CN-122008525-A - Plastic bottle production process parameter optimization control system based on artificial intelligence

CN122008525ACN 122008525 ACN122008525 ACN 122008525ACN-122008525-A

Abstract

The invention belongs to the technical field of plastic bottle manufacturing, and particularly discloses an artificial intelligence-based plastic bottle production process parameter optimization control system, which is characterized in that circumferential surface temperature is collected in real time and a three-dimensional dynamic temperature field is reconstructed by combining bottle blank continuous rotation in a reheating furnace, outlet temperature distribution is predicted based on the evolution trend of the temperature field along the conveying direction, a control target is converted into an actual bottle blank temperature state from furnace body parameters, a data basis is provided for follow-up zoned accurate thermal regulation, uniform and repeatable thermal history is ensured to obtain the bottle blank, so that consistency of material rheological response and deformation behavior in the stretch blow molding process is ensured, meanwhile, zoned deviation analysis is carried out by combining an ideal target temperature field after the predicted temperature field, and a power adjustment instruction of a downstream heating section is dynamically output according to the real-time axial position of the bottle blank in the furnace, so that accurate and differential thermal regulation of a specific area of the bottle blank is realized, and process requirements of complex bottle types on non-uniform temperature distribution are matched.

Inventors

LIU YANBING
ZHOU HONGBING
WANG FAZHANG
WU KAIXUAN

Assignees

昆明紫江包装有限公司

Dates

Publication Date: 20260512
Application Date: 20260319

Claims (10)

1. The plastic bottle production process parameter optimization control system based on artificial intelligence is characterized by comprising the following modules: The temperature field reconstruction module is used for setting at least one upstream observation position and one downstream observation position along a bottle blank conveying path in the reheating furnace, collecting infrared images of the whole outer surface of the rotary bottle blank at the upstream observation position and the downstream observation position, and reconstructing an upstream three-dimensional surface temperature field and a downstream three-dimensional surface temperature field which correspond to the bottle blank three-dimensional digital surface model respectively; The outlet temperature field prediction module is used for analyzing the change trend of the surface temperature of the bottle blank in the conveying direction based on the upstream three-dimensional surface temperature field and the downstream three-dimensional surface temperature field and predicting the temperature field distribution when the bottle blank reaches the outlet of the heating furnace; The temperature deviation analysis module is used for comparing the predicted temperature field distribution with an ideal target temperature field to generate temperature deviation vectors of different axial and circumferential areas of the bottle blank; And the power feedforward compensation module is used for calculating and outputting a power adjustment instruction of a downstream heating control sub-section which is not passed by the bottle blank according to the temperature deviation vector and combining with real-time tracking of the axial position of the bottle blank in the furnace.
2. The plastic bottle production process parameter optimization control system based on artificial intelligence as set forth in claim 1, wherein the temperature field reconstruction module comprises the following components: An observation window is respectively arranged on the furnace wall of the reheating furnace and corresponds to the upstream observation position and the downstream observation position of the bottle blank conveying path, and an infrared thermal imager is arranged outside each window; The bottle blank automatically advances in the furnace, and a rotary encoder is utilized to acquire a rotary angle signal of the bottle blank; Triggering each thermal imager by using a rotation angle signal, and collecting multi-frame two-dimensional thermal images of bottle blanks under different rotation angles; And mapping the multi-frame two-dimensional thermal images acquired by the same observing position and the same bottle blank in a complete rotation period to a three-dimensional digital surface model of the bottle blank according to the corresponding rotation angle, and respectively generating an upstream three-dimensional surface temperature field at an upstream observing position and a downstream three-dimensional surface temperature field at a downstream observing position by splicing.
3. The artificial intelligence-based plastic bottle production process parameter optimization control system as set forth in claim 1, wherein the three-dimensional digital surface model of the bottle blank is constructed by the following steps: Extracting geometric dimensions and curved surface characteristics of the outer surface of the bottle blank to be heated according to a product design drawing of the bottle blank to be heated; Constructing a continuous digital curved surface model of the outer surface of the bottle blank based on the geometric dimension and the curved surface characteristics by utilizing three-dimensional modeling software; Triangular gridding discrete processing is carried out on the digital curved surface model, and a bottle blank three-dimensional digital surface model is generated.
4. The plastic bottle production process parameter optimization control system based on artificial intelligence according to claim 1, wherein the outlet temperature field prediction module comprises the following components: regularly selecting a plurality of discrete temperature sampling points along the axial direction and the circumferential direction of the three-dimensional digital surface model of the bottle blank; extracting an upstream temperature value and a downstream temperature value of each temperature sampling point from the upstream three-dimensional surface temperature field and the downstream three-dimensional surface temperature field respectively; Calculating a difference value between a downstream temperature value and an upstream temperature value for each temperature sampling point to obtain a temperature increment of the sampling point between an upstream observation position and a downstream observation position; calculating the transmission time required by the bottle blank passing through the interval according to the known distance between the upstream observing position and the downstream observing position and the bottle blank conveying speed; Dividing the temperature increment of each temperature sampling point by the transmission time to obtain the average temperature change rate of the sampling point in the transmission interval; Based on the downstream temperature value of each sampling point in the downstream three-dimensional surface temperature field, the average temperature change rate, and the remaining heating time from the downstream observation position to the outlet of the heating furnace, the predicted temperature value of each sampling point at the outlet is calculated by linear extrapolation, thereby forming a predicted final temperature field distribution.
5. The artificial intelligence-based plastic bottle production process parameter optimization control system as set forth in claim 1, wherein the ideal target temperature field is obtained by the following construction process: Determining target temperature values of different areas of the bottle blank after heating according to the material characteristics, the geometric shape and the requirements of a subsequent stretch blow molding process of the bottle blank; The target temperature value is given to a corresponding area of the bottle blank three-dimensional digital surface model, and a digital temperature field template which is registered with the three-dimensional digital surface model in space and contains target temperature data of each point is generated as an ideal target temperature field.
6. The artificial intelligence-based plastic bottle production process parameter optimization control system according to claim 1, wherein the temperature deviation vector is obtained by the following analysis process: Subtracting the predicted temperature value of each temperature sampling point in the predicted temperature field distribution from the target temperature value of the corresponding sampling point in the ideal target temperature field point by point to obtain initial point temperature deviation; dividing a plurality of temperature sampling points into a plurality of logic partitions according to the axial and circumferential positions of the bottle blank, wherein each logic partition comprises a plurality of adjacent temperature sampling points; calculating the average value of the initial point temperature deviations of all the temperature sampling points in each logic partition to be used as the partition temperature deviation value of the logic partition; And arranging the partition temperature deviation values of all the logic partitions according to the spatial position sequence of the partition temperature deviation values on the bottle blank to form a temperature deviation vector.
7. The artificial intelligence-based plastic bottle production process parameter optimization control system as set forth in claim 6, wherein the logic partitioning is as follows: dividing the bottle blank into a plurality of axial areas along the axial direction of the bottle blank according to the change points of the geometric shape, wherein the axial areas comprise a bottle opening area, a shoulder transition area, a bottle body main body area and a bottom area; dividing bottle blanks into sectors with the same number along the circumferential area according to the number of heating units of the reheating furnace in the circumferential direction of the corresponding heating control subsection in each axial area; The intersection of each axial region with a circumferential sector is defined as a logical partition that contains all the temperature sampling points that fall within this intersection.
8. The plastic bottle production process parameter optimization control system based on artificial intelligence as set forth in claim 1, wherein the implementation of the power feedforward compensation module comprises the following steps: The reheating furnace is sequentially provided with a plurality of independent heating control subsections along the bottle blank conveying direction, each heating control subsection physically covers a section of axial heating area, and each heating control subsection is circumferentially composed of a plurality of heating units capable of independently controlling power; The system tracks the axial position of the bottle blank in the furnace in real time, and determines the downstream heating control sub-section which is about to enter and corresponds to each logic partition on the bottle blank currently based on the position; Reading partition temperature deviation values of all logic partitions in the temperature deviation vector; The partition temperature deviation value of each logic partition is directly converted into a power compensation instruction for a heating unit which is currently in the downstream heating control sub-section corresponding to the logic partition position and is the same as the logic partition circumferential position, wherein if the partition temperature deviation value is positive, the power compensation instruction is sent to the corresponding heating unit, and if the partition temperature deviation value is negative, the power compensation instruction is sent to the corresponding heating unit; the amplitude of the power compensation command and the absolute value of the partition temperature deviation value form a preset proportional relation.
9. The optimized control system for plastic bottle production process parameters based on artificial intelligence according to claim 8, wherein the tracking process of the downstream heating control sub-section is as follows: setting a reference position sensor at the downstream observation position, and recording the moment as a time reference zero point by the system when the bottle blank passes through the sensor; Calculating the axial position of the bottle blank head in the furnace in real time according to the known bottle blank conveying speed and the time counted from the time reference zero point; Calculating the current axial position of the center of each logical partition according to the geometric dimension of the bottle blank and the axial coordinate of each logical partition on the bottle blank; And comparing the current axial position of the center of each logic partition with the axial start-stop position range of each heating control sub-segment in the heating furnace, so as to determine the heating control sub-segment where each logic partition is currently located or is about to enter.
10. The system for optimizing and controlling plastic bottle production process parameters based on artificial intelligence according to claim 8, wherein the power feedforward compensation module comprises the following steps of superposition and arbitration processing of compensation instructions: When two or more logical partitions are determined to correspond to the same downstream heating control sub-segment, the power feed-forward compensation module receives a plurality of power compensation instructions for the heating control sub-segment; Calculating algebraic sum of the power compensation instructions as comprehensive power adjustment quantity for the heating control sub-section; Judging whether the comprehensive power adjustment quantity exceeds the safe power adjustment range allowed by the heating control sub-section; And if the integrated power adjustment quantity is not exceeded, issuing a final power control instruction based on the integrated power adjustment quantity, and if the integrated power adjustment quantity is exceeded, issuing the final power control instruction according to the upper limit or the lower limit of the safe power adjustment range.

Description

Plastic bottle production process parameter optimization control system based on artificial intelligence Technical Field The invention belongs to the technical field of plastic bottle manufacturing, and particularly discloses an artificial intelligence-based plastic bottle production process parameter optimization control system. Background In the production of plastic bottles, in particular in the process of poly stretch blow molding, reheating of bottle preforms is a key link connecting injection molding with blow molding. The purpose of this link is to heat the cooled and shaped preform uniformly and precisely to the desired temperature range for biaxial stretching, so that the preform material is in a highly elastic state, ensuring a uniform wall thickness distribution during the subsequent stretch blow molding process. At present, bottle blank reheating equipment widely adopted in industrial production is mainly a multi-section infrared heating furnace. The typical control mode is that an operator presets the furnace body temperature of each independent temperature zone of the heating furnace and the conveying time of bottle blanks in the furnace according to the bottle type, raw materials and experience, and when the system operates, the heating process is executed according to the fixed formula. However, this conventional control method has the following drawbacks: The control object of the traditional system is essentially the setting state of the heating furnace, but not the temperature state reached by the bottle blank, the control closed loop is established on the furnace body parameters, but not the real-time temperature field of the bottle blank, so that even if the furnace body parameters are constant, the actual heating effect of the bottle blank can be changed due to the interference, the thermal history of the bottle blank is inconsistent, and the problems of uneven material flow, uncontrolled wall thickness distribution, discrete mechanical properties of products and the like in the stretch blow molding process are caused. The traditional control method can only perform rough temperature setting of axial segmentation, and is difficult to realize fine and differential temperature regulation and control aiming at specific areas on the surface of the bottle blank, and the one-cut temperature strategy cannot match the technological requirements of complex bottle types on non-uniform temperature fields, so that local overheating or underheating is easy to cause, and the uniformity of the biaxial stretching ratio is influenced. More importantly, the existing system adopts a tail end detection mode to carry out quality verification, belongs to a typical hysteresis or post-regulation mechanism, and when the abnormal temperature of the bottle blank is identified at the outlet of the heating furnace, the heating process is irreversibly completed, and on-line correction cannot be carried out, so that unqualified products flow into subsequent procedures, the yield is reduced, and the waste cost is increased. Disclosure of Invention In view of the above, the invention aims to provide an artificial intelligence-based plastic bottle production process parameter optimization control system, which can perform dynamic feedforward compensation reheating control based on prediction by focusing on a bottle blank reheating link in plastic bottle production and taking a bottle blank real-time temperature field as a control object, thereby effectively solving the problems mentioned in the background art. The invention aims at realizing the technical scheme that the plastic bottle production process parameter optimization control system based on artificial intelligence comprises the following modules: And the temperature field reconstruction module is used for setting at least one upstream observation position and one downstream observation position along the bottle blank conveying path in the reheating furnace, collecting infrared images of the whole outer surface of the rotary bottle blank at the upstream observation position and the downstream observation position, and reconstructing an upstream three-dimensional surface temperature field and a downstream three-dimensional surface temperature field which correspond to the bottle blank three-dimensional digital surface model respectively. And the outlet temperature field prediction module is used for analyzing the change trend of the surface temperature of the bottle blank in the conveying direction based on the upstream three-dimensional surface temperature field and the downstream three-dimensional surface temperature field and predicting the temperature field distribution when the bottle blank reaches the outlet of the heating furnace. And the temperature deviation analysis module is used for comparing the predicted temperature field distribution with an ideal target temperature field to generate temperature deviation vectors of different axial and circumfere