Production method and system of polyvinyl chloride pipeline

Abstract

The invention relates to the technical field of pipeline production, and discloses a polyvinyl chloride pipeline production method which comprises the steps that S10, a plurality of weight sensors and component detectors are arranged on an extruder, the weight sensors are used for collecting raw material flow data when raw materials are located at different positions, and the component detectors are used for detecting the raw material flow data; the component detector is used for collecting the mass fractions of the raw materials at different positions; s20, conveying the raw materials into a hopper of an extruder, and heating and melting the raw materials through a screw rod; s30, the feeding amount and the rotating speed of a screw are adjusted according to the collected raw material flow data and the collected raw material mass fraction; s40, the molten raw materials are pushed to a mold through the screw rod, and a primarily-formed polyvinyl chloride pipeline is obtained; and S50, a cooling module is adopted for conducting water cooling on the primarily-formed polyethylene pipeline, the cooled pipeline is dragged through a dragger, and the stably-formed polyvinyl chloride pipeline is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pipe production, in particular to a production method and system of a polyvinyl chloride pipe. BACKGROUND

[0002] As one of the largest plastic pipe varieties in the world, polyvinyl chloride (PVC) pipes dominate in the fields of building water supply and drainage, municipal engineering, and agricultural irrigation. According to statistics, the global PVC pipe market size exceeded 60 billion US dollars in 2023, with an annual growth rate of about 5%. However, with the complexity of application scenarios and the improvement of environmental protection requirements such as renewable material utilization rate targets, the traditional PVC pipe production process has been difficult to meet the production demands of high performance, high efficiency, and low cost.

[0003] The current commonly used process route is based on extrusion molding technology. The core process of this method starts with the formulation and pretreatment of raw materials. According to the purpose and performance requirements of the final pipe, PVC resin powder and various additives are accurately measured and mixed in a specific ratio. The material is first heated to a set temperature (usually above 100℃) in a high-speed rotating hot mixer through strong shearing friction, so that the additives are preliminarily dispersed and partially penetrate or coat the surface of the PVC resin particles, forming a preliminary dry mixture. Then, the hot mixture is quickly transferred to a low-speed cold mixer for stirring and cooling to prevent the material from caking or degrading due to residual heat, and finally obtain a uniform, loose, and good flowing dry mixture. The molding stage mainly relies on single screw or parallel co-rotating twin screw extruders. The performance of the PVC pipe is determined by the final ratio of polyvinyl chloride resin, toughening agent, and antioxidant in the raw materials. However, during the production process, errors may occur in these ratios, for example, when the resin flow rate may be different from the pre-set ratio due to material caking, poor feeding, uneven density, etc. If the replenishment speed of the toughening agent and antioxidant cannot be adjusted synchronously at this time, it will cause an imbalance in the formula ratio, ultimately resulting in low-quality PVC pipes and waste of raw materials. The existing methods to eliminate ratio errors mainly rely on manual sampling or fixed threshold / PID control, which cannot dynamically adjust the process and optimize the process parameters in a timely manner. SUMMARY

[0004] The technical problem to be solved by the present application is to solve the problem of not being able to dynamically adjust the process and optimize the process parameters when there is an error in the pre-raw material ratio of the polyvinyl chloride pipe and the raw material ratio during the production process.

[0005] To solve the above technical problems, the present application provides a production method of a polyvinyl chloride pipe, which comprises:

[0006] S10, a plurality of weight sensors and component detectors are arranged on the extruder, the weight sensors are used to collect raw material flow data of the raw material at different positions, and the component detectors are used to collect raw material mass fraction of the raw material at different positions;

[0007] S20, the raw material is transported into the hopper of the extruder, and the raw material is heated and melted by the screw, the raw material includes polyvinyl chloride resin, toughening agent and antioxidant, the inside of the hopper is integrally provided with an infrared heating plate, and the heating plate is used for preheating the raw material to eliminate crystal water;

[0008] S30, the feeding amount and the rotating speed of the screw are adjusted according to the collected raw material flow data and raw material mass fraction, specifically:

[0009] A raw material consumption prediction model is established based on a neural network, when the polyvinyl chloride resin flow deviation is greater than a preset value, the replenishment speed of the toughening agent and the antioxidant is automatically adjusted, and the replenishment amount is automatically calculated according to a preset formula, and when the raw material mass fraction deviates from a preset value, the rotating speed of the screw is adjusted;

[0010] S40, the melted raw material is pushed to the mold by the screw to obtain a preliminarily formed polyvinyl chloride pipeline;

[0011] S50, a cooling module is used for water cooling of the preliminarily formed polyethylene pipeline, and the cooled pipeline is pulled by a traction machine to obtain a stably formed polyvinyl chloride pipeline;

[0012] The cooling module includes a plurality of cold water tanks, the temperatures of the plurality of cold water tanks are different, and the water temperatures of the plurality of cold water tanks are adjusted according to the raw material flow data.

[0013] Further, the infrared heating plate is made of silicon carbide, and the heating power of the infrared heating plate is adjusted according to the content of the crystal water.

[0014] Further, it further includes steel belt reinforcing composite of the stably formed polyvinyl chloride pipeline, and the specific steps are:

[0015] The surface of the steel belt is pretreated to obtain a steel belt with a locking structure;

[0016] The pretreated steel belt is preheated to make the steel belt have thermal adhesion;

[0017] The preheated steel belt is pressed with the preliminarily formed polyvinyl chloride pipeline to obtain a pressed polyvinyl chloride pipeline;

[0018] The pressed polyvinyl chloride pipeline is water cooled to obtain a polyvinyl chloride pipeline with a steel belt.

[0019] Further, the method further comprises:

[0020] An ultrasonic vibration device is arranged at the pressing position of the steel strip and the preliminarily formed polyvinyl chloride pipe to remove interface bubbles through high-frequency vibration and improve the bonding density.

[0021] Further, after step S450, the method further comprises:

[0022] After the polyvinyl chloride pipe is generated, the leftover material of the polyvinyl chloride pipe is put into a double-roller crusher for crushing treatment;

[0023] The crushed leftover material is allowed to enter an ultrasonic cleaning tank to remove surface impurities;

[0024] The cleaned leftover material is transferred to a vacuum drying box;

[0025] The near-infrared spectrometer is used to detect the mixing uniformity of the leftover material in real time, when the measured uniformity is lower than the preset value, the amount of supplementary material is calculated according to the formula and the amount of feeding is adjusted, and when the mixing uniformity of the leftover material is greater than the preset value, the leftover material is recycled as the raw material.

[0026] Further, the raw material consumption prediction model is a raw material consumption prediction model constructed based on a LSTM-CNN hybrid neural network, and the formula is:

[0027] Q t = LSTM(Q t-1, h t-1 ) + CNN(F t )W + b

[0028] wherein Q t is the raw material consumption prediction value (kg / h) at time step t, Q t-1 is the actual raw material flow (kg / h) at time step t-1, h t-1 is the LSTM hidden state (representing historical time sequence features), F t is the raw material feature at the current time, W is the weight, and b is the correction prediction bias.

[0029] According to another aspect of the present application, a polyvinyl chloride pipe production system is provided, and the polyvinyl chloride pipe production system includes the above-mentioned polyvinyl chloride pipe production method when manufacturing.

[0030] Further, the system comprises:

[0031] A raw material processing module, an extruder hopper integrated with an infrared heating plate, multiple weight sensors built-in for collecting real-time flow data of PVC resin, toughening agent and antioxidant, the extruder hopper is located at the starting end of the production system, the infrared heating plate is attached to the inner wall of the hopper, the screw is located below the hopper, the mold is located at the end of the extruder, and the mold is connected with a cooling module;

[0032] A data processing module, the data processing module is based on a raw material consumption prediction model of an LSTM-CNN hybrid neural network, the input is historical flow data and current raw material characteristics, and the output is a predicted consumption value;

[0033] A PLC control module, the PLC controller is used to adjust the replenishment amount of the toughening agent and the antioxidant, and the rotating speed of the screw;

[0034] A cooling forming module, the cooling forming module is used to adopt the cooling module to water-cool the preliminarily formed polyethylene pipeline, the cooled pipeline is pulled by a pulling machine to obtain a stably formed PVC pipeline, and the pulling machine is located at the end of the last-stage cold water tank.

[0035] Further, the system further comprises:

[0036] A steel belt reinforcing module, the steel belt reinforcing module comprises a steel belt pretreatment unit and an ultrasonic vibration pressing device, the steel belt pretreatment unit is independent of the production system, and the ultrasonic vibration pressing device is embedded between the mold and the pulling machine.

[0037] Compared with the prior art, the polyvinyl chloride pipeline production method has the following beneficial effects:

[0038] The application realizes real-time collection and multi-position monitoring of raw material flow and mass fraction by arranging weight sensors and component detectors, and can identify PVC resin flow deviation and component deviation by combining with a raw material consumption prediction model established by a neural network, automatically adjusts the replenishment speed of toughening agent and antioxidant and calculates the replenishment amount when detecting abnormal proportioning, eliminates the influence of proportioning error on product performance from the source by adjusting the screw speed to control the melting state, realizes automatic adjustment of the feed amount and screw speed by combining with the raw material consumption prediction model established by the neural network, and automatically calculates the replenishment amount and adjusts the replenishment speed while dynamically correcting the screw speed when detecting that the raw material flow deviation exceeds the preset value, thereby fundamentally solving the limitation of static adjustment in the prior art. The prediction model based on the neural network can monitor the PVC resin flow deviation in real time, and automatically link the adjustment of the replenishment speed of the toughening agent and the antioxidant to ensure that the proportion of each component strictly meets the preset formula. This intelligent replenishment mechanism avoids the hysteresis and error of manual adjustment, and significantly improves the product performance. The application realizes real-time optimization of the production process by dynamic process adjustment, raw material pretreatment optimization, intelligent replenishment control and segmented dynamic cooling, effectively solves the problem that the process parameters cannot be dynamically adjusted in the prior art, and significantly improves the production efficiency and product quality of the PVC pipe. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 is a flow chart of the production method of the PVC pipe provided by the embodiment of the application;

[0040] Figure 2 is a schematic diagram of the generation system of the PVC pipe provided by the embodiment of the application. DETAILED DESCRIPTION

[0041] The exemplary embodiments of the application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the application to help understanding, and should be considered as merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the application. Also, for the sake of clarity and conciseness, the description below omits the description of well-known functions and structures.

[0042] As shown in Figure 1 , in an optional embodiment of the application, a production method of a PVC pipe includes:

[0043] S10, arranging a plurality of weight sensors and component detectors on the extruder, the weight sensors being used to collect raw material flow data of the raw material at different positions, and the component detectors being used to collect raw material mass fraction of the raw material at different positions;

[0044] S20, conveying raw materials into the hopper of the extruder, and heating and melting the raw materials including polyvinyl chloride resin, toughening agent and antioxidant by the screw, and the inside of the hopper is integrally provided with an infrared heating plate for preheating the raw materials to eliminate crystal water;

[0045] S30, adjusting the feeding amount and the rotating speed of the screw according to the collected raw material flow data and the raw material mass fraction, specifically:

[0046] The raw material consumption prediction model is established based on a neural network, when it is detected that the polyvinyl chloride resin flow deviation is greater than a preset value, the replenishment speed of the toughening agent and the antioxidant is automatically adjusted, and the replenishment amount is automatically calculated according to a preset formula, when it is detected that the raw material mass fraction deviates from a preset value, the rotating speed of the screw is adjusted;

[0047] S40, pushing the melted raw materials to the mold through the screw to obtain a preliminarily formed polyvinyl chloride pipeline;

[0048] S50, using a cooling module to water-cool the preliminarily formed polyethylene pipeline, and the cooled pipeline is pulled by a traction machine to obtain a stably formed polyvinyl chloride pipeline; wherein the cooling module comprises a plurality of cold water tanks, the temperatures of the plurality of cold water tanks are different, and the water temperatures of the plurality of cold water tanks are adjusted according to the raw material flow data.

[0049] The weight sensor is a device for measuring the mass of an object, which is arranged at different positions of the extruder in the application, and is used for collecting the flow data of raw materials such as polyvinyl chloride resin and toughening agent in real time, so as to monitor the dynamic change of raw material conveying. The component detector is a detection device for analyzing the composition of a substance, such as a near-infrared spectrometer and an X-ray fluorescence spectrometer, which is used for detecting the mass fraction of each component of the raw materials, i.e. polyvinyl chloride resin, toughening agent and antioxidant, in real time in the application, so as to ensure that the ratio meets the preset requirements. The neural network raw material consumption prediction model is a mathematical model constructed based on an artificial neural network algorithm, which can predict the raw material consumption under different production conditions such as feeding amount and screw rotating speed by training and learning the raw material consumption law through historical data, and provide a basis for dynamic adjustment. The infrared heating plate is a device for heating by infrared radiation, which is integrated in the hopper of the extruder, and is used for preheating the raw materials, and the main purpose is to eliminate the crystal water in the polyvinyl chloride resin. The cooling module is a cooling system composed of multiple stages of cold water tanks, which realizes gradient cooling of the preliminarily formed pipeline by controlling the water temperature of different cold water tanks (adjusted dynamically according to the raw material flow), so as to avoid deformation or cracking caused by sudden temperature change.

[0050] In S10, by multi-position arrangement of weight sensors and component detectors, real-time monitoring of raw material flow and mass fraction throughout the process is realized, providing a data basis for subsequent dynamic adjustment. The weight sensor collects flow data at different positions (such as hopper inlet, screw conveying section), and the component detector detects the real-time mass fraction of each component in the raw material, ensuring that the proportioning error can be accurately captured. In S20, the polyvinyl chloride resin, toughening agent, and antioxidant in the raw material are transported to the extruder and preheated by an infrared heating plate to eliminate crystallization water and optimize melting effect. Preheating can reduce the energy required for raw material melting, while avoiding melt defects (such as bubbles) caused by crystallization water gasification at high temperatures. In S30, based on real-time monitoring data and neural network model, the feed amount, replenishment speed, and screw speed are automatically adjusted to correct the proportioning deviation and ensure process parameter optimization. When the polyvinyl chloride resin flow deviation exceeds the preset value, the model automatically calculates the replenishment amount of toughening agent and antioxidant (according to the preset formula) to avoid excessive or insufficient single component; when the component mass fraction deviates from the preset value, the screw speed is adjusted to control the melt viscosity and ensure the stability of extrusion. In S40, the molten raw material is pushed to the mold by the screw to form a preliminary shaped polyvinyl chloride pipe. The speed and temperature control of the screw directly affect the flowability of the melt and the forming precision, and the dynamically adjusted parameters can ensure the dimensional stability of the pipe. In S50, the preliminary shaped pipe is gradient cooled by multiple stages of cold water tank to avoid deformation or cracking caused by sudden temperature change, and the pipe shape is stabilized by a traction machine. The water temperature of the cold water tank is dynamically adjusted according to the raw material flow, such as lowering the water temperature to accelerate cooling at high flow, and increasing the water temperature to prevent stress concentration at low flow, ensuring that the cooling rate matches the production rhythm.

[0051] The application realizes real-time collection and multi-position monitoring of raw material flow and mass fraction by arranging weight sensors and component detectors, and can identify PVC resin flow deviation and component deviation by combining with a raw material consumption prediction model established by a neural network, automatically adjusts the replenishment speed of toughening agent and antioxidant and calculates the replenishment amount when detecting abnormal proportioning, eliminates the influence of proportioning error on product performance from the source by adjusting the screw rotation speed to control the melting state, realizes automatic adjustment of the feed amount and screw rotation speed by combining with the raw material consumption prediction model established by the neural network, and automatically calculates the replenishment amount and adjusts the replenishment speed while dynamically correcting the screw rotation speed when detecting that the raw material flow deviation exceeds the preset value, thereby fundamentally solving the limitations of static adjustment in the prior art. The prediction model based on the neural network can monitor the PVC resin flow deviation in real time, and automatically link the adjustment of the replenishment speed of the toughening agent and the antioxidant to ensure that the proportion of each component strictly meets the preset formula. This intelligent replenishment mechanism avoids the hysteresis and errors of manual adjustment, and significantly improves the product performance. The application realizes real-time optimization of the production process by dynamic process adjustment, raw material pretreatment optimization, intelligent replenishment control and segmented dynamic cooling, effectively solves the problem that the process parameters cannot be dynamically adjusted in the prior art, and significantly improves the production efficiency and product quality of the PVC pipe.

[0052] In an optional embodiment of the application, the infrared heating plate is made of silicon carbide, and the heating power of the infrared heating plate is adjusted according to the content of the crystallization water.

[0053] Specifically, the infrared heating plate made of silicon carbide improves the heating efficiency, high-temperature resistance and durability to reduce the equipment maintenance frequency, and ensures the stability of the preheating process. The heating power is adjusted according to the actual content of the crystallization water in the raw material to avoid energy waste caused by "overheating" or incomplete elimination of crystallization water caused by "underheating". For example, when it is detected that the content of the crystallization water in the raw material is high, the heating power is automatically increased to quickly eliminate the moisture; when the content is low, the power is reduced to save energy and reduce the risk of overheating of the raw material.

[0054] In an optional embodiment of the application, the application further comprises steel belt reinforcing and compounding of the stably formed PVC pipe, and the specific steps are as follows:

[0055] The surface of the steel belt is pretreated to obtain a steel belt with a locking structure;

[0056] The pretreated steel belt is preheated to make the steel belt have thermal adhesion;

[0057] The preheated steel belt is pressed with the preliminarily formed PVC pipe to obtain a pressed PVC pipe;

[0058] The polyvinyl chloride pipe after pressing is water-cooled to obtain a polyvinyl chloride pipe with a steel strip.

[0059] The locking structure refers to the concave-convex lines or micro-rough structure formed on the surface of the steel strip through pretreatment (such as rolling, etching or embossing), which is used to increase the contact area and mechanical interlocking force of the steel strip and the polyvinyl chloride matrix, and improve the interfacial bonding strength after compounding. The hot adhesion refers to that when the steel strip is preheated to a certain temperature, the surface molecular activity is enhanced, and the characteristics similar to viscous fluid are presented, so that the molecular level penetration and adhesion with the surface of the PVC pipe in a molten or semi-molten state can be achieved, and the delamination after compounding can be avoided. The pressing refers to that the preheated steel strip is tightly attached to the surface of the preliminarily formed PVC pipe by mechanical pressure, so that the locking structure of the steel strip is embedded into the PVC matrix, and the chemical bonding and mechanical interlocking are realized at the same time by the hot adhesion.

[0060] The steel strip surface pretreatment is to manufacture micro-rough structure on the surface of the steel strip by physical or chemical methods, so as to increase the contact area and mechanical interlocking force with the PVC, and provide a structural basis for subsequent hot adhesion pressing. The pretreatment methods include rolling concave-convex lines, laser etching or chemical etching, so as to form regular or irregular locking structures, and ensure the interfacial bonding strength of the steel strip and the PVC. The function of the steel strip preheating control is to heat the steel strip to a certain temperature (usually slightly lower than the PVC melting temperature), so that the surface molecular activity is enhanced, and the hot adhesion is presented, so that the molecular penetration and adhesion with the surface of the PVC pipe can be achieved. The preheating temperature needs to be accurately controlled (such as 120-180℃), so as to avoid oxidation of the steel strip or overheating decomposition of the PVC, and at the same time, ensure that the hot adhesion is sufficient to realize firm compounding. The pressing function of the steel strip and the PVC pipe is to press the preheated steel strip to the surface of the preliminarily formed PVC pipe by mechanical pressure, so as to realize the firm compounding of the steel strip and the PVC by the mechanical interlocking of the locking structure and the molecular penetration of the hot adhesion. The pressing pressure needs to be uniformly distributed (such as by rolling or molding), so as to ensure that the steel strip is completely embedded into the PVC matrix, and avoid local delamination or air bubbles. The function of the water cooling (determining the steel strip reinforced pipe) is to quickly cool the compounded pipe after pressing, so as to fix the bonding state of the steel strip and the PVC, and avoid shrinkage deformation or interfacial stress concentration due to temperature drop. The cooling water temperature, flow and spraying mode need to be matched with the pipe size and the steel strip thickness, so as to ensure that the cooling rate is uniform, and avoid cracking caused by local overcooling.

[0061] In the embodiment of the present application, through the processes of steel strip surface pretreatment, hot adhesion preheating, pressing and compounding and water cooling and determining, the high-strength combination of the steel strip and the PVC pipe is realized, the mechanical properties of the pipe are improved, the production efficiency and cost are optimized, and the technical and economic value is remarkable.

[0062] In an optional embodiment of the present application, after step S40, the method further comprises:

[0063] After the polyvinyl chloride pipe is generated, the leftover material of the polyvinyl chloride pipe is put into a double-roller crusher for crushing treatment;

[0064] The crushed leftover material is allowed to enter an ultrasonic cleaning tank to remove surface impurities;

[0065] The cleaned leftover material is transferred to a vacuum drying box;

[0066] The uniformity of the leftover material is detected in real time by a near-infrared spectrometer, when the measured uniformity is lower than a preset value, the amount of supplementary material is calculated according to a formula and the amount of feeding material is adjusted, when the uniformity of the leftover material is greater than the preset value, the leftover material is recycled as the raw material.

[0067] In an optional embodiment of the present application, the raw material consumption prediction model is a raw material consumption prediction model constructed based on a LSTM-CNN hybrid neural network, and the formula is

[0068] Q t = LSTM(Q t-1, h t-1 ) + CNN(F t ) * W + b

[0069] wherein Q t is a raw material consumption prediction value (kg / h) at a time step t, Q t-1 is an actual raw material flow (kg / h) at a time step t-1, h t-1 is an LSTM hidden state (representing historical time sequence features), F t is a raw material feature at the current time, W is a weight, and b is a correction prediction bias.

[0070] A specific embodiment is further described as follows:

[0071] The polyvinyl chloride resin (PVC, 85wt%), toughening agent (CPE, 10wt%), antioxidant (1010, 5wt%) are weighed by mass ratio, mixed and put into the hopper of the extruder. Turn on the infrared heating plate in the hopper, set the temperature to 90℃, the preheating time is 15 minutes, and the target water content is ≤0.3%. At the inlet of the hopper, the screw conveying section and the melting zone, respectively arrange weight sensors (range 0-500kg / h, accuracy ±0.1%) and component detectors (near infrared spectrometer, detection wavelength 1200-2500nm). Load the pre-trained LSTM-CNN model (LSTM layer 2, hidden unit 128; CNN convolution kernel 3x3, channel number 64), input historical data (past 1 hour flow sequence) for online learning. When the PVC flow deviation ΔQ_pvc>5% is detected (such as Q_pvc=200kg / h, the allowed range is 190-210kg / h), the model calculates the toughening agent supplement amount: when the component detector detects that the mass fraction of antioxidant 1010 is <4.5%, the screw speed is reduced to 120rpm, and the melting time is prolonged to improve the dispersion uniformity. The screw speed is 180rpm, the melting zone temperature is 180℃, the homogenization zone temperature is 190℃, and the mold temperature is 200℃. The preliminary formed pipeline successively passes through three stages of cold water tanks, and the water temperature is set to 15℃ for the first stage, 20℃ for the second stage, and 25℃ for the third stage, and the pulling speed V=2m / min.

[0072] According to another aspect of the present application, a polyvinyl chloride pipe production system is provided, which uses the polyvinyl chloride pipe production method described above when manufacturing.

[0073] As shown in Figure 2 The system comprises: a raw material processing module, an extruder hopper integrated with an infrared heating plate, which is internally provided with a plurality of weight sensors for collecting real-time flow data of polyvinyl chloride resin, toughening agent and antioxidant, the extruder hopper is located at the starting end of the production system, the infrared heating plate is attached to the inner wall of the hopper, the screw is located below the hopper, and the mold is located at the end of the extruder, and the mold is connected with a cooling module;

[0074] A data processing module, which is based on a raw material consumption prediction model of an LSTM-CNN hybrid neural network, and inputs historical flow data and current raw material characteristics, and outputs predicted consumption values;

[0075] A PLC control module, which is used to adjust the supplement amount of the toughening agent and the antioxidant, and the speed of the screw;

[0076] A cooling forming module for water cooling the preliminarily formed polyethylene pipe by using the cooling module, the cooled pipe will be pulled by a pulling machine to obtain a stably formed polyvinyl chloride pipe, and the pulling machine is located at the end of the last stage of the cold water tank.

[0077] Specifically, PVC, CPE and antioxidant are put into the hopper according to the proportion, and the infrared heating plate is preheated to 90°C to eliminate crystal water. The weight sensor and the component detector collect data in real time and input the LSTM-CNN model. The model predicts the consumption value Qt, and the PLC adjusts the feeding amount and the screw speed according to the prediction result. The screw pushes the molten raw material to the mold to form a preliminary pipe. The pipe passes through the three-stage cold water tank in turn, and the water temperature is dynamically adjusted according to the flow, and the pulling machine is shaped and output. The raw material processing module function: complete raw material pretreatment, flow acquisition and preliminary melting. The infrared heating plate is integrated with the inner wall of the hopper, and the preheating eliminates the crystal water. The built-in weight sensor collects the flow data of PVC, toughening agent and antioxidant in real time. The component detector such as near-infrared spectrometer is installed in the screw conveying section to monitor the mass fraction of PVC, CPE and antioxidant in real time. Based on the LSTM-CNN hybrid model, the raw material consumption is predicted, and the dynamic adjustment instruction is output. Input layer: historical flow data (PVC, CPE and antioxidant flow sequence in the past 1 hour, time step 1 minute). Current raw material characteristics (component mass fraction, temperature, pressure, dimension 1x6). LSTM layer: 2 layers of LSTM, with 128 hidden units, to capture the time sequence rules of flow changes (such as periodic fluctuations and trend changes). CNN layer: 2 layers of convolution (3x3 convolution kernel, 64 channels), to extract the local spatial correlation of component mass fraction. Fusion output layer: fully connected layer (128→1), to output the predicted consumption value Qt (unit: kg / h), formula:

[0078] Q t = LSTM(Q t-1 ,h t-1 )+CNN(F t )*W+b

[0079] Wherein, Q t is the raw material consumption prediction value (kg / h) at time step t, Q t-1 is the actual raw material flow (kg / h) at time step t-1, h t-1 is the LSTM hidden state (representing historical time sequence features), F t is the raw material feature at the current time, W is the weight, and b is the correction prediction bias. PLC control module function: execute model prediction results, adjust feeding amount and screw speed. Control logic: feeding amount adjustment: when the PVC flow deviation ΔQPVC is greater than the preset value, calculate CPE and antioxidant supply. Screw speed adjustment: adjust the speed when the component mass fraction deviates from the preset value.

[0080] The present application realizes real-time collection and multi-position monitoring of raw material flow and mass fraction by arranging weight sensors and component detectors, and can identify PVC resin flow deviation and component deviation by combining with a raw material consumption prediction model established by a neural network, automatically adjusts the replenishment speed of toughening agent and antioxidant and calculates the replenishment amount when detecting abnormal proportioning, eliminates the influence of proportioning error on product performance from the source by adjusting and controlling the melting state through screw speed, realizes automatic adjustment of feed amount and screw speed by combining with a raw material consumption prediction model established by a neural network, and automatically calculates the replenishment amount and adjusts the replenishment speed while dynamically correcting the screw speed when detecting that the raw material flow deviation exceeds the preset value, so that the process parameters are always matched with the actual production demand, and the limitation of static adjustment in the prior art is fundamentally solved.

[0081] The prediction model based on the neural network can realize real-time monitoring of PVC resin flow deviation, and automatically link and adjust the replenishment speed of toughening agent and antioxidant, so that the proportion of each component strictly meets the preset formula, and the intelligent replenishment mechanism avoids the hysteresis and error of manual adjustment, and significantly improves the product performance. The present application realizes real-time optimization of the production process by dynamic process adjustment, raw material pretreatment optimization, intelligent replenishment control and segmented dynamic cooling, effectively solves the problem that the process parameters cannot be dynamically adjusted in the prior art, and significantly improves the production efficiency and product quality of the PVC pipe.

[0082] The above specific embodiments do not constitute a limitation on the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for producing a polyvinyl chloride pipe, characterized by, The method comprises: S10, arranging a plurality of weight sensors and component detectors on an extruder, the weight sensors being used to collect raw material flow data of raw materials at different positions, and the component detectors being used to collect raw material mass fractions of raw materials at different positions; S20, conveying raw materials into a hopper of the extruder and heating and melting the raw materials by a screw, the raw materials comprising polyvinyl chloride resin, toughening agent and antioxidant, an internal part of the hopper being integrally provided with an infrared heating plate, the heating plate being used to preheat the raw materials to eliminate crystal water; S30, adjusting the feeding amount and the rotating speed of the screw according to the collected raw material flow data and raw material mass fractions, specifically: establishing a raw material consumption prediction model based on a neural network, automatically adjusting the replenishment speed of the toughening agent and the antioxidant and automatically calculating the replenishment amount according to a preset formula when it is detected that the polyvinyl chloride resin flow deviation is greater than a preset value, and adjusting the rotating speed of the screw when it is detected that the raw material mass fraction deviates from a preset value; S40, pushing the melted raw materials to a mold by the screw to obtain a preliminarily formed polyvinyl chloride pipeline; S50, using a cooling module to water-cool the preliminarily formed polyethylene pipeline, and using a traction machine to pull the cooled pipeline to obtain a stably formed polyvinyl chloride pipeline; The cooling module comprises a plurality of cold water tanks, the temperatures of the plurality of cold water tanks are different, and the water temperatures of the plurality of cold water tanks are adjusted according to the raw material flow data.

2. The method of producing a polyvinyl chloride pipe according to claim 1, characterized by, The infrared heating plate is made of silicon carbide, and the heating power of the infrared heating plate is adjusted according to the content of the crystal water.

3. The method of producing a polyvinyl chloride pipe according to claim 1, characterized by, Further comprising steel belt reinforcing and compounding of the stably formed polyvinyl chloride pipeline, and the specific steps are: pretreating a steel belt surface to obtain a steel belt with a locking structure; preheating the pretreated steel belt to make the steel belt have thermal adhesion; pressing the preheated steel belt and the preliminarily formed polyvinyl chloride pipeline to obtain a pressed polyvinyl chloride pipeline; water-cooling the pressed polyvinyl chloride pipeline to obtain a polyvinyl chloride pipeline with a steel belt.

4. The method of producing a polyvinyl chloride pipe according to claim 3, characterized by, The method further comprises setting an ultrasonic vibration device at the pressing position of the steel belt and the preliminarily formed polyvinyl chloride pipeline to remove interface bubbles through high-frequency vibration and improve the bonding density.

5. The method of producing a polyvinyl chloride pipe according to claim 1, characterized by, After step S50, the method further comprises: after the polyvinyl chloride pipeline is generated, feeding the edge and corner scraps of the polyvinyl chloride pipeline into a double-roller crusher for crushing treatment; letting the crushed edge and corner scraps enter an ultrasonic cleaning tank to remove surface impurities; transferring the cleaned edge and corner scraps to a vacuum drying box; using a near-infrared spectrometer to detect the mixing uniformity of the edge and corner scraps in real time, calculating the replenishment amount according to a formula and adjusting the feeding amount when the measured uniformity is lower than a preset value, and using the edge and corner scraps as the raw materials for recycling treatment when the mixing uniformity of the edge and corner scraps is greater than the preset value.

6. The method of producing a polyvinyl chloride pipe according to claim 1, characterized by, The raw material consumption prediction model is a raw material consumption prediction model constructed based on a LSTM-CNN hybrid neural network, and the formula is Q t = LSTM(Q t-1, h t-1 )+ CNN(F t )W + b where Q t is the raw material consumption prediction value (kg / h) at time step t, Q t-1 is the actual raw material flow (kg / h) at time step t-1, h t-1 is the LSTM hidden state (characterizing historical time series features), F t is the raw material feature at the current time, W is the weight, and b is the correction prediction bias.

7. A production system for polyvinyl chloride (PVC) pipes, characterized in that, The production system using the polyvinyl chloride pipeline comprises the polyvinyl chloride pipeline production method in any one of claims 1-7.

8. A production system of polyvinyl chloride pipe, comprising a plurality of weight sensors for collecting raw material flow data of raw materials at different positions, and component detectors for collecting mass fractions of raw materials at different positions, the weight sensors and component detectors being respectively located at a hopper and a screw conveying end, characterized in that, The system comprises: a raw material processing module, an extruder hopper integrated with an infrared heating plate, the extruder hopper being located at the starting end of the production system, the infrared heating plate being attached to the inner wall of the hopper, the screw being located below the hopper, the mold being located at the end of the extruder, the mold being connected with a cooling module; a data processing module, the data processing module being based on a raw material consumption prediction model of a mixed neural network of LSTM-CNN, the input being historical flow data and current raw material mass fraction, and the output being predicted consumption value and screw speed; a PLC control module, the PLC controller being used to adjust the replenishment amount of toughening agent and antioxidant and the speed of the screw; a cooling and molding module, the cooling and molding module being used to adopt the cooling module to water-cool the preliminarily molded polyethylene pipeline, the cooled pipeline being pulled by a pulling machine to obtain a stably molded polyvinyl chloride pipeline, the pulling machine being located at the end of the last-stage cooling water tank.

9. The polyethylene pipe production system of claim 8, wherein: The system further comprises: a steel belt reinforcing module, the steel belt reinforcing module comprising a steel belt pretreatment unit and an ultrasonic vibration pressing device, the steel belt pretreatment unit being independent of the production system, and the ultrasonic vibration pressing device being embedded between the mold and the pulling machine.

Images