Preparation method of modular junction box suitable for various detonation networks

Abstract

The invention provides a method suitable for various detonation network modular junction boxes, and the method comprises the steps: obtaining specific demand parameters in a pre-established detonation network demand information library, the demand parameters comprising a needed power supply type, a voltage range, a current range, a control mode and a safety protection level; the demand parameters are matched with preset functional module attributes, and a functional module combination scheme needed for achieving the detonation demand is determined; aiming at the functional module combination scheme, obtaining a corresponding three-dimensional virtual prototype model from a pre-established modularized junction box virtual prototype library, verifying the feasibility of the functional module combination scheme by adopting a virtual assembly simulation technology, and optimizing the assembly process to obtain an optimized assembly scheme; according to the optimized assembly scheme, a physical connection relation table of each functional module is generated, the physical connection relation table comprises the interface type, the interface position and the connection sequence of each module, and the physical connection relation table is used as guidance for actual junction box assembly.

Description

Technical Field

[0001] The invention relates to the technical field of information technology, and in particular to a method for preparing a modular junction box suitable for a variety of initiation networks. Background Art

[0002] In mine blasting operations, the junction box is a key device for connecting the blasting network. Traditional junction box designs usually adopt a fixed structure and layout, which is difficult to meet the needs of different blasting scenarios. For example, in large-scale open-pit mine blasting, the blasting network is large in scale and requires multiple junction boxes to work together. Traditional designs make it difficult to achieve flexible networking and control. In tunnel blasting, the blasting network is relatively simple, but the waterproof, dustproof, and anti-interference capabilities of the junction box are required to be high. Traditional designs are difficult to take into account various working conditions.

[0003] How to improve the adaptability and flexibility of the junction box while ensuring its reliability and safety is a technical problem that needs to be solved urgently. In response to this problem, the idea of ​​modular design of the junction box is proposed. By dividing the various functional components of the junction box into independent modules, such as power module, control module, safety module, communication module, etc., and defining a unified interface standard, different modules can be flexibly combined and interchanged, thereby realizing the customizability and scalability of the junction box function. This design idea can greatly improve the adaptability of the junction box and meet the needs of different detonation environments and working conditions. However, modular design also brings new technical challenges. How to ensure the compatibility and interoperability between modules, how to achieve the standardization and generalization of modules, and how to ensure the reliability and safety of modular junction boxes all require further research and development. Summary of the invention

[0004] The present invention provides a method for preparing a modular junction box applicable to a variety of detonation networks, which mainly comprises:

[0005] Obtaining specific demand parameters in a pre-established detonation network demand information database, the demand parameters including the required power type, voltage range, current range, control mode and safety protection level, matching the demand parameters with preset functional module attributes, and determining the functional module combination scheme required to achieve the detonation demand;

[0006] For the functional module combination scheme, a corresponding three-dimensional virtual prototype model is obtained from a pre-established modular junction box virtual prototype library, and the feasibility of the functional module combination scheme is verified by using virtual assembly simulation technology, and the assembly process is optimized to obtain an optimized assembly scheme;

[0007] According to the optimized assembly scheme, a physical connection relationship table of each functional module is generated, wherein the physical connection relationship table includes the interface type, interface position and connection sequence of each module, and the physical connection relationship table is used as a guide for the actual assembly of the junction box;

[0008] During the assembly process of the junction box, the assembly status image is acquired in real time through machine vision technology, and the image recognition algorithm is used to determine whether the assembly position and sequence of each functional module are consistent with the physical connection relationship table. If not, a warning signal is sent to the assembly control center to trigger an assembly pause instruction, and the inconsistent position is marked on the display screen to guide the operator to make corrections;

[0009] After assembly is completed, the electrical connection automatic detection technology is used to determine whether the electrical connection between the modules inside the junction box is correct. If there is a connection error, the error information is fed back to the pre-established fault diagnosis knowledge base. The knowledge base matches the corresponding troubleshooting steps according to the error type, generates a fault diagnosis and troubleshooting plan, and displays it through the maintenance terminal to guide maintenance personnel to troubleshoot the problem;

[0010] After the junction box is put into use, the operating parameters are collected in real time through the built-in status monitoring module. The operating parameters include voltage, current and temperature. The operating parameters are analyzed using a machine learning algorithm to build a multi-parameter fusion model under normal working conditions. When the collected real-time data deviates from the preset threshold of the predicted value of the multi-parameter fusion model, it is judged as a precursor to a fault, and an early warning message is generated and sent to the management platform. After receiving the early warning message, the management platform automatically triggers the preventive maintenance process.

[0011] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0012] The present invention discloses a method for preparing a modular junction box suitable for a variety of initiation networks, aiming to solve the problems of functional module combination, assembly optimization, assembly monitoring and fault prevention in the initiation network requirements. First, by obtaining the required parameters of the initiation network, matching the preset functional modules, determining the best combination scheme, and verifying its feasibility through virtual assembly simulation technology, the assembly process is optimized. Then, a physical connection relationship table is generated to guide the actual assembly. During the assembly process, the assembly state is monitored in real time by machine vision technology to ensure that the module position and sequence are correct. If they are not consistent, an alarm is issued and the assembly is suspended to guide the operator to correct. After the assembly is completed, the electrical connection automatic detection technology is used to determine whether the module connection is correct. If there is an error, the system will feedback to the fault diagnosis knowledge base, generate a troubleshooting plan, and guide the maintenance personnel to repair. After the junction box is put into use, the built-in status monitoring module collects the operating parameters in real time, uses the machine learning algorithm to analyze the data, and constructs a multi-parameter fusion model. If the data deviates from the model prediction value, the system will generate early warning information and trigger the preventive maintenance process. Through the above process, the present invention realizes the full process intelligent management from demand matching, assembly optimization, real-time monitoring to fault prevention, and improves the assembly efficiency and operation reliability of the junction box. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0014] Figure 1 The present invention is a flow chart of a method for preparing a modular junction box suitable for a variety of detonation networks.

[0015] Figure 2 It is a schematic diagram of a method for preparing a modular junction box suitable for various detonation networks according to the present invention.

[0016] Figure 3 It is another schematic diagram of a method for preparing a modular junction box suitable for various detonation networks according to the present invention. DETAILED DESCRIPTION

[0017] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0018] like Figure 1-3In this embodiment, a method for preparing a modular junction box applicable to multiple detonation networks may specifically include:

[0019] S101, obtaining specific demand parameters in a pre-established detonation network demand information database, wherein the demand parameters include the required power supply type, voltage value range, current value range, control mode and safety protection level, etc.; matching the demand parameters with preset function module attributes to determine the function module combination scheme required to realize the detonation demand.

[0020] When the acquired demand parameters are matched and compared with the preset functional module attributes, the attribute similarity is calculated.

[0021] Specifically, first, the system obtains specific demand parameters from the detonation network demand information library. For example, the required power type is lithium battery, the operating voltage range is 6V-2V, the maximum operating current is 2A, the control method is wireless remote control, and the safety protection level is IP65. The system matches and compares these parameters with the preset functional module attributes, and uses the cosine similarity algorithm to calculate the cosine value of the angle between the attribute vectors. The closer the cosine value is to 1, the higher the similarity. Through calculation, it is found that the similarity between this demand and the power module A, control module B, and safety protection module C are 92, 87, and 95 respectively, and the required functional module combination is determined to be A+B+C.

[0022] In another embodiment, according to the specific demand parameters in the acquired detonation network demand information library, semantic analysis technology is used to parse the demand parameters, extract keywords and construct a demand vector; the demand vector is similar to the preset function module attribute vector, and the cosine similarity algorithm is used to determine the most matching function module combination scheme. Specifically, first, the system uses natural language processing technology to perform semantic analysis on the acquired detonation network demand parameters, and extracts keywords such as "lithium battery", "6V-12V", "2A", "wireless remote control", "IP65", etc. through operations such as word segmentation, part-of-speech tagging, and named entity recognition. Then, these keywords are mapped to the preset parameter vector space to construct a 15-dimensional demand vector. Next, the system calculates the cosine similarity between the demand vector and the preset function module attribute vector, and by setting the similarity threshold, the power module A, control module B, and safety protection module C that best match the demand are screened out to form the optimal module combination scheme.

[0023] S102. For the functional module combination scheme, a corresponding three-dimensional virtual prototype model is obtained from a pre-established modular junction box virtual prototype library; virtual assembly simulation technology is used to perform assembly simulation on the obtained three-dimensional virtual prototype model, and the feasibility of the functional module combination scheme is verified, and the assembly process is optimized to obtain an optimized assembly scheme.

[0024] Specifically, in step S101, it has been determined that the required functional module combination is A+B+C. Next, the system extracts the 3D virtual prototype model data of the three modules A, B, and C from the modular junction box virtual prototype library.

[0025] In one implementation, virtual assembly simulation technology is used to simulate the assembly of modules based on a physical engine, and interference checks are performed on the combination scheme by calculating the minimum distance between models and collision detection. If it is found that there is a 5mm interference between modules B and C, the assembly optimization process is triggered, and the system automatically adjusts the installation position of module B, translates it 5mm along the positive direction of the X axis, and re-performs assembly verification until the assembly scheme passes the inspection.

[0026] Next, in one embodiment, the system uses an assembly process optimization method based on a genetic algorithm, with the shortest assembly time as the optimization goal, and evolves the optimal assembly process by designing the chromosome encoding of the module assembly sequence and installation position, using genetic operations such as selection, crossover, and mutation. After optimization, the best process route with a 15% increase in assembly efficiency is: assemble the power module A first, then assemble the control module B, and finally assemble the safety protection module C.

[0027] In another embodiment, a heuristic optimization algorithm, such as a genetic algorithm, may be used to optimize the module positions and sequences during the assembly process to obtain an optimal assembly solution.

[0028] In another specific implementation, according to the functional module combination determined in step S101: A+B+C. According to the combination scheme, the system calls the CAD model data of the three modules A, B, and C from the modular junction box virtual prototype library, and builds a virtual assembly environment through the Unity 3D engine. In this environment, the system combines the virtual prototype models of the three modules to simulate the assembly process. At the same time, the hierarchical bounding box algorithm based on the octree is used to perform collision detection on the geometric boundaries of each module, and it is found that there is a 5mm interference between modules B and C. Therefore, the system automatically triggers the assembly optimization process and uses a multi-objective genetic algorithm to solve the interference problem. By designing the chromosome encoding of the module position and assembly order, with the minimum assembly time and interference as the optimization goal, a Pareto optimal solution set is obtained after 500 generations of iteration. The system selects the solution with the shortest assembly time as the optimal assembly solution, translates module B along the X-axis in the positive direction by 5mm, and adjusts the assembly order to ABC.

[0029] S103, generating a physical connection relationship table of each functional module according to the optimized assembly scheme, wherein the physical connection relationship table includes the interface type, interface position and connection sequence of each module, and using the physical connection relationship table as a guide for actual junction box assembly.

[0030] In one implementation, a physical connection relationship table of each functional module is automatically generated by using a graphic recognition technology, including interface type, location, and connection sequence.

[0031] In one embodiment, the generated physical connection relationship table is converted into a machine-readable instruction set, such as G code, and transmitted to the industrial robot control system of the assembly workshop via Ethernet. The automated assembly equipment replays the virtual assembly process according to the received instruction set and the optimized assembly plan, processes and analyzes the key frame image through the OpenCV library, automatically identifies the connection position and interface type of each module, and generates a physical connection relationship table. For example, the output port of the power module A is XT30, and its position coordinates are (20, 30, 15), which needs to be connected to the input port (XT30, position (20, 38, 15)) of the control module B through a harness. After the system encapsulates the table in XML format, it is transmitted to the MES system of the assembly workshop via industrial Ethernet as an electronic process file to guide the assembly operation. Finally, during the actual assembly, the worker scans the QR code of each functional module through the PDA device and compares the scanned information with the physical connection relationship table. If the scanned module model does not match the relationship table, the system will promptly issue a warning to remind the worker to reselect the correct module; if the assembly sequence is incorrect, the system will also guide the worker to adjust the assembly steps through the PDA interface. Through this intelligent assembly process control and quality traceability, the assembly quality of each junction box can be effectively ensured and safety hazards caused by human errors can be prevented.

[0032] In one embodiment, a semantic-based interface matching algorithm is used to determine whether the interface types of each functional module are compatible. If not, the interface conversion module is triggered to automatically convert the interface type until the matching condition is met. The three-dimensional spatial positioning technology is used to determine the optimal installation position of each functional module inside the junction box, and the corresponding three-dimensional assembly coordinates are generated. For example, in the case where the XT30 output port of module A and the RS485 input port of module B do not match, the interface conversion module is triggered to automatically generate an XT30 to RS485 adapter board model. The SLAM algorithm is used to scan the internal space of the existing junction box, determine the optimal installation position of each module, and generate assembly coordinates (20,30,15), (20,38,15), and (85,30,15).

[0033] Finally, based on the physical connection relationship table and the three-dimensional assembly coordinates, a realistic three-dimensional animation of the assembly process is generated using virtual reality technology to simulate the assembly operation and optimize the assembly process; the optimized physical connection relationship table and the three-dimensional animation of the assembly process are uploaded to the digital assembly guidance system as a guide for the actual assembly of the junction box, thereby improving assembly efficiency and quality. In one embodiment, a realistic three-dimensional animation of the assembly process is generated by a VR engine and uploaded to a digital assembly guidance system to guide workers to perform actual assembly operations, thereby increasing assembly efficiency by 30% and achieving a first-time pass rate of 98%.

[0034] S104. During the assembly process of the junction box, the assembly status image is acquired in real time through machine vision technology; the image recognition algorithm is used to determine whether the assembly position and sequence of each functional module are consistent with the physical connection relationship table. If not, a warning signal is sent to the assembly control center to trigger an assembly pause instruction, generate a coordinate information signal of the inconsistent position, and mark the inconsistent position on the display screen to guide the operator to make corrections.

[0035] Specifically, the assembly state image of the junction box assembly process is obtained in real time by machine vision technology, and the image preprocessing algorithm is used to denoise and enhance the obtained image to obtain a high-quality assembly state image. According to the pre-established physical connection relationship table, the standard assembly position and sequence information of each functional module are obtained, and it is used as a reference standard for the image recognition algorithm. The target detection algorithm based on deep learning is used to locate and identify the functional modules in the assembly state image, and the actual assembly position and sequence information of each module are obtained. By comparing the actual assembly information with the standard assembly information, it is determined whether the assembly position and sequence of each functional module are consistent with the physical connection relationship table. If the judgment result is inconsistent, the warning signal of the assembly control center is triggered, and an assembly pause instruction is generated at the same time, and a coordinate information signal of the inconsistent position is generated and sent to the control module of the assembly equipment. According to the coordinate information of the inconsistent position, the wrong assembly position is marked on the display screen, and the operator is prompted to correct it by highlighting it. The operator adjusts and corrects the wrong assembly position according to the prompt information on the display screen, and sends a correction completion signal through the human-computer interaction interface after completion. After receiving the calibration completion signal, the machine vision technology is re-triggered to detect the assembly status to determine whether the assembly position and sequence after calibration meet the requirements. If the assembly status after calibration meets the requirements of the physical connection relationship table, the assembly pause command is released to allow the assembly process to continue; otherwise, the above steps are repeated until the calibration is successful.

[0036] In one embodiment, during the assembly process of the junction box, the YOLOv5 target detection algorithm based on deep learning is used to perform real-time analysis of the assembly state image. First, a high-definition image of the assembly site is collected by an industrial camera, with an image resolution of 1920×1080 and a frame rate of 30fps. Then, the image is denoised using a median filter algorithm, and image enhancement is achieved through a histogram equalization algorithm to make the outline of the functional module clearer. Next, the processed image is input into the YOLOv5 network for reasoning, which uses the CSPDarknet53 backbone network and PANet network structure to achieve real-time detection of multi-scale targets. By pre-training on the COCO dataset and fine-tuning with 5,000 junction box assembly images, a high-precision functional module detection model is finally obtained, with an mAP of 95%. By using this model to detect the assembly state image, information such as the category, position coordinates, and detection confidence of each functional module can be obtained. The detection results are compared with the standard information in the physical connection relationship table to determine whether the assembly position and sequence meet the requirements. When it is found that the requirements are not met, the system automatically generates an alarm signal and sends the control command to the PLC through the OPC UA protocol to suspend the operation of the assembly equipment. At the same time, the wrong assembly position is marked in the form of a red rectangular frame on the human-computer interaction interface, and the operator is prompted to make corrections in the form of voice and text. The operator adjusts the position of the functional module according to the prompt information, and clicks the "Reset" button after the correction is completed. The system re-detects the assembly status. If it meets the requirements, the alarm is lifted and the assembly task continues; otherwise, the above process is repeated until the correction is successful. By combining machine vision technology and automated control methods, real-time monitoring and quality control of the junction box assembly process are achieved, effectively improving assembly efficiency and product qualification rate.

[0037] In another embodiment, after step S103, the robot first grabs the power module A with an electric gripper and installs it on the base plate according to the received instructions, then grabs the control module B, achieves precise docking with A through visual servo and force control, and finally grabs the C module and installs it. The entire assembly process is fully automated, the assembly accuracy reaches 1mm, and the efficiency is improved by 20%.

[0038] S105. After assembly is completed, the electrical connection automatic detection technology is used to determine whether the electrical connection between the modules inside the junction box is correct; if there is a connection error, the error information is fed back to a pre-established fault diagnosis knowledge base, and the knowledge base matches the corresponding troubleshooting steps according to the error type, generates a fault diagnosis and troubleshooting plan, and displays it through a maintenance terminal to guide maintenance personnel to troubleshoot the problem.

[0039] Specifically, according to the functional module combination scheme of the junction box, the electrical connection relationship data of each module is obtained, and the internal electrical connection model of the junction box is constructed. Through the electrical connection automatic detection technology, the electrical connection data of the junction box after actual assembly is collected to obtain the actual electrical connection model. The actual electrical connection model is compared with the theoretical electrical connection model to determine whether there is a connection error. If there is a connection error, the error information is extracted to form structured error description data. The error description data is input into a pre-established fault diagnosis knowledge base, which uses a rule-based reasoning method to match the corresponding fault cause and troubleshooting steps. After the knowledge base reasoning obtains the fault cause and troubleshooting solution, a fault diagnosis report is generated in the form of text and image. The fault diagnosis report is transmitted to the maintenance terminal through the network and presented to the maintenance personnel on the display screen. The maintenance personnel inspect and repair the junction box according to the fault location and troubleshooting method prompted by the diagnosis report. After the repair is completed, it is verified again by the electrical connection automatic detection technology to ensure that the internal electrical connection of the junction box is completely correct, and the final quality inspection report is formed for archiving.

[0040] In a specific implementation, according to the functional module combination scheme of the junction box, the electrical schematic design software E-Plan is used to extract the electrical connection relationship data of each module, and a theoretical electrical connection model in XML format containing information such as components, wires, and terminals is constructed. By installing an inductive sensor on the junction box assembly tooling, the actual connection current signal of the terminal and the wire is collected, and transmitted to the PLC for A / D conversion and digital filtering processing, characteristic quantities such as current amplitude and phase angle are obtained, mapped to the actual electrical connection relationship, and an actual electrical connection model in JSON format is formed. The two model files are parsed using Python's json and xml libraries, and the differences between nodes and edges are compared through the graph isomorphism algorithm to identify typical connection errors such as missing, dislocation, and short circuit, extract parameters such as error type, location, and device information, and generate structured error description data in key-value format. The error description data is inserted into the fault diagnosis knowledge base using SQL statements. The knowledge base is based on the Drools rule engine and uses a forward reasoning algorithm to match 500 preset fault rules to infer the corresponding fault causes and troubleshooting steps containing text and images. The diagnostic report is transmitted to an Android tablet via the WebSocket protocol and displayed using native UI controls. Maintenance personnel locate faulty components based on the 3D model animation and text description in the report, use multimeters, oscilloscopes and other tools to check wires and terminals, and repair and replace defective parts. After repair, the automated electrical inspection process is performed again. When the operation coverage reaches 100% and the error rate is less than 1%, a final quality inspection report in PDF format is generated for archiving.

[0041] S106. After the junction box is put into use, the operating parameters are collected in real time through the built-in status monitoring module, and the operating parameters include voltage, current and temperature. The operating parameters are analyzed using a machine learning algorithm to build a multi-parameter fusion model under normal working conditions. When the collected real-time data deviates from the predicted value of the multi-parameter fusion model by more than a preset threshold, it is judged as a precursor to a fault, and an early warning message is generated and sent to the management platform. After receiving the early warning message, the management platform automatically triggers the preventive maintenance process.

[0042] Specifically, the built-in status monitoring module collects the running parameters of the junction box in real time, including voltage, current, temperature and other data, to form multi-dimensional time series data. The collected multi-dimensional time series data is preprocessed, including data cleaning, normalization and feature extraction, to obtain a standardized feature vector set. According to the feature vector set, a multi-parameter fusion model under normal working conditions is constructed using machine learning algorithms such as support vector machines or neural networks to achieve modeling of the equipment operating status. The sliding time window mechanism is used to dynamically update the multi-parameter fusion model to adapt to changes in the equipment operating conditions and improve the robustness and accuracy of the model. The running parameters of the junction box are monitored online in real time, and the real-time feature vector is obtained through the data preprocessing module. The real-time feature vector is input into the trained multi-parameter fusion model to obtain the corresponding predicted value. The deviation between the real-time feature vector and the model predicted value is calculated. If the deviation exceeds or falls below the preset threshold, it is judged as a fault precursor and an early warning message is generated. The early warning message is sent to the management platform through the communication module to trigger the preventive maintenance process, generate a maintenance work order and send it to the relevant department. The management platform receives early warning information from multiple junction boxes, comprehensively analyzes the health status of equipment, optimizes maintenance strategies, and realizes full life cycle management of equipment.

[0043] In a specific embodiment, in the state monitoring system of the junction box, key operating parameters are first collected in real time through built-in sensors, such as voltage may be 220V±5V, current is 5A±5A, and temperature varies between -10℃ and 50℃. These data are converted into digital signals through high-speed ADC (analog-to-digital converter) to form a multi-dimensional time series data stream. The data preprocessing module first cleans these data to remove abnormal values, such as obvious erroneous data where the current suddenly jumps to 100A. Then, the Z-score standardization method is used to normalize the data so that its mean is 0 and the standard deviation is 1 to eliminate the influence of different dimensions. In the feature extraction link, the sliding window technology (the window size is set to 30 seconds and the step size is 5 seconds) is used to calculate the statistical features such as the mean, maximum, minimum and standard deviation in the window to form a normalized feature vector set. Using these feature vectors, a neural network model (such as a three-layer fully connected neural network, the number of input layer nodes corresponds to the number of features, the number of hidden layer nodes is set to 64, and the output layer is a binary classification, indicating normal and abnormal states) is used for training to establish a multi-parameter fusion model under normal working conditions of the equipment. During the model training process, the cross-validation method is used to optimize the hyperparameters, such as setting the learning rate to 0.1, the batch size to 32, and the number of iterations to 1000. In order to adapt to the changes in the operating conditions of the equipment, the model adopts an online learning method, and the model parameters are updated with the latest data at regular intervals (such as every day). In actual operation, the real-time monitored data undergoes the same preprocessing and feature extraction process to obtain real-time feature vectors. These feature vectors are input into the trained neural network model, and the model outputs the predicted value. By calculating the deviation between the real-time feature vector and the model prediction value, such as using the Euclidean distance, if the distance exceeds the preset threshold (such as 5 times the standard deviation), it is judged as a precursor to a fault, and the system automatically generates early warning information. These early warning messages are sent to the management platform through the communication module (such as using the MQTT protocol), triggering the preventive maintenance process, automatically generating maintenance work orders and sending them to the maintenance department. The management platform will also comprehensively analyze the data from multiple junction boxes, optimize the maintenance strategy, and realize intelligent management of the entire life cycle of the equipment.

[0044] The above is only a specific implementation method of this specification. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system, module and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. It should be understood that the protection scope of this specification is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this specification, and these modifications or replacements should be covered within the protection scope of this specification.

Claims

1. A method for preparing a modular junction box suitable for a variety of detonation networks, characterized in that: The method comprises: Obtain the demand parameters in the detonation network demand information database, match them with the preset function module attributes, and determine the function module combination scheme; Obtain the corresponding three-dimensional virtual prototype model from the modular junction box virtual prototype library, use virtual assembly simulation technology to perform assembly simulation on the obtained three-dimensional virtual prototype model, verify the aforementioned module combination scheme, and optimize and form an assembly scheme; Generate a physical connection relationship table based on the assembly plan to guide assembly, use machine vision technology to monitor the assembly process, issue a warning and mark the non-conforming position when there is a discrepancy; After assembly is completed, the correctness of the electrical connection is determined by the electrical connection automatic detection technology, and the fault diagnosis and correction plan is generated using the fault diagnosis knowledge base.

2. The method for preparing a modular junction box applicable to a variety of detonation networks according to claim 1, characterized in that: The step of obtaining the demand parameters in the detonation network demand information database includes the required power type, voltage value range, current value range, control mode and safety protection level; In the step of matching the acquired demand parameters with the preset functional module attributes, the functional module combination scheme required to realize the detonation demand is determined by calculating the attribute similarity; In the step of assembling the acquired three-dimensional virtual prototype model using virtual assembly simulation technology, the feasibility of the functional module combination scheme is verified by an interference checking method; If the solution verification fails, return to the requirement matching step, redesign the functional module combination solution and re-verify until the verification passes; For the verified functional module combination scheme, the assembly process optimization algorithm is used to adjust the module assembly sequence and position relationship to obtain the assembly scheme with the best assembly efficiency; The physical connection relationship table includes information such as the module name, module interface type, interface location and connection sequence; The step of generating a physical connection relationship table according to the assembly scheme to guide assembly specifically includes transmitting the generated physical connection relationship table to the production control system of the junction box assembly workshop as an electronic process file to guide the assembly operation; Before using machine vision technology to monitor the assembly process, the following steps are also implemented: scan the unique identification code of each functional module with a barcode scanner, compare the scanned information with the physical connection relationship table, determine whether the module selection and assembly sequence are correct, and if there is any deviation, promptly warn and guide the operator to correct it.

3. The method for preparing a modular junction box applicable to a variety of detonation networks according to claim 1, characterized in that: In the step of obtaining the demand parameters in the detonation network demand information library, semantic analysis technology is used to parse the demand parameters, extract keywords and construct a demand vector, the demand vector and the preset function module attribute vector are similarly calculated, and the cosine similarity algorithm is used to determine the most matching function module combination scheme; In the step of assembling the acquired three-dimensional virtual prototype model using virtual assembly simulation technology, a collision detection algorithm is used to determine whether there is interference between the functional modules. If there is interference, an assembly optimization process is triggered, and a heuristic optimization algorithm, such as a genetic algorithm, is used to optimize the module position and sequence during the assembly process to obtain an optimal assembly solution; In the step of generating a physical connection relationship table according to the assembly plan to guide the assembly, the physical connection relationship table of each functional module is automatically generated through graphic recognition technology, including the interface type, position and connection order, and the generated physical connection relationship table is converted into a machine-readable instruction set, such as G code, and transmitted to the automated assembly equipment control system.

4. The method for modularizing a junction box applicable to a variety of detonation networks according to claim 3, characterized in that: The step of generating a physical connection relationship table according to the assembly scheme to guide assembly includes: Through the interface matching algorithm, it is determined whether the interface types of each functional module are compatible. If not, the interface conversion module is triggered to automatically convert the interface type until the matching conditions are met. Using 3D spatial positioning technology, determine the best installation position of each functional module inside the junction box and generate the corresponding 3D assembly coordinates; According to the physical connection relationship table and 3D assembly coordinates, virtual reality technology is used to generate realistic 3D animation of the assembly process, simulate assembly operations, and optimize the assembly process; Upload the optimized physical connection relationship table and the three-dimensional animation of the assembly process to the digital assembly guidance system.

5. The method for preparing a modular junction box applicable to a variety of detonation networks according to claim 1, characterized in that: The step of using machine vision technology to monitor the assembly process includes: The assembly status image of the junction box during assembly is acquired in real time through machine vision technology. The acquired image is denoised and enhanced using an image preprocessing algorithm. The functional modules in the processed assembly status image are located and identified using a target detection algorithm based on deep learning to obtain the actual assembly position and sequence information of each module. According to the physical connection relationship table, obtain standard assembly position and sequence information of each functional module; Using an image recognition algorithm to compare the actual assembly position and sequence information with the standard assembly position and sequence information, and determine whether the assembly position and sequence of each functional module is consistent with the physical connection relationship table; If the judgment result is non-conformity, a warning signal of the assembly control center is triggered, and an assembly pause instruction and a coordinate information signal of the non-conformity position are generated and sent to the control module of the assembly equipment; According to the coordinate information of the inconsistent position, the wrong assembly position is marked on the display screen, and the operator is prompted to make corrections by highlighting it; The operator adjusts and corrects the incorrect assembly position according to the prompt information on the display screen, and sends a correction completion signal through the human-computer interaction interface after completion; After receiving the calibration completion signal, the machine vision technology is re-triggered to detect the assembly status to determine whether the assembly position and sequence after calibration meet the requirements; If the corrected assembly state meets the requirements of the physical connection relationship table, the assembly pause instruction is released to allow the assembly process to continue; Otherwise, repeat the above steps until the calibration is successful.

6. The method for modular junction boxes applicable to multiple detonation networks according to claim 1, characterized in that: The step of "judging the correctness of the electrical connection by the electrical connection automatic detection technology, and generating a fault diagnosis and correction plan by using the fault diagnosis knowledge base" includes: According to the functional module combination scheme of the junction box, the electrical connection relationship data of each module is obtained, and the internal electrical connection model of the junction box is constructed; Through the electrical connection automatic detection technology, the electrical connection data of the junction box after actual assembly is collected to obtain the actual electrical connection information; Compare the actual electrical connection information with the electrical connection model to determine whether there is a connection error; If there is a connection error, the error information is extracted to form structured error description data; Input the structured error description data into the fault diagnosis knowledge base, which uses rule-based reasoning to match the corresponding fault causes and troubleshooting steps, and generates a fault diagnosis report in the form of text and images; The fault diagnosis report is transmitted to the maintenance terminal through the network and presented on the display screen to the maintenance personnel for inspection and repair; After the repair is completed, it is verified again through the electrical connection automatic detection technology until the electrical connection inside the junction box is completely correct.

7. The method for preparing a modular junction box applicable to a variety of detonation networks according to claim 1, characterized in that: Also includes: After the junction box is put into use, the built-in status monitoring module collects operating parameters in real time, including voltage, current and temperature data, to form multi-dimensional time series data; Preprocess the collected multi-dimensional time series data, including data cleaning, normalization and feature extraction, to obtain a standardized feature vector set. Use machine learning algorithms, such as support vector machines or neural networks, to build a multi-parameter fusion model under normal working conditions, to achieve modeling of the running status of the junction box, and obtain the real-time prediction value of the junction box under the working status based on the multi-parameter fusion model. Using the sliding time window mechanism, the multi-parameter fusion model is dynamically updated; When the collected real-time data deviates from the predicted value of the multi-parameter fusion model and exceeds or falls below the preset threshold, it is judged as a precursor to a fault, and an early warning message is generated and sent to the management platform. After receiving the early warning message, the management platform automatically triggers the preventive maintenance process, generates a maintenance work order and sends it to the relevant department.

Images