Automatic filling control method and device for large-diameter emulsion explosive

Abstract

The application provides an automatic filling control method and device for large-diameter emulsion explosive, and relates to the technical field of intelligent control. The method comprises the following steps: detecting the state of the emulsion explosive to be filled in an emulsion explosive filling machine to obtain an explosive state detection result; a predetermined packaging bag is sleeved to a filling interface through a bag taking unit and a bag sleeving unit; filling unit of the emulsion explosive filling machine is subjected to filling control decision optimization according to explosive basic information, explosive filling environment information and a target charge quantity to obtain a filling optimization strategy; the filling unit fills the emulsion explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and the filled explosive is sealed through an aluminum wire stamping unit of the emulsion explosive filling machine. The application solves the technical problem of low filling efficiency caused by low automation degree of filling control of large-diameter emulsion explosive, realizes automatic filling of large-diameter emulsion explosive through filling decision space loss inspection and optimization, and improves the filling efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, and particularly relates to an automatic filling control method and device for large-diameter emulsion explosive. BACKGROUND

[0002] At present, the automatic filling technology of emulsion explosive has made progress to a certain extent, especially for conventional-diameter emulsion explosive, the filling process has achieved a certain degree of automation. In the filling of conventional-diameter emulsion explosive, a rotary type carding, a spray type filling or other mechanical operation modes are usually adopted, which greatly improves the efficiency and precision of filling. However, due to the particularity of large-diameter emulsion explosive, the traditional equipment cannot quickly and accurately complete the filling task of large-diameter emulsion explosive. For the filling equipment of large-diameter emulsion explosive, the degree of automation is low, and the filling process may rely on manual intervention to adjust the packaging bag, the filling amount and other parameters, thereby causing the technical problems of low filling efficiency, poor precision and poor production line collaboration.

[0003] In summary, in the prior art, there is a technical problem of low filling efficiency due to the low degree of automation of the filling control of large-diameter emulsion explosive. SUMMARY

[0004] The purpose of the present application is to provide an automatic filling control method and device for large-diameter emulsion explosive, so as to solve the technical problem of low filling efficiency due to the low degree of automation of the filling control of large-diameter emulsion explosive in the prior art.

[0005] In view of the above problems, the present application provides an automatic filling control method and device for large-diameter emulsion explosive.

[0006] The first aspect provides an automatic filling control method of large-diameter emulsion explosive, which is realized by an automatic filling control device of large-diameter emulsion explosive, and the method comprises the following steps: detecting the state of emulsion explosive to be filled in an emulsion explosive filling machine to obtain an explosive state detection result; when the explosive state detection result is normal, a predetermined packaging bag meeting a target charge quantity is sleeved to a filling interface of the emulsion explosive filling machine through a bag taking unit and a bag sleeving unit of the emulsion explosive filling machine; a filling unit of the emulsion explosive filling machine is controlled and decided according to explosive basic information, explosive filling environment information and the target charge quantity, a first filling decision space is established, which comprises the following steps: a filling control scheme retrieval is performed on the filling unit according to the explosive basic information, the explosive filling environment information and the target charge quantity, a retrieval filling scheme set is obtained, a trigger feature is combed according to the retrieval filling scheme set, a multi-element filling control domain corresponding to a multi-element filling control variable is established, the target charge quantity is taken as a filling control target, control parameter decision combination is performed on the multi-element filling control variable according to the multi-element filling control domain, and the first filling decision space is generated; a loss verification optimization is performed on the first filling decision space according to an explosive filling loss factor, a filling decision second space is obtained, which comprises the following steps: the explosive filling loss factor is set according to the explosive filling loss factor, wherein the explosive filling loss factor comprises explosive filling wall sticking loss, explosive filling cavity loss and explosive filling broken charge loss, a filling loss record learning is performed according to the explosive filling loss factor, an explosive filling loss prediction model is constructed, a loss verification is performed on the first filling decision space according to the explosive filling loss prediction model and the explosive filling loss constraint, a plurality of filling loss verification results are obtained, the first filling decision space is optimized and selected according to the plurality of filling loss verification results, and the filling decision second space is generated; a filling optimization strategy is obtained by performing variation expansion optimization on the filling decision second space according to an explosive filling loss cost function and a filling variation quantitative function, which comprises the following steps: the explosive filling loss factor is weighted and distributed to obtain the explosive filling loss cost function, a filling loss cost calculation is performed on each filling decision in the filling decision second space according to the explosive filling loss cost function to obtain a filling loss cost first distribution, the filling decision second space is optimized and selected according to a predetermined filling loss cost coefficient based on the filling loss cost first distribution to obtain a filling decision third space, the filling decision third space is expanded and varied according to the explosive filling loss cost function, the filling variation quantitative function and the explosive filling loss factor to construct a filling decision fourth space, and the filling loss cost minimization optimization is performed according to the filling decision fourth space to generate the filling optimization strategy.The filling unit fills the explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and seals the filled explosive through the aluminum wire clamping unit of the emulsion explosive filling machine.

[0007] Optionally, the real-time status data of the explosive to be filled is obtained; standard status data of the explosive is matched according to the specifications and model of the explosive to be filled; the real-time status data and the standard status data of the explosive are input into a Siamese neural network to obtain the standard coefficient of the explosive status; it is determined whether the standard coefficient of the explosive status is greater than or equal to the standard threshold of the explosive status, and the explosive status detection result is generated.

[0008] Optionally, if the standard coefficient of explosive state is greater than or equal to the standard threshold of explosive state, the explosive state detection result is normal; if the standard coefficient of explosive state is less than the standard threshold of explosive state, the explosive state detection result is abnormal.

[0009] Optionally, based on the first filling decision space, a first filling decision is extracted; the explosive basic information, the explosive filling environment information, and the first filling decision are input into the explosive filling loss model to obtain a first explosive filling loss prediction result; it is determined whether the first explosive filling loss prediction result meets the explosive filling loss constraint, a first filling loss inspection result is generated, and the first filling loss inspection result is added to the plurality of filling loss inspection results.

[0010] Optionally, the number of variations in the third space of the filling decision is calculated according to the filling variation quantification function to obtain a variation distribution; the third space of the filling decision is mutated according to the variation distribution to obtain a first variation domain of the filling decision; the first variation domain of the filling decision is optimized by loss verification according to the explosive filling loss factor to obtain a second variation domain of the filling decision; the second variation domain of the filling decision is calculated by filling loss cost according to the explosive filling loss cost function to obtain a second distribution of filling loss cost; based on the second distribution of filling loss cost, the second variation domain of the filling decision is optimized and selected according to the predetermined filling loss cost coefficient to obtain a third variation domain of the filling decision; the third space of the filling decision is expanded according to the third variation domain of the filling decision to generate the fourth space of the filling decision.

[0011] Optionally, the filling variation quantification function is: Among them, R i The coefficient representing the filling loss cost corresponding to the i-th filling decision, where i is a positive integer, S i The variable represents the number of variations corresponding to the i-th filling decision, Floor refers to rounding down, R0 represents the predetermined filling loss cost coefficient, and S maxS represents the upper limit of the number of variations in filling decisions. min Characterizes the lower limit of the number of variations in filling decisions.

[0012] Secondly, this application also provides an automatic filling control device for large-diameter emulsion explosives, used to execute an automatic filling control method for large-diameter emulsion explosives as described in the first aspect. The automatic filling control device for large-diameter emulsion explosives includes: a status detection module for detecting the status of the explosive to be filled in the emulsion explosive filling machine and obtaining an explosive status detection result; a packaging and fitting module for fitting a predetermined packaging bag meeting the target charge amount to the filling interface of the emulsion explosive filling machine through the bag-taking unit and bag-fitting unit of the emulsion explosive filling machine when the explosive status detection result is normal; and a decision space construction module for making filling control decisions for the filling unit of the emulsion explosive filling machine based on the explosive basic information, explosive filling environment information, and the target charge amount, establishing a first filling decision space, including: searching for filling control schemes for the filling unit based on the explosive basic information, the explosive filling environment information, and the target charge amount, obtaining a searched filling scheme set, and performing... The process involves: identifying triggering features and establishing a multi-dimensional filling control domain corresponding to the multi-dimensional filling control variables; using the target charge amount as the filling control objective; combining control parameters for the multi-dimensional filling control variables according to the multi-dimensional filling control domain to generate a first filling decision space; and a decision space optimization module, used to perform loss verification and optimization on the first filling decision space based on explosive filling loss factors to obtain a second filling decision space. This includes: setting explosive filling loss constraints based on the explosive filling loss factors, wherein the explosive filling loss factors include explosive filling wall adhesion loss, explosive filling void loss, and explosive filling breakage loss; learning filling loss records based on the explosive filling loss factors to construct an explosive filling loss prediction model; performing loss verification on the first filling decision space based on the explosive filling loss prediction model and the explosive filling loss constraints to obtain multiple filling loss verification results; and optimizing and selecting the first filling decision space based on the multiple filling loss verification results to generate the second filling decision space.The optimization strategy determination module is used to perform variation expansion optimization on the second space of the filling decision based on the explosive filling loss cost function and the filling variation quantitative function to obtain a filling optimization strategy. This includes: weighting the explosive filling loss factors to obtain the explosive filling loss cost function; calculating the filling loss cost for each filling decision in the second space of the filling decision based on the explosive filling loss cost function to obtain a first distribution of the filling loss cost; and, based on the first distribution of the filling loss cost, performing a predetermined filling loss cost coefficient on the second space of the filling decision. The process involves optimization and selection to obtain a third space for filling decisions. This third space is then expanded and modified based on the explosive filling loss cost function, the filling variation quantification function, and the explosive filling loss factor to construct a fourth space for filling decisions. The filling loss cost is minimized using this fourth space to generate the optimal filling strategy. A filling and sealing module is used by the filling unit to fill the explosive to be filled into the predetermined packaging bag according to the optimal filling strategy, and the filled explosive is sealed using the aluminum wire clamping unit of the emulsion explosive filling machine.

[0013] One or more technical solutions provided in this application have at least the following beneficial effects:

[0014] The state of the explosive to be filled in the emulsion explosive filling machine is detected to obtain the explosive state detection result. When the explosive state detection result is normal, the bag picking unit and bag putting unit of the emulsion explosive filling machine put a predetermined packaging bag that meets the target charge amount onto the filling interface of the emulsion explosive filling machine. Based on the basic explosive information, explosive filling environment information, and the target charge amount, the filling unit of the emulsion explosive filling machine makes filling control decisions to establish a first filling decision space, including: searching for filling control schemes for the filling unit based on the basic explosive information, explosive filling environment information, and target charge amount to obtain a searched filling scheme set, and then proceeding according to the searched filling scheme set... The process involves: analyzing triggering features to establish a multi-dimensional filling control domain corresponding to multiple filling control variables; using the target charge amount as the filling control objective; combining control parameters of the multi-dimensional filling control variables according to the multi-dimensional filling control domain to generate a first filling decision space; and performing loss verification and optimization on the first filling decision space based on explosive filling loss factors to obtain a second filling decision space. This includes: setting explosive filling loss constraints based on the explosive filling loss factors, wherein the explosive filling loss factors include explosive filling wall adhesion loss, explosive filling void loss, and explosive filling breakage loss; and performing filling loss record learning based on the explosive filling loss factors to construct an explosive filling loss prediction model. Based on the explosive filling loss prediction model, the first filling decision space is tested for loss according to the explosive filling loss constraint to obtain multiple filling loss test results. The first filling decision space is then optimized and selected based on these multiple filling loss test results to generate the second filling decision space. The second filling decision space is then expanded and optimized based on the explosive filling loss cost function and the filling variation quantitative function to obtain a filling optimization strategy. This strategy includes: weighting the explosive filling loss factors to obtain the explosive filling loss cost function; calculating the filling loss cost for each filling decision in the second filling decision space based on the explosive filling loss cost function; and obtaining... A first distribution of filling loss cost is obtained. Based on the first distribution of filling loss cost, the second filling decision space is optimized and selected according to a predetermined filling loss cost coefficient to obtain a third filling decision space. The third filling decision space is then expanded by mutation according to the explosive filling loss cost function, the filling variation quantification function, and the explosive filling loss factor to construct a fourth filling decision space. The filling loss cost is minimized according to the fourth filling decision space to generate the filling optimization strategy. The filling unit fills the explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and seals the filled explosive through the aluminum wire clipping unit of the emulsion explosive filling machine.In other words, by detecting the state of the explosive, once the detection is normal, the packaging bag is placed onto the interface. Based on the explosive information, environmental information, and target quantity, the optimal filling strategy is determined, the explosive is filled, and the sealing is completed. This achieves automated filling of large-diameter emulsion explosives, improves filling efficiency, and reduces production costs.

[0015] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an automatic filling control method for large-diameter emulsion explosives according to this application.

[0018] Figure 2 This is a schematic diagram of the structure of an automatic filling control device for large-diameter emulsion explosives according to this application.

[0019] Figure labeling: 11 State detection module, 12 Packaging module, 13 Decision space construction module, 14 Decision space optimization module, 15 Optimization strategy determination module, 16 Filling and sealing module. Detailed Implementation

[0020] This application provides an automated filling control method and equipment for large-diameter emulsion explosives, solving the technical problem of low filling efficiency due to the low degree of automation in the filling control of large-diameter emulsion explosives in the prior art. By detecting the state of the explosive, and upon confirming normal operation, a packaging bag is placed onto the interface. Based on explosive information, environmental information, and target quantity, a filling optimization strategy is determined, the explosive is filled, and sealed. This achieves automated filling of large-diameter emulsion explosives, improving filling efficiency and reducing production costs.

[0021] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0022] Example 1, please refer to the appendix. Figure 1 This application provides an automatic filling control method for large-diameter emulsion explosives. The automatic filling control method for large-diameter emulsion explosives is executed by an automatic filling control device for large-diameter emulsion explosives. The automatic filling control method for large-diameter emulsion explosives specifically includes the following steps:

[0023] S100: Performs condition detection on the explosives to be filled in the emulsion explosive filling machine and obtains the explosive condition detection results.

[0024] Furthermore, this application S100 includes:

[0025] Obtain real-time status data of the explosive to be filled; match standard status data of the explosive according to the specifications and model of the explosive to be filled; input the real-time status data and the standard status data of the explosive into a twin neural network to obtain the standard coefficient of the explosive status; determine whether the standard coefficient of the explosive status is greater than or equal to the standard threshold of the explosive status, and generate the explosive status detection result.

[0026] Furthermore, this application also includes the following steps:

[0027] If the standard coefficient of explosive state is greater than or equal to the standard threshold of explosive state, the explosive state detection result is normal; if the standard coefficient of explosive state is less than the standard threshold of explosive state, the explosive state detection result is abnormal.

[0028] Specifically, this involves acquiring real-time status data of the explosives to be filled, that is, monitoring and collecting parameters of the explosives in real time during the production process, including temperature, humidity, viscosity, density, particle size, and flowability. Based on the specifications and model of the explosives to be filled, standard state data of the explosives is determined by matching them in the database of the production control center. Standard state data refers to the ideal physical and chemical parameters of the explosives as specified according to their standard specifications and design requirements. Different models of explosives will have different standard state data. Explosive specifications and models refer to the product type determined according to the design, purpose, and production standards of the explosives.

[0029] Real-time monitored data of the explosive's current state and matching standard state data are input into a Siamese neural network. The Siamese network compares these two sets of data and calculates the standard coefficients of the explosive's state. A Siamese neural network is a neural network architecture that uses two identical neural networks to process two input data sets and outputs the similarity between them. The Siamese neural network consists of two identical subnetworks with shared parameters, and is trained using a contrastive learning algorithm to determine the similarity or difference between the two inputs.

[0030] The Siamese neural network compares these two sets of data, calculating the differences or similarities between them using similarity calculation methods within the neural network (typically Euclidean distance or cosine similarity). For example, if the real-time temperature of the explosive is 22°C, the humidity is 3%, and the particle size is 1.8 mm, while the standard temperature is 22.7°C, the humidity is 3.6%, and the particle size is 1.5 mm, the calculated temperature difference is 1, the humidity difference is 0.6, and the particle size difference is 0.3. Therefore, the standard coefficient of the explosive state calculated using Euclidean distance is 0.96.

[0031] The explosive condition standard threshold is a preset limit value, representing the minimum acceptable value of the explosive condition standard coefficient. If the explosive condition standard coefficient is higher than or equal to this threshold, it indicates that the explosive condition meets the standard requirements, and the next step can be performed. If it is lower than this threshold, it indicates that the explosive condition is abnormal and needs adjustment or reprocessing. In this case, an early warning mechanism needs to be triggered to remind the operator to take appropriate measures. When the control system detects that the explosive condition standard coefficient is less than the threshold, it will automatically trigger the early warning mechanism. Early warning can be implemented through audible and visual alarms, sending notifications to operators, or suspending the production line. For example, if the coefficient is less than 0.9, an alarm will sound and the warning information will be displayed on the control interface. Based on the comparison between the explosive condition standard coefficient and the explosive condition standard threshold, an explosive condition detection result is generated, indicating whether the explosive is in a normal production state.

[0032] S200: When the explosive status detection result is normal, the bag taking unit and bag putting unit of the emulsion explosive filling machine will put the predetermined packaging bag that meets the target charge amount onto the filling interface of the emulsion explosive filling machine.

[0033] Specifically, when the standard coefficient of the explosive state is greater than or equal to the set standard threshold, the explosive state is judged to be normal, and subsequent filling operations can continue. An emulsion explosive filling machine is a device used in the automated production process of emulsion explosives to fill explosives into predetermined packaging bags. Through the coordinated work of multiple units, it completes the automated steps of bag removal, bagging, filling, and sealing. To ensure safe production, emulsion explosive filling machines are typically equipped with explosion-proof devices, including explosion-proof motors, explosion-proof solenoid valves, and explosion-proof power distribution cabinets.

[0034] The bag-retrieving unit, part of the emulsion explosive filling machine, automatically retrieves a predetermined number of bags from the bag stack, preparing them for the next step and ensuring the correct number of bags are used for each filling. The bag-attaching unit places the retrieved bags onto the filling interface, ready to receive the explosives, ensuring the bags are correctly positioned for subsequent filling. The filling unit, the core of the emulsion explosive filling machine, is responsible for conveying the explosives from the filling system into the bags, ensuring precise control to guarantee the correct amount of explosive in each bag. The aluminum wire sealing unit seals the filled explosive bags using aluminum wire or other sealing materials to prevent leakage or environmental contamination. The aluminum wire sealing device provides secure sealing to ensure the stability of the explosives during storage and transportation.

[0035] Explosion-proof motors, explosion-proof solenoid valves, and explosion-proof distribution cabinets are used to improve equipment safety, especially in highly hazardous environments. The production process of emulsion explosives may generate explosive gases or dust; therefore, equipment must be equipped with explosion-proof devices to prevent electrical equipment from causing sparks or malfunctions, thereby improving operational safety.

[0036] The bag-picking unit automatically retrieves a predetermined number of bags from the bag stack, ensuring accurate bag quantity for each filling operation. This process is automated using sensors and a robotic arm. Simultaneously, it must ensure the bags contain the target explosive charge. The bag-fitting unit is responsible for fitting the bags from the bag-picking unit onto the filling interface, ensuring a perfect fit for the subsequent explosive filling process.

[0037] S300: Based on the basic information of the explosive, the information of the explosive filling environment, and the target charge amount, the filling unit of the emulsion explosive filling machine makes a filling control decision and establishes a first filling decision space, including: searching for filling control schemes for the filling unit based on the basic information of the explosive, the information of the explosive filling environment, and the target charge amount to obtain a searched filling scheme set; sorting out the trigger features based on the searched filling scheme set to establish a multi-dimensional filling control domain corresponding to the multi-dimensional filling control variables; taking the target charge amount as the filling control target; and combining control parameters for the multi-dimensional filling control variables based on the multi-dimensional filling control domain to generate the first filling decision space.

[0038] Specifically, this involves acquiring basic information about the explosives to be filled, including fundamental data related to the explosives themselves, such as their type, specifications, model, density, composition, and sensitivity. Environmental information related to the filling process, such as temperature, humidity, air pressure, and equipment status, is also obtained through sensors and other equipment on the production line. Environmental factors directly affect the filling effect, especially in the production of sensitive substances like explosives; changes in environmental factors can lead to errors or safety hazards during the filling process. The target charge amount refers to the amount of explosives that need to be loaded.

[0039] Based on the input basic information of the explosive, the filling environment information, and the target charge amount, the system searches the pre-stored filling control scheme data in the filling unit to find a matching set of filling schemes. For example, for a specific type and specification of explosive, under specific environmental conditions, multiple different filling schemes will be retrieved for different target charge amounts. Retrieving the set of filling schemes refers to using the basic information of the explosive, the filling environment information, and the target charge amount to query the database and select a set of control schemes suitable for the current filling task.

[0040] The retrieved filling scheme set was analyzed, and key variables affecting the filling results were extracted through trigger feature analysis. Trigger feature analysis refers to identifying and extracting key features affecting filling accuracy, efficiency, and quality by analyzing multiple retrieved filling scheme sets. Through analysis of the filling scheme set, it was identified which variables have a critical impact on the accuracy, stability, and filling speed of the final charge. For example, changes in temperature may affect the flowability of the explosive, while changes in pressure may affect the density of the explosive.

[0041] Based on the identified key variables, a multivariate filling control domain is established, which is a multi-dimensional control space that defines the optimal value range of each control variable and their interrelationships. For example, the optimal temperature may be between 18-22℃, and the optimal flow rate may be between 3L / min and 5L / min. By defining the value ranges of these variables, they can be dynamically adjusted in actual production according to different environmental conditions and target charge amounts.

[0042] Using the target dosage as the control objective, the system performs control parameter decision combinations on the multivariate filling control variables according to the multivariate filling control domain. This adjusts the values ​​of the multivariate filling control variables, generating a series of filling parameter combinations, thus constructing the first filling decision space. This space represents all possible parameter combinations and their corresponding filling results. Each combination in the first filling decision space represents a different control scheme. By selecting the decision space most suitable for the current conditions, accurate filling of the target dosage can be ensured. Optimizing the combination of control variables and automatically adjusting the filling control scheme enhances the flexibility of the production process.

[0043] S400: Optimize the first filling decision space based on the explosive filling loss factor to obtain a second filling decision space, including: setting explosive filling loss constraints based on the explosive filling loss factor, wherein the explosive filling loss factor includes explosive filling wall adhesion loss, explosive filling void loss, and explosive filling breakage loss; learning filling loss records based on the explosive filling loss factor to construct an explosive filling loss prediction model; based on the explosive filling loss prediction model, performing loss checks on the first filling decision space according to the explosive filling loss constraints to obtain multiple filling loss check results; and optimizing and selecting the first filling decision space based on the multiple filling loss check results to generate the second filling decision space.

[0044] Specifically, corresponding loss constraints are set according to different explosive filling loss factors (such as wall adhesion loss, void loss, and explosive breakage loss). During the filling process of explosives, some material loss may occur due to factors such as equipment, material characteristics, and operating conditions.

[0045] Explosive wall adhesion loss refers to the loss of explosives during filling due to their physical properties (such as strong adhesion), causing them to adhere to the internal surfaces of equipment (such as pipes, containers, feeding devices, etc.) during the filling process, resulting in some explosives failing to enter the packaging bag or target location. Explosive cavity loss refers to the material loss caused by gaps or uneven distribution of explosives within the packaging bag or container during the filling process, resulting in insufficient filling. For example, some areas of bagged explosives may remain unfilled during the filling process, especially when the filling volume is large. Explosive interruption loss refers to the loss caused by interruptions in explosive filling or damage to explosive particles during the filling process due to equipment failure, operational errors, or other external factors. This typically occurs when equipment operation is unstable or operating conditions do not meet requirements.

[0046] Based on these loss factors, maximum allowable values ​​are set for each type of loss. For example, the maximum allowable percentage of wall adhesion loss is 2% of the total filling volume, the void loss per bag of explosive must not exceed 3%, and the allowable breakage loss in each batch of production is 1% of the total production volume. These constraints are set based on numerous trials and production experience, with the aim of minimizing waste and losses while ensuring filling accuracy and explosive quality. The filling control strategy is automatically optimized using these established loss constraints.

[0047] Collecting historical filling data requires including records of various loss factors during the filling process, i.e., losses incurred during filling operations over a past period. The collected data undergoes preprocessing to remove missing values, outliers, or inconsistent data. All input variable values ​​are normalized or standardized to ensure that all features are on the same scale.

[0048] The neural network model architecture is constructed, including an input layer, hidden layers, and an output layer. The input layer contains all feature data, such as explosive type, temperature, flow rate, and equipment status. Each input feature corresponds to a node in the network. The number of hidden layers and nodes needs to be adjusted according to the actual situation; the purpose is to perform non-linear transformations on the input data through inter-layer weight connections, thereby extracting patterns from the data. The nodes in the output layer represent the filling loss predicted by the model. Multiple output nodes can be set to predict different types of losses (such as wall adhesion loss, cavitation loss, and explosive depletion loss).

[0049] The preprocessed data is divided into training and testing sets. The training set is used to train the model, and the testing set is used to validate the model's generalization ability; typically, the training set comprises 80% and the testing set 20%. An appropriate loss function (such as Mean Squared Error, MSE) is chosen to measure the model's prediction error, calculating the difference between the true and predicted values ​​using MSE. Based on the loss, the weights of the neural network are adjusted so that the model's output is as close as possible to the actual loss data. During training, the network weights are iteratively optimized multiple times, allowing the model to gradually reduce its prediction error.

[0050] After training, the neural network model is evaluated using a test set to check its predictive ability. Convergence criteria are set for the model, such as a validation set loss variation of less than 0.01 for five consecutive rounds or a training set accuracy reaching 95%. Training stops when the model reaches convergence, and the explosive filling loss prediction model is then applied to predict filling losses. Input data includes various filling conditions (such as temperature, flow rate, and pressure), while output data consists of the corresponding loss values ​​(wall adhesion loss, cavitation loss, and explosive breakage loss).

[0051] A loss prediction model for explosive filling is used to test the first filling decision space. Each control scheme in the first filling decision space is input into the model to predict the potential losses for each scheme during actual filling, including the filling wall adhesion loss coefficient, filling void loss coefficient, and filling interruption loss coefficient for each scheme. The model then determines whether the prediction results of each scheme meet the explosive filling loss constraints, identifying all schemes that meet these constraints and eliminating those that do not. The remaining schemes constitute the second filling decision space.

[0052] The second space for filling decisions encompasses all filling control decisions that meet loss constraints, ensuring that explosive loss is minimized while achieving the target charge amount with precision. By predicting and verifying losses for each filling scheme, explosive loss during the filling process is minimized, improving production efficiency and economic benefits.

[0053] Furthermore, this application also includes the following steps:

[0054] Based on the first filling decision space, a first filling decision is extracted; the explosive basic information, the explosive filling environment information, and the first filling decision are input into the explosive filling loss model to obtain a first explosive filling loss prediction result; it is determined whether the first explosive filling loss prediction result meets the explosive filling loss constraint, a first filling loss inspection result is generated, and the first filling loss inspection result is added to the plurality of filling loss inspection results.

[0055] Specifically, a decision is randomly selected from the first filling decision space as the first filling decision. This decision includes the specific values ​​of the explosive filling operation parameters, such as temperature, flow rate, and pressure, determining how to perform the actual filling. All decisions in the first filling decision space satisfy the target charge amount. The first filling decision, along with the basic information of the explosive and the filling environment information, is input into the explosive filling loss model. The output of the model is the predicted result of the first explosive filling loss, that is, the estimated possible loss under given operating conditions.

[0056] The predicted loss results for the first explosive filling process include the filling wall adhesion loss coefficient, the filling void loss coefficient, and the filling interruption loss coefficient corresponding to the first filling decision. In other words, during the filling process of the first filling decision, the loss is due to the portion of explosive material adhering to the equipment wall and not being effectively filled, the loss due to the presence of air bubbles or voids that prevent the explosive from completely filling the packaging bag, and the loss due to the breakage or damage of the explosive.

[0057] Explosive filling loss constraints are pre-defined standards used to limit the maximum allowable loss in each filling process. The predicted filling loss result of the first explosive is compared with the explosive filling loss constraints to verify whether production requirements are met. For example, if the predicted filling wall adhesion loss coefficient is 2.5%, it means that 2.5% of the explosive cannot be filled due to adhesion to the equipment wall; the filling void loss coefficient is 1.2%, it means that 1.2% of the explosive fails to completely fill the packaging bag; and the filling breakage loss coefficient is 0.3%, representing losses due to explosive breakage or fracture. The wall adhesion loss coefficient of 2.5% is higher than the 2% limit, the void loss coefficient of 1.2% is lower than the 3% limit, and the breakage loss coefficient of 0.3% is lower than the 1% limit. Since the wall adhesion loss coefficient exceeds the allowable range, it is determined that the constraint conditions are not met. The first filling loss inspection result (not exceeding) is added to multiple filling loss inspection results.

[0058] The same steps are performed on other decision schemes in the first filling decision space to obtain the loss test result for each scheme, resulting in multiple filling loss test results. Decision schemes with multiple unsatisfactory filling loss test results are eliminated, leaving only those that satisfy the explosive filling loss constraints, forming the second filling decision space. Through accurate loss prediction and optimization, schemes that do not meet the loss constraints are eliminated, narrowing the scope of optimization and ensuring that the amount of explosive in each package meets the target requirements and quality standards.

[0059] S500: The second space of the filling decision is expanded and optimized based on the explosive filling loss cost function and the filling variation quantification function to obtain a filling optimization strategy. This includes: weighting the explosive filling loss factor to obtain the explosive filling loss cost function; calculating the filling loss cost for each filling decision in the second space of the filling decision based on the explosive filling loss cost function to obtain a first distribution of filling loss costs; optimizing and selecting the second space of the filling decision based on the first distribution of filling loss costs and a predetermined filling loss cost coefficient to obtain a third space of filling decisions; expanding and mutating the third space of the filling decision based on the explosive filling loss cost function, the filling variation quantification function, and the explosive filling loss factor to construct a fourth space of filling decisions; and minimizing the filling loss cost based on the fourth space of the filling decision to generate the filling optimization strategy.

[0060] Specifically, weights are assigned to the explosive filling loss factors to determine the impact of each loss on overall production efficiency and cost. The explosive filling loss cost function is L = α·C + β·H + Υ·B, where L is the total filling loss cost, C is the wall adhesion loss coefficient, representing the actual proportion of wall adhesion loss, H is the void loss coefficient, representing the actual proportion of void loss, B is the interruption loss coefficient, representing the actual proportion of interruption loss, and α, β, and Υ are the corresponding loss weights, typically between 0 and 1, representing the contribution ratio of each loss factor to the overall loss cost. For example, wall adhesion loss may have a significant impact on overall production, therefore it should be assigned a higher weight, while interruption loss is relatively small and can be assigned a lower weight. For instance, the weight of wall adhesion loss is 0.4, the weight of void loss is 0.3, and the weight of interruption loss is 0.3.

[0061] The filling loss cost is calculated for each filling decision in the second filling decision space using the explosive filling loss cost function. Specifically, the filling wall adhesion loss coefficient, filling void loss coefficient, and filling interruption loss coefficient are extracted from the explosive filling loss prediction results obtained by inputting the scheme into the explosive filling loss model, and substituted into the explosive filling loss cost function to obtain the first distribution of the filling loss cost. For example, the first distribution of the filling loss cost in the second filling decision space is shown in Table 1:

[0062] Table 1. First Distribution of Filling Loss Costs in the Second Filling Decision Space

[0063]

[0064] The first distribution of filling loss costs includes the loss costs of all filling decisions in the second decision space. The predetermined filling loss cost coefficient is a pre-set standard used to guide the tolerance for loss during the optimization process; it represents the upper limit of loss. Only when the loss cost of a decision is less than or equal to this coefficient is it considered acceptable and proceeds to the next optimization stage.

[0065] Select all filling decisions in the second filling decision space that correspond to filling loss cost coefficients less than the predetermined filling loss cost coefficient, and eliminate schemes that are greater than or equal to the predetermined filling loss cost coefficient, thus forming the third filling decision space.

[0066] The third space of filling decisions is expanded by the third mutation domain of the filling decision, resulting in the fourth space of filling decisions. This expanded decision space contains all filling decisions after mutation, optimization, and screening. The optimization strategy for minimizing filling loss cost is then performed within the fourth space. This involves calculating the loss cost of all filling decisions in the fourth space and finding the solution with the minimum loss cost, which is then used as the filling optimization strategy.

[0067] Furthermore, this application also includes the following steps:

[0068] The number of variations in the third space of the filling decision is calculated according to the filling variation quantification function to obtain the variation distribution; the third space of the filling decision is mutated according to the variation distribution to obtain the first variation domain of the filling decision; the first variation domain of the filling decision is optimized by loss test according to the explosive filling loss factor to obtain the second variation domain of the filling decision; the filling loss cost is calculated according to the explosive filling loss cost function to obtain the second distribution of the filling loss cost; the second variation domain of the filling decision is optimized and selected according to the predetermined filling loss cost coefficient based on the second distribution of the filling loss cost to obtain the third variation domain of the filling decision; the third space of the filling decision is expanded according to the third variation domain of the filling decision to generate the fourth space of the filling decision.

[0069] Furthermore, this application also includes the following steps:

[0070] The quantitative function for filling variation is: Among them, R i The coefficient representing the filling loss cost corresponding to the i-th filling decision, where i is a positive integer, S i The variable represents the number of variations corresponding to the i-th filling decision, Floor refers to rounding down, R0 represents the predetermined filling loss cost coefficient, and S max S represents the upper limit of the number of variations in filling decisions. min Characterizes the lower limit of the number of variations in filling decisions.

[0071] Specifically, the filling variation quantification function is: Among them, R i The filling loss cost coefficient, representing the filling decision at the i-th bottling stage, is a value calculated during the filling process. It represents the loss cost of a given decision. A larger loss cost coefficient indicates a larger loss for that decision, while a smaller coefficient indicates a smaller loss. i is a positive integer, and S... i The number of mutations represents the number of variations corresponding to the i-th filling decision. The number of mutations measures the degree to which the filling decision needs adjustment during the optimization process; a higher number of mutations means a larger adjustment is required. `Float` refers to rounding down to the nearest integer. `R0` represents the predetermined filling loss cost coefficient, a preset threshold that controls the maximum acceptable loss cost. If R... i A value greater than or equal to R0 indicates that the cost of the proposed solution exceeds the predetermined tolerance range. However, this value does not exist in the third space of the filling decision, because such solutions have already been eliminated as mentioned above. max S represents the upper limit of the number of variations in filling decisions. min Characterizes the lower limit of the number of variations in filling decisions.

[0072] The number of mutations for all filling decisions in the third space of the filling decision process is calculated to obtain the number of mutations for each filling decision, thus yielding a mutation distribution. This distribution includes the number of mutations for each filling decision within the third space. For example, assuming there are 5 filling decisions, the calculated number of mutations might be [2, 3, 4, 5, 6]. This distribution represents how many mutations are required for each filling decision. The filling decisions are then adjusted based on this mutation distribution. For instance, if the number of mutations for a filling decision is 3, the operational parameters of that decision (such as filling speed, pressure, time, etc.) will be adjusted 3 times. The final mutated filling decisions generate the first mutation domain of the filling decisions, containing all mutated schemes.

[0073] Repeat the aforementioned steps, and perform loss verification and optimization on the first variation domain of the filling decision based on the explosive filling loss factor. That is, input all the schemes in the first variation domain of the filling decision into the explosive filling loss model for loss prediction, and determine whether the prediction results meet the explosive filling loss constraints. Select the schemes that meet the explosive filling loss constraints to obtain the second variation domain of the filling decision.

[0074] All schemes in the second variation domain of the filling decision are input into the explosive filling loss cost function, and the filling scheme with a loss cost coefficient less than the predetermined filling loss cost coefficient is found to obtain the third variation domain of the filling decision. In other words, the same optimization process is performed on the mutated filling decision as on the optimization space before the mutation.

[0075] The third space of filling decisions is expanded by extending the third mutation domain of the filling decision process, resulting in the fourth space of filling decisions. This expanded decision space contains all filling decisions after mutation, optimization, and screening. Minimizing the filling loss cost is then performed within the fourth space. This involves calculating the loss cost of all filling decisions in the fourth space and finding the solution with the minimum loss cost, which serves as the filling optimization strategy. By minimizing the filling loss cost within the fourth space, an optimized filling optimization strategy is generated. This strategy maximizes production efficiency and minimizes various losses through precise control of key parameters in the filling process, ultimately achieving a more efficient and precise explosive filling process.

[0076] S600: The filling unit fills the explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and seals the filled explosive through the aluminum wire clipping unit of the emulsion explosive filling machine.

[0077] Specifically, the filling unit fills the explosives into predetermined packaging bags according to the previously optimized filling optimization strategy, i.e., the optimal filling control parameters (such as filling speed, pressure, time, etc.). The filling unit controls the filling operation according to the filling optimization strategy. When the filling volume reaches the predetermined target, the filling process automatically stops, ensuring that the amount of explosives in each packaging bag meets the standard. After the explosives are filled, the next step is to seal the packaging bags using the aluminum wire clamping unit of the emulsion explosive filling machine to ensure that the explosives do not leak or become contaminated by external sources.

[0078] The aluminum wire securing unit uses specialized equipment to fix aluminum wire to the opening of the packaging bag. Aluminum wire is a common material used to secure packaging bags, effectively preventing the entry of external substances or leakage of explosives. After sealing, the airtightness of the packaging bag needs to be tested. Sensors embedded in the filling machine detect whether the seal is secure, ensuring that the packaging bag is free of air leaks or openings. Through an optimization strategy for filling, the efficiency and accuracy of the explosive filling process are ensured, reducing errors and waste caused by manual operation. The use of the aluminum wire securing unit guarantees the airtightness of the explosive packaging, avoiding potential leakage or contamination problems during transportation and storage.

[0079] In summary, the automatic filling control method for large-diameter emulsion explosives provided in this application has the following beneficial effects:

[0080] The state of the explosive to be filled in the emulsion explosive filling machine is detected to obtain the explosive state detection result. When the explosive state detection result is normal, the bag picking unit and bag putting unit of the emulsion explosive filling machine put a predetermined packaging bag that meets the target charge amount onto the filling interface of the emulsion explosive filling machine. Based on the basic explosive information, explosive filling environment information, and the target charge amount, the filling unit of the emulsion explosive filling machine makes filling control decisions to establish a first filling decision space, including: searching for filling control schemes for the filling unit based on the basic explosive information, explosive filling environment information, and target charge amount to obtain a searched filling scheme set, and then proceeding according to the searched filling scheme set... The process involves: analyzing triggering features to establish a multi-dimensional filling control domain corresponding to multiple filling control variables; using the target charge amount as the filling control objective; combining control parameters of the multi-dimensional filling control variables according to the multi-dimensional filling control domain to generate a first filling decision space; and performing loss verification and optimization on the first filling decision space based on explosive filling loss factors to obtain a second filling decision space. This includes: setting explosive filling loss constraints based on the explosive filling loss factors, wherein the explosive filling loss factors include explosive filling wall adhesion loss, explosive filling void loss, and explosive filling breakage loss; and performing filling loss record learning based on the explosive filling loss factors to construct an explosive filling loss prediction model. Based on the explosive filling loss prediction model, the first filling decision space is tested for loss according to the explosive filling loss constraint to obtain multiple filling loss test results. The first filling decision space is then optimized and selected based on these multiple filling loss test results to generate the second filling decision space. The second filling decision space is then expanded and optimized based on the explosive filling loss cost function and the filling variation quantitative function to obtain a filling optimization strategy. This strategy includes: weighting the explosive filling loss factors to obtain the explosive filling loss cost function; calculating the filling loss cost for each filling decision in the second filling decision space based on the explosive filling loss cost function; and obtaining... A first distribution of filling loss cost is obtained. Based on the first distribution of filling loss cost, the second filling decision space is optimized and selected according to a predetermined filling loss cost coefficient to obtain a third filling decision space. The third filling decision space is then expanded by mutation according to the explosive filling loss cost function, the filling variation quantification function, and the explosive filling loss factor to construct a fourth filling decision space. The filling loss cost is minimized according to the fourth filling decision space to generate the filling optimization strategy. The filling unit fills the explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and seals the filled explosive through the aluminum wire clipping unit of the emulsion explosive filling machine.In other words, by detecting the state of the explosive, once the detection is normal, the packaging bag is placed onto the interface. Based on the explosive information, environmental information, and target quantity, the optimal filling strategy is determined, the explosive is filled, and the sealing is completed. This achieves automated filling of large-diameter emulsion explosives, improves filling efficiency, and reduces production costs.

[0081] Example 2: Based on the same inventive concept as the automatic filling control method for large-diameter emulsion explosives in Example 1, this application also provides an automatic filling control device for large-diameter emulsion explosives. Please refer to the appendix. Figure 2 The automatic filling and control equipment for large-diameter emulsion explosives includes:

[0082] The status detection module 11 is used to detect the status of the explosive to be filled in the emulsion explosive filling machine and obtain the explosive status detection result; the packaging and fitting module 12 is used to fit a predetermined packaging bag that meets the target charge amount to the filling interface of the emulsion explosive filling machine through the bag taking unit and bag fitting unit when the explosive status detection result is normal; the decision space construction module 13 is used to make filling control decisions for the filling unit of the emulsion explosive filling machine based on the explosive basic information, explosive filling environment information and the target charge amount, and establish a first filling decision space, including: making filling control decisions for the filling unit based on the explosive basic information, explosive filling environment information and the target charge amount. The unit performs a filling control scheme retrieval to obtain a set of retrieved filling schemes. Based on the retrieved filling scheme set, it sorts out trigger features and establishes a multi-dimensional filling control domain corresponding to the multi-dimensional filling control variables. Taking the target charge amount as the filling control target, it performs control parameter decision combination on the multi-dimensional filling control variables according to the multi-dimensional filling control domain to generate the first filling decision space. The decision space optimization module 14 is used to perform loss verification and optimization on the first filling decision space according to the explosive filling loss factor to obtain a second filling decision space. This includes: setting explosive filling loss constraints according to the explosive filling loss factor, wherein the explosive filling loss factor includes explosive filling wall adhesion loss, explosive filling loss, and explosive filling loss. The system addresses cavitation loss and explosive filling interruption loss. It learns from filling loss records based on the explosive filling loss factors to construct an explosive filling loss prediction model. Based on this model, it performs loss checks on the first filling decision space according to the explosive filling loss constraints, obtaining multiple filling loss check results. Based on these results, it optimizes and selects the first filling decision space to generate the second filling decision space. An optimization strategy determination module 15 is used to perform mutation expansion optimization on the second filling decision space based on the explosive filling loss cost function and the filling variation quantitative function to obtain a filling optimization strategy, including: performing... Weight allocation is performed to obtain the explosive filling loss cost function. The filling loss cost is calculated for each filling decision in the second filling decision space based on the explosive filling loss cost function, resulting in a first distribution of filling loss costs. Based on the first distribution of filling loss costs, the second filling decision space is optimized according to a predetermined filling loss cost coefficient to obtain a third filling decision space. The third filling decision space is then expanded by mutation according to the explosive filling loss cost function, the filling variation quantification function, and the explosive filling loss factor to construct a fourth filling decision space. The filling loss cost is minimized based on the fourth filling decision space to generate the filling optimization strategy.The filling and sealing module 16 is used by the filling unit to fill the explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and to seal the filled explosive through the aluminum wire clamping unit of the emulsion explosive filling machine.

[0083] Furthermore, the status detection module 11 in the automatic filling control equipment for large-diameter emulsion explosives is also used for:

[0084] Obtain real-time status data of the explosive to be filled; match standard status data of the explosive according to the specifications and model of the explosive to be filled; input the real-time status data and the standard status data of the explosive into a twin neural network to obtain the standard coefficient of the explosive status; determine whether the standard coefficient of the explosive status is greater than or equal to the standard threshold of the explosive status, and generate the explosive status detection result.

[0085] Furthermore, the status detection module 11 in the automatic filling control equipment for large-diameter emulsion explosives is also used for:

[0086] If the standard coefficient of explosive state is greater than or equal to the standard threshold of explosive state, the explosive state detection result is normal; if the standard coefficient of explosive state is less than the standard threshold of explosive state, the explosive state detection result is abnormal.

[0087] Furthermore, the decision space optimization module 14 in the automatic filling control equipment for large-diameter emulsion explosives is also used for:

[0088] Based on the first filling decision space, a first filling decision is extracted; the explosive basic information, the explosive filling environment information, and the first filling decision are input into the explosive filling loss model to obtain a first explosive filling loss prediction result; it is determined whether the first explosive filling loss prediction result meets the explosive filling loss constraint, a first filling loss inspection result is generated, and the first filling loss inspection result is added to the plurality of filling loss inspection results.

[0089] Furthermore, the optimization strategy determination module 15 in the automatic filling control equipment for large-diameter emulsion explosives is also used for:

[0090] The number of variations in the third space of the filling decision is calculated according to the filling variation quantification function to obtain the variation distribution; the third space of the filling decision is mutated according to the variation distribution to obtain the first variation domain of the filling decision; the first variation domain of the filling decision is optimized by loss test according to the explosive filling loss factor to obtain the second variation domain of the filling decision; the filling loss cost is calculated according to the explosive filling loss cost function to obtain the second distribution of the filling loss cost; the second variation domain of the filling decision is optimized and selected according to the predetermined filling loss cost coefficient based on the second distribution of the filling loss cost to obtain the third variation domain of the filling decision; the third space of the filling decision is expanded according to the third variation domain of the filling decision to generate the fourth space of the filling decision.

[0091] Furthermore, the optimization strategy determination module 15 in the automatic filling control equipment for large-diameter emulsion explosives is also used for:

[0092] The quantitative function for filling variation is: Among them, R i The coefficient representing the filling loss cost corresponding to the i-th filling decision, where i is a positive integer, S i The variable represents the number of variations corresponding to the i-th filling decision, Floor refers to rounding down, R0 represents the predetermined filling loss cost coefficient, and S max S represents the upper limit of the number of variations in filling decisions. min Characterizes the lower limit of the number of variations in filling decisions.

[0093] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1 The automatic filling control method and specific examples of a large-diameter emulsion explosive in Example 1 are also applicable to the automatic filling control device of a large-diameter emulsion explosive in this embodiment. Through the foregoing detailed description of the automatic filling control method of a large-diameter emulsion explosive, those skilled in the art can clearly understand the automatic filling control device of a large-diameter emulsion explosive in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0094] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0095] Obviously, those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. An automatic filling control method for large-diameter emulsion explosives, characterized in that, include: The condition of the explosives to be filled in the emulsion explosive filling machine is detected to obtain the condition detection results of the explosives; When the explosive status detection result is normal, the bag taking unit and bag putting unit of the emulsion explosive filling machine will put the predetermined packaging bag that meets the target charge amount onto the filling interface of the emulsion explosive filling machine. Based on the basic information of the explosive, the information of the explosive filling environment, and the target charge amount, the filling unit of the emulsion explosive filling machine makes filling control decisions and establishes a first filling decision space. This includes: searching for filling control schemes for the filling unit based on the basic information of the explosive, the information of the explosive filling environment, and the target charge amount to obtain a set of searched filling schemes; sorting out trigger features based on the set of searched filling schemes to establish a multi-dimensional filling control domain corresponding to the multi-dimensional filling control variables; taking the target charge amount as the filling control target; and combining control parameters for the multi-dimensional filling control variables based on the multi-dimensional filling control domain to generate the first filling decision space. The process of obtaining a second filling decision space by performing loss verification and optimization based on explosive filling loss factors includes: setting explosive filling loss constraints based on the explosive filling loss factors, wherein the explosive filling loss factors include explosive filling wall adhesion loss, explosive filling void loss, and explosive filling interruption loss; performing filling loss record learning based on the explosive filling loss factors to construct an explosive filling loss prediction model; performing loss verification on the first filling decision space based on the explosive filling loss prediction model and the explosive filling loss constraints to obtain multiple filling loss verification results; and optimizing and selecting the first filling decision space based on the multiple filling loss verification results to generate the second filling decision space. The filling decision second space is expanded and optimized based on the explosive filling loss cost function and the filling variation quantification function to obtain a filling optimization strategy. This includes: weighting the explosive filling loss factor to obtain the explosive filling loss cost function; calculating the filling loss cost for each filling decision in the filling decision second space based on the explosive filling loss cost function to obtain a first distribution of filling loss costs; optimizing and selecting the filling decision second space based on the first distribution of filling loss costs and a predetermined filling loss cost coefficient to obtain a third filling decision space; expanding and mutating the filling decision third space based on the explosive filling loss cost function, the filling variation quantification function, and the explosive filling loss factor to construct a fourth filling decision space; and minimizing the filling loss cost based on the fourth filling decision space to generate the filling optimization strategy. The filling unit fills the explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and seals the filled explosive through the aluminum wire clipping unit of the emulsion explosive filling machine; The filling variation quantification function is as follows: ； Among them, R i The coefficient representing the filling loss cost corresponding to the i-th filling decision, where i is a positive integer, S i The variable represents the number of variations corresponding to the i-th filling decision, Floor refers to rounding down, R0 represents the predetermined filling loss cost coefficient, and S max S represents the upper limit of the number of variations in filling decisions. min Characterizes the lower limit of the number of variations in filling decisions.

2. The automatic filling control method for large-diameter emulsion explosives as described in claim 1, characterized in that, Based on the explosive filling loss prediction model, loss checks are performed on the first filling decision space according to the explosive filling loss constraints to obtain multiple filling loss check results, including: Based on the first filling decision space, extract the first filling decision; The explosive basic information, the explosive filling environment information, and the first filling decision are input into the explosive filling loss prediction model to obtain the first explosive filling loss prediction result. Determine whether the first explosive filling loss prediction result meets the explosive filling loss constraint, generate a first filling loss inspection result, and add the first filling loss inspection result to the plurality of filling loss inspection results.

3. The automatic filling control method for large-diameter emulsion explosives as described in claim 1, characterized in that, Based on the explosive filling loss cost function, the filling variation quantification function, and the explosive filling loss factor, the third space of the filling decision is expanded by variation to construct a fourth space of the filling decision, including: The number of variations in the third space of the filling decision is calculated based on the filling variation quantification function to obtain the variation distribution. Based on the distribution of the number of mutations, the third space of the filling decision is mutated to obtain the first mutation domain of the filling decision. Based on the explosive filling loss factor, the first variation domain of the filling decision is optimized by loss test to obtain the second variation domain of the filling decision. The filling loss cost is calculated in the second variation domain of the filling decision based on the explosive filling loss cost function to obtain the second distribution of filling loss cost; Based on the second distribution of filling loss cost, the second variation domain of filling decision is optimized and selected according to the predetermined filling loss cost coefficient to obtain the third variation domain of filling decision. The third space of filling decisions is expanded according to the third variation domain of the filling decision to generate the fourth space of filling decisions.

4. The automatic filling control method for large-diameter emulsion explosives as described in claim 1, characterized in that, The condition of the explosives to be filled in the emulsion explosive filling machine is monitored to obtain the explosive condition monitoring results, including: Obtain real-time status data of the explosive to be filled; Match the standard state data of the explosives according to the specifications and models of the explosives to be filled; The real-time state data and standard state data of the explosive are input into a twin neural network to obtain the standard coefficient of the explosive state. Determine whether the standard coefficient of the explosive state is greater than or equal to the standard threshold of the explosive state, and generate the explosive state detection result.

5. The automatic filling control method for large-diameter emulsion explosives as described in claim 4, characterized in that, Determine whether the standard coefficient of the explosive state is greater than or equal to the standard threshold of the explosive state, and generate the explosive state detection result, including: If the standard coefficient of explosive state is greater than or equal to the standard threshold of explosive state, the explosive state detection result is normal. If the standard coefficient of explosive state is less than the standard threshold of explosive state, the explosive state detection result is abnormal.

6. An automatic filling and control device for large-diameter emulsion explosives, characterized in that, The step of implementing the automatic filling control method for a large-diameter emulsion explosive according to any one of claims 1 to 5, wherein the automatic filling control device for the large-diameter emulsion explosive comprises: The status detection module is used to detect the status of the explosives to be filled in the emulsion explosive filling machine and obtain the status detection results of the explosives. The packaging and fitting module is used to fit a predetermined packaging bag that meets the target charge amount to the filling interface of the emulsion explosive filling machine through the bag taking unit and bag fitting unit of the emulsion explosive filling machine when the explosive state detection result is normal. The decision space construction module is used to make filling control decisions for the filling unit of the emulsion explosive filling machine based on the explosive basic information, explosive filling environment information, and the target charge amount, and to establish a first filling decision space. This includes: retrieving filling control schemes for the filling unit based on the explosive basic information, explosive filling environment information, and the target charge amount to obtain a set of retrieved filling schemes; sorting trigger features based on the retrieved filling scheme set to establish a multi-dimensional filling control domain corresponding to the multi-dimensional filling control variables; taking the target charge amount as the filling control target; and combining control parameters for the multi-dimensional filling control variables based on the multi-dimensional filling control domain to generate the first filling decision space. The decision space optimization module is used to perform loss verification and optimization on the first filling decision space based on explosive filling loss factors to obtain a second filling decision space. The module includes: setting explosive filling loss constraints based on the explosive filling loss factors, wherein the explosive filling loss factors include explosive filling wall adhesion loss, explosive filling void loss, and explosive filling breakage loss; performing filling loss record learning based on the explosive filling loss factors to construct an explosive filling loss prediction model; performing loss verification on the first filling decision space based on the explosive filling loss prediction model and the explosive filling loss constraints to obtain multiple filling loss verification results; and performing optimization and selection on the first filling decision space based on the multiple filling loss verification results to generate the second filling decision space. The optimization strategy determination module is used to perform mutation expansion optimization on the second space of the filling decision based on the explosive filling loss cost function and the filling variation quantitative function to obtain the filling optimization strategy. This includes: weighting the explosive filling loss factor to obtain the explosive filling loss cost function; calculating the filling loss cost for each filling decision in the second space of the filling decision based on the explosive filling loss cost function to obtain a first distribution of filling loss costs; optimizing and selecting the second space of the filling decision based on the first distribution of filling loss costs and a predetermined filling loss cost coefficient to obtain a third space of filling decisions; expanding the third space of the filling decision based on the explosive filling loss cost function, the filling variation quantitative function, and the explosive filling loss factor to construct a fourth space of filling decisions; and minimizing the filling loss cost based on the fourth space of the filling decision to generate the filling optimization strategy. A filling and sealing module is used by the filling unit to fill the explosive to be filled into the predetermined packaging bag according to the filling optimization strategy, and to seal the filled explosive through the aluminum wire clipping unit of the emulsion explosive filling machine.

Images