CN-121989421-A - Pipe molding system based on real-time defect feedback

CN121989421ACN 121989421 ACN121989421 ACN 121989421ACN-121989421-A

Abstract

The invention belongs to the technical field of pipe manufacture and intelligent control, and solves the problems of longitudinal cracks, scratches, pits, uneven size and overlarge residual stress on the surface caused by thermodynamic coupling unbalance in the existing pipe extrusion molding. The invention discloses a pipe molding system based on real-time defect feedback, which comprises an execution device, a central control device, a multi-mode sensing device and a three-section temperature control device, wherein a three-section temperature control structure of a primary cooling section, a constant temperature section and a final cooling section is adopted, the multi-mode sensing device is distributed at the outlet of each section to collect surface images, wall thickness and temperature data in real time, a defect characteristic map is constructed, a thermal coupling regulation and control instruction is output by matching a process knowledge base, stress relaxation and shaping control is realized through constant temperature section partition heating and final cooling section gradient cooling, and meanwhile, closed loop feedback compensation is combined to dynamically correct process parameters. The invention can effectively inhibit the surface defect of the pipe, reduce the residual stress, improve the uniformity of the wall thickness and the mechanical property, and improve the production stability and the product qualification rate.

Inventors

LI CHENG
WANG ZEMING
ZHAO CHENG
WU JIAWEI
CHEN CHEN

Assignees

公元管道(天津)有限公司

Dates

Publication Date: 20260508
Application Date: 20260409

Claims (9)

1. A pipe molding system based on real-time defect feedback comprises an execution device, a central control device and a multi-mode sensing device, and is characterized in that the temperature control device comprises a constant temperature section arranged between a primary cooling section and a final cooling section, and the multi-mode sensing device is circumferentially arranged at the outlet of the primary cooling section, the outlet of the constant temperature section and the outlet of the final cooling section; The central control device is configured with central control instructions, comprising: s1, acquiring multi-element data of a pipe in real time through a multi-mode sensing device to generate acquisition information; s2, constructing a pipe defect characteristic map according to the acquired information; s3, according to the defect characteristic map, a thermodynamic coupling regulation instruction set is called from a pre-constructed process knowledge base; s4, executing the thermodynamic coupling regulation instruction set; S5, configuring a feedback compensation strategy, wherein the feedback compensation strategy is used for generating a traceability analysis result, generating compensation parameters according to the traceability analysis result, and correcting a thermal coupling regulation instruction set in a process knowledge base; The multi-element data comprise pipe outer surface images, wall thickness profile data and infrared temperature data; The construction of the pipe defect characteristic map comprises the steps of extracting spatial frequency characteristics of appearance defects, generating a temperature gradient vector field along the axial direction and the circumferential direction of the pipe, extracting low-frequency fluctuation components of wall thickness profile data as wall thickness unevenness indexes, fusing the spatial frequency characteristics, the temperature gradient vector field and the wall thickness unevenness indexes into defect characterization vectors, and inputting the defect characterization vectors into a pre-trained lightweight convolutional neural network model to output the defect characteristic map.
2. The real-time defect feedback based tubing forming system of claim 1, wherein the defect signature comprises a defect classification label and a corresponding confidence score; The thermodynamic coupling regulation instruction set comprises an optimization regulation instruction set and an emergency regulation instruction set, wherein the optimization regulation instruction set is called according to the defect classification label when the confidence score is higher than the confidence threshold value, and the emergency regulation instruction set is called when the confidence score is lower than the confidence threshold value.
3. The pipe molding system based on real-time defect feedback according to claim 2, wherein the S4 further comprises a parameter optimization sub-strategy, the thermal coupling regulation instruction set comprises joint regulation parameters corresponding to different execution devices, the parameter optimization sub-strategy is used for optimizing the joint regulation parameters in the optimized regulation instruction set, the parameter optimization sub-strategy configures corresponding quality optimization weights and matches corresponding defect classification labels according to different time window periods, and an update factor is generated through a preset multi-objective optimization algorithm to update the corresponding joint regulation parameters.
4. The pipe forming system based on real-time defect feedback of claim 3, wherein S4 further comprises an emergency response sub-strategy, the emergency response sub-strategy being executed in response to an emergency regulation instruction set, the emergency response sub-strategy comprising returning to step S1 until the confidence score meets a preset stable achievement condition.
5. The pipe molding system based on real-time defect feedback according to claim 4, wherein S5 further comprises comparing collected data corresponding to the multi-mode sensing devices with preset relations to generate feedback results, and carrying the feedback results into preset influence rules in a process knowledge base to match, so as to generate traceability analysis results.
6. The pipe molding system based on real-time defect feedback according to claim 5, wherein the upstream multi-mode sensing device is arranged at the primary cooling section outlet and the constant temperature section outlet; The method comprises the steps of synchronously executing the collection actions of all the multi-mode sensing devices based on a uniform time stamp, establishing a mapping relation between collection time and the axial position of the pipe by taking the pipe production speed as a reference, generating a feedback result by calculating the feature similarity of an upstream defect characterization vector and a downstream defect severity degree change rate, quantitatively calculating the defect severity degree change rate by longitudinal crack density reduction proportion, temperature gradient standard deviation reduction amplitude and wall thickness fluctuation variance change quantity, and adjusting a weight coefficient or amplifying the adjustment amplitude of effective regulation parameters or redesigning parameter combinations proportionally according to the feedback result.
7. The pipe forming system based on real-time defect feedback according to claim 1, wherein the executing device comprises a constant temperature heating body arranged in an array in a constant temperature cavity; The thermal coupling regulation and control instruction set comprises a constant temperature device control instruction, wherein the constant temperature device control instruction comprises an axial heating sub-instruction and a circumferential heating sub-instruction, the axial heating sub-instruction is used for controlling the relation between heating powers of constant temperature heating bodies positioned on the same axis, and the circumferential heating sub-instruction is used for controlling the relation between heating powers of constant temperature heating bodies positioned on the same circumference.
8. The pipe molding system based on real-time defect feedback according to claim 1, wherein the final cooling section adopts a gradient cooling structure and comprises a first-stage atomizing spray area, a second-stage aerosol mixing area and a third-stage immersed water tank area, the cooling rate of each area meets a piecewise function relation along the axial direction of the pipe, the cooling coefficients of each area decrease in sequence, and the cooling intensity ratio of the first-stage atomizing spray area to the third-stage immersed water tank area is dynamically adjusted according to real-time wall thickness profile data acquired by a multi-mode sensing device.
9. The pipe forming system based on real-time defect feedback according to claim 1, wherein the executing device comprises a spraying compensation device, the corresponding thermal coupling regulation instruction set comprises a spraying compensation sub-instruction, the spraying compensation sub-instruction is used for controlling the spraying compensation device to work, the spraying compensation sub-instruction is configured with a trigger condition, when the compensation trigger condition is met, the spraying compensation sub-instruction is output, the compensation trigger condition is a value, and the traction speed of the traction device is lower than the highest speed allowed by spraying.

Description

Pipe molding system based on real-time defect feedback Technical Field The invention belongs to the technical field of pipe manufacturing and intelligent control, and particularly relates to a pipe molding system based on real-time defect feedback. Background In the field of pipe production, processing and manufacturing, an extrusion, stretching and orientation forming process is one of the most widely used core processing technologies at present, and has been widely applied to pipe forming processes of thermoplastic materials such as polyethylene, polyvinyl chloride, random copolymer polypropylene and the like, and the basic flow of the process generally comprises: firstly heating a thermoplastic raw material to a molten state through an extruder, extruding the thermoplastic raw material through an annular machine head to form a continuous tubular blank, carrying out directional stretching on the tubular blank in a high-temperature state under the action of axial tension applied by a traction device, and simultaneously carrying out rapid cooling shaping on the tubular blank by a cooling system to finally obtain a finished product tubular product with specific outer diameter, wall thickness and mechanical properties; the process is widely used in projects such as municipal heating, building heating, centralized cooling and the like due to the advantages of relatively simple equipment structure, high production efficiency, controllable cost and the like. However, although this process is mature in industrial applications, during processing, the outer surface of the pipe occasionally develops longitudinal cracks or irregular scratches or pits resembling a stretched surface; The pipe with the condition has the problem of poor appearance consistency, is extremely easy to cause brittle fracture, cracking and leakage and the like due to factors such as external impact, medium pressure, environmental corrosion and the like in the subsequent installation and construction and long-term use processes, greatly shortens the design service life of the pipe body, and even causes potential safety hazards. Disclosure of Invention Aiming at the defects existing in the prior art, the invention aims to provide a pipe molding system and a control method based on real-time defect feedback, which are characterized in that a three-zone structure comprising cooling, constant temperature and re-cooling is introduced in an extrusion stretching orientation molding process, and the decoupling control of thermal stress and mechanical stretching force is realized by combining multi-mode online defect detection, so that the generation of pipe surface defects is effectively restrained, the mechanical property of the pipe is improved, and the qualification rate and the production stability are remarkably improved. The pipe molding system based on real-time defect feedback comprises an execution device, a central control device and a multi-mode sensing device, wherein the temperature control device comprises a constant temperature section arranged between a primary cooling section and a final cooling section, and the multi-mode sensing device is circumferentially arranged at the outlet of the primary cooling section, the outlet of the constant temperature section and the outlet of the final cooling section; The central control device is configured with central control instructions, comprising: s1, acquiring multi-element data of a pipe in real time through a multi-mode sensing device to generate acquisition information; s2, constructing a pipe defect characteristic map according to the acquired information; s3, according to the defect characteristic map, a thermodynamic coupling regulation instruction set is called from a pre-constructed process knowledge base; s4, executing the thermodynamic coupling regulation instruction set; And S5, configuring a feedback compensation strategy, wherein the feedback compensation strategy is used for generating a traceability analysis result, generating compensation parameters according to the traceability analysis result, and correcting a thermal coupling regulation instruction set in a process knowledge base. The multi-element data comprise pipe outer surface images, wall thickness profile data and infrared temperature data; The construction of the pipe defect characteristic map comprises the steps of extracting spatial frequency characteristics of appearance defects, generating a temperature gradient vector field along the axial direction and the circumferential direction of the pipe, extracting low-frequency fluctuation components of wall thickness profile data as wall thickness unevenness indexes, fusing the spatial frequency characteristics, the temperature gradient vector field and the wall thickness unevenness indexes into defect characterization vectors, and inputting the defect characterization vectors into a pre-trained lightweight convolutional neural network model to output the defect charact