A flexible scheduling system for the whole process of manufacturing cable protection pipes

Abstract

This invention discloses a flexible scheduling system for the entire manufacturing process of cable protection pipes, relating to the field of industrial automation control. The system includes: a real-time data sensing system for the entire process, a digital twin mapping module, a multi-objective constraint flexible scheduling engine, a disturbance and anomaly real-time monitoring module, and a closed-loop feedback execution control interface. The sensing system captures real-time production element data; the twin module evolves synchronously in virtual space; the scheduling engine generates an optimal task schedule based on multi-objective constraints; the monitoring module captures anomalies in real time and triggers rescheduling decisions; and the control interface executes closed-loop feedback instructions. This invention aims to address the shortcomings of insufficient overall planning of the entire production process and the inability of static scheduling to cope with complex disturbances. It achieves full-process collaborative optimization, significantly enhancing the flexible response capability, resource allocation efficiency, and quality control level of the production system, and effectively reducing production energy consumption.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control, and in particular to a flexible scheduling system for the entire manufacturing process of cable protection pipes. Background Technology

[0002] With the rapid development of intelligent manufacturing and industrial internet technologies, the production model of the cable protection pipe manufacturing industry is evolving from the traditional process-driven model to digitalization and intelligence. As a core component of modern industrial systems, end-to-end manufacturing scheduling encompasses the collaborative management of all production elements from raw material entry to finished product warehousing, playing a crucial role in improving overall factory operational efficiency, reducing energy consumption, and ensuring product quality consistency.

[0003] Among them, the flexible scheduling system for the entire manufacturing process of cable protection pipes aims to achieve optimal matching between production tasks and underlying hardware resources by comprehensively coordinating multiple process links such as mixing, extrusion, shaping, cooling, cutting, and testing. The key technology in this field lies in using industrial sensing and communication networks to build a dynamic perception and collaborative control model covering the entire production chain, so as to respond in real time to various internal and external demand changes and environmental disturbances during the manufacturing process.

[0004] Current technologies for cable protection pipe production management still face significant challenges. Some technical solutions overemphasize localized optimization of single physical processes, such as improving mixing efficiency or reducing energy consumption, but fail to achieve digital modeling and integrated scheduling of all process nodes from batching and molding to warehousing, lacking overall coordination capabilities for the entire production process. Although some management systems have introduced upper-level management logic such as Enterprise Resource Planning (ERP), the lack of a closed-loop linkage mechanism with lower-level production equipment results in extremely weak capabilities for acquiring and feeding back real-time data on equipment status, work-in-process location, and process parameters. Scheduling decisions are often in a static or semi-dynamic state. This rigid scheduling model struggles to effectively cope with complex disturbances such as switching between multiple product types and small batches of orders, emergency order insertions, or occasional equipment failures, leading to a severe disconnect between production planning and execution, low resource utilization, and slow response times. It fails to build a flexible system that organically integrates perception, decision-making, and execution. Summary of the Invention

[0005] The purpose of this invention is to provide a flexible scheduling system for the entire manufacturing process of cable protection pipes, in order to solve the problems of insufficient overall planning of the entire production process of cable protection pipes in the existing technology, the inability of static scheduling to cope with complex production disturbances, and the data disconnect between the manufacturing execution and management levels.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A flexible scheduling system for the entire manufacturing process of cable protection pipes includes:

[0008] The system includes a full-process real-time data perception system, a digital twin mapping module, a multi-objective constraint flexible scheduling engine, a disturbance and anomaly real-time monitoring module, and a closed-loop feedback execution control interface.

[0009] A real-time data sensing system is deployed at each key physical node in the manufacturing of cable protection pipes to achieve comprehensive digital capture of production factors. This sensing system is integrated into the mixing, extrusion molding, shaping and cooling, length cutting, and finished product inspection processes. In the mixing process, high-precision weighing sensors and industrial communication gateways collect the instantaneous flow rate and cumulative feed amount of each component raw material. In the extrusion molding process, pressure sensors and thermocouples integrated into the extruder barrel acquire the melt pressure and temperature curves of each heating section in real time, while a power monitoring module collects the operating current and power consumption data of the main drive motor. In the shaping and cooling process, ultrasonic flow meters and temperature transmitters monitor the flow rate and temperature rise of the cooling circulating water. In the length cutting process, rotary encoders and photoelectric sensors precisely lock the displacement and real-time length of the pipe. In the finished product inspection process, industrial cameras and laser diameter gauges acquire images of pipe appearance defects and geometric dimensional deviations. All sensor data is aggregated to a data integration gateway via the industrial Ethernet protocol, where protocol conversion and timestamp alignment are performed to generate a structured, real-time dataset of the entire production process.

[0010] The digital twin mapping module is used to construct a digital model that corresponds one-to-one with the physical production line in a virtual space. Based on a hybrid architecture of mechanistic and data-driven models, this module establishes the dynamic logical relationships of the entire cable protection pipe production chain. Internally, it includes an equipment state sub-model, a process mechanism sub-model, and a work-in-process flow rotor model. The equipment state sub-model updates the health status and remaining service life predictions of key equipment such as extruders and cutting machines in real time based on equipment operating parameters obtained from the full-process real-time data sensing system. The process mechanism sub-model simulates the influence of process variables such as temperature, pressure, and speed on the physical properties of the pipe. The work-in-process flow rotor model uses RFID technology to identify the physical location of each batch of pipes on the production line. The digital twin mapping module receives the data stream output from the full-process real-time data sensing system in real time and uses this to drive the synchronous evolution of the virtual model, achieving high-fidelity representation and performance prediction of the physical production process in the digital space.

[0011] The multi-objective constrained flexible scheduling engine is the core decision-making hub of the system, used to seek the optimal balance between production efficiency, energy efficiency, and product quality while meeting hard constraints such as delivery time, equipment capacity, and process continuity. Internally, the engine constructs a full-process production scheduling model, transforming orders into discrete task sequences and dynamically matching them with resource status in the digital twin mapping module. During task allocation, the scheduling engine employs a logic combining heuristic search algorithms and deep reinforcement learning to perform load balancing calculations for multiple parallel extrusion units. Furthermore, the engine optimizes the production sequence based on the die-changing time and temperature adjustment delay of different specifications of cable protection pipes to reduce wasted time caused by process switching. The output of the scheduling engine is a task scheduling table covering all workstations throughout the entire process, clearly defining the start time, planned end time, material quota, and recommended process baseline parameters for each process.

[0012] The real-time disturbance anomaly monitoring module is used to capture various deviations and emergencies during production execution around the clock. This module compares the actual production data fed back by the end-to-end real-time data sensing system with the planned data output by the multi-objective constraint flexible scheduling engine in real time. When the system detects material shortages, unexpected equipment downtime, continuous product quality drift, or urgent order requests, the real-time disturbance anomaly monitoring module immediately triggers a rescheduling decision procedure. This procedure calculates the impact of the current disturbance on subsequent production plans and automatically determines whether to implement a local fine-tuning strategy or initiate a global rescheduling plan.

[0013] The closed-loop feedback execution control interface is used to issue scheduling commands generated by the multi-objective constraint flexible scheduling engine to the underlying control equipment. This interface directly intervenes in the feeding frequency of the mixer, the screw speed of the extruder, and the traction speed of the traction machine through a programmable logic controller (PLC) communication protocol. Simultaneously, the execution control interface synchronizes the latest scheduling commands to the workshop handheld terminal and the production site display panel, enabling flexible human-machine collaborative operations.

[0014] The end-to-end real-time data sensing system also includes environmental monitoring units. These units are distributed throughout the workshop and are used to collect atmospheric temperature, humidity, and airborne dust concentration. These environmental parameters are introduced as correction terms into the digital twin mapping module to compensate for the impact of environmental factors on the cooling rate and surface morphology of the cable protection pipes.

[0015] The digital twin mapping module also includes an energy efficiency assessment submodule. This submodule calculates the unit product energy efficiency index based on power data and real-time output of each process. By analyzing the correlation function between extrusion pressure, traction speed, and energy consumption, it provides the scheduling engine with optimization constraints aimed at low-carbon manufacturing.

[0016] The multi-objective constrained flexible scheduling engine employs a dynamic evaluation algorithm based on priority weights when handling urgent order insertions. This algorithm automatically reconstructs the task sequence based on customer credit rating, order profit margin, delivery urgency, and current production progress. Furthermore, the engine utilizes idle time window filling logic to embed small-batch orders into the production gaps of large orders, thereby improving equipment utilization.

[0017] The real-time disturbance anomaly monitoring module integrates a device fault early warning mechanism based on long short-term memory networks. This mechanism analyzes historical time-series characteristics such as vibration and current, and issues an early warning signal within a preset time before a substantial equipment failure occurs. Based on the early warning information, the real-time disturbance anomaly monitoring module proactively coordinates with the scheduling engine to adjust subsequent task allocation, thereby achieving a flexible scheduling shift from reactive response to proactive prevention.

[0018] In one embodiment of the present invention, the closed-loop feedback execution control interface has an adaptive adjustment function. When the melt pressure fluctuation in the extrusion process exceeds a preset threshold, the interface automatically adjusts the speed of the feeding motor to achieve closed-loop correction of process parameters. Simultaneously, the interface transmits execution feedback information back to the scheduling engine for online correction of prediction deviations in the digital twin model.

[0019] The system also includes a top-level collaborative command center. This center centrally displays real-time dynamics of the entire process, order completion rates, overall equipment efficiency, and key process indicators on a large screen. The center supports remote intervention, allowing process engineers to adjust the weighting preferences of scheduling strategies in real time through a virtual interactive interface.

[0020] In this invention, each step in the manufacturing process of cable protection pipes is abstracted as a logical unit with specific attributes and constraints. The mixing process is defined as a discrete proportioning unit, the extrusion molding process is defined as a continuous flow unit, and the shaping, cooling, and cutting processes are defined as time-coupled units. Through this unified logical modeling approach, the system can handle the alternating manufacturing processes of cable protection pipes of different specifications and materials on the same physical production line.

[0021] Furthermore, the multi-objective constraint flexible scheduling engine also considers the physical wear and tear of the molds when generating production plans. Whenever a specific type of pipe completes its production task, the system automatically increments the corresponding mold's lifespan count. When the count reaches a preset maintenance threshold, the scheduling engine will forcibly insert a mold repair and maintenance task into the task queue, thereby ensuring long-term production stability through a flexible scheduling mechanism.

[0022] As one embodiment of the present invention, the end-to-end real-time data sensing system also includes a raw material traceability unit. This unit uses barcode technology to register each batch of resin particles, additives, etc., and links the batch information with process data during production and the final quality inspection report throughout the entire lifecycle. This enables the scheduling system to accurately pinpoint whether quality fluctuations are caused by process variations or raw material defects.

[0023] As one embodiment of the present invention, the multi-objective constrained flexible scheduling engine also introduces a resource conflict resolution strategy based on game theory. When multiple extrusion production lines compete for the same automatic feeding system or shared cutting equipment, the engine dynamically allocates access permissions and execution time by calculating the priority payoff function of each production line, thereby avoiding the generation of production bottlenecks.

[0024] The real-time disturbance anomaly monitoring module and the digital twin mapping module share a high-speed data bus. This ensures that the anomaly characteristics acquired by the monitoring module can be fed back to the virtual model for evolutionary deduction within 100 milliseconds. Through this millisecond-level response mechanism, the system can calculate the post-reordered system state in advance before a chain reaction occurs in the physical system, thereby ensuring the stability of scheduling instructions.

[0025] The closed-loop feedback execution control interface is seamlessly integrated with the factory's production execution system (MES). The system automatically receives the monthly production outline from the MES and breaks it down into fine-grained execution plans at the daily or even hourly level. Simultaneously, the system reports the actual quantity of qualified products, the proportion of defective products, and the true loss of raw materials to the MES in real time, achieving complete transparency of information between the manufacturing site and enterprise management.

[0026] As one embodiment of the present invention, the system adopts a distributed architecture based on edge computing. The end-to-end real-time data perception system has multiple edge computing nodes located near the device to process high-frequency sampled data and execute basic anomaly identification logic. Only cleaned and compressed key state features are transmitted to the cloud-based multi-objective constraint flexible scheduling engine via wireless or wired networks. This architecture significantly reduces the computational burden on the core scheduling server and improves the system's real-time anti-interference capability.

[0027] Following the finished product inspection process, the system also includes an intelligent sorting unit. Based on the quality level feedback from the inspection process, this unit automatically controls a sorting robot to deliver the pipes to the qualified products warehouse, rework area, or waste disposal station. The sorting results are fed back to the scheduling engine in real time, serving as the basis for generating subsequent replenishment tasks.

[0028] As one embodiment of the present invention, the multi-objective constrained flexible scheduling engine also possesses self-learning and evolutionary capabilities. The system periodically extracts historical scheduling data, corresponding execution feedback, and final production efficiency indicators, and optimizes the heuristic rule weights in the production scheduling logic through machine learning algorithms. As the system's running time accumulates, the scheduling engine will become increasingly suited to the specific factory's process characteristics and production rhythm.

[0029] In this invention, through the collaborative work of the aforementioned modules, a flexible closed-loop system for cable protection pipe manufacturing with sensing, analysis, decision-making, and execution capabilities is constructed. This system breaks through the deadlock of isolated operation of each process step in the traditional manufacturing model, endowing the physical production line with intelligent attributes to autonomously respond to external changes through digital technology, and significantly improving the level of intelligence in cable protection pipe manufacturing.

[0030] Compared with the prior art, the beneficial technical effects of the present invention are as follows:

[0031] This invention achieves deep integration and digital transparency of the entire cable protection pipe manufacturing process. By constructing a real-time data perception system and a digital twin mapping module for the entire process, it completely breaks down the information silos between mixing, extrusion, molding, cutting and testing. The system can perceive the production status and process environment of each meter of pipe in real time, providing a precise and real-time data foundation for global optimization scheduling, and effectively solving the problem of poor overall process coordination caused by local optimization in the existing technology.

[0032] This invention greatly improves the flexibility and anti-interference capability of the production system. By introducing a multi-objective constraint flexible scheduling engine and a real-time disturbance anomaly monitoring module, it can provide the optimal rescheduling solution in a very short time for uncertainties in the manufacturing process, such as emergency order insertion, material fluctuations, and equipment failures. This dynamic and closed-loop scheduling mechanism changes the production line stagnation and plan disconnection phenomenon in the traditional static production scheduling mode, enabling the production system to efficiently cope with the modern market demand of multiple varieties and small batches.

[0033] This invention achieves closed-loop optimization control of production process, quality and efficiency. Through the closed-loop feedback execution control interface, it realizes the linkage between scheduling commands and underlying equipment parameters. The system can automatically correct core process parameters such as extrusion pressure and speed based on real-time detected quality deviations, which significantly reduces the product scrap rate. Through energy efficiency assessment and load balancing strategies, it effectively reduces production energy consumption while ensuring production capacity, thereby improving the company's overall competitiveness and green manufacturing level.

[0034] This invention constructs an intelligent decision-making system with self-evolution capabilities. By utilizing a distributed edge computing architecture and machine learning self-learning logic, the scheduling system has the ability to continuously evolve from historical production experience. The system can automatically correct equipment status prediction models and process mechanism models. As the operating cycle increases, its scheduling accuracy and prediction accuracy continue to improve, providing solid technical support for the cable protection pipe manufacturing industry to move towards smart factories. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the overall technical solution architecture of a flexible scheduling system for the entire manufacturing process of cable protection pipes proposed in this invention;

[0036] Figure 2 This is a schematic diagram of the core principle framework of the multi-objective constraint flexible scheduling engine in this invention. Detailed Implementation

[0037] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely intended to explain the present invention and not to limit the present invention. For those skilled in the art, the present invention can be practiced without some of these specific details. The following description of the embodiments is merely to provide a better understanding of the present invention by illustrating examples of the invention.

[0038] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0039] In the embodiments of the present invention, the same reference numerals denote the same components, and for the sake of brevity, detailed descriptions of the same components are omitted in different embodiments. It should be understood that the thickness, length, width, and other dimensions of various components in the embodiments of the present invention shown in the accompanying drawings, as well as the overall thickness, length, width, and other dimensions of the integrated device, are merely illustrative and should not constitute any limitation on the present invention; the term "multiple" in the present invention refers to two or more (including two).

[0040] Example 1

[0041] Please refer to the attached document. Figure 1 This embodiment discloses a flexible scheduling system for the entire manufacturing process of cable protection pipes, built on an industrial internet architecture. It aims to address core pain points in cable protection pipe production, such as isolated processes, rigid scheduling, and data distortion, through comprehensive digitalization. The system consists of a real-time data perception system, a digital twin mapping module, a multi-objective constraint flexible scheduling engine, a real-time disturbance and anomaly monitoring module, a closed-loop feedback execution control interface, a collaborative command center, a raw material traceability unit, and an intelligent sorting execution unit.

[0042] The end-to-end real-time data sensing system, acting as the tentacles of the entire system, is deeply embedded in every physical node of cable protection pipe manufacturing. In the mixing process, this sensing system is equipped with multiple sets of high-precision weighing sensors. These sensors possess a linear error accuracy of 0.01% and can capture instantaneous mass changes in PVC resin, stabilizers, lubricants, and fillers in real time. Data is reported to the central database 10 times per second via an industrial Ethernet gateway, ensuring the accuracy of raw material proportions. In the extrusion molding process, the end-to-end real-time data sensing system integrates 5 to 8 sets of thermocouple temperature sensors and melt pressure transmitters distributed throughout the extruder barrel. These sensors can monitor temperature fluctuations of 0.1 degrees Celsius within the 200-250 degree Celsius range, as well as dynamic melt pressures up to 40 MPa. Simultaneously, through the power monitoring module, the system acquires the current waveform and active power of the main drive motor at a sampling rate of 1000 Hz, providing fundamental data for subsequent energy efficiency analysis.

[0043] Please refer to the appendix during the shaping and cooling process. Figure 1The entire process utilizes a real-time data sensing system to monitor the circulating water flow rate in the cooling water tank in real time using an ultrasonic flow meter, covering a flow range of 10 to 50 cubic meters per hour. This is combined with a temperature transmitter to monitor the temperature difference between the inlet and outlet water, ensuring that the heat exchange efficiency of the pipes during the shaping process meets preset standards. The fixed-length cutting process relies on an incremental rotary encoder installed at the output end of the traction machine. This encoder outputs 5000 pulses per revolution, combined with a photoelectric sensor to achieve millimeter-level locking of the pipe length. The finished product inspection process consists of an industrial camera and a dual-axis laser diameter gauge. The industrial camera uses a 40-megapixel high-resolution sensor and employs machine vision algorithms to identify scratches, pits, and color differences on the pipe surface. The laser diameter gauge acquires the outer diameter and wall thickness data of the pipe at a scanning speed of 500 scans per second, with all measurement accuracy maintained within 0.02 millimeters.

[0044] The digital twin mapping module receives structured, real-time production datasets from the end-to-end real-time data sensing system and constructs a high-fidelity model in virtual space. The integrated equipment state sub-model uses a Markov chain algorithm to predict the wear state of the extruder screw, while its process mechanism sub-model simulates the shear rate and pressure distribution of the melt within the die flow channel using finite element analysis. The in-process flow rotor model uses RFID readers placed at key workstations in the workshop to read RFID tag information affixed to the pipe supports in real time, ensuring that the physical position of each pipe segment is updated synchronously in the digital space within seconds. The digital twin mapping module not only visualizes the production process but also constructs a correlation model between ambient temperature, humidity, and product shrinkage rate through deep learning of historical data.

[0045] The multi-objective constrained flexible scheduling engine serves as the core of the system's decision-making process, combined with... Figure 2 As shown, its operating logic is based on an improved multi-objective optimization algorithm. When processing order scheduling, the engine sets delivery achievement rate, overall equipment efficiency, unit product energy consumption, and production changeover cost as core optimization objectives. When generating the initial scheduling sequence, the engine first calls the real-time resource profile provided by the digital twin mapping module to determine the available capacity of four or more parallel production lines. The scheduling engine automatically combines and optimizes large-volume standard-specification orders with small-volume irregular-specification orders by calculating the urgency factor of each pending order.

[0046] To achieve precision in the production scheduling process, the multi-objective constrained flexible scheduling engine introduces the following scheduling evaluation function:

[0047]

[0048] In the above formula, Represents the overall scheduling cost evaluation index. , , These are the weights for delivery delay, energy consumption, and process changeover costs, respectively. Let i be the actual completion time of the i-th order. Let the contractually agreed delivery time be for the i-th order. Let be the predicted total power consumption of the j-th extruder unit during the production cycle. This represents the wasted time cost incurred by the k-th mold change or significant adjustment of process parameters. Through heuristic search within the solution space, the scheduling engine can output the optimal production plan for the next 72 hours within 120 seconds.

[0049] The real-time disturbance monitoring module monitors any minute deviations on the production line around the clock. This module integrates a fault warning mechanism based on a long short-term memory network. Through time-series analysis of extruder vibration frequency and motor current characteristics, it can issue a warning two hours before bearing damage or heating coil burnout. When the actual extrusion speed is detected to be lower than 95% of the planned value, or the scrap rate reported by the finished product inspection process exceeds 2% for three consecutive minutes, the real-time disturbance monitoring module immediately activates the rescheduling procedure. This procedure determines the response level based on the severity of the disturbance. If it is a localized, small-scale fluctuation, online compensation is performed by fine-tuning the traction speed and feeding frequency; if it is a major equipment failure or an urgent order, a global rescheduling is triggered, recalculating the priority and resource allocation paths of all online tasks.

[0050] The closed-loop feedback execution control interface acts as a bridge between the decision-making and execution layers. It communicates with the underlying programmable logic controller (PLC) via standard industrial protocols, directly issuing setpoints for speed, temperature, and pressure. For example, when the multi-objective constraint flexible scheduling engine detects fluctuations in the melt flow index of the current raw material batch, the closed-loop feedback execution control interface automatically increases the setpoint temperature of the extruder's third zone by 2 degrees Celsius and simultaneously increases the screw speed by 0.5 Hz to maintain a constant melt output. This adjustment logic is adaptive, automatically correcting PID control parameters based on feedback deviations to ensure the physical system's execution accuracy remains within the optimal range.

[0051] The collaborative command center is located at the top level of the system's architecture. Its core is a large-screen visualization system integrating big data analytics and virtual simulation. The center uses 3D modeling technology to recreate the real-world logic of the entire manufacturing workshop, allowing managers to view the real-time load rates of each process—mixing, extrusion, cooling, and cutting—on the screen. The collaborative command center also supports remote distribution and auditing of process parameters. Any adjustment to the scheduling strategy weights requires dual authentication from the center to ensure the authority and security of production instructions.

[0052] The raw material traceability unit achieves a closed-loop data chain covering the entire lifecycle from particle warehousing to pipe product delivery. Using barcode technology, this unit records the supplier, batch number, and warehousing test data for each bag of resin and each barrel of stabilizer. During manufacturing, the traceability unit strongly correlates raw material batches with the extrusion process curves and environmental monitoring data at that time. If the finished product inspection process finds that the ring stiffness of a certain batch of pipes does not meet the standard, the traceability unit can quickly trace back to the specific raw material component or a specific production period to pinpoint the root cause of the failure.

[0053] The intelligent sorting execution unit, deployed after the finished product inspection process, consists of a high-speed pneumatic sorting mechanism and a six-axis industrial robot. This unit classifies and processes the pipes based on the quality grade signals transmitted in real time from the inspection process. Qualified products are automatically guided to the packaging area by a conveyor chain, while defective products with outdated dimensions or surface defects are precisely sorted into the rework area or the crushing and reuse area. The sorting results are fed back to the scheduling engine in real time via a high-speed bus. If the number of qualified products produced does not meet the order requirements, the system will automatically add supplementary production tasks to the subsequent production plan, thereby ensuring the integrity of order delivery.

[0054] In this embodiment, the full-process real-time data sensing system is additionally equipped with an environmental monitoring unit to monitor subtle environmental changes within the workshop. Because the physical properties of the cable protection pipe material are highly sensitive to ambient temperature, when the workshop temperature rises above 35 degrees Celsius in summer, the environmental monitoring unit feeds this information back to the digital twin mapping module. The module automatically corrects the target setpoint of the cooling water system, compensating for the impact of ambient heat by increasing the circulating water volume. This refined compensation mechanism enables the system to produce cable protection pipes of uniform quality in different seasons and geographical environments.

[0055] During system operation, the digital twin mapping module also includes an energy efficiency assessment submodule. This submodule not only calculates electrical energy consumption but also comprehensively accounts for compressed air consumption and cooling water loss. By analyzing the relationship between extrusion pressure and unit output energy consumption, the submodule suggests that the scheduling engine prioritize the extruder unit with the best current energy efficiency to perform high-load tasks. When handling alternating production of multiple product types, the scheduling engine utilizes idle time window filling logic to cleverly insert small-diameter, easily processed pipe orders into the preparation gaps of large-specification, high-difficulty orders. This flexible strategy improves the overall utilization rate of the equipment by more than 15%.

[0056] The multi-objective constraint flexible scheduling engine also demonstrates a high level of intelligence in mold management. The system records the cumulative online running time and total amount of material processed for each mold. When the processing mileage of a mold reaches the preset 50,000-meter maintenance threshold, the scheduling engine automatically locks the mold in the task flow and sends a maintenance work order to the collaborative command center. During maintenance, the engine automatically finds alternative molds with similar processing capabilities or adjusts the production sequence to ensure that the production process does not experience prolonged shutdowns due to mold maintenance.

[0057] To ensure real-time data transmission, the system employs a distributed architecture based on edge computing. Edge computing nodes are deployed alongside each extruder unit, handling over 90% of the high-frequency data cleaning and feature extraction tasks. For example, to address millisecond-level fluctuations in extruder current, the edge computing nodes first perform a Fast Fourier Transform to extract characteristic frequencies before sending the compressed status packet to the cloud server. This architecture effectively reduces the computational load on the core scheduling engine and keeps the system's response latency to sudden anomalies within 50 milliseconds.

[0058] In real-world production scenarios, when an urgent order is inserted, the multi-objective constrained flexible scheduling engine activates a resource conflict resolution strategy based on game theory. The system automatically assesses the impact of the inserted order on the delivery time of existing orders and dynamically adjusts the priority weights of each production line to achieve a smooth insertion of the inserted order. For example, if a power engineering project urgently needs a batch of flame-retardant cable protection pipes of special specifications, the scheduling engine will immediately analyze the load of the current four production lines, identify one of the production lines that is producing non-urgent stock orders, and automatically insert an urgent order at the minimum process switch point after the current task is completed. At the same time, it will automatically allocate inventory from the raw material traceability unit, achieving a minute-level response from order receipt to production scheduling.

[0059] The flexible scheduling system for the entire cable protection pipe manufacturing process in this embodiment constructs a highly autonomous manufacturing closed loop through close collaboration of perception, mapping, decision-making, monitoring, and execution. It no longer relies on manual experience for production scheduling, but instead makes optimal decisions based on objective reflections of physical laws and real-time data. As the running time increases, the multi-objective constraint flexible scheduling engine continuously absorbs experience from handling historical anomalies through its self-learning and evolutionary capabilities, resulting in a sustained increase in the accuracy of its production scheduling plan and the robustness of the production system.

[0060] Example 2

[0061] This embodiment focuses on describing the system collaboration details in the parallel production of high-strength polyethylene cable protection pipes by multiple units, especially the deep scheduling strategy of the multi-objective constrained flexible scheduling engine when facing complex raw material fluctuations.

[0062] In Example 2, the production line consists of six high-power single-screw extruders, each equipped with an independent real-time data sensing system. The raw material traceability unit within this sensing system plays a more crucial role in this example, using a near-infrared spectroscopy analyzer to perform real-time component detection on the polyethylene particles at the feed inlet. Because different batches of resin exhibit slight differences in density and melt flow index, the raw material traceability unit transmits the detected rheological parameters to the digital twin mapping module in real time.

[0063] After receiving these raw material parameters, the digital twin mapping module immediately performs dynamic simulation calculations within its internal process mechanism sub-model to predict the pressure change trend at the extrusion die under the current raw material conditions. If the prediction indicates that the melt pressure may exceed the safety threshold of 35 MPa, the multi-objective constraint flexible scheduling engine will pre-adjust the extrusion screw speed through a closed-loop feedback execution control interface and simultaneously adjust the linear speed of the traction machine to ensure the uniformity of the pipe wall thickness.

[0064] In this embodiment, the multi-objective constrained flexible scheduling engine employs a reinforcement learning-based search mechanism to optimize load allocation across multiple generating units. The scheduling engine constructs a deep neural network model whose input vector includes the current operating status of each generating unit, the material requirements of pending orders, current energy efficiency indicators, and environmental parameters.

[0065]

[0066] In the reinforcement learning logic described above, Q(s,a) represents the expected reward of taking scheduling action a in state s, R is the immediate feedback, which includes the output achievement rate and energy efficiency score, and s is the current state. For the actions currently being taken, For the next state, For possible actions in the next state, This serves as a discount factor. Through continuous iteration, the scheduling engine learns to adopt the optimal dispatch strategy at different production stages. For example, when Unit 1 experiences a slight vibration anomaly but does not trigger the on-site shutdown threshold, the engine identifies its potential risk through the real-time disturbance anomaly monitoring module and automatically reduces the high-intensity task allocation for that unit, switching it to a low-pressure, low-speed production mode, thereby extending the effective operating time of the equipment.

[0067] In this embodiment, the real-time disturbance anomaly monitoring module integrates a multi-sensor fusion algorithm. By jointly analyzing extruder torque fluctuations, heating current imbalances, and cooling water pressure drops, it achieves precise classification of production anomalies. The module's response matrix categorizes anomalies into three levels: Level 3 anomalies only record data and trigger a yellow flashing alert in the collaborative command center; Level 2 anomalies trigger local scheduling corrections, with the closed-loop feedback execution control interface adjusting the parameters of relevant actuators; Level 1 anomalies, such as main motor overload or severe pipe cracking, immediately shut down the machine and trigger a global emergency production scheduling plan initiated by the multi-objective constraint flexible scheduling engine.

[0068] In this embodiment, the closed-loop feedback execution control interface demonstrates extremely high real-time collaborative capabilities. When the multi-objective constraint flexible scheduling engine decides to switch production to Unit 3, the interface automatically controls the automatic feeding system of the mixing process to change the formula, and accurately calculates the critical point for the alternation of new and old materials based on the cleaning time calculated by the digital twin mapping module. When the new material reaches the die head position, the execution control interface automatically switches the fixed-length parameters of the cutting process and marks the transition material as awaiting recycling, which is then automatically rejected by the intelligent sorting execution unit.

[0069] In this embodiment, the collaborative command center provides more in-depth decision support functions. The large screen in the command center not only displays the current real-time output but also presents a comprehensive indicator called the production stability index. This index is calculated by the real-time disturbance anomaly monitoring module based on the variance of fluctuations from sensors across the entire production line, and can intuitively reflect the current operational quality of the flexible scheduling system. Managers can adjust the search depth of the multi-objective constraint flexible scheduling engine in real time through the command center's virtual interactive interface, finding the optimal balance between pursuing maximum efficiency and ensuring system robustness.

[0070] In Example 2, the raw material traceability unit further expands its functionality by deeply integrating with the factory's warehouse management system, enabling dynamic prediction of material consumption. Once the scheduling engine generates a one-week production schedule, the raw material traceability unit automatically calculates the total amount of various additives and resins required and checks the real-time inventory in the current warehouse. If it finds that the inventory of a specific pigment cannot support the planned production tasks, the system will issue a replenishment warning to the purchasing personnel 48 hours in advance and suggest that the scheduling engine adjust the production schedule for that specification of pipe.

[0071] In Example 2, the intelligent sorting execution unit adds an online weighing compensation function. Using the weight data per meter of pipe fed back by the sensing system, the sorting unit can determine in real time whether there are any internal cavities invisible to the naked eye. For pipes with a weight deviation exceeding 2%, the robotic arm will sort them to the degraded product area, and the system will automatically record the process coordinates of the defect, allowing the digital twin mapping module to perform logical backtracking and parameter correction.

[0072] During the operational cycle of this embodiment, the system can simultaneously handle more than 20 different specifications of cable protection pipes. Because the multi-objective constraint flexible scheduling engine incorporates a mold thermal expansion compensation algorithm, during the preheating stage after mold replacement, the system can precisely control the temperature rise curve of the mold temperature controller through a closed-loop feedback execution control interface, reducing the warm-up time by more than 30%. This extreme control over production details demonstrates the significant advantages of the flexible scheduling system in improving manufacturing flexibility.

[0073] Example 3

[0074] This embodiment explores the system performance under extreme production loads and highly uncertain environments, particularly the response mechanism of the flexible scheduling system for the entire cable protection pipe manufacturing process when the workshop faces frequent power peak-shaving demands and sudden order changes.

[0075] In Example 3, the cable protection pipe manufacturing system needs to be connected to the factory's smart grid management terminal. The full-process real-time data sensing system increases the ability to sense external electricity price signals. When the power peak-shaving signal indicates that the electricity price will be in the peak period within the next 4 hours, the multi-objective constraint flexible scheduling engine will automatically retrieve the current production plan, identify the high-energy-consuming extrusion process, and, under the premise of meeting the delivery constraints, attempt to migrate some high-load tasks to the off-peak electricity price period for execution.

[0076] At this point, the digital twin mapping module plays a crucial simulation role. It simulates the balance between the thermal energy loss after task migration and the benefits from the electricity price difference. If the model shows that reducing the extrusion speed and maintaining a warm state is more economical than a complete shutdown and restart, the multi-objective constraint flexible scheduling engine will issue an instruction to switch the unit to energy-saving standby mode through the closed-loop feedback execution control interface. In this mode, the extruder maintains an extremely low speed to prevent melt degradation in the barrel, while precisely maintaining the temperature of the heating zone to ensure a rapid return to full-load production after the peak electricity price period.

[0077] The real-time disturbance anomaly monitoring module faced significant challenges in this embodiment. Due to frequent adjustments in production load, traditional threshold-based alarm methods are prone to false alarms. Therefore, this module employs an anomaly detection algorithm based on an autoencoder. By learning the characteristic distribution under normal production load fluctuations, it constructs a dynamic monitoring envelope. The system only identifies an anomaly when actual sensor data deviates from the envelope. This method enables the system to maintain an anomaly identification accuracy rate of over 98% even during complex peak-shaving production processes.

[0078] In handling the emergency order insertion in Example 3, the multi-objective constrained flexible scheduling engine introduced virtual production scheduling simulation technology. When the collaborative command center receives an urgent order for nuclear power engineering cable protection pipes, the scheduling engine does not immediately modify the current physical production instructions. Instead, it creates a virtual copy in the digital twin mapping module and performs a simulation run that is 50 times faster. By simulating the impact of different order insertion times on the total 48-hour production capacity, the engine selects the entry point that minimizes the impact on existing orders and carries the lowest quality risk.

[0079] At the closed-loop feedback execution control interface level, this embodiment adds wireless sensor network access functionality to address the data transmission needs of some mobile production modules. For example, in the finished product yard, the closed-loop feedback execution control interface can acquire deformation monitoring data of pipes during storage through low-power wide-area network technology. If the sensing system detects that a batch of pipes has undergone geometric deformation due to excessive stacking height, the system will activate the intelligent sorting execution unit to re-inspect the batch, ensuring the pass rate of outgoing products.

[0080] In this embodiment, the collaborative command center integrates augmented reality collaboration functionality. When the real-time disturbance anomaly monitoring module reports an electrical fault in the extruder's main drive, the center automatically retrieves the digital twin model of the equipment and sends the fault location information to the maintenance engineer's portable terminal. The engineer can then scan a QR code on the equipment to view an internal structural diagram and maintenance instructions superimposed on the physical entity on a screen. Simultaneously, the multi-objective constraint flexible scheduling engine automatically redirects the tasks of the faulty unit to the standby unit, enabling collaborative maintenance and production.

[0081] In Example 3, the raw material traceability unit achieved data interoperability with upstream suppliers. When a batch of raw materials is consumed too quickly in the system, the traceability unit automatically retrieves the supplier's logistics information and displays the real-time transit status of the raw materials on the large screen of the collaborative command center. This cross-enterprise collaborative capability enables the flexible scheduling system to anticipate potential material shortage disturbances earlier and adjust production scheduling strategies 12 hours in advance.

[0082] When dealing with extreme loads, the intelligent sorting execution unit employs dynamic priority logic. When the output rate of qualified products exceeds the processing capacity of subsequent packaging processes, the sorting unit automatically redirects some non-urgent orders to a temporary buffer area based on their delivery urgency, prioritizing the smooth flow of high-priority orders. The sorting unit's execution data is synchronously fed back to the scheduling engine for online correction of material flow rate parameters in the digital twin model.

[0083] In this embodiment, the flexible scheduling system for the entire cable protection pipe manufacturing process demonstrates its autonomous decision-making and resilience in complex and dynamic environments. Through the refined calculations of the multi-objective constrained flexible scheduling engine, the factory maintained an on-time delivery rate of over 95% while coping with power peak shaving, and the average energy consumption per ton of product was reduced by more than 12%.

[0084] Example 4

[0085] This embodiment further refines the manufacturing process details and scheduling logic of the system when processing cable protection pipes made of special materials, such as high-performance flame-retardant reinforced pipes. These types of pipes, due to the addition of a high proportion of flame retardants and reinforcing fibers, are extremely sensitive to processing temperature and shear rate, and are highly prone to process instability.

[0086] The end-to-end real-time data sensing system enhances the monitoring of extruder torque signals for this type of material. Using a high-sampling-rate current transmitter, the system can capture torque pulsations on the order of 10 milliseconds caused by uneven material mixing. These minute pulsations are transmitted in real-time to the digital twin mapping module, serving as a basis for correcting the material viscosity parameters in the process mechanism model.

[0087] The digital twin mapping module has constructed a specialized material degradation model for such high-performance materials. The model calculates the residence time distribution of the material in the barrel in real time and, combined with the real-time measured melt temperature, assesses its thermal degradation risk. If the calculation shows that the residence time is too long, resulting in a material degradation probability exceeding 5%, the multi-objective constraint flexible scheduling engine will forcibly increase the minimum extrusion speed, even if this means that subsequent cutting and packaging processes will have to withstand higher workloads.

[0088] When scheduling such challenging orders, the multi-objective constrained flexible scheduling engine reserves a dedicated process stability window. Before production begins, the engine automatically issues instructions, requiring the closed-loop feedback execution control interface to execute a preheating and material feeding procedure for a specific period. During production, to prevent fiber accumulation at the die head, which can lead to surface roughness, the scheduling engine periodically induces minute fluctuations in extrusion pressure. These pressure pulses achieve self-cleaning of the die head, a process entirely controlled automatically by the system without manual intervention.

[0089] In Example 4, the real-time disturbance anomaly monitoring module employs higher-dimensional feature analysis. It monitors not only the values ​​of individual sensors but also the correlations between them. For example, when the melt pressure increases but the motor torque remains constant, the module determines it as sensor failure rather than a genuine process anomaly, thus avoiding invalid rescheduling triggers. This intelligent judgment logic significantly improves the system's anti-interference resilience.

[0090] In the shaping process of this type of pipe, a multi-level closed-loop control strategy is employed in the closed-loop feedback execution control interface. By controlling the vacuum level within the vacuum shaping chamber and the distribution of the spray cooling water, the interface can accurately compensate for the non-uniform shrinkage of the pipe during the cooling process. The diameter measurement data fed back by the sensing system is directly used as the feedback signal for vacuum level adjustment, forming a closed loop.

[0091] In Example 4, the collaborative command center added a subsystem called the Process Knowledge Base. This system automatically records successful parameter configurations and troubleshooting experiences from each high-performance pipe production run. When receiving an order for similar materials, the multi-objective constraint flexible scheduling engine prioritizes retrieving the optimal parameters from the knowledge base as the initial search point, significantly shortening the debugging cycle of the new process.

[0092] The raw material traceability unit focuses on strengthening the audit of additive ratios. By integrating the data from each feeding scale in the mixing process, the traceability unit can accurately calculate the true flame retardant content in each meter of pipe and include this data as part of the electronic quality certificate. This deep data linkage provides an unalterable basis for the quality assurance of high-end cable protection pipes.

[0093] In Example 4, the intelligent sorting unit is equipped with an ultrasonic flaw detector to detect voids or foreign objects within the pipe wall. For such high-performance pipes, even minor internal defects can lead to failure under high pressure. Based on the flaw detection results, the sorting unit precisely rejects pipes with potential risks.

[0094] Through the synergy of the aforementioned modules, this system not only handles flexible manufacturing of conventional products but also possesses the professional capability to manage complex processes and ensure the quality of high-end products. This end-to-end, all-element digital collaborative scheduling represents a significant technological advancement in the cable protection pipe manufacturing industry, moving towards high-end and intelligent manufacturing.

[0095] In summary, the flexible scheduling system for the entire cable protection pipe manufacturing process proposed in this invention achieves a self-perceiving, autonomous decision-making, and automatic execution intelligent manufacturing closed loop through the organic integration of core modules such as a full-process real-time data perception system, a digital twin mapping module, a multi-objective constraint flexible scheduling engine, a disturbance anomaly real-time monitoring module, and a closed-loop feedback execution control interface. This system significantly improves the flexibility of cable protection pipe production, reduces material and energy losses during production, and enhances product consistency and on-time delivery rates. Whether facing daily multi-variety, small-batch order switching or sudden equipment failures and external environmental changes, the system demonstrates excellent resilience and efficiency. By introducing cutting-edge technologies such as machine learning, deep reinforcement learning, and edge computing, the system possesses continuous evolution capabilities, constantly optimizing its scheduling strategy as factory operation data accumulates, thus providing cable protection pipe manufacturers with a solid technological barrier and management support in fierce market competition. The application of this system is not limited to a single cable protection pipe production line; its layered and decoupled architecture design allows for easy expansion to other similar plastic pipe extrusion manufacturing fields, demonstrating extremely high industrial application potential and socio-economic value.

[0096] Throughout the description, all physical parameters, control frequencies, and calculation accuracy involved have been verified through engineering practice, and the interaction logic between modules conforms to standard protocols for industrial control and manufacturing execution systems. This invention, by constructing a seamless mapping between the physical and digital worlds, connects previously fragmented process islands into an organic whole, greatly promoting the digital transformation of the cable protection pipe manufacturing industry. In future practical applications, this system will further integrate more advanced cognitive computing and collaborative optimization algorithms to address the needs of more complex and extreme manufacturing scenarios.

[0097] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape, and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A flexible scheduling system for the entire manufacturing process of cable protection pipes, characterized in that, include: The full-process real-time data perception system is deployed at physical nodes in the manufacturing of cable protection pipes, including mixing, extrusion molding, shaping and cooling, fixed-length cutting, and finished product inspection. It collects production element data and aggregates the data to the data integration gateway via the industrial Ethernet protocol for protocol conversion and timestamp alignment to generate a structured full-process real-time production dataset. The digital twin mapping module is used to build digital models based on a hybrid architecture of mechanistic models and data-driven models; The multi-objective constraint flexible scheduling engine, as the decision-making center of the system, is used to construct a full-process production scheduling model and transform orders to be produced into discrete task sequences under the premise of meeting the constraints of delivery time, equipment capacity and process continuity. By combining the resource status in the digital twin mapping module, heuristic search algorithm and deep reinforcement learning algorithm are used to perform load balancing calculation and production sequence optimization, so as to output a task scheduling table covering all workstations in the entire process. The real-time disturbance anomaly monitoring module is used to capture deviations and emergencies during the production execution process around the clock. By comparing the actual production data fed back from the real-time dataset of the entire production process with the planned data output from the task scheduling table, a rescheduling judgment program is triggered when an anomaly is detected to automatically determine whether to execute a local fine-tuning strategy or initiate a global rescheduling plan. The closed-loop feedback execution control interface is used to issue the scheduling instructions to the underlying control equipment to intervene in production parameters through the programmable logic controller communication protocol, and synchronize the latest scheduling instructions to the production site display board. At the same time, the execution feedback information is passed back to the multi-objective constraint flexible scheduling engine to correct the prediction deviation in the digital twin mapping module online.

2. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, The full-process real-time data sensing system also includes an environmental monitoring unit; the environmental monitoring unit is distributed inside the production workshop and is used to collect atmospheric temperature, humidity and air dust concentration; the production factor data is collected through high-precision weighing sensors, pressure sensors, thermocouples, power monitoring modules, ultrasonic flow meters, rotary encoders and laser diameter measuring instruments; the digital twin mapping module receives the atmospheric temperature, humidity and air dust concentration as correction terms to compensate for the influence of environmental factors on the cooling rate and surface morphology of the cable protection pipe.

3. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, The digital model internally includes an equipment state sub-model, a process mechanism sub-model, and a work-in-process flow rotor model. By receiving real-time datasets from the entire production process, the digital model evolves synchronously, achieving a high-fidelity representation and performance prediction of the physical production process in virtual space. The digital twin mapping module also includes an energy efficiency assessment sub-module. Based on the power data and real-time output of each process, the energy efficiency assessment sub-module calculates the unit product energy efficiency index. The multi-objective constraint flexible scheduling engine receives optimization constraints provided by the energy efficiency assessment sub-module by analyzing the correlation function between extrusion pressure, traction speed, and energy consumption.

4. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, When handling urgent order insertions, the multi-objective constrained flexible scheduling engine performs the following steps: it adopts a dynamic evaluation algorithm based on priority weights to reconstruct the task sequence according to customer credit rating, order profit margin, delivery urgency, and current production progress; and it uses idle time window filling logic to embed small-batch orders into the production gaps of large orders to improve equipment utilization. Based on the mold change time and temperature adjustment delay of different specifications of cable protection pipes, the production sequence is optimized to reduce the invalid working time caused by process switching.

5. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, The task scheduling table clearly defines the start time, planned end time, material quota, and process baseline parameters for each process; the multi-objective constraint flexible scheduling engine introduces a scheduling evaluation function for calculation when generating the task scheduling table; The scheduling evaluation function includes the total scheduling cost evaluation index, delivery delay weight, energy consumption weight, process switching cost weight, actual order completion time, order contractual delivery time, predicted total power consumption of the extruder unit during the production cycle, and invalid working time cost generated by process parameter adjustment; the multi-objective constrained flexible scheduling engine performs heuristic search in the solution space to minimize the total scheduling cost evaluation index, so as to output the optimal production schedule.

6. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, The real-time disturbance anomaly monitoring module integrates an equipment fault early warning mechanism based on a long short-term memory network. The equipment fault early warning mechanism analyzes the vibration and current historical time sequence characteristics of the extruder unit and issues an early warning signal within a preset time before a substantial equipment failure occurs. The real-time disturbance anomaly monitoring module, based on the early warning signal, links with the multi-objective constraint flexible scheduling engine to adjust the subsequent task allocation.

7. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, The closed-loop feedback execution control interface has an adaptive adjustment function; when the melt pressure fluctuation in the extrusion process exceeds the preset threshold, the closed-loop feedback execution control interface automatically adjusts the speed of the feeding motor to achieve closed-loop correction of process parameters; the closed-loop feedback execution control interface is integrated with the production execution system, receives the monthly production outline and decomposes the outline into fine-grained execution plans, and reports the number of qualified products, the proportion of waste products and the actual loss of raw materials to the production execution system.

8. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, It also includes a collaborative command center; the collaborative command center displays the real-time dynamics of the entire process, order completion rate, overall equipment efficiency and key process indicators on a large screen; the collaborative command center receives the adjustment instructions for scheduling strategy weights input by process engineers through a virtual interactive interface and issues them to the multi-objective constraint flexible scheduling engine.

9. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, It also includes a raw material traceability unit and an intelligent sorting execution unit; the raw material traceability unit uses barcode technology to register each batch of raw materials and associates the batch information with the process data and quality inspection reports during the production process; the intelligent sorting execution unit is deployed after the finished product inspection process, and controls the sorting robot to send the pipes to different areas according to the quality level signal, and feeds back the sorting results to the multi-objective constraint flexible scheduling engine as the basis for generating replenishment tasks.

10. The flexible scheduling system for the entire manufacturing process of cable protection pipes according to claim 1, characterized in that, The system adopts a distributed architecture based on edge computing; the full-process real-time data perception system has multiple edge computing nodes set up near the device to process high-frequency sampled data and execute anomaly identification logic. The edge computing nodes transmit the cleaned and compressed state features to the multi-objective constraint flexible scheduling engine through the communication network, so as to reduce the computing pressure on the core scheduling server and improve the real-time anti-interference capability.

Images