An internet of things interworking pc component digitization assembly line process coordination system

Abstract

The application relates to the technical field of industrial control systems, in particular to a kind of interworking PC component digital assembly line process coordination system of Internet of Things, the PC component assembly line is equipped with mould table transverse moving vehicle and central mobile transport platform, comprising: data perception layer, deployment in each process equipment, real-time acquisition operating state parameter;Process dependency graph construction module, each process equipment is taken as node, material dependency and time sequence dependency relationship between processes are taken as directed edge, and each directed edge is configured with propagation weight, and a process dependency graph is constructed;The application identifies abnormal signs when the equipment is not completely faulted, predicts fault propagation path, impact strength and arrival time by constructing process dependency graph, and generates hierarchical blocking instructions according to the influence range level, realizes the change from passive response to active prior blocking, and solves the technical problem that single-point fault is amplified along the process dependency chain to cause the whole line to stop production from the data model layer and the control strategy layer.

Description

Technical Field

[0001] This invention relates to the field of industrial control system technology, and in particular to an Internet of Things (IoT) interconnected PC component digital production line process coordination system. Background Technology

[0002] PC (Precast Concrete) components are the main structure of prefabricated buildings. Their production typically employs an assembly line approach, with the molds undergoing a series of processes including cleaning, spraying, marking, installation of reinforcing bars and embedded parts, concrete placement, vibration, pre-curing, curing, and demolding to complete the component formation. With the rapid development of prefabricated buildings, the automation and digitalization levels of PC component production lines are continuously improving. Some production lines have achieved automatic scheduling and intelligent operation through integration with MES systems, as well as dynamic tracking of the flow based on sensors and the Internet of Things.

[0003] Regarding the control and scheduling of PC component production lines, existing technologies have proposed production sequencing methods based on order and production line data, as well as dynamic tracking systems based on detection sensors. In terms of fault handling, existing technologies have proposed diagnostic systems that enable real-time monitoring of the operating status of automated PC production lines and rapid fault location and alarm functions. Furthermore, to reduce the impact of single-point failures on the entire line, some production lines have adopted a design scheme that integrates series and parallel structures.

[0004] However, existing technologies still have shortcomings in the following aspects: While existing scheduling and control systems can monitor the operating status of each process in the production line, there is a lack of unified modeling based on process dependencies between equipment in each process, making it difficult to quantify the propagation path and impact intensity of a single point of failure along the process chain. Furthermore, existing fault diagnosis solutions are all reactive, locating and alarming after equipment failure occurs, failing to identify abnormal signs before equipment fully fails, and even less able to predict the propagation path of the fault to downstream processes. Moreover, there is a lack of a tiered proactive intervention mechanism based on the impact level after a fault occurs. When an anomaly occurs in a process, it is impossible to proactively block its propagation chain before the fault's impact spreads to downstream processes, causing the impact of a single point of failure to amplify tier by tier along the process dependency chain, potentially causing the entire production line to shut down. Summary of the Invention

[0005] To address the technical problems existing in the prior art, this invention provides an IoT-interconnected digital assembly line process coordination system for PC components. The PC component assembly line is equipped with a mold table traversing vehicle and a central mobile transport platform, including: The data sensing layer is deployed on equipment in each process to collect operating status parameters in real time. The process dependency graph construction module uses each process equipment as nodes, the material dependencies and time dependencies between processes as directed edges, and configures propagation weights for each directed edge to construct the process dependency graph. The fault symptom identification module performs time-series analysis based on the operating status parameters, identifies abnormal symptoms of the target process equipment, and outputs an abnormal symptom signal when the target process equipment has not completely failed. The propagation path prediction module, in response to the abnormal symptom signal, predicts the path of potential fault propagation along the directed edge, the expected impact intensity of downstream nodes, and the arrival time based on the process dependency graph. The proactive blocking decision module generates proactive blocking decision instructions according to the impact range level before the actual impact of the potential fault reaches the downstream node, based on the expected impact intensity and arrival time. The execution control module converts the active blocking decision command into a control signal and sends it to the actuator of the corresponding process equipment via the Internet of Things to complete the blocking operation before the impact of the fault arrives.

[0006] As a further aspect of the present invention, the data sensing layer is deployed in at least one of the following process equipment in the PC component production line: cleaning station equipment, spraying station equipment, marking station equipment, rebar and embedded part installation station equipment, quality inspection station equipment, side mold installation station equipment, material placement station equipment, mold lateral movement station equipment, leveling station equipment, pre-curing station equipment, polishing station equipment, curing kiln equipment, stacking station equipment, mold removal station equipment, and hoisting station equipment; the operating status parameters include at least one of the vibration data, temperature data, current data, and rotational speed data of each process equipment.

[0007] As a further embodiment of the present invention, the process dependency graph construction module is also configured with a dynamic dependency update interface, which is used to receive externally input process arrangement change instructions when the process arrangement of the PC component pipeline changes, and add, delete or modify the nodes and directed edges and their propagation weights in the process dependency graph according to the process arrangement change instructions.

[0008] As a further aspect of the present invention, the propagation path prediction module includes: The influence propagation model submodule is configured with a multi-level propagation algorithm based on the process dependency graph, which is used to calculate the expected influence intensity attenuation coefficient at each downstream node when a potential fault propagates along each directed edge; The time window calculation submodule is used to calculate the expected arrival time of the potential fault along each propagation path to each downstream node based on the current operating cycle time of each process equipment and the mold table turnover time between each process. The path enumeration submodule is used to traverse all reachable paths from the target process equipment in the process dependency graph, and output each reachable path and its corresponding node sequence, expected influence intensity sequence, and expected influence arrival time sequence.

[0009] As a further aspect of the present invention, the active blocking decision module includes: The level determination submodule is configured with a first influence range threshold and a second influence range threshold, wherein the first influence range threshold is less than the second influence range threshold. It is used to determine the influence range level corresponding to the current abnormal symptom as Level 1 blocking, Level 2 blocking, or Level 3 blocking based on the comparison result between the influence range output by the propagation path prediction module and the first influence range threshold and the second influence range threshold. The strategy generation submodule is used to generate an active blocking decision instruction corresponding to the level of influence determined by the level determination submodule. The execution mechanism response strategy for the first-level interruption is as follows: notify the upstream process equipment of the target process equipment to reduce the production cycle time and reserve a buffer time window for the downstream process equipment of the target process equipment; the execution mechanism response strategy for the second-level interruption is as follows: temporarily switch the affected downstream process of the target process equipment to a parallel channel or a spare workstation; the execution mechanism response strategy for the third-level interruption is as follows: activate the cross-line mold transfer plan and transfer the molds that are in circulation to other production lines for continued production through the mold traversing vehicle and the central mobile transport platform.

[0010] As a further aspect of the present invention, a blocking effect verification module is also included, which is used to collect the actual propagation path of the actual fault along each directed edge and the actual impact intensity of each downstream node after the actual fault occurs in the target process equipment, compare the actual propagation path and actual impact intensity with the predicted propagation path and expected impact intensity output by the propagation path prediction module, generate model correction parameters based on the comparison results, and feed the model correction parameters back to the propagation path prediction module to update the prediction model.

[0011] As a further aspect of the present invention, when the blocking effect verification module determines that the target process equipment has a repetitive failure mode, the cross-production line process capacity coordination module is triggered. The cross-production line process capacity coordination module includes: The real-time capacity quantification submodule is used to obtain the real-time idle capacity of each process on each production line in multiple PC component production lines, and to standardize and quantify the real-time idle capacity of each process. The capacity storage submodule is used to summarize and store the real-time idle capacity information of each process on each production line according to the process type, forming a process capacity pool across production lines. The capacity scheduling decision submodule is used to retrieve the corresponding process of other production lines with idle capacity that is the same type as the target process from the process capacity pool when the real-time available capacity of the target process of the target production line is lower than the preset capacity threshold, and generate a cross-line capacity borrowing instruction. The cross-line mold table scheduling and execution submodule is used to convert the cross-line capacity borrowing instruction into a mold table transfer control signal, control the mold table traversing vehicle and the central mobile transport platform to transfer the mold table carrying the semi-finished components from the target production line to the corresponding process station of the production line whose capacity is borrowed, and transfer the mold table back to the target production line after the process is completed.

[0012] As a further aspect of the present invention, the cross-production line process capacity coordination module further includes: The capacity credit rating submodule is used to record the borrower production line identifier, lender production line identifier, borrowed capacity quantity, and borrowing start and end time for each cross-line capacity borrowing, and to calculate the capacity credit score of each production line based on the number of times capacity is borrowed and borrowed. The scheduling priority adjustment submodule is used to adjust the scheduling priority of each production line in the capacity scheduling decision submodule according to the capacity credit score, wherein the production line with the capacity credit score higher than the preset credit threshold is given priority matching right when requesting to borrow capacity.

[0013] As a further aspect of the present invention, it also includes a process time sensing and dynamic cycle adjustment module, wherein the process time sensing and dynamic cycle adjustment module includes: The real-time process time recording submodule is used to record the actual time spent on each process after it is completed, and to calculate the deviation between the actual time spent and the standard time for that process. The cumulative waiting time prediction submodule is used to calculate the predicted cumulative waiting time value at each downstream process after the deviation value of the current process and the standard working hours and process dependencies of each downstream process. The dynamic beat adjustment submodule is used to generate a start timing adjustment instruction and a task diversion instruction for a downstream process before the actual occurrence of the cumulative waiting time when the predicted value of the cumulative waiting time of a certain downstream process exceeds the preset waiting threshold. The cycle instruction execution submodule is used to convert the start timing adjustment instruction and task diversion instruction into control signals for the corresponding process equipment and send them out for execution.

[0014] As a further aspect of the present invention, the process time sensing and dynamic cycle adjustment module further includes: The prediction self-learning submodule is used to compare the predicted cumulative waiting time with the actual waiting time after the actual waiting time of each process occurs, calculate the prediction error, and adjust the prediction model parameters of the cumulative waiting time prediction submodule according to the prediction error.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: This invention transforms from passive response to proactive pre-emptive prevention by constructing a process dependency graph, identifying abnormal signs when equipment is not completely faulty, predicting the fault propagation path, impact intensity, and arrival time, and generating graded blocking commands according to the impact range level. It addresses the technical problem of single-point faults amplifying along the process dependency chain, leading to production line shutdowns, at both the data model and control strategy levels. The invention uses a blocking effect verification module to compare the actual propagation path with the predicted value, generating model correction parameters to update the prediction model, thus continuously optimizing prediction accuracy during operation. Vibration, temperature, current, and speed sensors deployed in each process piece collect operating status parameters at a 100Hz sampling frequency. All modules share the same data perception layer and issue control signals through a unified execution control module, forming a complete process coordination architecture with unified perception, distributed decision-making, and collaborative execution. This invention triggers cross-production line capacity coordination when equipment exhibits a recurring failure mode. It acquires real-time idle capacity of each production line at the process level and aggregates it into a capacity pool based on process type. Idle capacity of the same type is retrieved, and the traverse car and central mobile transport platform are controlled to implement cross-line transfer, enabling process-level complementarity of idle capacity for the same process across different production lines. This solves the structural capacity mismatch problem caused by asymmetric process capabilities across production lines. Furthermore, this invention records the deviation between actual time consumption and standard working hours, calculates the predicted cumulative waiting time at each downstream process after this deviation propagates along the process chain, and generates start-up timing adjustment and task diversion instructions before the actual cumulative waiting time occurs. This eliminates the invalid waiting time caused by the cumulative amplification of process time deviation along the dependency chain. By comparing the predicted value with the actual value to calculate the prediction error and adjusting the model parameters, the prediction model is continuously optimized, improving the accuracy of dynamic cycle time adjustment. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a block diagram of the overall system architecture of the present invention; Figure 2 This is a schematic diagram of the process dependency diagram of the present invention; Figure 3 This is a flowchart of the fault propagation prediction and graded blocking process of the present invention; Figure 4 This is a flowchart illustrating the cross-production line process capacity coordination of the present invention. Figure 5 This is a flowchart of the dynamic beat adaptive adjustment process of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0020] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0021] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0022] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0023] Please see Figures 1 to 5 As shown, this embodiment of the invention provides an IoT-interconnected digital assembly line process coordination system for PC components. The PC component assembly line is equipped with a mold table traversing vehicle and a central mobile transport platform, including: The data sensing layer is deployed on equipment in each process to collect operating status parameters in real time. The process dependency graph construction module uses each process equipment as nodes, the material dependencies and time dependencies between processes as directed edges, and configures propagation weights for each directed edge to construct the process dependency graph. The fault symptom identification module performs time-series analysis based on the operating status parameters, identifies abnormal symptoms of the target process equipment, and outputs an abnormal symptom signal when the target process equipment has not completely failed. The propagation path prediction module, in response to the abnormal symptom signal, predicts the path of potential fault propagation along the directed edge, the expected impact intensity of downstream nodes, and the arrival time based on the process dependency graph. The proactive blocking decision module generates proactive blocking decision instructions according to the impact range level before the actual impact of the potential fault reaches the downstream node, based on the expected impact intensity and arrival time. The execution control module converts the active blocking decision command into a control signal and sends it to the execution mechanism of the corresponding process equipment through the Internet of Things, so as to complete the blocking operation before the impact of the fault arrives; The factory is equipped with two PC component production lines. Each production line is equipped with a mold traverse car and a central moving transport platform. The mold traverse car is installed in the transverse passage of the production line and is used to move the mold trolley from one production line to another. The central moving transport platform runs along the longitudinal track of the production line and is used to transfer the mold trolley from the current workstation to the designated workstation downstream. The mold traverse car is responsible for transverse cross-line transfer, and the central moving transport platform is responsible for longitudinal transfer between workstations. The two work together to complete the cross-line mold trolley transfer.

[0024] The system comprises a data perception layer, a process dependency graph construction module, a fault symptom identification module, a propagation path prediction module, an active blocking decision module, and an execution control module. Upon initial system startup, the process dependency graph construction module automatically generates an initial process dependency graph based on a pre-set process arrangement configuration file. During system operation, all modules continuously collaborate, forming a complete closed loop of perception → identification → prediction → blocking → execution.

[0025] The data sensing layer is deployed on various process equipment on the production line, including vibration and temperature sensors on the cloth placing machine, temperature and humidity sensors in the curing kiln, current and speed sensors on the traverse car drive motor, and displacement sensors on the central mobile transport platform. Each sensor reports the collected data to the central control system in real time via an Internet of Things communication protocol (such as MQTT).

[0026] The process dependency graph construction module uses equipment for each process as nodes (such as cleaning machine node, spraying machine node, concrete placing machine node, curing kiln node, traverse car node, etc.), and directive edges representing material dependencies (e.g., the mold must pass through cleaning → spraying → concrete placing → curing) and temporal dependencies (e.g., curing can only begin after concrete placing is completed). Each directed edge is assigned a propagation weight. The direction of the directed edges strictly follows the material and temporal dependencies and exists only between processes with direct upstream and downstream relationships. For example, there is a direct material flow relationship between the concrete placing machine node and the vibrating station node (concrete placing → vibrating), so a higher propagation weight is assigned. Although there is an indirect connection between the spraying machine node and the concrete placing machine node, they are not directly upstream and downstream, so no directed edge is assigned between them, and no propagation weight is set. Each propagation weight is pre-configured based on the tightness of the direct material flow dependencies between processes.

[0027] The fault symptom identification module performs time-series analysis on the operating status parameters collected by the data sensing layer. For example, when the vibration data of the concrete placing machine continuously rises above the normal threshold and the current data shows abnormal fluctuations, it is identified as an abnormal symptom of the concrete placing machine, and an abnormal symptom signal is output when the equipment has not completely failed.

[0028] After receiving the abnormal symptom signal, the propagation path prediction module predicts the propagation path of the potential fault of the concrete placing machine based on the process dependency graph: concrete placing machine (fault source) → vibrating station (expected impact intensity 0.7, arrival time 3min) → pre-curing station (expected impact intensity 0.5, arrival time 8min) → curing kiln (expected impact intensity 0.3, arrival time 15min).

[0029] Based on the prediction results, the active blocking decision module determines that the affected area covers 3 downstream nodes, corresponding to a secondary blocking, and generates an active blocking decision instruction: temporarily switch the operation of the vibration station and the pre-curing station to the backup channel.

[0030] The execution control module converts the blocking decision command into control signals for each process equipment, and sends them to the controllers of the vibration station and pre-curing station via the Internet of Things to execute the blocking operation. At this time, the concrete placing machine is not completely faulty, but the blocking measures have been completed before the impact of the fault reaches each downstream node.

[0031] Furthermore, the data sensing layer is deployed in at least one of the following process equipment in the PC component production line: cleaning station equipment, spraying station equipment, marking station equipment, rebar and embedded part installation station equipment, quality inspection station equipment, side mold installation station equipment, material placement station equipment, mold lateral movement station equipment, leveling station equipment, pre-curing station equipment, polishing station equipment, curing kiln equipment, stacking station equipment, mold removal station equipment, and hoisting station equipment; the operating status parameters include at least one of the vibration data, temperature data, current data, and rotational speed data of each process equipment; The data sensing layer is deployed in the following process equipment of the PC component production line: proximity sensors and photoelectric sensors are installed on the cleaning station equipment to detect the mold table's positioning status and the cleaning brush's wear status; liquid level sensors and pressure sensors are installed on the spraying station equipment to detect the release agent level and spraying pressure; vision sensors are installed on the marking station equipment to detect the marking position accuracy; weight sensors are installed on the rebar and embedded part installation station equipment to detect the weight of the rebar cage; vibration sensors (to detect the vibration of the concrete placing machine), temperature sensors (to detect the concrete temperature), and weighing sensors (to detect the amount of concrete placed) are installed on the placing station equipment; temperature sensors and humidity sensors are installed on the curing kiln equipment to detect the curing temperature and humidity; and current sensors and speed sensors are installed on the mold table traverse trolley drive motor.

[0032] Operating status parameters include vibration data (such as the vibration acceleration of the fabric feeder), temperature data (such as the temperature of the curing kiln), current data (such as the current of the traverse car drive motor), and speed data (such as the speed of the traverse car drive motor) of each process equipment. Each sensor collects data in real time at a sampling frequency of 100Hz and transmits it to the central control system via industrial Ethernet.

[0033] Furthermore, the process dependency graph construction module is also configured with a dynamic dependency update interface, which is used to receive external process arrangement change instructions when the process arrangement of the PC component pipeline changes, and add, delete or modify the nodes and directed edges and their propagation weights in the process dependency graph according to the process arrangement change instructions. The original production line of a PC component factory was arranged as follows: cleaning → spraying → marking → rebar installation → material placement → vibration → pre-curing → curing → demolding. Later, due to the production of new Diego-style composite slab components, an insulation board placement step needed to be added after the material placement step. The original production line was then arranged as follows: cleaning → spraying → marking → rebar installation → material placement → insulation board placement → vibration → pre-curing → curing → demolding.

[0034] Factory managers input process arrangement change instructions through the operation interface of the central control system. The instructions include labels for adding insulation board placement equipment and its type, adding directional edge fabric → insulation board placement (propagation weight 0.85) and insulation board placement → vibration (propagation weight 0.9), and deleting the original directional edge fabric → vibration.

[0035] After receiving the instruction, the dynamic update interface for dependency relationships in the process dependency graph construction module adds a node for the insulation board placement equipment, adds two new directed edges and their propagation weights to the process dependency graph, deletes the original directed edge for material placement and vibration, and recalculates the topological sorting of the affected nodes. After the update is completed, the system automatically verifies the integrity of the dependency graph - confirms that all nodes are still connected and there are no circular dependencies. After the verification is passed, the new process dependency graph is enabled.

[0036] Furthermore, the propagation path prediction module includes: The influence propagation model submodule is configured with a multi-level propagation algorithm based on the process dependency graph, which is used to calculate the expected influence intensity attenuation coefficient at each downstream node when a potential fault propagates along each directed edge; The time window calculation submodule is used to calculate the expected arrival time of the potential fault along each propagation path to each downstream node based on the current operating cycle time of each process equipment and the mold table turnover time between each process. The path enumeration submodule is used to traverse all reachable paths from the target process equipment in the process dependency graph, and output each reachable path and its corresponding node sequence, expected impact intensity sequence and expected impact arrival time sequence; The propagation path prediction module includes an impact propagation model submodule, a time window calculation submodule, and a path enumeration submodule. The impact propagation model submodule is configured with a multi-level propagation algorithm based on a process dependency graph. This algorithm uses an attenuation coefficient model: when a fault propagates from the current node to the next adjacent node, the impact intensity is multiplied by the propagation weight of the directed edge; when it propagates to the second downstream node, it is multiplied by the propagation weight of the next directed edge, and so on. For example, the initial impact intensity of a concrete placing boom fault is 1.0, which becomes 0.7 after propagating to the vibrating station (propagation weight 0.7), then to the pre-curing station (propagation weight 0.6), becoming 0.42, and finally to the curing kiln (propagation weight 0.5), becoming 0.21.

[0037] The time window calculation submodule calculates the expected impact arrival time based on the current operating cycle time of each process equipment and the mold table transfer time between processes. For example, if the current operating cycle time of the material placing station is 12 minutes per mold table, and it takes 2 minutes for the mold table to transfer from the material placing station to the vibrating station, 3 minutes from the vibrating station to the pre-curing station, and 5 minutes from the pre-curing station to the curing kiln, then the expected impact arrival time for a material placing machine failure to propagate to the vibrating station is 2 minutes, to the pre-curing station is 5 minutes, and to the curing kiln is 10 minutes.

[0038] The path enumeration submodule traverses all reachable paths from the target process equipment (such as a concrete placing boom) in the process dependency graph. For example, reachable paths from the concrete placing boom include: Path 1: Concrete placing boom → Vibration station → Pre-curing station → Curing kiln; Path 2: Concrete placing boom → Vibration station → Backup passage → Curing kiln (if a backup passage exists). Each path outputs the corresponding node sequence, expected influence intensity sequence, and expected influence arrival time sequence.

[0039] Furthermore, the active blocking decision module includes: The level determination submodule is configured with a first influence range threshold and a second influence range threshold, wherein the first influence range threshold is less than the second influence range threshold. It is used to determine the influence range level corresponding to the current abnormal symptom as Level 1 blocking, Level 2 blocking, or Level 3 blocking based on the comparison result between the influence range output by the propagation path prediction module and the first influence range threshold and the second influence range threshold. The strategy generation submodule is used to generate an active blocking decision instruction corresponding to the level of influence determined by the level determination submodule. The execution mechanism response strategy for the first-level interruption is as follows: notify the upstream process equipment of the target process equipment to reduce the production cycle time and reserve a buffer time window for the downstream process equipment of the target process equipment; the execution mechanism response strategy for the second-level interruption is as follows: temporarily switch the affected downstream process of the target process equipment to a parallel channel or a spare workstation; the execution mechanism response strategy for the third-level interruption is as follows: activate the cross-line mold transfer plan and transfer the molds in circulation to other production lines for continued production through the mold traversing vehicle and the central mobile transport platform; The level determination submodule is configured with a first range influence threshold and a second range influence threshold, where the first range influence threshold is smaller than the second range influence threshold. Taking a PC component production line with a total of 15 process nodes, including 3 key nodes (central controller node, main conveyor line node, and curing kiln node), as an example, the first range influence threshold is configured for 2 nodes, and the second range influence threshold is configured for 5 nodes. These thresholds are pre-configured in the system based on the total number of processes and the distribution of key equipment in the production line, and can be adjusted by administrators through the configuration interface.

[0040] The level determination submodule determines the level based on the impact range (i.e., the total number of affected upstream and downstream nodes) output by the propagation path prediction module: if the failure of the concrete placing boom only affects one node of its directly downstream vibrating station (impact range = 1 ≤ first threshold 2), it is determined as a level 1 blockage; if the failure of the concrete placing boom affects a total of 3 nodes, including the vibrating station, pre-curing station, and curing kiln (first threshold 2 < impact range = 3 ≤ second threshold 5), it is determined as a level 2 blockage; if the failure of the concrete placing boom affects more than 5 nodes (impact range > second threshold 5), or if the impact range does not exceed 5 nodes but includes critical nodes (such as the central controller node or curing kiln node included in the impact range), it is determined as a level 3 blockage.

[0041] The strategy generation submodule generates corresponding proactive blocking decision instructions based on the determined impact level: Level 1 Blocking: The strategy generation submodule first generates an instruction to notify the upstream process equipment of the concrete placing boom (such as the rebar installation station) to reduce the production cycle from 12 minutes / former to 15 minutes / former. After the upstream cycle is adjusted, the submodule then reserves a former buffer time window (two empty formers as buffer) for the downstream process equipment (vibration station) of the concrete placing boom. The two actions are performed sequentially to avoid conflicts between the upstream deceleration and the downstream buffer reservation. Level 2 Blocking: The strategy generation submodule generates an instruction to temporarily switch the affected downstream process of the concrete placing boom (vibration station, pre-curing station) to a parallel channel or a backup station. For example, if the factory is equipped with parallel vibration channels A and B, when the vibration station in the main channel cannot work normally due to the risk of upstream concrete placing boom failure, the former will be switched to vibration channel B via a traverse vehicle to continue production. Level 3 Interruption: The strategy generation submodule generates an instruction to activate the cross-line mold transfer plan. The molds currently in operation are transferred to other production lines via a mold traversing vehicle and a central mobile transport platform to continue production. For example, three molds that have already been laid out on production line 1 are transferred to their corresponding workstations on production line 2 via the mold traversing vehicle, where production line 2 continues to complete the subsequent vibration, pre-curing, and curing processes.

[0042] Furthermore, it also includes a blocking effect verification module, which is used to collect the actual propagation path of the actual fault along each directed edge and the actual impact intensity of each downstream node after the actual fault occurs in the target process equipment. The actual propagation path and actual impact intensity are compared with the predicted propagation path and expected impact intensity output by the propagation path prediction module. Based on the comparison results, model correction parameters are generated and the model correction parameters are fed back to the propagation path prediction module to update the prediction model. During a certain operation, the propagation path prediction module predicted the propagation path of the concrete placing machine failure as follows: concrete placing machine → vibratory compaction station (expected impact intensity 0.7) → pre-curing station (expected impact intensity 0.5) → curing kiln (expected impact intensity 0.3), with expected impact arrival times of 3 min, 8 min and 15 min respectively.

[0043] The concrete placing boom experienced a complete failure after 2 minutes of actual operation. The blocking effect verification module collected data on the actual propagation path of the failure along each directed edge and the actual impact intensity of each downstream node after the failure: The actual propagation path was: concrete placing boom → vibratory compaction station (actual impact intensity 0.65) → pre-curing station (expected impact intensity 0.48) → curing kiln (expected impact intensity 0.28); the actual impact time for the vibratory compaction station was 2.5 minutes after the failure, for the pre-curing station it was 7 minutes, and for the curing kiln it was 13 minutes.

[0044] The verification module compares the actual propagation path and actual impact with the predicted values: the influence intensity deviation of the vibration station is 0.05, the deviation of the pre-curing station is 0.02, and the deviation of the curing kiln is 0.02; the arrival time deviations are -0.5 min, -1 min, and +2 min, respectively.

[0045] Based on the comparison results, the model correction parameters are generated: the propagation weight of the directed edge from the concrete placing boom to the vibrating station is adjusted from 0.70 to 0.68, and the time transfer coefficient from the concrete placing boom to the vibrating station is adjusted from 1.0 to 1.2. The correction parameters are then fed back to the propagation path prediction module to update the prediction.

[0046] Furthermore, when the blocking effect verification module determines that the target process equipment has a repetitive failure mode, the cross-production line process capacity coordination module is triggered. The cross-production line process capacity coordination module includes: The real-time capacity quantification submodule is used to obtain the real-time idle capacity of each process on each production line in multiple PC component production lines, and to standardize and quantify the real-time idle capacity of each process. The capacity storage submodule is used to summarize and store the real-time idle capacity information of each process on each production line according to the process type, forming a process capacity pool across production lines. The capacity scheduling decision submodule is used to retrieve the corresponding process of other production lines with idle capacity that is the same type as the target process from the process capacity pool when the real-time available capacity of the target process of the target production line is lower than the preset capacity threshold, and generate a cross-line capacity borrowing instruction. The cross-line mold table scheduling and execution submodule is used to convert the cross-line capacity borrowing instruction into a mold table transfer control signal, control the mold table traversing vehicle and the central mobile transport platform to transfer the mold table carrying the semi-finished components from the target production line to the corresponding process station of the production line whose capacity is borrowed, and transfer the mold table back to the target production line after the process is completed. A PC component factory has two production lines, Production Line 1 and Production Line 2. After multiple verifications, the blocking effect verification module determines that the material placing machine of Production Line 1 has a repetitive failure mode (the repetitive failure mode threshold on which this determination is based - such as more than a preset number of occurrences within 30 days - is preset by the management personnel through the system configuration interface. In this embodiment, the preset threshold is triggered when more than 3 similar failures occur within 30 days), thus triggering the cross-production line process capacity coordination module.

[0047] The real-time process capacity quantification submodule obtains the real-time idle capacity of each process on the two production lines. Taking the fabric feeding process of production line 1 as an example, the designed total capacity of this process is 20 pieces / hour, the current capacity utilization rate is 60%, and the idle capacity is 8 pieces / hour. The curing process of production line 2 has a designed total capacity of 40 pieces / hour, the current capacity utilization rate is 55%, and the idle capacity is 18 pieces / hour.

[0048] The capacity storage submodule summarizes and stores the real-time idle capacity information of each process on each production line according to the process type, forming a process capacity pool across production lines: the fabric process capacity pool includes 8 pieces / hour of idle capacity of production line 1 and 4 pieces / hour of idle capacity of production line 2; the curing process capacity pool includes 2 pieces / hour of idle capacity of production line 1 and 18 pieces / hour of idle capacity of production line 2.

[0049] The capacity scheduling decision submodule detected that the real-time available capacity of the fabric process of production line 1 was 12 pieces / hour, which was lower than the preset capacity threshold of 15 pieces / hour. It searched the process capacity pool for other production lines with the same type of fabric process and with idle capacity. Production line 2 had an idle capacity of 4 pieces / hour for the fabric process and generated a cross-line capacity borrowing instruction: borrow 4 pieces / hour of fabric capacity from production line 2.

[0050] The cross-line mold table scheduling execution submodule converts the cross-line capacity borrowing instruction into a mold table transfer control signal, which controls the mold table lateral movement vehicle to move the four empty mold tables on production line 1, which have been cleaned, sprayed, marked, and reinforced with steel bars, laterally to the material placement station on production line 2. After the material placement is completed on production line 2, the central mobile transport platform transports the mold table carrying the semi-finished components back to the vibration station on production line 1 along the longitudinal track to continue the subsequent processes.

[0051] Furthermore, the cross-production line process capacity coordination module also includes: The capacity credit rating submodule is used to record the borrower production line identifier, lender production line identifier, borrowed capacity quantity, and borrowing start and end time for each cross-line capacity borrowing, and to calculate the capacity credit score of each production line based on the number of times capacity is borrowed and borrowed. The scheduling priority adjustment submodule is used to adjust the scheduling priority of each production line in the capacity scheduling decision submodule according to the capacity credit score, wherein the production line with the capacity credit score higher than the preset credit threshold is given priority matching right when requesting to borrow capacity.

[0052] The capacity credit evaluation submodule of the cross-production line process capacity coordination module records detailed information for each cross-production line capacity borrowing as shown in the table below:

[0053] The capacity credit rating submodule calculates the capacity credit score based on the number of times each production line lends out and borrows its capacity: Production line 1: 2 times of lending out (2nd and 4th times), 2 times of borrowing in (1st and 3rd times), credit score = number of times of lending out / (number of times of lending out + number of times of borrowing in) × 100 = 50 points; Production Line 2: 2 times of lending out (1st and 3rd times), 2 times of borrowing in (2nd and 4th times), credit score = 50 points.

[0054] If the preset credit threshold is 40 points, both production lines are above the threshold.

[0055] The scheduling priority adjustment submodule adjusts the scheduling priority of each production line in the capacity scheduling decision submodule based on its capacity credit score. Production lines with a capacity credit score higher than a preset credit threshold (40 points) receive priority matching when requesting to borrow capacity. For example, when production lines 1 and 2 simultaneously request to borrow fabric capacity from a third production line, the system prioritizes responding to the production line with the higher credit score.

[0056] Furthermore, it also includes a process time sensing and dynamic cycle adjustment module, which includes: The real-time process time recording submodule is used to record the actual time spent on each process after it is completed, and to calculate the deviation between the actual time spent and the standard time for that process. The cumulative waiting time prediction submodule is used to calculate the predicted cumulative waiting time value at each downstream process after the deviation value of the current process and the standard working hours and process dependencies of each downstream process. The dynamic beat adjustment submodule is used to generate a start timing adjustment instruction and a task diversion instruction for a downstream process before the actual occurrence of the cumulative waiting time when the predicted value of the cumulative waiting time of a certain downstream process exceeds the preset waiting threshold. The cycle time instruction execution submodule is used to convert the start timing adjustment instruction and task diversion instruction into control signals for the corresponding process equipment and send them out for execution; The standard process times for a certain PC component production line are as follows: cleaning 8 min / mold, spraying 5 min / mold, marking 6 min / mold, rebar installation 15 min / mold, material placement 12 min / mold, vibration 4 min / mold, pre-curing 20 min / mold, curing 480 min / mold.

[0057] The real-time process time recording submodule records the actual time spent on each process after its completion. For example, if the actual time spent on the rebar installation process for a batch of components is 22 minutes, the deviation from the standard time of 15 minutes for this process is +7 minutes.

[0058] The cumulative waiting time prediction submodule calculates the predicted waiting time at each downstream process after the deviation is propagated along the process chain, based on the deviation value of the current process (rebar installation) + 7 minutes, the standard working hours of each downstream process, and the process dependencies. This post-propagation waiting time refers to the remaining impact value after the deviation value is propagated step by step along the process chain, rather than the sum of the waiting times of each downstream process. Its physical meaning is: the delay effect that is still retained when the delay of the upstream process is finally propagated to the downstream process after being buffered and absorbed by each intermediate process.

[0059] The attenuation coefficient of each process reflects the buffering and absorption capacity of that process to upstream delays. The smaller the attenuation coefficient, the stronger the buffering and absorption capacity of that process. An attenuation coefficient of 1.0 indicates that the process has no buffering and absorption capacity and the upstream delay is completely transmitted. In this embodiment, the attenuation coefficients are pre-configured based on the buffer capacity of the mold table and the flexibility of each process: Fabric laying process: 7 × 0.8 = 5.6 min (attenuation coefficient 0.8, because a buffer mold table is configured at the entrance of the fabric laying process, which can absorb 20% of the delay); Vibration process: 7 × 0.8 × 0.9 = 5.0 min (attenuation coefficient 0.9, because there is no additional buffer before the vibration process, it only absorbs 10% of the delay); Pre-curing process: 7 × 0.8 × 0.9 × 1.0 = 5.0 min (attenuation coefficient 1.0, there is no buffer at the entrance of the pre-curing process, and the delay is completely transmitted); Curing process: 7 × 0.8 × 0.9 × 1.0 × 1.0 = 5.0 min (attenuation coefficient 1.0, there is also no buffer at the entrance of the curing process, and the delay is completely transmitted).

[0060] The dynamic beat adjustment submodule detects that the predicted waiting time after the transmission of the curing process is 5.0 min, which exceeds the preset waiting threshold (3 min). Before the actual waiting time after transmission occurs, it generates a start-up timing adjustment instruction for the curing process: delaying the kiln entry time of the curing kiln by 5 min, and generates a task diversion instruction: prioritizing the two molds in the current batch that have completed material placement and vibration to enter the curing kiln to avoid conflicts with subsequent batches.

[0061] The cycle instruction execution submodule converts the start timing adjustment instruction into the kiln entry delay signal of the curing kiln controller, and converts the task diversion instruction into the scheduling signal of the mold table transverse moving car and sends it out for execution.

[0062] Furthermore, the process time sensing and dynamic cycle adjustment module also includes: The prediction self-learning submodule is used to compare the predicted value of the cumulative waiting time with the actual waiting time after the actual waiting time of each process occurs, calculate the prediction error, and adjust the prediction model parameters of the cumulative waiting time prediction submodule according to the prediction error. The predictive self-learning submodule of the process time perception and dynamic rhythm adjustment module compares the predicted waiting time after transmission with the actual waiting time after each process's actual waiting time occurs. For example, in the scenario of the above embodiment, after a 7-minute delay in the rebar installation process: the system predicts a 5.0-minute waiting time for the curing process; in actual operation, the actual waiting time for the curing process is 4.2 minutes (because there is an empty formwork before the curing process as a buffer, absorbing part of the delay).

[0063] The prediction self-learning submodule calculates the prediction error as follows: Predicted value - Actual value = 5.0 - 4.2 = 0.8 min, and the relative error is approximately 0.8 / 4.2 ≈ 9%.

[0064] Based on the prediction error, the prediction model parameters of the cumulative waiting time prediction submodule were adjusted: the attenuation coefficient for rebar installation → concrete placement was adjusted from 0.80 to 0.85 (because the actual propagation attenuation was less than expected), and the attenuation coefficient for concrete placement → vibration compaction was adjusted from 0.90 to 0.92. The adjusted model recalculated the predicted waiting time after transmission in subsequent batches, gradually approaching the actual value. The system continuously accumulates prediction-actual comparison data during continuous operation, enabling continuous optimization of the prediction model parameters.

[0065] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An IoT-interconnected digital assembly line process coordination system for PC components, wherein the PC component assembly line is equipped with a mold table traversing vehicle and a central mobile transport platform, characterized in that, include: The data sensing layer is deployed on equipment in each process to collect operating status parameters in real time. The process dependency graph construction module uses each process equipment as nodes, the material dependencies and time dependencies between processes as directed edges, and configures propagation weights for each directed edge to construct the process dependency graph. The fault symptom identification module performs time-series analysis based on the operating status parameters, identifies abnormal symptoms of the target process equipment, and outputs an abnormal symptom signal when the target process equipment has not completely failed. The propagation path prediction module, in response to the abnormal symptom signal, predicts the path of potential fault propagation along the directed edge, the expected impact intensity of downstream nodes, and the arrival time based on the process dependency graph. The proactive blocking decision module generates proactive blocking decision instructions according to the impact range level before the actual impact of the potential fault reaches the downstream node, based on the expected impact intensity and arrival time. The execution control module converts the active blocking decision command into a control signal and sends it to the actuator of the corresponding process equipment via the Internet of Things to complete the blocking operation before the impact of the fault arrives.

2. The IoT-interoperable PC component digital production line process coordination system according to claim 1, characterized in that, The data sensing layer is deployed in at least one of the following process equipment in the PC component production line: cleaning station equipment, spraying station equipment, marking station equipment, rebar and embedded part installation station equipment, quality inspection station equipment, side mold installation station equipment, material placement station equipment, mold lateral movement station equipment, leveling station equipment, pre-curing station equipment, polishing station equipment, curing kiln equipment, stacking station equipment, mold removal station equipment, and hoisting station equipment; the operating status parameters include at least one of the vibration data, temperature data, current data, and rotational speed data of each process equipment.

3. The IoT-interoperable PC component digital production line process coordination system according to claim 2, characterized in that, The process dependency graph construction module is also configured with a dynamic dependency update interface, which is used to receive external process arrangement change instructions when the process arrangement of the PC component pipeline changes, and add, delete or modify the nodes and directed edges and their propagation weights in the process dependency graph according to the process arrangement change instructions.

4. The IoT-interoperable PC component digital production line process coordination system according to claim 3, characterized in that, The propagation path prediction module includes: The influence propagation model submodule is configured with a multi-level propagation algorithm based on the process dependency graph, which is used to calculate the expected influence intensity attenuation coefficient at each downstream node when a potential fault propagates along each directed edge; The time window calculation submodule is used to calculate the expected arrival time of the potential fault along each propagation path to each downstream node based on the current operating cycle time of each process equipment and the mold table turnover time between each process. The path enumeration submodule is used to traverse all reachable paths from the target process equipment in the process dependency graph, and output each reachable path and its corresponding node sequence, expected influence intensity sequence, and expected influence arrival time sequence.

5. The IoT-interoperable PC component digital production line process coordination system according to claim 4, characterized in that, The active blocking decision module includes: The level determination submodule is configured with a first influence range threshold and a second influence range threshold, wherein the first influence range threshold is less than the second influence range threshold. It is used to determine the influence range level corresponding to the current abnormal symptom as Level 1 blocking, Level 2 blocking, or Level 3 blocking based on the comparison result between the influence range output by the propagation path prediction module and the first influence range threshold and the second influence range threshold. The strategy generation submodule is used to generate an active blocking decision instruction corresponding to the level of influence determined by the level determination submodule. The execution mechanism response strategy for the first-level interruption is as follows: notify the upstream process equipment of the target process equipment to reduce the production cycle time and reserve a buffer time window for the downstream process equipment of the target process equipment; the execution mechanism response strategy for the second-level interruption is as follows: temporarily switch the affected downstream process of the target process equipment to a parallel channel or a spare workstation; the execution mechanism response strategy for the third-level interruption is as follows: activate the cross-line mold transfer plan and transfer the molds that are in circulation to other production lines for continued production through the mold traversing vehicle and the central mobile transport platform.

6. The IoT-interoperable PC component digital production line process coordination system according to claim 5, characterized in that, It also includes a blocking effect verification module, which is used to collect the actual propagation path of the actual fault along each directed edge and the actual impact intensity of each downstream node after the actual fault occurs in the target process equipment. The actual propagation path and actual impact intensity are compared with the predicted propagation path and expected impact intensity output by the propagation path prediction module. Based on the comparison results, model correction parameters are generated and the model correction parameters are fed back to the propagation path prediction module to update the prediction model.

7. The IoT-interoperable PC component digital production line process coordination system according to claim 6, characterized in that, When the blocking effect verification module determines that the target process equipment has a repetitive failure mode, the cross-production line process capacity coordination module is triggered. The cross-production line process capacity coordination module includes: The real-time capacity quantification submodule is used to obtain the real-time idle capacity of each process on each production line in multiple PC component production lines, and to standardize and quantify the real-time idle capacity of each process. The capacity storage submodule is used to summarize and store the real-time idle capacity information of each process on each production line according to the process type, forming a process capacity pool across production lines. The capacity scheduling decision submodule is used to retrieve the corresponding process of other production lines with idle capacity that is the same type as the target process from the process capacity pool when the real-time available capacity of the target process of the target production line is lower than the preset capacity threshold, and generate a cross-line capacity borrowing instruction. The cross-line mold table scheduling and execution submodule is used to convert the cross-line capacity borrowing instruction into a mold table transfer control signal, control the mold table traversing vehicle and the central mobile transport platform to transfer the mold table carrying the semi-finished components from the target production line to the corresponding process station of the production line whose capacity is borrowed, and transfer the mold table back to the target production line after the process is completed.

8. The IoT-interoperable PC component digital production line process coordination system according to claim 7, characterized in that, The cross-production line process capacity coordination module also includes: The capacity credit rating submodule is used to record the borrower production line identifier, lender production line identifier, borrowed capacity quantity, and borrowing start and end time for each cross-line capacity borrowing, and to calculate the capacity credit score of each production line based on the number of times capacity is borrowed and borrowed. The scheduling priority adjustment submodule is used to adjust the scheduling priority of each production line in the capacity scheduling decision submodule according to the capacity credit score, wherein the production line with the capacity credit score higher than the preset credit threshold is given priority matching right when requesting to borrow capacity.

9. The IoT-interoperable PC component digital production line process coordination system according to claim 8, characterized in that, It also includes a process time sensing and dynamic cycle adjustment module, which includes: The real-time process time recording submodule is used to record the actual time spent on each process after it is completed, and to calculate the deviation between the actual time spent and the standard time for that process. The cumulative waiting time prediction submodule is used to calculate the predicted cumulative waiting time value at each downstream process after the deviation value of the current process and the standard working hours and process dependencies of each downstream process. The dynamic beat adjustment submodule is used to generate a start timing adjustment instruction and a task diversion instruction for a downstream process before the actual occurrence of the cumulative waiting time when the predicted value of the cumulative waiting time of a certain downstream process exceeds the preset waiting threshold. The cycle instruction execution submodule is used to convert the start timing adjustment instruction and task diversion instruction into control signals for the corresponding process equipment and send them out for execution.

10. The IoT-interoperable PC component digital production line process coordination system according to claim 9, characterized in that, The process time sensing and dynamic rhythm adjustment module also includes: The prediction self-learning submodule is used to compare the predicted cumulative waiting time with the actual waiting time after the actual waiting time of each process occurs, calculate the prediction error, and adjust the prediction model parameters of the cumulative waiting time prediction submodule according to the prediction error.

Images