Production device of wear-resistant plastic tray

Abstract

The invention discloses a wear-resistant plastic tray production device which is used for tray production and comprises a demolding module and an intelligent monitoring module, the demolding module comprises a mold, an electromagnetic induction unit, a temperature sensing unit and an ejection unit, and a super-hydrophobic nano coating is prepared on the surface of the mold; for the electromagnetic induction unit, an electromagnetic induction coil is embedded in the surface of a cavity of the mold in a spiral and array combined mode; the intelligent monitoring module comprises an edge computing node processor, a cloud server and a plurality of monitoring sensors. According to the production device for the wear-resistant plastic tray, linkage is conducted through the demolding module and the intelligent monitoring module, on one hand, damage to the surface of the tray can be reduced, and the demolding efficiency is improved; on the other hand, intelligent monitoring and predictive maintenance of production equipment are achieved, the equipment failure shutdown time is shortened, and production continuity is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of pallet production, and in particular relates to a production device for a wear-resistant plastic pallet. Background Art

[0002] Plastic pallets are a crucial logistics tool, widely used in cargo storage and transportation. However, many existing pallet production systems lack intelligent monitoring during the production process, lacking real-time health diagnostics for each piece of equipment. This prevents immediate diagnosis and analysis of faults, impacting production efficiency. Furthermore, pallets are prone to sticking to molds during the demolding process, compromising product quality.

[0003] Therefore, further improvements are made to the above problems. Summary of the Invention

[0004] The main purpose of the present invention is to provide a production device for wear-resistant plastic pallets, which is linked by a demoulding module and an intelligent monitoring module. On the one hand, it can reduce damage to the surface of the pallet and improve demoulding efficiency; on the other hand, it can realize intelligent monitoring and predictive maintenance of production equipment, reduce equipment downtime, and improve production continuity.

[0005] To achieve the above objectives, the present invention provides a production device for wear-resistant plastic pallets, which is used for the production of pallets and includes a demoulding module and an intelligent monitoring module, wherein:

[0006] The demoulding module includes a mold, an electromagnetic induction unit, a temperature sensing unit and an ejection unit, wherein:

[0007] For the mold, a superhydrophobic nanocoating is prepared on the surface;

[0008] For the electromagnetic induction unit, an electromagnetic induction coil is embedded in the cavity surface of the mold in a combination of spiral and array types, and an electromagnetic induction heating controller is used to control the electromagnetic induction coil to generate an electromagnetic field;

[0009] For the temperature sensing unit, a high-precision temperature sensor is installed to monitor the surface temperature of the mold in real time and work in conjunction with the electromagnetic induction heating controller. When the temperature sensor detects that the surface temperature of the mold reaches the preset demoulding temperature;

[0010] The ejection unit includes a main ejection cylinder and an auxiliary ejection cylinder. The main ejection cylinder is used to initially eject the pallet from the mold cavity. The auxiliary ejection cylinder is distributed at the key supporting parts of the pallet. It is activated after the main ejection cylinder has ejected a certain distance to provide auxiliary support and further eject the pallet.

[0011] The intelligent monitoring module includes an edge computing node processor, a cloud server and multiple monitoring sensors. The edge computing node processor is connected to multiple monitoring sensors to realize real-time collection and preliminary processing of monitoring data and transmit the processed monitoring data to the cloud server for fault diagnosis.

[0012] As a further preferred technical solution of the above technical solution, the workflow of the demoulding module is as follows:

[0013] When the pallet is formed in the mold, the electromagnetic induction heating controller starts, the electromagnetic induction coil generates an electromagnetic field, the mold surface temperature rises rapidly, and the temperature sensor monitors the temperature in real time. When the preset demoulding temperature is reached, the temperature is maintained stable for a preset time to fully reduce the adhesion between the plastic pallet and the mold;

[0014] The main ejection cylinder starts to eject the pallet from the mold cavity to a preset distance at the first speed. At this time, the auxiliary ejection cylinder starts to cooperate with the main ejection cylinder to continue ejecting the pallet. During the ejection process, the pneumatic proportional valve adjusts the ejection speed in real time according to the preset program until the pallet is completely out of the mold.

[0015] After demolding is completed, the surface of the mold is cleaned to remove residual plastic debris and impurities, the integrity of the super-hydrophobic nano-coating is checked, and maintenance treatment is performed if necessary before the next round of production.

[0016] As a further preferred technical solution of the above technical solution, for cloud servers, a distributed storage architecture is adopted to classify and store uploaded data, including real-time monitoring data, historical data, and equipment parameter data. Big data analysis technology is used to clean, integrate and analyze the collected data, and extract key feature parameters. At the same time, machine learning algorithms are used to deeply mine the data, establish an evaluation model for the equipment operation status, and realize real-time evaluation and diagnosis of the equipment health status.

[0017] As a further preferred technical solution of the above technical solution, the evaluation model is specifically implemented as follows:

[0018] First, collect equipment operation data from various monitoring sensors installed on production equipment to form raw time series data. Remove outliers and missing values ​​from the data, fill in missing values ​​using linear interpolation or time series-based prediction methods, and normalize the data.

[0019] Second, divide the preprocessed data into training set, validation set and test set;

[0020] Randomly initialize the weight matrix Wf, WC, Wi, Wo and bias items bf, bC, bi, bo of the LSTM network;

[0021] The training set data is input into the LSTM network in sequence according to the time step, and calculation is performed to obtain the hidden state ht and memory unit Ct of each time step;

[0022] The mean square error is used as the loss function to measure the difference between the model prediction value and the true value. The formula is: Where n is the number of samples, y i is the true value, is the model prediction value;

[0023] The gradient of the loss function with respect to the network parameters is calculated through the back-propagation algorithm, and the weight matrix and bias terms are updated using stochastic gradient descent or its improved algorithm, continuously reducing the loss function value until the model achieves optimal performance on the validation set.

[0024] Third, the pre-processed real-time monitoring data is input into the trained LSTM model. The model predicts the future operating status of the equipment based on the patterns learned from historical data. When the prediction results show that the equipment operating parameters exceed the normal range and reach the preset fault threshold, the equipment is judged to be at risk of failure and a maintenance warning is triggered.

[0025] As a further preferred technical solution of the above technical solution, maintenance warning information is sent to corresponding maintenance personnel, so that they can log in to the cloud management platform to view the detailed operation data and fault diagnosis report of the equipment, and formulate a maintenance plan based on the fault diagnosis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0027] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are for illustrative purposes only, and those skilled in the art will readily appreciate other obvious variations. The basic principles of the present invention defined in the following description may be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the present invention.

[0028] The present invention discloses a production device for a wear-resistant plastic pallet. The specific embodiments of the invention will be further described below in conjunction with preferred embodiments.

[0029] In the embodiments of the present invention, those skilled in the art will note that the tray and the like involved in the present invention may be considered as prior art.

[0030] Preferred embodiment.

[0031] like Figure 1As shown, the present invention discloses a production device for wear-resistant plastic pallets, which is used for the production of pallets and includes a demoulding module and an intelligent monitoring module, wherein:

[0032] The demoulding module includes a mold, an electromagnetic induction unit, a temperature sensing unit and an ejection unit, wherein:

[0033] For the mold, a super-hydrophobic nano-coating is prepared on the surface (the mold is placed in a vacuum reactor and a mixed gas containing raw materials such as silane and fluoride is introduced. Under the action of high temperature (about 200-300°C) and catalyst, the gas molecules react chemically on the mold surface to form a uniform and dense nano-scale coating, which greatly reduces the friction between the plastic and the mold surface);

[0034] For the electromagnetic induction unit, an electromagnetic induction coil is embedded in the mold cavity surface using a combination of spiral and array methods. (For areas with regular tray shapes and uniform force, a spiral coil layout is used to generate a uniform electromagnetic field, causing the mold surface temperature to rise evenly. For areas with complex shapes and greater demolding resistance, such as the edges and corners of the tray, an array coil layout is used. The heating intensity of different areas can be flexibly adjusted according to actual needs to ensure that the adhesion between the plastic and the mold can be effectively reduced in all areas.) The electromagnetic induction heating controller is used to control the electromagnetic induction coil to generate an electromagnetic field.

[0035] For the temperature sensing unit, a high-precision temperature sensor is installed to monitor the surface temperature of the mold in real time and work in conjunction with the electromagnetic induction heating controller. When the temperature sensor detects that the surface temperature of the mold has reached the preset demoulding temperature (set according to the characteristics of the plastic material, such as 80-100°C for polypropylene plastic), the electromagnetic induction heating controller automatically adjusts the current of the electromagnetic induction coil and maintains a stable temperature (to avoid damage to the mold or pallet due to excessive temperature, and to prevent demoulding difficulties due to insufficient temperature).

[0036] The ejection unit includes a main ejection cylinder and an auxiliary ejection cylinder. The main ejection cylinder is used to initially eject the pallet from the mold cavity. The auxiliary ejection cylinder is distributed at the key supporting parts of the pallet. After the main ejection cylinder has ejected a certain distance, it is activated to provide auxiliary support and further eject the pallet (to prevent deformation or breakage of the pallet due to uneven force during demolding).

[0037] The intelligent monitoring module includes an edge computing node processor, a cloud server and multiple monitoring sensors (installed in various production equipment, for example, a high-precision vibration sensor (such as a piezoelectric vibration sensor) is installed at the bearing of the extruder screw to monitor the vibration frequency and amplitude of the screw during rotation in real time, and determine whether the screw has problems such as wear and eccentricity; a pressure sensor is installed at the inlet and outlet of the hydraulic cylinder of the mold clamping mechanism to monitor the pressure changes during the clamping process to ensure that the clamping force meets the process requirements; a torque sensor is installed at the driving roller and tensioning roller of the conveyor belt to monitor the torque changes during the transmission process and promptly detect problems such as conveyor belt slippage and load abnormality). The edge computing node processor is connected to multiple monitoring sensors to realize real-time collection and preliminary processing of monitoring data and transmit the processed monitoring data to the cloud server for fault diagnosis.

[0038] Specifically, the workflow of the demoulding module is:

[0039] When the pallet is formed in the mold, the electromagnetic induction heating controller starts, the electromagnetic induction coil generates an electromagnetic field, the mold surface temperature rises rapidly, and the temperature sensor monitors the temperature in real time. When the preset demoulding temperature is reached, the temperature is maintained stable for a preset time to fully reduce the adhesion between the plastic pallet and the mold;

[0040] The main ejector cylinder starts and ejects the pallet from the mold cavity to a preset distance (5-10mm) at a first speed (slower). At this time, the auxiliary ejector cylinder starts and cooperates with the main ejector cylinder to continue ejecting the pallet. During the ejection process, the pneumatic proportional valve adjusts the ejection speed in real time according to the preset program until the pallet is completely separated from the mold (the pneumatic proportional valve accurately controls the air flow of the ejector cylinder to adjust the ejection speed. In the initial stage of demolding, the ejection speed is slow (such as 5-10mm / s) to avoid the impact force caused by the instantaneous separation of the pallet and the mold due to excessive speed; when the pallet is mostly out of the mold cavity, the ejection speed is appropriately increased (such as 15-20mm / s) to improve the demolding efficiency. The ejection speed can be flexibly adjusted according to the different specifications and thicknesses of pallets to ensure a smooth and efficient demolding process).

[0041] After demolding is completed, the surface of the mold is cleaned to remove residual plastic debris and impurities, and the integrity of the super-hydrophobic nano-coating is checked. If necessary, maintenance treatment is performed (clean the mold surface to remove surface dirt and impurities, and then locally repair the coating by spraying to ensure that the coating always maintains good hydrophobic properties, extending the demolding effect and service life of the mold), and then the next round of production is carried out.

[0042] More specifically, for cloud servers, a distributed storage architecture is adopted to classify and store uploaded data, including real-time monitoring data, historical data, and equipment parameter data. Big data analysis technology is used to clean, integrate and analyze the collected data, and extract key characteristic parameters (such as vibration characteristic values ​​of equipment operation, temperature change trends, pressure fluctuation range, etc.). At the same time, machine learning algorithms are used to deeply mine the data, establish an evaluation model for equipment operation status, and realize real-time evaluation and diagnosis of equipment health status.

[0043] Furthermore, the evaluation model is specifically implemented as follows:

[0044] First, collect equipment operation data from various monitoring sensors installed on production equipment (such as vibration sensors, temperature sensors, etc.) to form raw time series data. Remove outliers and missing values ​​from the data, fill in missing values ​​using linear interpolation or time series-based prediction methods, and normalize the data.

[0045] Second, divide the preprocessed data into training set, validation set and test set;

[0046] Randomly initialize the weight matrix Wf, WC, Wi, Wo and bias items bf, bC, bi, bo of the LSTM network;

[0047] The training set data is input into the LSTM network in sequence according to the time step, and calculation is performed to obtain the hidden state ht and memory unit Ct of each time step;

[0048] The mean square error (MSE) is used as the loss function to measure the difference between the model prediction value and the true value. The formula is: Where n is the number of samples, y i is the true value, is the model prediction value;

[0049] The gradient of the loss function with respect to the network parameters is calculated through the back-propagation algorithm, and the weight matrix and bias terms are updated using stochastic gradient descent (SGD) or its improved algorithm (such as the Adam optimization algorithm), continuously reducing the loss function value until the model achieves optimal performance on the validation set.

[0050] Third, the pre-processed real-time monitoring data is input into the trained LSTM model. The model predicts the future operating status of the equipment based on the patterns learned from historical data. When the prediction results show that the equipment operating parameters exceed the normal range and reach the preset fault threshold, the equipment is judged to be at risk of failure and a maintenance warning is triggered.

[0051] Furthermore, maintenance warning information will be sent to the corresponding maintenance personnel, who will then log in to the cloud management platform to view the detailed operating data and fault diagnosis reports of the equipment, and formulate maintenance plans based on the fault diagnosis results.

[0052] It is worth mentioning that the technical features such as the pallet involved in the patent application of this invention should be regarded as prior art. The specific structure, working principle and possible control method and spatial layout method of these technical features can be selected by conventional means in the field and should not be regarded as the inventive point of this patent. This patent will not be further elaborated.

[0053] For those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned embodiments, or to make equivalent replacements for some of the technical features therein. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A production device for wear-resistant plastic pallets, used for the production of pallets, characterized in that: It includes demoulding module and intelligent monitoring module, including: The demoulding module includes a mold, an electromagnetic induction unit, a temperature sensing unit and an ejection unit, wherein: For the mold, a superhydrophobic nanocoating is prepared on the surface; For the electromagnetic induction unit, an electromagnetic induction coil is embedded in the cavity surface of the mold in a combination of spiral and array types, and an electromagnetic induction heating controller is used to control the electromagnetic induction coil to generate an electromagnetic field; For the temperature sensing unit, a high-precision temperature sensor is installed to monitor the surface temperature of the mold in real time and work in conjunction with the electromagnetic induction heating controller. When the temperature sensor detects that the surface temperature of the mold reaches the preset demoulding temperature; The ejection unit includes a main ejection cylinder and an auxiliary ejection cylinder. The main ejection cylinder is used to initially eject the pallet from the mold cavity. The auxiliary ejection cylinder is distributed at the key supporting parts of the pallet. It is activated after the main ejection cylinder has ejected a certain distance to provide auxiliary support and further eject the pallet. The intelligent monitoring module includes an edge computing node processor, a cloud server and multiple monitoring sensors. The edge computing node processor is connected to multiple monitoring sensors to realize real-time collection and preliminary processing of monitoring data and transmit the processed monitoring data to the cloud server for fault diagnosis.

2. The production device of a wear-resistant plastic pallet according to claim 1, characterized in that: The workflow of the demoulding module is as follows: When the pallet is formed in the mold, the electromagnetic induction heating controller starts, the electromagnetic induction coil generates an electromagnetic field, the mold surface temperature rises rapidly, and the temperature sensor monitors the temperature in real time. When the preset demoulding temperature is reached, the temperature is maintained stable for a preset time to fully reduce the adhesion between the plastic pallet and the mold; The main ejection cylinder starts to eject the pallet from the mold cavity to a preset distance at the first speed. At this time, the auxiliary ejection cylinder starts to cooperate with the main ejection cylinder to continue ejecting the pallet. During the ejection process, the pneumatic proportional valve adjusts the ejection speed in real time according to the preset program until the pallet is completely out of the mold. After demolding is completed, the surface of the mold is cleaned to remove residual plastic debris and impurities, the integrity of the super-hydrophobic nano-coating is checked, and maintenance treatment is performed if necessary before the next round of production.

3. The production device of a wear-resistant plastic pallet according to claim 2, characterized in that: For cloud servers, a distributed storage architecture is adopted to classify and store uploaded data, including real-time monitoring data, historical data, and equipment parameter data. Big data analysis technology is used to clean, integrate and analyze the collected data, and extract key feature parameters. At the same time, machine learning algorithms are used to deeply mine the data, establish an evaluation model for the equipment operating status, and realize real-time evaluation and diagnosis of the equipment health status.

4. The production device of a wear-resistant plastic pallet according to claim 3, characterized in that: The specific implementation of the evaluation model is as follows: First, collect equipment operation data from various monitoring sensors installed on production equipment to form raw time series data. Remove outliers and missing values ​​from the data, fill in missing values ​​using linear interpolation or time series-based prediction methods, and normalize the data. Second, divide the preprocessed data into training set, validation set and test set; Randomly initialize the weight matrix Wf, WC, Wi, Wo and bias items bf, bC, bi, bo of the LSTM network; The training set data is input into the LSTM network in sequence according to the time step, and calculation is performed to obtain the hidden state ht and memory unit Ct of each time step; The mean square error is used as the loss function to measure the difference between the model prediction value and the true value. The formula is: Where n is the number of samples, y i is the true value, is the model prediction value; The gradient of the loss function with respect to the network parameters is calculated through the back-propagation algorithm, and the weight matrix and bias terms are updated using stochastic gradient descent or its improved algorithm, continuously reducing the loss function value until the model achieves optimal performance on the validation set. Third, the pre-processed real-time monitoring data is input into the trained LSTM model. The model predicts the future operating status of the equipment based on the patterns learned from historical data. When the prediction results show that the equipment operating parameters exceed the normal range and reach the preset fault threshold, the equipment is judged to be at risk of failure and a maintenance warning is triggered.

5. The production device of a wear-resistant plastic pallet according to claim 4, characterized in that: The maintenance warning information is sent to the corresponding maintenance personnel, who can then log in to the cloud management platform to view the detailed operating data and fault diagnosis report of the equipment, and formulate a maintenance plan based on the fault diagnosis results.

Images