stability environment field provides a scientific basis for simulation stacking storage, the stacking simulation module effectively reflects the bearing pressure under different stacking conditions in combination with the environment field, the deduced stacking bearing pressure data provides a key parameter for safety evaluation, the real-time safety evaluation module performs comprehensive stacking safety evaluation in combination with the working condition data and the bearing pressure, generates a real-time safety report, the calculation of the stacking cumulative bearing pressure provides data support for long-term safety, the safety joint evaluation module can perform continuous pressure deformation prediction, provides a warning for long-term storage and use of the explosive, the formed comprehensive safety evaluation system promotes the intelligentization and modernization of explosive management, significantly improves the safety standards and technical level of the industry, ensures the stability and safety of the explosive under various environments, provides strong technical support for future safety management of explosive application, promotes the technical progress and standardized development of the explosive industry, and improves the overall safety management capability and efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0020] Fig. 1A step flow diagram of a kind of explosive safety prediction method based on digital twin model is shown in the figure;

[0021] Fig. 2 A detailed implementation step flow diagram of step S2 is shown in the figure;

[0022] The implementation of the present application, functional features and advantages will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0023] The technical method of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0024] In addition, the accompanying drawings are only schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated description thereof will be omitted. Some block diagrams shown in the drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0025] It should be understood that although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be called a second element, and similarly a second element can be called a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0026] To achieve the above-mentioned purpose, please refer to Figs. 1-2 A kind of explosive safety prediction method based on digital twin model, comprising the following steps:

[0027] Step S1: obtaining original explosive data; extracting particle size distribution data in original explosive data; constructing original explosive data into a simulation explosive model with equal proportion based on particle size distribution data;

[0028] In this embodiment, by setting up a multi-channel laser particle size analyzer in the production line, the particle size data of the explosive raw materials passing above the conveying belt is continuously collected at a scanning frequency of 10 Hz. The single sampling result is calibrated by three multi-section screening. First, the particle size measurement range is set between 0.5 microns and 1500 microns, and a total of 300 particle size channels are constructed with every 5 microns as a particle size interval. Then, the original data collected is fitted with multi-interval distribution, and the error range is controlled within 2% by using the back projection calibration strategy. Then, based on the obtained particle size distribution data, a three-dimensional modeling module is used to construct a simulated explosive model using the OpenCASCADE modeling kernel driven by Python script. During modeling, the model is scaled at a volume ratio of 1:100. The particles in each particle size section are stacked in the simulation space according to the Poisson distribution rule, and the contact detector based on the face normal coincidence constraint is used to correct the particle contact boundary in the modeling process. Finally, the three-dimensional simulation explosive model data is obtained, which is stored in STL (Stereolithography) format with a face number of not less than 100000, ensuring that the simulation accuracy reaches the level of 0.1 mm.

[0029] Step S2: Record the explosive forming process; extract the working condition characteristics in the explosive forming process, and label the explosive working condition data of each stage of the explosive forming process based on the working condition characteristics;

[0030] In this embodiment, the full-process process data acquisition platform configured in the pressurized filling link is used to collect five core data parameters including the pressure rate of the pressure head, the filling mass change rate, the internal pressure of the mold cavity, the mold temperature and the vibration amplitude, with the pressure rate of the pressure head limited to 0.3 to 1.2 mm / s, the filling mass change rate limited to not more than 1.8 kg / min, and the mold cavity pressure maintained at 0.6 to 2.5 MPa. The data acquisition lasts for more than 180 minutes. On this basis, an explosive process time axis is established, each process node is divided into 15 seconds as the minimum granularity, the time axis is segmented by using a labeling algorithm, the whole process time is divided into four stages of pre-filling, steady-state pressurization, end compaction and stability maintenance by using a threshold difference segmentation method, then the working condition data of each stage is extracted, the extracted features include stability threshold variation rate, pressure fluctuation frequency and filling rate gradient, the stage process data is labeled by setting the sensitive range of each index and constructing an index space vector labeling template, and finally the explosive working condition data set with a unique working condition identification for each stage is generated.

[0031] Step S3: Perform real-time stability analysis of the explosive using explosive working condition data to obtain real-time stability data of the explosive; construct the explosive stability environment field based on the real-time stability data of the explosive;

[0032] In this embodiment, the explosive working condition data is reorganized into a structured time series matrix according to stages. The temporal states are extracted using gating units in an LSTM (Long Short-Term Memory) neural network architecture. The model input is a 20-dimensional working condition vector sequence with a step size of 10. The model is trained for 200 epochs using the Adam optimizer with a learning rate of 0.001. The model output is then mapped to a real-time stability score for the explosive. The stability score ranges from 0 to 1, and data with scores less than 0.35 are marked as unstable. All stability scores are input into a three-dimensional stability environment field construction module in the form of a structure. This module generates an environmental field tension domain based on the spatial gradient and temporal variation amplitude of the stability scores. The tension domain is divided using a three-dimensional mesh, with each cell being 1 cm in size. 3 And insert local stability vectors in each cell to form a complete explosive stability environment field.

[0033] Step S4: Simulate the storage of explosives in a simulated explosive stacking environment based on the explosive stability environment to obtain simulated explosive stacking data; infer the pressure data that the explosive stack will withstand based on the simulated explosive stacking data;

[0034] In this embodiment, the environmental field is used as the simulation background. The previously constructed simulated explosive model data is imported, and a physics engine (such as Bullet Physics) is used to perform a stacking simulation. The stacking rule is based on the decreasing particle size hierarchy. The simulation area is set to a three-dimensional space of 30cm×30cm×50cm. The total number of models is controlled between 1200 and 1600. The stacking order is arranged from low to high according to the stability field tension gradient. The simulation time is 30 minutes, and a snapshot of the stacking structure is output every 2 seconds. All simulation snapshots are combined to form simulated explosive stacking data. The local contact force on each explosive model is calculated by the contact point data recorded in the simulation. The pressure superposition structure formed by each stacking layer is deduced by combining the stacking sequence order. The pressure distribution curve is constructed using the Force Chain Propagation Model and archived as explosive stack pressure data.

[0035] Step S5: Perform a real-time stacking safety assessment based on the pressure borne by the explosive stack and the explosive working condition data to generate real-time explosive stacking safety; calculate the cumulative pressure borne by the stack based on the explosive stacking pressure data;

[0036] In this embodiment, according to the explosive stack pressure bearing data and explosive working condition data, the cross pressure mapping structure is used to evaluate the safety of the real-time explosive stack state, the pressure peak-working condition sensitive threshold mapping model is used for coupling processing, the stack stress threshold surface and the working condition disturbance response surface are established respectively, the intersection area is defined as the safety critical surface, the real-time explosive stack safety judgment standard is constructed, the real-time safety is scored once every 3 seconds, the real-time explosive stack safety score data is generated, and at the same time, the stack bearing pressure at all time points is vertically stacked and accumulated, the stress integral algorithm is used to accumulate every unit thickness in the stack height direction, the integral step is 2 mm, and the maximum cumulative layer is 25, so as to generate the stack cumulative bearing pressure data.

[0037] Step S6: continuously predicting the pressure deformation of the simulation explosive model based on the stack cumulative bearing pressure to obtain predicted explosive deformation data; continuously predicting the stack safety based on the predicted explosive deformation data, and combining the real-time explosive stack safety to evaluate the safety.

[0038] In this embodiment, based on the stack cumulative bearing pressure data, the static finite element structure analysis module is used to simulate the continuous pressure loading of the simulation explosive model, the simulation continuous loading time is 3600 seconds, the pressure loading is taken as a period every 60 seconds, the pressure value is increased by 0.1 MPa, the stress-strain coupling modeling is carried out by using the SolidWorks Simulation module, the simulation explosive model is divided into hexahedral element grid, the length of each element is not greater than 0.5 mm, the loading direction is adjusted in real time through the coupling change of the pressure loading path and the gravity center trajectory, so as to obtain the continuous pressure simulation data. On this basis, the foregoing explosive pressure response data is imported into the deformation response prediction network, the deformation conduction prediction algorithm based on graph convolutional neural network (GCN) is used to input the continuous pressure simulation data, the input dimension of each node is three, i.e. applied pressure, direction angle and material initial density, and the output is the deformation rate change in unit time. Through 10 rounds of prediction iteration, the predicted explosive deformation data is generated. Finally, the predicted explosive deformation data and the real-time explosive stack safety data are fused and compared in two axes, the final evaluation value is corrected based on the hyperbolic stability mapping formula, the stack safety grade of the simulation explosive model under the current working condition is evaluated, and the complete continuous stack safety prediction output data is formed.

[0039] Preferably, step S1 comprises the following steps:

[0040] Step S11: obtaining original explosive data; extracting particle basic physical property parameters in the original explosive data;

[0041] Step S12: Particle size classification measurement is performed according to the particle basic physical property parameters to obtain particle size distribution data, wherein a particle classification device is used, and the particle size screen specifications are set to 8 screen sections between 10-500 μm;

[0042] Step S13: Particle frequency ratio is constructed based on the particle size distribution data, wherein the frequency ratio value is limited within a ±5% error range, and the distribution ratio range of each particle size section is controlled to be 5%-30%;

[0043] Step S14: Three-dimensional packing simulation is performed on the original explosive data through the particle frequency ratio, so as to generate a simulated packing configuration, wherein the compaction degree of the simulated packing is limited to be between 62%-76%, and the simulation unit size is controlled to be within 10 cm 3

[0044] Step S15: A simulated explosive model of equal proportion is restored based on the simulated packing configuration combined with the particle basic physical property parameters.

[0045] In this embodiment, the physical property measurement operation is performed on the original explosive sample, a thermogravimetric analyzer (TGA) is used to obtain the data of the pyrolysis starting temperature, mass retention ratio, residual carbon component proportion and the like of the explosive within the range of 25-300 ℃ under normal pressure, a laser particle size analyzer (Malvern Mastersizer3000) is used to measure that the average particle size of the particles is 186 μm, the particle density is 1.73 g / cm 3 , and the static specific surface area is 0.98 m 2 ​ / cm, static friction coefficient 0.52, moisture content 1.8%, particle size distribution was measured by YBS-200 ultrasonic vibration sieve machine, the standard sieve mesh was 10 μm, 30 μm, 50 μm, 75 μm, 125 μm, 200 μm, 350 μm and 500 μm, the single sample feeding amount was 150 g, the amplitude was set to 3.5 mm, the sieving time was set to 15 min, the motor frequency was 45 Hz, after sieving, the residual particles on each screen were weighed and recorded, the statistical results showed that the particle size was mainly distributed between 50 μm and 350 μm, the highest proportion was 22.4% in the 125 μm segment, and the lowest proportion was 5.1% in the 30 μm segment, all the sieving data was standardized and converted into a frequency distribution curve, the SciPy library in Python environment was used for numerical interpolation fitting, the eight particle size segments were fitted with a continuous curve by using a cubic spline interpolation function, the error floating range was limited to ± 5%, the distribution limit interval of each particle size segment was set to 5% to 30%, and the weight was automatically adjusted to make the total particle size frequency sum equal to 100%, the completed frequency distribution model was drawn by Matplotlib to visualize the accuracy and smoothness of the distribution, the distribution value was stored as a structured JSON format as the input template for subsequent filling simulation particle size, the open source particle packing simulation tool Package Itasca PFC3D (Particle Flow Code in 3 Dimensions) was used, the simulation unit size was set to 5 cm x 2 cm x 1 cm, which was about 10 cm 3 The total number of particles was set to 3000 according to the internal control requirements, the particle size frequency data was input and the compaction degree parameter was set to 0.62 to 0.76 to automatically fit the compaction state, periodic boundary conditions were used in the simulation process to simulate the uniformity of the packing environment, the particle collision parameters were set to elastic modulus 2 GPa, Poisson's ratio 0.25 and contact friction coefficient 0.5, the simulation time was set to 6000 steps, the final compaction uniformity was 0.69, the space packing uniformity error was controlled within 3%, and the output configuration file was an.obj three-dimensional grid model, the.obj file was imported into the ANSYS SpaceClaim modeling environment, the particles in the configuration were marked as definable material domains and revalued according to the original particle basic physical parameters, including density 1.73 g / cm 3, the thermal conductivity is set to 0.24 W / (m·K), the elastic modulus is set to 1.3 GPa, the inter-particle bonding force is set to 0.2 N, and the inter-particle thermal resistance is set to 2.1 K / W. After the construction is completed, the model is transferred to the Digital Twin Studio platform environment, a coordinate system and structure mapping relationship consistent with the real explosive sample are established therein, the twin mapping and debugging of the model are completed through the VRML interface, the simulation explosive model has the ability of real mapping in the same proportion corresponding to the physical, geometric and thermal performance, and thus the digital twin explosive model data structure that can be used for subsequent forming process simulation and mechanical prediction is output.

[0046] Preferably, step S2 comprises the following steps:

[0047] Step S21: record the explosive forming process; sort the process nodes in the explosive forming process by time to generate an explosive process timeline, wherein the minimum granularity unit of each node is 15 seconds, and the total cycle is not less than 180 minutes;

[0048] Step S22: segment and mark the nodes of the explosive process timeline to generate a plurality of stage division data, wherein the stage segmentation length is set to be between 5-30 minutes;

[0049] Step S23: identify the working condition characteristics of each stage in the stage division data;

[0050] Step S24: label the explosive working condition data of each stage in the explosive forming process according to the working condition characteristics of each stage, wherein the data sampling frequency is limited to 1-10 Hz.

[0051] In this embodiment, Moxa ioLogik E1260 industrial remote data acquisition module is selected, which supports Modbus TCP communication protocol and can be real-time connected with PLC. It is installed in the key control nodes of the molding production line, including the feeding device, mixing device, mold system, vacuum treatment cabin and temperature control module. Process parameters are recorded every 15 seconds, including pressure (unit: MPa), temperature (unit: ℃), speed (unit: RPM), humidity (unit: %RH) and mold time (unit: s). Synchronous timestamp method is used for unified time marking. The recording period lasts for 192 minutes, covering the entire process of explosive molding. All data are connected to the OPC UA server interface of the central processing control unit, using a unified clock reference to form a process original time sequence matrix, with a sampling interval of 15 seconds for each time node. The time series data is structured processed by pandas and numpy libraries in Python. First, the time axis is tiled according to the time sequence, and then it is marked according to the set stage length range. The stage length is set to a fixed value of 20 minutes. In the 192-minute whole process, there are 9 complete stages and 1 tail stage, each containing 80 15-second time points. The tail stage is 12 minutes long and contains 48 time points. Each piece of data is named stage_1 to stage_10, and the original time sequence label is retained inside each piece of data. The division does not do interpolation or reconstruction processing, only adds boundary identifiers to the segmentation position for subsequent working condition recognition and calling. Statistical analysis is performed on the time period of each stage. Standard deviation, maximum value, minimum value and mean value are used as primary statistical features, and time domain change rate is used as a dynamic feature index. Temperature change rate ΔT is the temperature difference between adjacent time points divided by time interval 15 seconds, and pressure change rate ΔP is processed similarly. The threshold recognition criteria are set as follows: if ΔT exceeds 0.3 ℃ / s, it is marked as high temperature variable working condition; if ΔP exceeds 0.02 MPa / s, it is marked as high pressure variable working condition; if temperature stability σT is lower than 0.5 ℃ and pressure stability σP is lower than 0.01MPa, the label is constant stable working condition, the time points in each section are marked according to the above rules and output as stage working condition label sequence, the stage working condition label is divided into five categories, including high temperature change, high pressure change, mixed vibration, vacuum preservation and constant stable pressure mode, the label structure keeps one-to-one correspondence with the time points, high-frequency data acquisition is performed by using NI cDAQ-9178 data acquisition case and four-channel voltage input module NI-9215, the sampling frequency is set to 5Hz in each stage, that is, 5 time points are recorded per second, five indexes of temperature, pressure, mold position, mixed motor current and molding cavity pressure are recorded continuously, all data sampling is set through LabVIEW environment Data flow path, and the foregoing stage working condition label is introduced synchronously, each sampling point is associated with the stage and working condition type, data recording lasts for 192 minutes, a total of 10 stage blocks, and finally a 5Hz sampling frequency explosive process data set is formed, containing more than 57,600 process feature records, each record contains a timestamp, process segment number, working condition type, five numerical indicators and corresponding change rate indicators, the multi-stage working condition data labeling of the whole explosive forming process is completed, the measurement error of all acquisition modules is not more than ±0.5%, the channel response time is less than 1ms, and the sampling consistency is controlled within ±2μs.

[0052] Preferably, step S3 comprises the following steps:

[0053] Step S31: cutting the explosive working condition data by time window to obtain explosive working condition time sequence fragments; and constructing a continuous working condition sequence according to the explosive working condition time sequence fragments;

[0054] Step S32: identifying a filling rate change sequence in the continuous working condition sequence; performing explosive stability analysis based on the filling rate change sequence and extracting real-time stability data, so as to obtain explosive real-time stability data;

[0055] Step S33: determining a steady-state region based on the explosive real-time stability data; and parameterizing the steady-state region to obtain field parameter data;

[0056] Step S34: reconstructing a scene simulation according to the field parameters to build an explosive stability environment field.

[0057] In this embodiment, the explosive forming condition data with a sampling frequency of 5 Hz after pre-processing is selected, which contains six dimensions of time stamp, condition label, temperature, pressure, mixing motor current, cavity pressure, and mold displacement. In the process of constructing the time window, a sliding window mechanism is adopted, and the length of each window is set to 10 seconds, containing 5 sampling points per second. Therefore, each window contains 50 data points, and the sliding step is set to 2 seconds, that is, each 10-second window slides backward by 2 seconds to reconstruct the next window. After sliding cutting the whole 192-minute explosive condition data, about 5,700 explosive condition time sequence segments are constructed, each segment structure is unified as a matrix form with 50 rows and 6 columns, and the corresponding time period information and condition label information of each segment are retained. Then, the condition label splicing operation is performed on all segments, and the adjacent segments with the same condition label in the continuous segments are combined as continuous condition sequences in the time dimension. If the labels of two adjacent segments are the same and the time interval is less than 5 seconds, they are merged into a continuous condition sequence, otherwise, they are marked as the starting point of a new sequence. Finally, 1,240 non-repeated and overlapping explosive continuous condition sequences are generated, each of which contains multiple time window segments. Based on the mold displacement data and time stamp data in each time sequence segment, the mold feeding distance change per unit time is calculated using the finite difference method, and then the corresponding explosive filling rate is estimated, with the unit set to mm / s. The filling rate change sequence of each segment is formed into a change curve with equal interval time as the horizontal axis and the current segment filling rate as the vertical axis. Then, the weighted smoothing filter algorithm is used to process the filling rate sequence, and the filter window is set to 5 points, and the filter function is the weighted average weight function w=[1,2,3,2,1], to enhance the continuity of the change trend and eliminate the pseudo fluctuation interference. In the stability analysis, the filling rate first-order difference value less than 0.1 mm / s is set as the stable section, and the value exceeding 0.3mm / s is the fluctuation section, the middle value is the transition section, and each section of the change sequence is marked according to the above rules. Finally, the real-time stability data of the explosive is formed, which contains the filling rate value of each time point, the category label of the fluctuation section, and the stability level identification. The stability level identification is divided into three categories, namely stable, critical and unstable. The stability level is divided by the average value and the maximum amplitude range of the change amplitude. The stability level continuous section extraction operation is performed on the real-time stability data of the explosive. The sliding scanning mechanism is adopted, and the window length is set to 60 seconds. In each continuous time slice, the stability level label frequency is counted. If the proportion of the stable level is more than 80%, it is determined as a steady-state region. The boundary information of each steady-state region is recorded according to the starting time and the ending time, and all the stability characteristic data in the boundary are statistically extracted. The specific extraction indexes include the average filling rate of the stable period, the pressure stability σP, the temperature stability σT and the current fluctuation rate αI. The stability is defined as the standard deviation of the index in the steady-state section, and the current fluctuation rate is defined as the ratio of the maximum difference to the average value of the current. After the extraction indexes are normalized, the K-means clustering algorithm is used to cluster and classify the steady-state regions. The cluster number K is set to 3. Finally, each steady-state region is classified into the corresponding stability field category, and three types of field parameter templates are constructed on this basis to form the field parameter data used for simulation environment construction, which includes the average characteristic value of the steady-state, the stability change range, the boundary duration and the corresponding working condition category. Unity3D simulation engine is used as the scene reconstruction platform, and parameter injection module is constructed through C# script language. Each type of field parameter data is injected into the digital twin scene model. The digital twin model is imported into the skeleton based on the three-dimensional model of the explosive forming equipment constructed by SolidWorks. The model is provided with a cavity volume change mechanism, a mold feeding system and a dynamic temperature and pressure sensing module. The simulation running script is built for each type of stability field. The time step of each frame is set to 0.2 seconds, and the simulation period is uniformly set to 90 seconds. The filling state, pressure distribution and mold response delay parameters in the cavity are collected in real time during the simulation, and the field atlas is output in real time. The field atlas is constructed based on the three-dimensional heat map. The filling density is used as the heat distribution benchmark, and the color interval is from blue (0 g / cm. 3 ) to red (maximum filling density). The three-dimensional atlas of the stability environment is presented. All simulation results are recorded as discrete frame sequence data and spatial distribution matrix, which constitute the stability environment modeling basic data required for the subsequent digital twin prediction model.

[0058] Preferably, the step S4 of simulating the explosive stacking storage of the simulation explosive model based on the explosive stability environment field comprises:

[0059] mapping the simulation explosive model based on the explosive stability environment field to obtain an explosive stacking unit;

[0060] mapping a space layout according to the explosive stacking unit to obtain a space layout structure;

[0061] determining an explosive model size of the simulation explosive model;

[0062] planning a stacking path based on the space layout structure and the explosive model size to obtain stacking path data;

[0063] stacking and storing simulation of the simulation explosive model based on the stacking path data and the explosive model size, thereby obtaining simulation explosive stacking data.

[0064] In this embodiment, the constructed explosive stability environment field is taken as the background data source. The environment field is a three-dimensional continuous tensor with a size of 200x200x150 and a unit of mm. The explosive simulation model entity is embedded in the space. The size of the simulation model is set as 50 mm per explosive unit. The simulation explosive model is based on a C4 explosive entity three-dimensional model with a regular hexahedral structure. The field domain is divided into identical stacking units by three-dimensional space block mapping. Each stacking unit is a 50mmx50mmx50mm cubic unit. The mapping algorithm is based on spatial orthogonal partitioning. The three-dimensional right-angle coordinate axis starting point is taken as the origin. The spatial index table is constructed in sequence along the x, y, and z axis directions. All non-boundary units participate in the formation of stacking units. A field stability label is added to each stacking unit. The label is derived from the stability grade value at the corresponding position in the environment field domain parameter tensor. The grade is represented by 0 for extreme instability and 3 for high stability. Finally, the explosive stacking unit cluster data containing the three-dimensional mapping of spatial position, unit size, and stability label are formed. The spatial layout mapping operation is carried out. The spatial layout mapping is based on the above-mentioned stacking unit cluster data. The three-dimensional stability priority algorithm is used for layout region extraction. First, scanning starts from the lowest layer of the z axis and moves upwards layer by layer. All stacking units with a stability grade of 3 form a stable layout candidate block. Each candidate block is area clustered according to its x-y plane projection range. The minimum clustering area threshold is set to 0.25m2, the spatial closure correction processing is performed on the clustering results, all isolated stacked units that cannot be connected are excluded by the eight-neighborhood connectivity rule, only continuous stacked blocks are reserved, then all stackable regions are numbered, and three-dimensional space layout structure data is generated in the order of numbering. The structure data contains the volume information, starting coordinates, maximum stacking height, and average internal stability of each stackable unit block. This structure data will be used as the spatial input basis for subsequent stacking path planning and body size matching calculations. According to the initial geometric structure of the explosive model, the body size parameters of the simulated explosive model are extracted, including the base size, total height, center of gravity position, outer box volume, and minimum circumscribed cube shape. The base size is automatically identified by the CAD modeling system by combining the longest side and the widest side of the lower surface of the explosive model. The maximum base length is set to no more than 3 times the length of the stacking unit, i.e. 150mm. The total height is measured by the maximum z-axis distance and cannot exceed the allowed stacking height of the stable region. The center of gravity position is calculated by voxelizing the three-dimensional model and calculating the three-dimensional coordinate point based on uniform mass density. The outer box volume is obtained by the boundary envelope algorithm. The minimum cuboid is constructed by the outermost point of the explosive model. The minimum circumscribed cube is obtained by identifying the optimal rotation angle of the rotating envelope method. Based on the existing space layout structure and the body size parameters of the explosive model, the three-dimensional Boolean mapping algorithm is used to calculate whether each stackable region can completely accommodate the unit body size of the explosive model. If the stacking region size meets the outer box volume and shape requirements, it is included in the path candidate set. In the path planning, the three-dimensional path search algorithm is used, the initial point is set as the environment field feeding channel position, and the target point is set as the center point of the stacking region. In the search process, the composite weighting of the path cost function is performed for each step. The cost function f(n)=g(n)+h(n)+s(n), where g(n) is the current path cumulative cost, h(n) is the Manhattan distance estimation cost, and s(n) is the stability penalty term. The stability penalty term is calculated by the average stability level of the region passed by the path. The lower the stability level, the higher the penalty value. The final selected path with the minimum total cost is the recommended stacking path. The search result is output as a point sequence to form the stacking path data. Each path node contains spatial coordinates and path sequence number. The Unity3D three-dimensional simulation platform is used to build the explosive stacking evolution environment, import the stacking path data and explosive model body size parameters, and set the rigid body properties of the explosive model using the physical engine module RigidBody, including the density set to 1.4g / cm. 3 , and the elastic modulus set to 1.2x10 7Pa, the friction coefficient is set to 0.35, the model placement process takes the stacking path sequence as input to control the model movement trajectory, moves frame by frame in the order of path nodes, and performs displacement of no more than 10mm per frame, while enabling the collision detection module to prevent interpenetration between models, after the stacking is completed, the final center point coordinates of each model, the placement order and the stability level of the lower contact surface are recorded, the simulation platform simultaneously outputs the stacking structure image and the dynamic data frame of the stacking process, and finally forms complete simulation explosive stacking data.

[0065] Preferably, the step S4 of inferring the explosive stacking pressure data from the simulation explosive stacking data comprises:

[0066] Confirming the gravity center of the simulation explosive model;

[0067] Calculating the interaction force data between adjacent simulation explosive models based on the simulation explosive stacking data and the gravity center of the simulation explosive model;

[0068] Determining the explosive stacking frame according to the simulation explosive stacking data;

[0069] Constructing the stacking pressure conduction chain through the explosive stacking frame and the interaction force data;

[0070] Inferring the explosive stacking pressure data of each simulation explosive model according to the stacking pressure conduction chain.

[0071] In this embodiment, the gravity center calculation operation is performed based on the constructed three-dimensional simulation explosive model data, the simulation model takes hexahedral C4 explosive unit as the structural reference, each model is divided into voxel unit bodies with a side length of 5mm through three-dimensional grid division, and the voxel density is set to 1.59g / cm 3 , all voxel units are assigned with equal mass values, and then the spatial mass gravity center of the three-dimensional structure is solved by using the voxel gravity center superposition method, in the calculation process, the OpenVDB voxel data processing tool is selected, VDBVolumeToPoints module is used to convert all voxel positions and mass into dense point cloud format, then a three-dimensional floating point coordinate is generated by using the self-defined C++ program to multiply the x, y, z coordinates in all point cloud data by the mass value and then perform weighted average, which represents the mass gravity center of the simulation explosive model, the gravity center coordinates are represented in the form of three-axis floating point numbers, for example (25.6, 31.2, 49.8) units of millimeters, the center coordinates, bottom contact surface index and gravity center position of each model entity in the simulation explosive stacking data are extracted, the number of support units and their spatial positions are determined through the surface contact identification algorithm for the bottom contact surface, the contact threshold is set to less than 1mm and the normal direction angle is less than 5°, which is the effective support surface, based on the identification of the support surface, the static pressure force value of each explosive unit is calculated, and the gravitational acceleration is set to 9.8m / s2 The weight of the unit body is obtained by multiplying the unit mass by gravity, and the weight is equally divided to each support surface contact point, and the pressure value is taken as the force value applied by the support point to the lower explosive unit. This process is implemented through the computeContactForces module in the PyBullet physics simulation engine, and high-precision double-precision floating-point mode is used in the calculation to ensure data accuracy. Finally, the contact force data between each adjacent explosive unit is extracted as a triple structure, including the index of the force applying unit, the index of the force receiving unit, and the normal pressure value in Newton. According to the stacking data, a three-dimensional space stacking topology graph is constructed, with each explosive unit as a node in the graph, and all adjacent contact units are connected by edges to represent the stacking relationship. Each edge is assigned a contact type attribute, including "face-to-face", "edge-to-face", and "point-to-face". "Face-to-face" is defined as two explosive models with a contact area ≥300mm 2 "Edge-to-face" is defined as a single edge line segment length ≥20mm in contact with a plane, and "point-to-face" represents a single point support relationship. Based on the above structure, the topology graph data structure is constructed, and the adjacency list is used. The node records the model index, coordinate position, and model mass, and the edge records the contact type and contact area size. After construction, the connected_components algorithm in Boost Graph Library is used to verify the connectivity of the stacking frame, ensuring that all explosive units are in the same connected domain. Based on the three-dimensional stacking topology structure graph and the interaction force triple data constructed as described above, a directed graph model with direction and weight is used to model the pressure conduction path. Set all forces from the upper model node to the lower support node, and take the pressure size as the weight of the edge. If a node has multiple lower support units, the weight applied by it should be distributed in proportion to the support area. For example, the bottom surface of explosive model A is in contact with model B and model C with contact areas of 80cm 2 and 120cm 2, the distribution ratio is 2:3, and directed weighted edges are constructed for all support paths and added to the conduction graph. The DAG (directed acyclic graph) structure is used to maintain the conduction direction and prevent the existence of loops. After the construction is completed, a hierarchical traversal algorithm is executed, starting from the top model node and sequentially propagating the pressure value to the lower layer, recording the cumulative pressure value of each intermediate node. A stability label is added as a propagation damping factor in each propagation, and if the target node has a low stability level, a penalty coefficient is added to the corresponding path to reflect the pressure deformation sensitivity. The conduction chain eventually forms a complete pressure propagation graph data. A top-down propagation path accumulation algorithm is used to calculate the pressure accumulation of each path from the top of the stack to any node. The final bearing pressure value of the node is obtained by adding the total pressure of all paths to the node, with the unit being Newton. The conduction weight of each path is determined by the gravity of the starting node multiplied by the weighted coefficient of the path edge, and the weighted coefficient of the edge is determined by the contact area ratio and the stability level of the support node. For example, if the support surface of a support path accounts for 60%, and the stability level is 3, the edge weight is set to 0.6x1.0=0.6. If the stability level is 1, multiply by the stability decay factor 0.75 to get the edge weight of 0.45. All path calculations are performed through the custom_weighted_path_sum function in the networkx graph calculation library, ensuring that all conduction paths are free of repetition and loops. Finally, each explosive model entity corresponds to a set of bearing pressure data, which is composed of three-dimensional coordinates and pressure values, with the unit being Newton.

[0072] Especially important is that the interaction force data between adjacent simulation explosive models is calculated based on the simulated explosive stack data and the center of gravity of the explosive model, including:

[0073] Based on the simulated explosive stack data, the simulation explosive model arrangement position is located;

[0074] The simulation explosive model arrangement position is projected and transformed, and the center of gravity direction vector of the simulation explosive model is analyzed;

[0075] The adjacent boundary of the simulation explosive model arrangement position and the center of gravity direction vector of the simulation explosive model is extracted;

[0076] The contact surface geometry of the adjacent boundary is reconstructed, and the model boundary contact area is calculated;

[0077] The contact pressure is estimated based on the model boundary contact area, and the force vector is converted to obtain the interaction force data between adjacent simulation explosive models.

[0078] In this embodiment, the original three-dimensional model index data set in the simulated explosive stack data is called, which contains the model number, initial placement coordinates and model rotation matrix of all explosive units, which is derived from the three-dimensional explosive scene stack simulation environment. All explosive units are hexahedral C4 voxel structures, each with a side length of 5mm. Open3D (Open Three-Dimensional Data Processing Library) is used to construct a spatial point cloud distribution map, and the VoxelGrid filter in PCL (Point Cloud Library) is used to perform spatial regularization processing on the point cloud data. Then the local coordinates of each model are converted into spatial positions in the global three-dimensional coordinate system using the rotation matrix of each model. The conversion process is referenced to the simulation coordinate origin, and matrix multiplication is used to multiply the model local coordinates (x, y, z) with the rotation matrix to obtain the transformed position vector. The coordinate mapping is realized by adding the displacement vector. The positions of all explosive models are output in the form of a three-dimensional floating-point array, where the position of each explosive unit is represented by (xi, yi, zi) in millimeters. Taking the three-dimensional position coordinates of each explosive model as the center point, the mass center of gravity data of the model is called. The center of gravity position has been obtained by the voxel weighted average method. The center of gravity direction vector of each model is obtained by subtracting the model bottom reference point coordinates from the center of gravity coordinates. Then a three-dimensional projection transformation matrix is established with the vertical direction as the main axis. The model position and center of gravity coordinates are represented by the homogeneous coordinate system. The affine projection transformation is applied to each model to project it from the three-dimensional coordinate system to the two-dimensional plane, where the two-dimensional plane is taken as the horizontal bottom surface (x-z plane). The cv: projectPoints function in OpenCV is used for conversion during the transformation process to obtain the two-dimensional projection coordinates (x_proj, z_proj). The center of gravity direction vector is normalized to have a unit length of 1. The vector is decomposed in the two-dimensional plane to obtain the projection angles in the horizontal and vertical directions, with the unit being radians. Based on the two-dimensional projection coordinate set of all models, a spatial index tree is constructed. The k-d tree (K-Dimensional Tree) structure is used to accelerate the neighborhood search process. After construction, the neighborhood radius is set to 60mm. The neighborhood search within the radius range is performed for each model to extract all adjacent model indexes that are in contact or have a distance less than the set threshold. Then, for each pair of adjacent models, the included angle between their respective center of gravity direction vectors is analyzed. If the included angle is less than 15 degrees and there is an intersection voxel between the model boundaries, it is considered to have a real adjacent relationship. The boundary overlap detection is performed. The Boolean intersection operation in MeshLab is used to calculate the intersection area between the two model meshes. If the projected area of the intersection area is greater than 30mm 2If a boundary is confirmed as a valid adjacent boundary, it is composed of two model number pairs and their overlapping region indices. All boundary data is output as triplets, including model A number, model B number, and overlapping region index number. The overlapping voxel region of each confirmed adjacent boundary pair is extracted. The Polyhedron_3 mesh Boolean operation in CGAL (Computational Geometry Algorithms Library) is called to perform precise polyhedral reconstruction of the overlapping region. This operation voxels the overlapping mesh regions of the two models, finds their intersection, and reconstructs them into a set of regular polygonal faces. The areas of all faces in this set are then summed to calculate the contact area, in square millimeters. During processing, the minimum side length of each triangular face is set to be no less than 1 mm to avoid mesh fragmentation causing error amplification. The area calculation is performed using the Heron formula, with double-precision floating-point numbers used for face accumulation in each calculation. All contact area data is stored with a correspondence between the adjacent model pair number and the area value. The contact area between each pair of models must be no less than 30 mm². 2 If the value is below this threshold, it is considered an ineffective support contact, and the boundary pair is excluded. The mass value of each model and its applied gravity are then retrieved. The mass value is calculated by multiplying the number of voxels by the mass per voxel (the mass per voxel is 1.59g, and the voxel volume is 125mm²). 3 The result is obtained by multiplying by 9.8 m / s. 2 Obtain the weight, then distribute the weight evenly across all contact boundaries, weighting them according to the contact area ratio. For example, if the total contact area of ​​model A is 1800 mm²... 2 The contact area with model B is 600 mm². 2 The pressure allocated to model B is (600 / 1800) * the weight of model A. This pressure value is the total force acting on the contact boundary, in Newtons. Then, the force vector is projected by combining the model's center of gravity direction vector. The pressure value is multiplied by the unit direction vector to obtain the three-dimensional component form of the force vector (Fx, Fy, Fz). All contact force vectors are processed in batches using NumPy array calculation. Finally, the interaction force data between each pair of adjacent models consists of the model A number, the model B number, and the corresponding three-dimensional force vector, in Newtons.

[0079] Preferably, step S5 includes the following steps:

[0080] Step S51: Determine the completion rate of the explosive working condition data based on the preset standard working condition data to obtain the explosive working condition completion rate data;

[0081] Step S52: Confirm the explosive pressure threshold according to the explosive working condition completion degree data; real-time stack safety evaluation is carried out through the explosive stack pressure and the explosive pressure threshold to generate real-time explosive stack safety;

[0082] Step S53: Identify the pressure change data in the stacking process based on the explosive stack pressure data;

[0083] Step S54: Infer the segmented stack pressure of each simulation explosive model through the pressure change data; calculate the stack cumulative pressure of the simulation explosive model based on the simulation explosive model.

[0084] In this embodiment, a complete standard working condition data set is established, which should be derived from the five basic variables recorded in the standard detonation test experiment conducted at room temperature 23 degrees Celsius, relative humidity 40% environment, including detonation velocity, detonation pressure, temperature rise curve, stacking density and stress change. The standard data is grouped by explosive type as the classification standard. In each group of data, the detonation velocity is controlled in the range of 7000 to 8200 meters per second, the detonation pressure is 25 to 32 gigapascal (GPa), the detonation temperature rise is not more than 480 degrees Celsius, the stacking density standard error should not be greater than 0.2 grams per cubic centimeter, and the stress fluctuation frequency needs to be controlled within 200 Hz. The working condition data of the explosive to be evaluated is input into the tensor model with the same structure as the standard working condition data. The two tensors are checked using the symmetric cosine similarity matching method. The similarity threshold is set to 0.92. If the similarity of the explosive working condition data and the standard working condition data is greater than or equal to 0.92, it is determined that the working condition completion degree is 100%. If the similarity is less than 0.92, the actual working condition completion degree is calculated using the piecewise linear regression function. The final output is the explosive working condition completion degree data. The corresponding working condition matching level is extracted from the aforementioned explosive working condition completion degree data as the input condition. The corresponding pressure threshold of the explosive is obtained by setting the empirical mapping function, which is based on the empirical formula Pth=P0×(1+k×(1―Cm)), where Pth is the pressure threshold, P0 is the theoretical maximum static pressure of the material, and k is the empirical coefficient with a value of 0.4, Cm is a working condition completion degree proportional factor, the value range is 0 to 1, after calculating the pressure threshold value, it is compared with the real-time pressure data obtained in the explosive stacking process frame by frame, the process uses a laser interference pressure measurement sensor array (PDV array) to record the pressure change during the stacking process at an interval of 2 nanoseconds, a sliding window mechanism is used to perform moving average processing on the time series data, the window width is set to 200 nanoseconds and the step is 20 nanoseconds, if the average pressure of any window exceeds the pressure threshold value by more than 10%, it is determined that the stacking safety state is decreased, otherwise it is recorded as the stacking safety state is normal, finally the stacking safety state is output as real-time stacking safety data, the original pressure data sequence obtained by the above PDV array is processed by multi-resolution wavelet transform, db6 wavelet function is used for five-layer decomposition, approximation coefficients and detail coefficients are obtained at each layer, the third and fourth layers of detail coefficient sequences are analyzed, the critical point of pressure change is identified by variance mutation point extraction algorithm, the discrimination standard is that the mean difference of five consecutive data points is greater than 3 times the standard deviation of the previous mean, the data within 20 nanoseconds after each extracted mutation point is classified as a single pressure change event, the pressure change data in the entire stacking process is constructed into a pressure change map by rearranging and classifying these single events, each event in the map contains five key parameters: start time, end time, maximum pressure, change rate and change duration, the pressure change map is input into a structural mapping neural network model (SMNN), the model consists of three fully connected networks and a bidirectional long short-term memory network (BiLSTM), the input dimension is 5, the number of hidden layer neurons is 64, 128 and 64 respectively, and the output is the pressure bearing capacity value of each segmented structure unit in each simulated explosive model, each segment length is not more than 1 / 15 of the total length of the explosive, and each segment model number is consistent with the corresponding simulation geometric grid mapping, the inference result is the maximum pressure bearing capacity of each segment structure under the given stacking stress sequence, then each segment pressure bearing capacity data is input into the stacking path of the simulated explosive model, each segment data is aggregated upward according to the spatial hierarchical structure, and a weighted average formula is used, wherein the weight is set according to the material density and volume proportion of each segment, and finally the stacking cumulative bearing pressure data of the simulated explosive model under the complete stacking path is output, which can be used for fitting and comparison with the pressure monitoring data of the real explosive structure to complete the reverse optimization and calibration of the simulation model precision. All the calculation processes do not introduce thermal, entropy structure or dynamics methods, but only based on pressure, density, geometric structure and stress time sequence change for modeling and inference, and all parameters are clearly defined by experimental data and structure configuration.

[0085] Preferably, the step S6 of continuously predicting the pressure deformation of the simulated explosive model based on the stacking cumulative bearing pressure comprises:

[0086] material deconstruction data is obtained by deconstructing the simulation explosive model;

[0087] pressure response analysis is performed on the simulation explosive model based on the material deconstruction data to generate explosive pressure response data;

[0088] sustained pressure simulation data is generated by performing sustained pressure simulation on the simulation explosive model by stacking and accumulating pressure bearing;

[0089] predicted explosive deformation data is obtained by performing deformation reaction prediction on the sustained pressure simulation data according to the explosive pressure response data.

[0090] In this embodiment, the simulation explosive model is subjected to material deconstruction processing, which is performed using a layered mapping material identification mechanism. First, the three-dimensional simulation explosive model is voxelized and cut in the X, Y, and Z three-axis directions at 1 millimeter intervals to obtain a voxel cluster composed of three-dimensional grid elements. Then, the corresponding material composition table is retrieved through the known explosive original formula number. The material composition table is obtained through X-ray diffraction (XRD) and scanning electron microscope energy dispersive spectroscopy (SEM-EDS) experiments. With the mass fraction of RDX as the main variable, combined with the proportions of Al powder, paraffin, glass fiber, and other components, the material database is matched according to the density and texture parameters of different voxel regions. Each voxel region is assigned a specific material label and a set of microstructure parameters, including particle size distribution range (controlled within 5 to 80 microns), average particle spacing (less than 3 microns), microcrack density (not higher than 0.01 per cubic millimeter), and binder coverage (greater than 95%). The final output material deconstruction data includes the material number, microparameter group, and macroscopic density value of each voxel element, and a structure-material mapping relationship table is established. A microstructure response model of the explosive material is constructed using the multi-physics coupled finite element tool ANSYS Autodyn. The model takes the micro-particle structure in the material deconstruction data as the input basis, and constructs different pressure-strain curves and yield models for each material number. The RDX region is modeled using the Johnson-Cook yield criterion, with parameters set to A = 240 megapascals (MPa), B = 90 MPa, n = 0.26, C = 0.015, and m = 1.03. The glass fiber region uses a linear elastic-damage model, with Young's modulus set to 70 gigapascals (GPa) and Poisson's ratio set to 0.21. The binder region uses an elastic-viscoplastic constitutive model, with a viscosity modulus set to 25 megapascal-seconds. Different material regions are coupled and simulated on the interface through the multi-body interaction interface defined in the model. A 0 to 200 MPa proportional pressure curve is applied to the model using compression loading boundary conditions, with the simulation time controlled within 2 microseconds. The maximum plastic strain, local compression rate, and residual deformation of each material region are calculated. The stacking cumulative pressure data obtained in the previous stage is input into the material response model and subjected to time extension simulation. First, a continuous medium simulation model based on the Eulerian grid is constructed. The stacking pressure signal is discretized into 500 nanoseconds per frame in the time dimension, with a total time length of 5 milliseconds. The stacking pressure is applied to the contact surface area of the simulation explosive model using a continuous loading method. The pressure value of each frame is equal to the weighted update amount of the previous frame pressure value, with the weight set to the product ratio of the local voxel strain gradient and its corresponding material impedance. The minimum element size used in the simulation process is 0.25 millimeter grid is set up, and ALE (Arbitrary Lagrangian-Eulerian) algorithm is enabled to ensure that the grid does not collapse, and the model records four types of indexes, i.e. local displacement, stress distribution, volume change rate and strain rate, in each frame in real time, and the final output of the continuous pressure simulation data is composed of stress and strain data of each voxel node evolving over time, including a pressure field response sequence at continuous time points under three-dimensional spatial distribution, a spatial deformation propagation model based on GNN (Graph Neural Network) is constructed, the model takes the three-dimensional structure of the simulated explosive as a node graph structure, the edge connection represents the mechanical coupling relationship between adjacent voxel units, the node feature vector is composed of the strain difference, displacement difference and pressure difference of each node in the continuous pressure simulation data at any two adjacent time points, a three-layer graph convolution network structure is used, the ReLU activation is used in the node update function of each layer, the output dimension of the first layer is 64, the second layer is 128, and the third layer is 64, then the GAT (Graph Attention Network) is used to enhance the identification ability of the heterogeneous connection area, the response features of each node are aggregated into a global prediction function, and finally the deformation value prediction of each voxel in the future 1-10 milliseconds is output by the fully connected layer, and the complete predicted explosive deformation data is organized in the form of structural deformation field, which is a three-dimensional tensor format, wherein each voxel corresponds to the axial and radial deformation value in the future time period, and the unit is millimeter, in the model training stage, high-frequency mechanical experiment data is used as the true label for fitting optimization, the training batch is set to 300 rounds, the Adam optimizer is used, the initial learning rate is set to 0.0005, the batch size is 64, and the loss function is a dual constraint function of mean square error and structural deformation consistency.

[0091] Especially important is that the continuous pressure simulation on the simulated explosive model by stacking and accumulating pressure bearing includes:

[0092] The stacking pressure bearing data is linearly expanded in sections to obtain the stacking pressure bearing distribution;

[0093] According to the stacking pressure bearing distribution data, the surface pressure data of the simulated explosive model is generated by pressure mapping conversion;

[0094] The node pressure injection simulation is performed on the simulated explosive model through the surface pressure data of the simulated explosive model, so as to obtain the node pressure input data;

[0095] According to the node pressure input data, the time step loading is performed to generate the continuous pressure data of the simulated explosive model;

[0096] The pressure response trajectory of the continuous pressure data of the simulated explosive model is tracked, and the iterative superposition is performed based on the pressure response trajectory, so as to obtain the continuous pressure simulation data.

[0097] In this embodiment, the original pressure data is divided into multiple continuous segments according to the time axis, each segment corresponds to a physical loading cycle, and each loading cycle is set to 2 milliseconds in the experiment. The pressure data of the stack is obtained from multiple loading experiments. The piezoelectric sensor is used to record the force change curve in each experiment, and the time interval is 1 microsecond, forming a data segment of 500 points. The least square method is used to linearly fit each segment of data, and a linear function form P(t) = a*t + b is constructed for each segment, where a is the pressure growth rate and b is the initial pressure value. The linear expansion aims to minimize the square residual error, and the linear segment with a fitting residual less than 0.03 megapascal (MPa) is retained. For the paragraphs with residual error exceeding the threshold, double-interval splitting processing is performed, and linear approximation is performed again. The final stack pressure distribution is a sequence of linear increments of pressure under multiple time segments. Each sequence item contains the start point, end point, fitting parameters a and b, and its error limit. The entire linear segment is structured and output. The time-pressure linear function is mapped to the surface of the simulation explosive model. The operation first meshes the surface of the simulation explosive model into triangular facets. The Marching Cubes algorithm is used to extract the isosurface from the voxel structure to generate a shell model composed of facets. The size of the facet is about 0.5 millimeters, the number of grid points is controlled within 500,000 points, and the linear segment pressure function is assigned to the model shell. The loading value of each pressure segment is uniformly distributed to each triangular surface element of the model contact surface by area weighting method. The pressure increment per unit time is defined as ΔP = a × Δt, where a is the linear segment pressure growth rate and Δt is the time interval. In space, the normal vector direction of each surface patch is corrected, and only the surface patch elements pointing to the external loading direction are involved in the pressure calculation. The final generated surface pressure data of the simulated explosive model is a multi-dimensional array structure, which records the loading start time, termination time, pressure change function parameters and pressure direction vector information of each surface patch for each frame. The node pressure injection framework corresponding to the simulation model grid is constructed. First, the entire model is numbered, and each node records its spatial coordinates, surface patch number and material number. Through the surface pressure mapping relationship, the pressure function of each pressure surface patch is applied to the node set it covers. The local coordinate of each node is calculated to obtain the force area projection value. The real-time calculation value of the pressure function and the node area ratio are used for local pressure interpolation to obtain the pressure value P_node(t) = P_face(t) * A_node / A_face of the node at time t, where P_face is the unit area pressure of the surface patch, A_node is the node area projection value, and A_face is the entire surface patch area. Each node generates a set of pressure input time series, which is sampled at a time resolution of 1 microsecond. The node pressure input data finally represents a three-dimensional array, with each dimension being the node number, time step, and applied pressure value. The numerical unit is unified as megapascal. The finite element calculation platform ANSYS ExplicitDynamics module is called to input the node pressure sequence into the loading control module to construct the loading boundary condition. In this process, the time step control strategy is adopted, with each step being 1 microsecond and the total loading time being 10 milliseconds. For each microsecond step, the control module reads the corresponding node pressure value array and applies it to the simulation model node. The simulation calculation uses the self-adaptive time integration method to ensure numerical stability, and the Newmark-beta integration method is used for time advancement with the control parameter β set to 0.25 and γ set to 0.5, ensure that the numerical error is lower than 10^-5 in each loading process, enable the grid cell interpolation mapping method in the calculation process, map the node pressure value to the element stress field, output the node stress, strain and deformation data inside the model at each time through the structure solver, and the process is continuously executed until all time step loading is completed. Finally, the continuous pressure data of the simulated explosive model is output. The data structure includes the stress path, total strain increment and instantaneous displacement vector value of all nodes under the entire loading time sequence. A three-dimensional time-space pressure trajectory tensor is constructed, which has dimensions X x Y x Z x T, where X, Y and Z are the dimensions of the model voxel grid, and T is the total number of time frames. The pressure value of each voxel node during the simulation process is spliced into a complete sequence along the time axis, and the response trajectory of each voxel is defined as P(x, y, z, t). Every 10 frames, the local difference of the trajectory is calculated, and the impedance factor of the corresponding voxel in the material deconstruction data is multiplied to update the local pressure residual of the current node. Set the iteration increment as the difference between the last time and the current time ΔP = P(t)-P(t-1), and perform multiple iterations until the pressure change rate of all nodes is lower than 0.001 MPa per microsecond. Record the change path of each iteration, and accumulate the total pressure response value. Finally, the final pressure response of all voxel points is superimposed to form continuous pressure simulation data, which is a four-dimensional tensor with spatial coordinates X, Y, Z and time T. The corresponding unit is megapascal, and the data magnitude is not less than 10. 9 Each record represents the complete pressure response history of a specific spatial node at a single time step.

[0098] Preferably, the continuous stacking safety prediction according to the predicted explosive deformation data in step S6 is combined with the real-time explosive stacking safety to perform safety evaluation, which includes:

[0099] According to the accumulated bearing pressure of the stack, the continuous pressure application simulation of the simulated explosive model is performed, so as to observe the continuous deformation data of the explosive model;

[0100] Determine the explosive deformation threshold value through the continuous deformation data of the explosive model;

[0101] Determine the dangerous deformation distance based on the explosive deformation threshold value and the predicted explosive deformation data;

[0102] Perform continuous stacking safety prediction through the dangerous deformation distance, thereby generating continuous stacking safety;

[0103] Combine the real-time explosive stacking safety and the real-time explosive stacking safety to perform safety evaluation on the explosive, thereby obtaining explosive safety data.

[0104] In this embodiment, the finite element analysis (FEA) software is used to mesh the simulated explosive model, dividing it into small units to accurately simulate the stress distribution and deformation response of each unit. The input stacked cumulative pressure bearing value is used as an external load acting on the model surface and interior. When applying the load, the time step method is used to simulate the continuous application process of pressure through multiple time steps, recording the strain and displacement changes of the explosive model at each time step, and then obtaining the continuous deformation data of the explosive model during the entire pressure application period. This deformation data includes linear displacement, shear deformation, and volume change, among other multidimensional parameters. The deformation measurement accuracy is in the micrometer level, the pressure application time step is set to the millisecond level, and the pressure range is set to 10 to 100 megapascals based on previous working condition determination data. A statistical analysis method is used to analyze the peak value of the continuous deformation data, extract the maximum deformation value and its corresponding stress point, and use the critical deformation empirical value as the threshold reference. This threshold value is set based on the safety limit deformation obtained from previous laboratory blasting tests, and is usually set as a percentage of the maximum allowed deformation of the model. The specific threshold range is 85% to 95% of the elastic limit of the explosive material. The threshold determination process involves comparing multiple simulation results and excluding abnormal data points. A cubic spline interpolation method is used to fit the deformation curve to accurately locate the deformation limit. The maximum deformation in the predicted deformation data is compared with the deformation threshold, and the difference is calculated. The calculated difference is called the dangerous deformation distance, which represents the numerical range of the predicted deformation exceeding the safety limit. The Euclidean distance method is used to calculate the vector difference of the deformation variables of each node in the simulation explosive model space. The local deformation variable difference is accumulated by numerical integration method to obtain the dangerous deformation distance of the overall model. The numerical range of the dangerous deformation distance is limited to 0 to 5 millimeters, and the larger the value represents the higher the overrun. The calculation process strictly considers the material nonlinear characteristics and boundary condition constraints. The dangerous deformation distance is mapped to the safety level index, and the safety level uses a hierarchical numerical system with a value range from 0 to 1. A value of 0 represents extreme danger, and a value of 1 represents complete safety. An inverse mapping relationship is established between the dangerous deformation distance and the safety level, i.e., the larger the dangerous deformation distance, the lower the safety level. An exponential decay function is used in the mapping process, and the corresponding safety level value is output after inputting the dangerous deformation distance into the function. The parameters of the exponential decay function are adjusted based on historical explosive failure data, and the parameter range is set to a decay rate of 0.5 to 1.5, the mapping result is normalized to ensure that the output security level is in the interval of 0 to 1, and the security level is the continuous stacking security index, real-time explosive stacking security data is obtained, the data is collected by an online sensor system in real time during the explosive stacking process, and the data is calculated by a preset safety threshold model, the real-time security index value range is also between 0 and 1, then the continuous stacking security and the real-time security index are applied to a weighted fusion algorithm, the two are multiplied by weight values, the weight sum is 1, the weight is set to 0.6 for real-time security and 0.4 for continuous security, the fusion process adopts a weighted average method, the product results are added to form a comprehensive security index, and the final security data presents a continuous change curve, reflecting the security level of the current explosive stacking state, which is used to guide subsequent safety monitoring and risk management, the time sequence of each index data is aligned in the fusion process, the influence of data time delay on the accuracy of evaluation is eliminated, and the real-time and reliability of the comprehensive security evaluation result are ensured.

[0105] The application also provides an explosive security prediction system based on a digital twin model, which is used to execute the explosive security prediction method based on the digital twin model as described above, and comprises:

[0106] A model simulation module is configured to acquire original explosive data, extract particle size distribution data in the original explosive data, and construct the original explosive data into a simulation explosive model with equal proportions based on the particle size distribution data.

[0107] A working condition recording module is configured to record an explosive forming process, extract working condition characteristics in the explosive forming process, and label explosive working condition data of each stage of the explosive forming process based on the working condition characteristics.

[0108] A stability analysis module is configured to perform real-time stability analysis of explosives by using the explosive working condition data, so as to obtain real-time stability data of the explosives, and build an explosive stability environment field according to the real-time stability data of the explosives.

[0109] A stacking simulation module is configured to simulate explosive stacking and storage of the simulation explosive model based on the explosive stability environment field, so as to obtain simulation explosive stacking data, and deduce explosive stacking pressure data according to the simulation explosive stacking data.

[0110] A real-time safety evaluation module is configured to perform real-time stacking security evaluation according to the explosive stacking pressure and the explosive working condition data, so as to generate real-time explosive stacking security, and calculate stacking cumulative pressure by using the explosive stacking pressure data.

[0111] A safety joint evaluation module is used for continuously predicting pressure deformation of the simulation explosive model based on the stacking cumulative bearing pressure to obtain predicted explosive deformation data; continuously predicting the safety of the stacking based on the predicted explosive deformation data, and combining the real-time explosive stacking safety to evaluate the safety.

[0112] The model simulation module realizes comprehensive acquisition of original explosive data, ensures the integrity and accuracy of the data, and the extracted particle size distribution data provides a scientific basis for the construction of the simulation explosive model, the generated simulation explosive model realizes the real reproduction of the characteristics of the explosive, significantly improves the reliability and practicality of the simulation, the working condition recording module can comprehensively record the explosive forming process, and the extracted working condition characteristics provide important support for subsequent stability analysis, the real-time stability analysis module enhances the dynamic evaluation of the safety of the explosive through real-time monitoring of the working condition data of the explosive, the built stability environment field provides a scientific basis for the simulation of stacking storage, the stacking simulation module effectively reflects the bearing pressure under different stacking conditions in combination with the environment field, the inferred stacking bearing pressure data provides a key parameter for safety evaluation, the real-time safety evaluation module combines the working condition data and the bearing pressure to comprehensively evaluate the safety of the stacking, generates a real-time safety report, the calculation of the stacking cumulative bearing pressure provides data support for long-term safety, and the safety joint evaluation module can continuously predict pressure deformation, provides a warning for long-term storage and use of the explosive, and the comprehensive safety evaluation system formed promotes the intelligentization and modernization of explosive management, significantly improves the safety standards and technical level of the industry, ensures the stability and safety of the explosive in various environments, provides strong technical support for the safety management of future explosive applications, promotes the technical progress and standardized development of the explosive industry, and improves the overall safety management ability and efficiency.

[0113] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the application being defined by the attached claims rather than the above description, and it is therefore intended to encompass all variations falling within the meaning and scope of the equivalent elements of the application file.

[0114] The above description is merely one specific implementation of the application, which enables those skilled in the art to understand or implement the application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the application. Therefore, the application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting the safety of explosives based on a digital twin model, characterized in that, Includes the following steps: Step S1: Obtain the raw explosive data; extract the particle size distribution data from the raw explosive data; construct a scaled-down simulation explosive model based on the particle size distribution data. Step S2: Record the explosive forming process; Extract the working condition characteristics in the explosive forming process, and label the explosive working condition data of each stage of the explosive forming process based on the working condition characteristics. Step S3: Perform real-time stability analysis of explosives using explosive working condition data to obtain real-time stability data of explosives; construct an explosive stability environment field based on the real-time stability data of explosives, specifically: construct a digital spatial model based on the real-time stability data of explosives, which is divided into three-dimensional grids and assigned local stability vectors in each grid cell as the explosive stability environment field; Step S4: Simulate the stacking and storage of explosives on the simulated explosive model based on the explosive stability environment field to obtain simulated explosive stacking data; infer the explosive stacking pressure data based on the simulated explosive stacking data, specifically: perform dynamic simulation of explosive stacking and storage on the simulated explosive model in the explosive stability environment field, use a physics engine to calculate the stacking structure, and record the contact information between the models to generate simulated explosive stacking data; based on the contact information, infer the stacking pressure data borne by each simulated explosive model through contact mechanics analysis; Step S5: Perform a real-time stacking safety assessment based on the pressure borne by the explosive stack and the explosive working condition data to generate a real-time explosive stacking safety data. The cumulative pressure borne by the stack of explosives can be estimated by using data on the pressure borne by the stack of explosives. Step S6: Based on the cumulative pressure borne by the stack, predict the continuous pressure deformation of the simulated explosive model to obtain the predicted explosive deformation data; The safety of continuous stacking is predicted based on the predicted explosive deformation data, and a safety assessment is performed in combination with the real-time explosive stacking safety.

2. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain raw explosive data; extract the basic particle properties from the raw explosive data; Step S12: Perform particle size classification measurement based on the basic physical property parameters of the particles to obtain particle size distribution data. A particle sieving device is used, and the particle size screen specifications are set to a total of 8 sieve segments between 10-500μm. Step S13: Construct a particle frequency ratio based on particle size distribution data, wherein the frequency ratio value is limited to within ±5% error range, and the distribution ratio of each particle size segment is controlled within the range of 5%~30%; Step S14: Perform three-dimensional filling simulation on the original explosive data through particle frequency ratio to generate simulated filling configuration, wherein the compaction degree of simulated filling is limited to 62%~76%, and the size of the simulation unit is controlled within 10cm³. Step S15: Reconstruct a proportional simulated explosive model based on the simulated filler configuration and the basic physical properties of the particles.

3. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Record the explosive forming process; sort the process nodes in the explosive forming process by time to generate an explosive process timeline, where the smallest unit of each node is 15 seconds, and the total recording period is not less than 180 minutes. Step S22: Mark the nodes of the explosive process timeline to generate multiple stage division data, where the stage segment length is set between 5 and 30 minutes; Step S23: Identify the working condition characteristics of each stage in the stage zoning data; Step S24: Mark the explosive operating condition data of each stage of the explosive forming process according to the operating condition characteristics of each stage, wherein the data sampling frequency should be limited to 1-10Hz.

4. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Divide the explosive working condition data into time-series windows to obtain explosive working condition time-series segments; construct a continuous working condition sequence based on the explosive working condition time-series segments; Step S32: Identify the loading rate change sequence in the continuous working condition sequence; perform explosive stability analysis based on the loading rate change sequence, and extract real-time stability data to obtain real-time explosive stability data; Step S33: Determine the steady-state region based on the real-time stability data of the explosive; perform field parameterization on the steady-state region to obtain field parameter data; Step S34: Perform scene simulation reconstruction based on field parameters to build an explosive stability environment field.

5. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S4, which involves simulating the stacking and storage of explosives based on the explosive stability environment field, includes: Based on the explosive stability environment field, the simulated explosive model is mapped to stacked elements to obtain the explosive stacking elements; Spatial layout mapping is performed based on the explosive stacking units to obtain the spatial layout structure; Determine the size of the simulated explosive model; Stacking path planning is performed based on the spatial layout structure and the size of the explosive model to obtain stacking path data; Based on the stacking path data and the explosive model size, the simulated explosive model is stacked and stored to obtain the simulated explosive stacking data.

6. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S4 involves inferring the pressure data that the explosive stack can withstand based on the simulated explosive stack data, including: Confirm the center of gravity of the simulated explosive model; Based on the simulated explosive stacking data and the centroid of the explosive model, the interaction force data between adjacent simulated explosive models is calculated. Determine the explosive stacking framework based on simulated explosive stacking data; A stacked pressure conduction chain is constructed using the explosive stacking framework and the aforementioned interaction force data; The pressure data of the explosive stack under pressure in each simulated explosive model are inferred from the stack pressure transmission chain.

7. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S5 includes the following steps: Step S51: Determine the completion rate of the explosive working condition data based on the preset standard working condition data to obtain the explosive working condition completion rate data; Step S52: Confirm the explosive pressure threshold based on the explosive condition completion data; conduct a real-time stacking safety assessment by using the explosive stacking pressure and the explosive pressure threshold to generate real-time explosive stacking safety data. Step S53: Identify pressure change data during the stacking process based on the pressure data of the explosive stack; Step S54: Infer the stacked pressure of each simulated explosive model segment by using pressure change data; calculate the cumulative stacked pressure of the simulated explosive model based on the simulated explosive model.

8. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S6, which involves predicting the sustained pressure deformation of the simulated explosive model based on the cumulative pressure it withstands, includes: Material deconstruction is performed on the simulated explosive model to obtain material deconstruction data; Pressure response analysis of the simulated explosive model is performed based on material deconstruction data to generate explosive pressure response data; By accumulating pressure through stacking, the simulated explosive model is subjected to continuous pressure simulation, thereby generating continuous pressure simulation data. Deformation response is predicted based on the pressure response data of explosives and the continuous pressure simulation data, thereby obtaining the predicted explosive deformation data.

9. The method for predicting the safety of explosives based on a digital twin model according to claim 1, characterized in that, Step S6 involves predicting the continuous stacking safety based on the predicted explosive deformation data, and conducting a safety assessment in conjunction with the real-time explosive stacking safety, including: The simulated explosive model is subjected to continuous pressure based on the cumulative pressure of the stack, thereby observing the continuous deformation data of the explosive model. Determine the explosive deformation threshold using continuous deformation data from an explosive model; Determine the dangerous deformation distance based on the explosive deformation threshold and predicted explosive deformation data; Continuous stacking safety is generated by predicting the distance of dangerous deformation. The safety of explosives is assessed by combining real-time explosive stacking safety data, thereby obtaining explosive safety data.

10. A system for predicting the safety of explosives based on a digital twin model, characterized in that, For executing the explosive safety prediction method based on a digital twin model as described in claim 1, the explosive safety prediction system based on a digital twin model comprises: The model simulation module is used to acquire raw explosive data; extract particle size distribution data from the raw explosive data; and construct a scaled-down simulated explosive model based on the particle size distribution data. The working condition recording module is used to record the explosive forming process; extract the working condition characteristics in the explosive forming process, and annotate the explosive working condition data of each stage of the explosive forming process based on the working condition characteristics. The stability analysis module is used to perform real-time stability analysis of explosives using explosive operating condition data, thereby obtaining real-time stability data of the explosives; and to construct the stability environment field of the explosives based on the real-time stability data of the explosives. The stacking simulation module is used to simulate the stacking and storage of explosives based on the explosive stability environment field to obtain simulated explosive stacking data; and to infer the pressure data that the explosive stacks bear based on the simulated explosive stacking data. The real-time safety assessment module is used to perform real-time stack safety assessment based on the pressure borne by the explosive stack and the explosive working condition data, so as to generate real-time explosive stack safety; and to calculate the cumulative pressure borne by the stack through the explosive stack pressure data. The safety joint assessment module is used to predict the continuous pressure deformation of the simulated explosive model based on the cumulative pressure of the stack, so as to obtain the predicted explosive deformation data; based on the predicted explosive deformation data, the continuous stack safety is predicted, and the safety assessment is carried out in combination with the real-time explosive stack safety.

Images