A liquid silicone plastic molding system with micro-adjustment and method of operating the same

Abstract

The present application relates to liquid silicone plastic forming technology field, disclose a kind of liquid silicone plastic forming system with trace adjustment and its operating method, it includes: injection molding machine, upper forming die, lower forming die, adjusting knob, rotating rod, connecting rod gear set, cylinder, valve needle, glue guide pipe and controller, injection molding machine is equipped with controller, upper forming die and lower forming die are installed on injection molding machine, form forming die cavity, upper forming die is equipped with injection port, injection port is communicated with forming die cavity by glue guide pipe, rotating rod is rotatably installed in upper forming die, one end of rotating rod extends out of upper forming die and is equipped with adjusting knob, the other end of rotating rod is connected with one end of connecting rod gear set, the other end of connecting rod gear set is connected with piston in cylinder, cylinder is installed in upper forming die, valve needle is installed on cylinder, and valve needle is arranged in glue guide pipe inside;By the design of plastic forming system, intelligent control, self-adapting environment, strong perception, precision adjustment smooth trace injection molding processing can be realized.

Description

Technical Field

[0001] This invention relates to the field of plastic molding technology, and more specifically to a liquid silicone plastic molding system with micro-adjustment and its operating method. Background Technology

[0002] Plastic injection molding is a primary method for producing precision plastic products, widely used in electronics, medical, and automotive industries. In high-precision injection molding, micro-volume injection control is a key technology for ensuring product quality and for processing micro-sized plastic parts. Existing injection molding systems primarily regulate the injection volume into the mold's sprue using external injectors. However, the sprue size is relatively large, making it difficult to form micro-injection molded parts. The reasons for this include: limited adjustment precision (external injectors are large devices with standard large-size injection pipes, making it difficult for the control valves to achieve precise adjustments, resulting in significant errors); poor or nonexistent self-adaptability (inability to adapt to significant differences in the rheological properties of different plastic materials, hindering timely adjustments to the injection volume for better micro-injection processing); low or nonexistent intelligent control, especially for micro-injection process control, failing to learn and optimize process control from production data; and the lack of status sensing during micro-injection, making accurate status monitoring and fault prediction difficult.

[0003] Therefore, in order to address the shortcomings of existing technologies, it is necessary to develop a new type of plastic molding system that integrates precision mechanical adjustment, multimodal intelligent sensing, and autonomous learning control, which has significant technological value and broad market prospects. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing injection molding technologies, such as the inability to achieve micro-injection molding, limited adjustment precision in micro-injection molding, poor or no adaptive capability, low level of intelligent control or inability to achieve intelligent control, inability to perceive the state of micro-injection, and difficulty in achieving accurate state monitoring and fault prediction. Through the design of a plastic molding system, employing an injection molding machine, upper mold, lower mold, adjustment knob, rotating rod, connecting rod and gear set, cylinder, valve needle, guide tube, and controller, as well as a control system including a self-learning control module, drive control module, signal acquisition module, and control processing module, and corresponding control methods, this invention enables intelligent control, environmental adaptability, strong perception, and smooth precision adjustment in micro-injection molding.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A liquid silicone plastic molding system with micro-adjustment is disclosed. The system includes an injection molding machine, an upper mold, a lower mold, an adjustment knob, a rotating rod, a connecting rod and gear set, a cylinder, a valve needle, a guide tube, and a controller. The controller is mounted on the injection molding machine. Both the upper and lower molds are mounted on the injection molding machine and are mutually fitted to form multiple molding cavities. An injection port is provided on the top of the upper mold, and the injection port communicates with the molding cavities through the guide tube. The rotating rod is rotatably installed inside the upper molding mold, with one end of the rotating rod extending out of the upper molding mold and equipped with the adjusting knob. The other end of the rotating rod is connected to one end of the connecting rod gear set, and the other end of the connecting rod gear set is connected to the piston inside the cylinder. The cylinder is installed inside the upper molding mold, and the valve needle is installed on the cylinder. The valve needle is located inside the glue guide tube and is used to adjust the glue flow rate of the glue guide tube. The cylinder is used to drive the reciprocating motion of the valve needle, and the piston is used to adjust the space inside the cylinder. The control terminal of the cylinder is electrically connected to the control terminal of the controller, and the controller is equipped with a control system for controlling the cylinder. The rotating rod, connecting rod gear set, cylinder and valve needle are arranged in an inverted L-shape. The rotating rod is located on the horizontal side of the inverted L-shape, the connecting rod gear set is located at the corner of the inverted L-shape, and the cylinder and valve needle are located on the vertical side of the inverted L-shape. The number of the adjusting knob, rotating rod, connecting rod gear set, cylinder, valve needle and guide tube is the same as the number of molding cavity; The number of molding cavities can be selected as 16 or 32, etc.

[0006] The control system includes a self-learning control module, a drive control module, a signal acquisition module, and a control processing module. The self-learning terminal of the control processing module is electrically connected to the self-learning terminal of the self-learning control module. The drive terminal of the control processing module is electrically connected to the control terminal of the drive control module. The drive terminal of the drive control module is electrically connected to the control terminal of the cylinder. The acquisition terminal of the control processing module is electrically connected to the acquisition signal terminal of the signal acquisition module.

[0007] Furthermore, the connecting rod gear set includes a worm gear and a telescopic rod, the worm gear meshing with each other, the other end of the worm gear being connected to the rotating rod, the telescopic rod being connected to the worm gear, and the telescopic end of the telescopic rod being connected to the piston. The cooperation between the telescopic rod and the piston is used to control the valve needle stroke to control the amount of glue injected into each molding cavity. The worm gear is located on the horizontal side of the inverted L-shape, and the telescopic rod is located on the vertical end of the inverted L-shape.

[0008] Furthermore, the telescopic rod is a threaded telescopic rod, which includes a fixed part and a telescopic part. One end of the fixed part is connected to the worm gear, and the other end of the fixed part is threaded to the telescopic part. The other end of the telescopic part is connected to the piston, and the telescopic part is vertically slidably installed in the upper forming mold.

[0009] Furthermore, the piston includes a motorized piston and a valve needle piston, with a cylinder cavity space between the motorized piston and the valve needle piston. The motorized piston and the valve needle piston are disposed at both ends of the cylinder. The motorized piston is connected to the other end of the telescopic part, and the valve needle piston is connected to the valve needle.

[0010] Furthermore, the cylinder is selected as a dual-control cylinder or a single-control cylinder.

[0011] Furthermore, the self-learning control module is equipped with a convolutional neural network model, a recurrent neural network model, a deep reinforcement learning convolutional network model, a memory-enhanced network model, and a graph neural network model. The signal acquisition module transmits the acquired signals to the convolutional neural network model for feature vector processing through the control processing module. The convolutional neural network model then transmits the signals to the recurrent neural network model for context feature vector processing. Finally, the signals are processed by the deep reinforcement learning convolutional network model, which has a memory-enhanced network model and a graph neural network model working together, before being sent to the control processing module.

[0012] Furthermore, the signal acquisition module includes a position sensor, a temperature sensor, and a flow sensor. The position sensor is mounted on the cylinder, the temperature sensor is disposed on the side wall of the glue guide tube, and the flow sensor is disposed at the glue outlet end of the glue guide tube. The position sensor, temperature sensor, and flow sensor are electrically connected to the acquisition signal terminal of the control processing module.

[0013] Furthermore, the position sensor is selected as a magnetostrictive position sensor, which includes a measuring magnetic ring and a sensor body. The measuring magnetic ring is installed inside the piston of the cylinder, and the sensor body is installed outside the cylinder body. The position sensor is used to measure the precise position of the piston in real time.

[0014] The measuring magnetic ring is interference-fitted and installed inside the piston; the sensor body is installed in the cylinder mounting groove and maintains a gap of 1.5mm-3.5mm with the measuring magnetic ring; the position sensor measures the piston position in real time with an accuracy of ±0.005mm-±0.05mm.

[0015] Furthermore, the temperature sensor is a resistive temperature sensor, which is installed on the outer wall of the tube. The temperature sensor is used to monitor the working temperature of the tube.

[0016] The resistive temperature sensor is a PT100 platinum resistance sensor, which is installed in the middle of the tube via a threaded connection, and its measurement range is 0℃-280℃.

[0017] Furthermore, the flow sensor is selected as a Coriolis mass flow meter, and the flow sensor is installed at the outlet end of the adhesive tube. The flow sensor is used to detect the outflow status of liquid silicone plastic, etc.

[0018] Furthermore, the liquid silicone plastic molding system can achieve a manual control mode, which involves mechanical adjustment via the adjustment knob and linkage gear set to control the piston position within the cylinder, thereby changing the space within the cylinder. This is a coarse mechanical adjustment operation, allowing for quick adjustment of the space within the cylinder. If precise adjustment of the cylinder's regulating valve needle to release the plastic flow from the outlet of the guide tube is not required, coarse mechanical adjustment can be performed to change the cylinder space. The valve needle is slightly adjusted without fine adjustment, meaning the cylinder space is adjusted mechanically while the air volume remains almost constant, which is then transmitted to the valve needle to make corresponding changes, thereby adjusting the enlargement or reduction of the outlet. An automatic control mode is achieved through the self-learning control module, which enables intelligent adjustment. This means that the mechanical coarse adjustment remains constant, and the valve needle is adjusted according to the air volume difference in the cylinder to adjust the enlargement or reduction of the outlet. Fine adjustment can be achieved through the self-learning control module. A semi-automatic control mode is also available, combining manual setting and automatic fine adjustment. This means that mechanical coarse adjustment and cylinder fine adjustment work together to achieve both rapid and precise adjustment.

[0019] Furthermore, the liquid silicone plastic molding system also includes a safety protection module electrically connected to the control processing module. This safety protection module identifies position, temperature, and flow information thresholds from the position, temperature, and flow sensors. When it detects out-of-limit position, temperature exceeding the threshold, flow exceeding the limit, or an abnormality in the position, temperature, or flow sensors, it automatically shuts off the cylinder's air supply and issues an alarm. A buzzer alarm can be installed on the controller for this purpose.

[0020] Furthermore, a method for controlling a liquid silicone plastic molding system with micro-adjustment, the method comprising the following steps: Step 1: Formation of micro-molding mold The system comprises an injection molding machine, an upper mold, a lower mold, an adjustment knob, a rotating rod, a connecting rod gear set, a cylinder, a valve needle, a guide tube, and a controller. The controller is installed on the injection molding machine. Both the upper and lower molds are mounted on the injection molding machine and fit together to form multiple molding cavities. An injection port is provided on the top of the upper mold, which communicates with the molding cavities through the guide tube. The rotating rod is rotatably mounted inside the upper mold, with one end extending out of the upper mold and equipped with the adjustment knob. The other end of the rotating rod is connected to one end of the connecting rod gear set, and the other end of the connecting rod gear set is connected to a piston inside the cylinder. The cylinder is installed inside the upper mold and is equipped with the valve needle, which is located inside the guide tube and is used to adjust the flow rate of the adhesive through the guide tube. The control terminal of the cylinder is electrically connected to the control terminal of the controller, and the controller is equipped with a control system for controlling the cylinder. The rotating rod, connecting rod gear set, cylinder and valve needle are arranged in an inverted L-shape. The rotating rod is located on the horizontal side of the inverted L-shape, the connecting rod gear set is located at the corner of the inverted L-shape, and the cylinder and valve needle are located on the vertical side of the inverted L-shape. The number of the adjusting knob, rotating rod, connecting rod gear set, cylinder, valve needle and guide tube is the same as the number of molding cavity; Step 2: Installation of the signal acquisition module: Based on step 1, a signal acquisition module and a control processing module are selected. The signal acquisition module includes a position sensor, a temperature sensor, and a flow sensor. The position sensor is installed on the cylinder, the temperature sensor is set on the side wall of the cylinder, and the flow sensor is set at the dispensing end of the glue guide tube. The position sensor, temperature sensor, and flow sensor are electrically connected to the acquisition signal terminal of the control processing module. The position sensor is selected as a magnetostrictive position sensor, which includes a measuring magnetic ring and a sensor body. The measuring magnetic ring is installed inside the piston of the cylinder, and the sensor body is installed outside the cylinder body. The position sensor is used to measure the precise position of the piston in real time. The temperature sensor is a resistive temperature sensor, which is installed on the outer wall of the tube. The temperature sensor is used to monitor the working temperature of the tube. The flow sensor is a Coriolis mass flow meter, which is installed at the outlet end of the adhesive tube and is used to detect the outflow status of liquid silicone plastic. Step 3: Setting up the control system Based on steps 1 and 2, the controller is equipped with a control system, which is used to control the cylinder. The control system includes a self-learning control module, a drive control module, a signal acquisition module, and a control processing module. The self-learning terminal of the control processing module is electrically connected to the self-learning terminal of the self-learning control module. The drive terminal of the control processing module is electrically connected to the control terminal of the drive control module. The drive terminal of the drive control module is electrically connected to the control terminal of the cylinder. The acquisition terminal of the control processing module is electrically connected to the acquisition signal terminal of the signal acquisition module. The self-learning control module is equipped with a convolutional neural network model, a recurrent neural network model, a deep reinforcement learning convolutional network model, a memory-enhanced network model, and a graph neural network model. The signal acquisition module transmits the acquired signals to the convolutional neural network model for feature vector processing through the control processing module. The convolutional neural network model then transmits the signals to the recurrent neural network model for context feature vector processing. Finally, the signals are processed by the deep reinforcement learning convolutional network model, which has a memory-enhanced network model and a graph neural network model working together, before being sent to the control processing module. Beneficial effects

[0021] This invention organically combines precision mechanical design, multi-sensor fusion, and intelligent control. Through the design of a plastic molding system, it employs an injection molding machine, upper mold, lower mold, adjustment knob, rotating rod, connecting rod and gear set, cylinder, valve needle, guide tube, and controller. The control system includes a self-learning control module, drive control module, signal acquisition module, and control processing module, along with corresponding control methods. This enables intelligent control, environmental adaptability, strong sensing capabilities, and smooth precision adjustment in micro-injection molding. It achieves improvements in precision, efficiency, quality, and intelligence, solving key technical problems that have long plagued the precision injection molding industry. It provides reliable technical support for the manufacture of small plastic products and possesses significant technological advancement, economic practicality, and industry promotion value. Specifically, the piston stroke can be changed through mechanical external adjustment knobs, thereby altering the flow rate of plastic material at the valve needle via the cylinder, achieving precise injection molding. Furthermore, the use of multiple sensors and the control system enables real-time acquisition of the plastic material flow status, allowing for self-learning and self-adjustment, achieving intelligent control of plastic material flow during injection molding. It solves the long-standing industry pain points in precision injection molding, such as long debugging time, parameter dependence on experience, and difficulty in quality control. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the structure of a liquid silicone plastic molding system with micro-adjustment according to the present invention.

[0023] Figure 2 This is a partial structural schematic diagram of a liquid silicone plastic molding system with micro-adjustment according to the present invention.

[0024] Figure 3 This is a schematic diagram of the internal partial structure of a liquid silicone plastic molding system with micro-adjustment according to the present invention.

[0025] Figure 4 This is a schematic diagram of the cross-sectional structure of a cylinder in a liquid silicone plastic molding system with micro-adjustment according to the present invention.

[0026] Figure descriptions: 1. Injection gate; 2. Upper mold; 3. Lower mold; 01. Adjustment knob; 02. Rotating rod; 03. Connecting rod and gear set; 04. Cylinder; 05. Telescopic rod; 06. Glue guide tube; 07. Valve needle; 41. Motorized piston; 42. Valve needle piston. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] This invention provides a liquid silicone plastic molding system with micro-adjustment. The system includes an injection molding machine, an upper mold 2, a lower mold 3, an adjustment knob 01, a rotating rod 02, a connecting rod and gear set 03, a cylinder 04, a valve needle 07, a guide tube 06, and a controller. The controller is mounted on the injection molding machine. Both the upper mold 2 and the lower mold 3 are mounted on the injection molding machine and are mutually fitted, forming multiple molding cavities. These molding cavities receive micro-volume plastic material from the guide tube 06 and mold tiny plastic products within them. An injection port 1 is located at the top center of the upper molding mold 2, enabling the production of multiple plastic products from a single injection, facilitating mass production. The injection port 1 is connected to the molding cavity via a guide tube 06, forming a channel for the injection of plastic material into the molding cavity. A rotating rod 02 is rotatably mounted inside the upper molding mold 2, with one end extending out and equipped with an adjustment knob 01. The adjustment knob 01 has a scale for recording the rotation increments. The other end of the rotating rod 02 is connected to one end of a connecting rod gear set 03. The piston inside cylinder 04 is connected to the end of the cylinder. Cylinder 04 is installed inside the upper molding mold 2. A valve needle 07 is mounted on cylinder 04 and is located inside the glue guide tube 06. The valve needle 07 is used to adjust the glue flow rate of the glue guide tube 06. Cylinder 04 drives the reciprocating motion of valve needle 07, and the piston is used to adjust the internal space of cylinder 04. Valve needle 07 is made of YG8 cemented carbide and coated with a 3-4 μm thick DLC diamond-like carbon coating. The tip of the valve needle is either conical or elliptical. For conical tips, the included angle is 20°-30°. The included angle at the major axis of the elliptical conical needle tip is 30°-40°, and the included angle at the minor axis is 20°-30°. The elliptical conical needle tip optimizes the flow of the plastic fluid, ensuring proper dispersion of its components and preventing aggregation that could negatively impact the formation of the injection molded part. The outlet of the guide tube is also designed to be frustum-shaped, with its smaller end connecting to the mold cavity. The size of the valve needle is larger than the smaller end of the frustum-shaped outlet. The flow rate of the plastic material is controlled by adjusting the distance between the needle tip and the smaller end. The control terminal of cylinder 04 is electrically connected to the control terminal of the controller, which is equipped with a control system for controlling cylinder 04. The rotating rod 02, the connecting rod gear set 03, the cylinder 04 and the valve needle 07 are arranged in an inverted L-shape. The rotating rod 02 is located on the horizontal side of the inverted L-shape, the connecting rod gear set 03 is located on the corner of the inverted L-shape, and the cylinder 04 and the valve needle 07 are located on the vertical side of the inverted L-shape. The number of the adjusting knob 01, rotating rod 02, connecting rod gear set 03, cylinder 04, valve needle 07, and guide tube 06 is the same as the number of molding cavities; the number of molding cavities can be selected as 16 or 32, etc. The piston includes a motor piston 41 and a valve needle piston 42. There is a cylinder cavity space between the motor piston 41 and the valve needle piston 42. The motor piston 41 and the valve needle piston 42 are located at both ends of the cylinder. The motor piston 41 is connected to the other end of the telescopic part, and the valve needle piston 42 is connected to the valve needle. The motorized piston and valve needle 07 are independent components. The valve needle 07 has a valve needle piston at its end. The motorized piston and valve needle piston are located at both ends of the cylinder, forming a cylinder cavity between them. The end of the cylinder (at the valve needle piston) is sealed with a guide tube 06. The valve needle piston is connected to a valve needle. The area between the valve needle piston and the valve needle tip is where liquid silicone plastic flows. Liquid silicone plastic from the injection port flows into the guide tube below the valve needle piston through the injection guide tube. The cylinder cavity is changed by coarse mechanical adjustment of the motorized piston, and then the valve needle piston is driven by fine cylinder adjustment. This creates clearance channels of different sizes at the outlet end of the guide tube (at the valve needle tip), allowing liquid silicone plastic to flow into the molding cavity, thus achieving micro-control. The guide tube uses a T-shaped three-way configuration, meaning that the vertical end of the guide tube is open at both ends, with one end connected to the cylinder and the other end connected to the molding cavity. The horizontal end of the guide tube extends to the injection port. This horizontal end is located at the lowest point of the valve needle piston stroke at the entrance of the vertical guide tube, ensuring that the valve needle piston does not affect the flow of liquid silicone. The valve needle is located vertically within the guide tube. The connecting rod gear set 03 includes a worm gear and a telescopic rod 05. The worm gear and worm gear mesh with each other. The other end of the worm gear is connected to a rotating rod 02. The telescopic rod 05 is connected to the worm gear, and the telescopic end of the telescopic rod 05 is connected to a piston. That is, the telescopic rod 05 and the rotating rod 02 are set at a 90° angle. The telescopic rod 05 is a threaded telescopic rod, which includes a fixed part and a telescopic part. One end of the fixed part is connected to the worm gear, and the other end of the fixed part is threaded to the telescopic part. The other end of the telescopic part is connected to the piston. The telescopic part is slidably installed in the upper molding mold 2. That is, turning the adjustment knob will drive the rotating rod to rotate, the rotating rod will drive the worm gear to rotate, the worm gear meshes with the worm gear to rotate, changing from horizontal to vertical. The worm gear drives the fixed rod to rotate, and the fixed part performs the up-and-down reciprocating motion of the telescopic part through the thread. The cylinder 04 is a double-controlled cylinder or a single-controlled cylinder, which can realize precise and repeatable reciprocating linear motion, which helps to accurately adjust the valve needle so as to realize the micro-flow of plastic material into the molding cavity. The control system includes a self-learning control module, a drive control module, a signal acquisition module, and a control processing module. The self-learning end of the control processing module is electrically connected to the self-learning end of the self-learning control module, the drive end of the control processing module is electrically connected to the control end of the drive control module, the drive end of the drive control module is electrically connected to the control end of the cylinder 04, and the acquisition end of the control processing module is electrically connected to the acquisition signal end of the signal acquisition module.

[0029] The self-learning control module employs a hierarchical, cascaded multi-model fusion architecture. This module incorporates a convolutional neural network model with medium-sized kernels, a recurrent neural network model, a deep reinforcement learning convolutional network model, a memory-enhanced network model, and a graph neural network model. The signal acquisition module preprocesses and distributes the acquired signals through the control processing module, then transmits them to the convolutional neural network model with medium-sized kernels. This model performs medium-range local feature extraction in the temporal domain, obtaining local feature sequence representation information. For example, position information S acquired by a position sensor is a sequence of position points in the temporal domain. After encoding each position point into a feature vector, the signal is transmitted... A convolutional neural network model with medium-sized kernels can be used to obtain local feature sequence representations from the feature vectors of these location information. For example, temperature information T collected by a temperature sensor is a time-domain sequence of temperature points. After encoding each temperature point as a feature vector, a convolutional neural network model with medium-sized kernels can be used to obtain local feature sequence representations from these temperature information. Similarly, flow information Q collected by a flow sensor is a time-domain sequence of flow points. After encoding each flow point as a feature vector, a convolutional neural network model with medium-sized kernels can be used to obtain local feature sequence representations from these flow information. The S, T, and Q information can be encoded into an N-dimensional feature vector S. i T i Q i For example, N is 64-dimensional, and M time-domain sequences are selected (M can be 50, etc.) to form an M×N feature matrix. Each feature matrix is ​​then integrated into a convolutional neural network model with a medium-sized convolutional kernel to form a local feature sequence F. S F T F QThen, the data is processed by a recurrent neural network model and concatenated according to feature dimensions to form a fused feature sequence. The feature representation H with temporal context information is extracted. The recurrent neural network model captures the temporal dependencies and dynamic evolution in the signal through its recurrent structure, outputting a feature representation with temporal context information. These temporal feature representations are sent to a high-level decision module. This high-level decision module uses a deep reinforcement learning convolutional network model as its core, in conjunction with a memory enhancement network model and a graph neural network model, to output feature representations of control commands. The memory enhancement network model retrieves relevant historical experience by querying external and internal memory matrices and fuses the retrieval results with the current features to form enhanced features with historical context and reasoning capabilities. These enhanced features are input to a graph neural network model. In this model, the system constructs a graph structure of features and establishes interactive relationships between them, ultimately outputting a holistic representation with structured semantics for use by control commands.

[0030] The signal acquisition module includes a position sensor, a temperature sensor, and a flow sensor. The position sensor is mounted on the cylinder 04, the temperature sensor is located on the side wall of the glue guide tube 06, and the flow sensor is located at the glue outlet end of the glue guide tube 06. The position sensor, temperature sensor, and flow sensor are electrically connected to the acquisition signal terminal of the control processing module.

[0031] Among them, the position sensor is a magnetostrictive position sensor, which includes a measuring magnetic ring and a sensor body. The measuring magnetic ring is installed inside the piston of cylinder 04, and the sensor body is installed on the outside of the cylinder body of cylinder 04. The position sensor is used to measure the precise position of the piston in real time.

[0032] The measuring magnetic ring is interference-fitted and installed inside the piston; the sensor body is installed in the cylinder mounting slot of cylinder 04 and maintains a gap of 1.5mm-3.5mm with the measuring magnetic ring; the position sensor measures the piston position in real time with an accuracy of ±0.005mm-±0.05mm.

[0033] Among them, a resistive temperature sensor is selected. The temperature sensor is installed on the outer wall of the tube 06. The temperature sensor is used to monitor the working temperature of the tube 06 and also to monitor the temperature of the liquid silicone.

[0034] The resistive temperature sensor is a PT100 platinum resistance sensor, which is installed in the middle of the tube 06 via a threaded connection, and its measurement range is 0℃-280℃.

[0035] The flow sensor is a Coriolis mass flow meter, which is installed at the outlet end of the adhesive tube 06. The flow sensor is used to detect the outflow status of liquid silicone plastic.

[0036] Specifically, the position sensor collects the displacement stroke, displacement amount, rate of change, and start and end position information of the piston (including the motorized piston and the valve needle piston), and transmits this data to a convolutional neural network model with a medium-sized convolutional kernel to output a local position feature vector; the temperature sensor collects the temperature information data of the guide tube, and transmits this data to a convolutional neural network model with a medium-sized convolutional kernel to output a local temperature feature vector; the flow sensor collects the flow rate information of the liquid silicone at the outlet of the guide tube, and transmits this data to a convolutional neural network model with a medium-sized convolutional kernel to output a local flow rate feature vector; the local position feature vector, local temperature feature vector, and local flow rate feature vector are then transmitted to a recurrent neural network model to output a comprehensive contextual feature, which captures the cross-modal temporal dependencies and outputs a unified comprehensive contextual feature containing dynamic evolution information. This feature can characterize the current injection... The injection process is described in several ways. The piston position reflects the space of the liquid silicone within the guide tube, the guide tube temperature reflects the temperature of the liquid silicone, and the flow rate at the guide tube outlet reflects the flow rate of the liquid silicone injected into the molding cavity. The batch of comprehensive contextual features processed above are passed to a deep reinforcement learning convolutional network model, which, along with a memory reinforcement network model and a graph neural network model, performs the processing. By observing the current state, actions corresponding to the features are selected, and the feature data is executed. New feature data is generated through the physical environment. A convolutional neural network model with medium-sized convolutional kernels and a recurrent neural network model process this data to obtain the features of the new state. The features of the current state and the new state are stored in system memory for data aggregation, and then updated and adjusted. A deep network trained through reinforcement learning is formed, performing real-time calculations and inferences, ultimately generating the optimal control commands (how to adjust piston speed, piston position, temperature, flow rate, etc.) – a decision-making process. The liquid silicone plastic molding system can be configured in a manual control mode, where mechanical adjustment is achieved through the adjustment knob and linkage gear set. This controls the piston position within the cylinder, changing the space within the cylinder. This is a coarse mechanical adjustment, allowing for quick adjustment of the cylinder's space. If precise adjustment of the plastic flow at the outlet of the guide tube is not required, coarse mechanical adjustment can be performed to change the cylinder space. The valve needle is slightly adjusted, without fine adjustment. In this coarse mechanical adjustment, the cylinder's air volume remains almost constant, which is then transmitted to the valve needle to adjust the size of the outlet. The automatic control mode utilizes a self-learning control module for intelligent adjustment. While the coarse mechanical adjustment remains constant, the valve needle is adjusted based on differences in cylinder air volume to adjust the size of the outlet. This allows for fine adjustment. A semi-automatic control mode combines manual setting and automatic fine adjustment, allowing for both rapid and precise adjustments.

[0037] The liquid silicone plastic molding system also includes a safety protection module electrically connected to the control processing module. This module identifies position, temperature, and flow information thresholds from the position, temperature, and flow sensors. When it detects out-of-limit position, temperature exceeding a threshold, flow exceeding a limit, or an abnormality in the position, temperature, or flow sensors, it automatically shuts off the cylinder's air supply and triggers an alarm. A buzzer alarm can be installed on the controller for this purpose.

[0038] One method for controlling a liquid silicone plastic molding system with micro-adjustment includes the following steps: Step 1: Formation of micro-molding mold The system comprises an injection molding machine, an upper mold 2, a lower mold 3, an adjustment knob 01, a rotating rod 02, a connecting rod and gear set 03, a cylinder 04, a valve needle 07, a guide tube 06, and a controller. The controller is installed on the injection molding machine. Both the upper mold 2 and the lower mold 3 are mounted on the injection molding machine and are mutually fitted, forming multiple molding cavities. An injection port 1 for liquid silicone is provided on the top of the upper mold 2. The injection port 1 communicates with the molding cavities through the guide tube 06. A rotating rod 02 is rotatably installed inside the upper molding mold 2, with one end of the rotating rod 02 extending out of the upper molding mold 2 and equipped with the adjusting knob 01. The other end of the rotating rod 02 is connected to one end of the connecting rod gear set 03, and the other end of the connecting rod gear set 03 is connected to the piston inside the cylinder 04. The cylinder 04 is installed inside the upper molding mold 2, and a valve needle 07 is installed on the cylinder 04. The valve needle 07 is located inside the glue guide tube 06 and is used to adjust the glue flow rate of the glue guide tube 06. The control terminal of the cylinder 04 is electrically connected to the control terminal of the controller, and the controller is equipped with a control system, which is used to control the cylinder 04. The rotating rod 02, the connecting rod gear set 03, the cylinder 04 and the valve needle 07 are arranged in an inverted L-shape. The rotating rod 02 is located on the horizontal side of the inverted L-shape, the connecting rod gear set 03 is located at the corner of the inverted L-shape, and the cylinder 04 and the valve needle 07 are located on the vertical side of the inverted L-shape. The number of the adjusting knob 01, rotating rod 02, connecting rod gear set 03, cylinder 04, valve needle 07 and guide tube 06 is the same as the number of molding cavities; Step 2: Installation of the signal acquisition module: Based on step 1, a signal acquisition module and a control processing module are selected. The signal acquisition module includes a position sensor, a temperature sensor, and a flow sensor. The position sensor is installed on the cylinder 04, the temperature sensor is set on the side wall of the glue guide tube 06, and the flow sensor is set at the glue outlet end of the glue guide tube 06. The position sensor, temperature sensor, and flow sensor are electrically connected to the acquisition signal terminal of the control processing module. The position sensor is selected as a magnetostrictive position sensor, which includes a measuring magnetic ring and a sensor body. The measuring magnetic ring is installed inside the piston of the cylinder 04, and the sensor body is installed outside the cylinder body of the cylinder 04. The position sensor is used to measure the precise position of the piston in real time. The temperature sensor is a resistive temperature sensor, which is installed on the outer wall of the tube 06. The temperature sensor is used to monitor the working temperature of the tube 06. The flow sensor is a capacitive flow sensor, which is installed at the outlet end of the adhesive tube 06 and is used to detect the flow status of liquid silicone plastic. Step 3: Setting up the control system Based on steps 1 and 2, the controller is equipped with a control system, which is used to control cylinder 04. The control system includes a self-learning control module, a drive control module, a signal acquisition module, and a control processing module. The self-learning terminal of the control processing module is electrically connected to the self-learning terminal of the self-learning control module. The drive terminal of the control processing module is electrically connected to the control terminal of the drive control module. The drive terminal of the drive control module is electrically connected to the control terminal of the cylinder 04. The acquisition terminal of the control processing module is electrically connected to the acquisition signal terminal of the signal acquisition module. The self-learning control module is equipped with a convolutional neural network model, a recurrent neural network model, a deep reinforcement learning convolutional network model, a memory-enhanced network model, and a graph neural network model. The signal acquisition module transmits the acquired signals to the convolutional neural network model for feature vector processing through the control processing module. The convolutional neural network model then transmits the signals to the recurrent neural network model for context feature vector processing. Finally, the signals are processed by the deep reinforcement learning convolutional network model, which has a memory-enhanced network model and a graph neural network model working together, before being sent to the control processing module.

[0039] Specifically, for the processing and molding of micro-sized liquid silicone plastic products, the mold is placed on an injection molding machine, the injection nozzle is aligned with and connected to the plastic material output port, and mechanical transmission is achieved through an adjustment knob to the motorized piston. The motorized piston coarsely adjusts the internal cavity space of the cylinder, and the controller operates the cylinder to adjust the valve needle with a valve needle piston. The valve needle adjusts the space size at the outlet of the guide tube, which is equivalent to adjusting the flow rate of the liquid silicone plastic to meet the injection molding parameter requirements of the product. The position information of the motorized piston and the valve needle piston is collected and stored by a position sensor, the temperature information of the guide tube is collected and stored by a temperature sensor, and the flow rate information of the outlet is collected and stored by a flow sensor. The process involves obtaining suitable parameters for a given product, within a specific range. For another product specification, the same information gathering process is required to establish parameters within the same range. This process continues, collecting demand information for numerous products. Combined with self-learning capabilities, this information is stored and processed for rapid application later. Mechanical adjustment serves as coarse adjustment, while cylinders provide fine adjustment. When processing micro-sized products, both mechanisms work together to achieve the required processing parameters. Since different auxiliary fillers exist in liquid silicone plastic, they affect the liquid silicone plastic itself, resulting in different injection molding parameters. Intelligent processing of these parameters facilitates the injection molding of micro-sized products. For example, to injection mold a 3mm x 5mm silicone plastic product, with the injection mold and injection molding machine ready, the mechanical coarse adjustment is performed by adjusting the adjustment knob to the motorized piston. This coarse adjustment changes the cylinder cavity, and the controller then operates the cylinder's valve piston and valve needle. Here, three sensors collect information such as the position of the motorized piston, the position of the valve needle piston, the temperature of the guide tube, and the flow rate of the silicone material. If the injection molded product meets the specifications and quality, then this parameter is the optimal parameter. Changing to a different liquid silicone will produce another optimal parameter, and so on, forming a vast amount of data parameters. Furthermore, past experience can be stored in the controller's control system as a self-learning capability. Through this intelligent design, different mechanical coarse adjustments can optimize the processing of small products and improve their quality for different products.

[0040] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A liquid silicone plastic molding system with micro-adjustment, characterized in that, The liquid silicone plastic molding system includes an injection molding machine, an upper molding die (2), a lower molding die (3), an adjustment knob (01), a rotating rod (02), a connecting rod and gear set (03), a cylinder (04), a valve needle (07), a guide tube (06), and a controller. The controller is installed on the injection molding machine. The upper molding die (2) and the lower molding die (3) are both installed on the injection molding machine, and the upper molding die (2) and the lower molding die (3) fit together to form multiple molding cavities. An injection port (1) is provided on the top of the upper molding die (2). The injection port (1) is connected to the molding cavity through the guide tube (06). The rotating rod (02) is rotatably mounted on the upper molding die (2). Inside the upper molding mold (2), one end of the rotating rod (02) extends out of the upper molding mold (2) and is equipped with the adjusting knob (01). The other end of the rotating rod (02) is connected to one end of the connecting rod gear set (03). The other end of the connecting rod gear set (03) is connected to the piston inside the cylinder (04). The cylinder (04) is installed inside the upper molding mold (2). The valve needle (07) is installed on the cylinder (04). The valve needle (07) is located inside the glue guide tube (06). The valve needle (07) is used to adjust the amount of glue flowing from the glue guide tube (06). The cylinder (04) is used to drive the reciprocating motion of the valve needle (07). The piston is used to adjust the space inside the cylinder (04). The control terminal of the cylinder (04) is electrically connected to the control terminal of the controller, and the controller is equipped with a control system, which is used to control the cylinder (04). The rotating rod (02), the connecting rod gear set (03), the cylinder (04) and the valve needle (07) are arranged in an inverted L-shape. The rotating rod (02) is located on the horizontal side of the inverted L-shape, the connecting rod gear set (03) is located on the corner of the inverted L-shape, and the cylinder (04) and the valve needle (07) are located on the vertical side of the inverted L-shape. The number of the adjustment knob (01), rotating rod (02), connecting rod gear set (03), cylinder (04), valve needle (07) and guide tube (06) is the same as the number of molding cavities; The control system includes a self-learning control module, a drive control module, a signal acquisition module, and a control processing module. The self-learning end of the control processing module is electrically connected to the self-learning end of the self-learning control module. The drive end of the control processing module is electrically connected to the control end of the drive control module. The drive end of the drive control module is electrically connected to the control end of the cylinder (04). The acquisition end of the control processing module is electrically connected to the acquisition signal end of the signal acquisition module. The liquid silicone plastic molding system also includes a safety protection module, which is electrically connected to the control processing module. The safety protection module is used to identify the position information threshold, temperature information threshold, and flow information threshold from the position sensor, temperature sensor, and flow sensor. When the position exceeds the limit, the temperature exceeds the threshold, the flow exceeds the limit, or the position sensor, temperature sensor, and flow sensor are abnormal, the air supply to the cylinder will be automatically shut off and an alarm will be triggered. For alarms, a buzzer alarm can be installed on the controller to trigger an alarm. The connecting rod gear set (03) includes a worm gear and a telescopic rod (05). The worm gear meshes with each other. The other end of the worm is connected to the rotating rod (02). The telescopic rod (05) is connected to the worm gear. The telescopic end of the telescopic rod (05) is connected to the piston. The cooperation between the telescopic rod (05) and the piston is used to control the stroke of the valve needle (07) to control the amount of glue injected into each molding cavity. The worm gear is located on the horizontal side of the inverted L-shape, and the telescopic rod is located on the vertical end of the inverted L-shape. Among them, the telescopic rod (05) is selected as a threaded telescopic rod (05). The telescopic rod (05) includes a fixed part and a telescopic part. One end of the fixed part is connected to the worm gear, and the other end of the fixed part is threaded to the telescopic part. The other end of the telescopic part is connected to the piston. The telescopic part is vertically slidably installed in the upper forming mold (2). The valve needle (07) is made of YG8 cemented carbide and coated with a 3-4 μm thick DLC diamond-like carbon coating. The tip of the valve needle (07) is either conical or elliptical. Among them, the included angle for the conical needle tip is 20°-30°; Among them, the included angle of the major axis of the elliptical cone-shaped needle tip is 30°-40°, and the included angle of the minor axis is 20°-30°.

2. The liquid silicone plastic molding system with micro-adjustment according to claim 1, characterized in that, The cylinder (04) is selected as either a dual-control cylinder or a single-control cylinder.

3. The liquid silicone plastic molding system with micro-adjustment according to claim 1, characterized in that, The self-learning control module is equipped with a convolutional neural network model, a recurrent neural network model, a deep reinforcement learning convolutional network model, a memory-enhanced network model, and a graph neural network model. The signal acquisition module transmits the acquired signals to the convolutional neural network model for feature vector processing through the control processing module. The convolutional neural network model then transmits the signals to the recurrent neural network model for context feature vector processing. Finally, the signals are processed by the deep reinforcement learning convolutional network model, which has a memory-enhanced network model and a graph neural network model working together, before being sent to the control processing module.

4. The liquid silicone plastic molding system with micro-adjustment according to claim 1, characterized in that, The signal acquisition module includes a position sensor, a temperature sensor, and a flow sensor. The position sensor is installed on the cylinder (04), the temperature sensor is located on the side wall of the glue guide tube (06), and the flow sensor is located at the glue outlet end of the glue guide tube (06). The position sensor, temperature sensor, and flow sensor are electrically connected to the acquisition signal terminal of the control processing module.

5. A liquid silicone plastic molding system with micro-adjustment according to claim 4, characterized in that, The position sensor is selected as a magnetostrictive position sensor, which includes a measuring magnetic ring and a sensor body. The measuring magnetic ring is installed inside the piston of the cylinder (04), and the sensor body is installed on the outside of the cylinder (04). The position sensor is used to measure the precise position of the piston in real time.

6. A liquid silicone plastic molding system with micro-adjustment according to claim 4, characterized in that, The temperature sensor is a resistive temperature sensor, which is installed on the outer wall of the tube (06) and is used to monitor the working temperature of the tube (06).

7. A liquid silicone plastic molding system with micro-adjustment according to claim 4, characterized in that, The flow sensor is a Coriolis mass flow meter, which is installed at the outlet end of the adhesive tube (06) and is used to detect the outflow status of liquid silicone plastic.

8. A method for controlling a liquid silicone plastic molding system with micro-adjustment, characterized in that, The operation method steps are as follows: Step 1: Formation of micro-molding mold Select an injection molding machine, an upper mold (2), a lower mold (3), an adjustment knob (01), a rotating rod (02), a connecting rod gear set (03), a cylinder (04), a valve needle (07), a guide tube (06), and a controller. The controller is installed on the injection molding machine. The upper mold (2) and the lower mold (3) are both installed on the injection molding machine, and the upper mold (2) and the lower mold (3) fit together to form multiple molding cavities. An injection port (1) is provided on the top of the upper mold (2). The injection port (1) is connected to the molding cavity through the guide tube (06). The rotating rod... (02) Rotary rod (02) is rotatably installed in the upper molding mold (2), and one end of the rotating rod (02) extends out of the upper molding mold (2) and is equipped with the adjusting knob (01). The other end of the rotating rod (02) is connected to one end of the connecting rod gear set (03). The other end of the connecting rod gear set (03) is connected to the piston in the cylinder (04). The cylinder (04) is installed in the upper molding mold (2). The valve needle (07) is installed on the cylinder (04). The valve needle (07) is located inside the glue guide tube (06). The valve needle (07) is used to adjust the glue flow rate of the glue guide tube (06). The control terminal of the cylinder (04) is electrically connected to the control terminal of the controller, and the controller is equipped with a control system, which is used to control the cylinder (04). The rotating rod (02), the connecting rod gear set (03), the cylinder (04) and the valve needle (07) are arranged in an inverted L-shape. The rotating rod (02) is located on the horizontal side of the inverted L-shape, the connecting rod gear set (03) is located on the corner of the inverted L-shape, and the cylinder (04) and the valve needle (07) are located on the vertical side of the inverted L-shape. The number of the adjustment knob (01), rotating rod (02), connecting rod gear set (03), cylinder (04), valve needle (07) and guide tube (06) is the same as the number of molding cavities; Step 2: Installation of the signal acquisition module: Based on step 1, a signal acquisition module and a control processing module are selected. The signal acquisition module includes a position sensor, a temperature sensor and a flow sensor. The position sensor is installed on the cylinder (04), the temperature sensor is set on the side wall of the glue guide tube (06), and the flow sensor is set at the glue outlet end of the glue guide tube (06). The position sensor, temperature sensor and flow sensor are electrically connected to the acquisition signal terminal of the control processing module. The position sensor is selected as a magnetostrictive position sensor, which includes a measuring magnetic ring and a sensor body. The measuring magnetic ring is installed inside the piston of the cylinder (04), and the sensor body is installed outside the cylinder (04). The position sensor is used to measure the precise position of the piston in real time. The temperature sensor is a resistive temperature sensor, which is installed on the outer wall of the tube (06) and is used to monitor the working temperature of the tube (06). The flow sensor is a Coriolis mass flow meter, which is installed at the outlet end of the adhesive tube (06) and is used to detect the outflow status of liquid silicone plastic. Step 3: Setting up the control system Based on steps 1 and 2, the controller is equipped with a control system, which is used to control the cylinder (04); The control system includes a self-learning control module, a drive control module, a signal acquisition module, and a control processing module. The self-learning end of the control processing module is electrically connected to the self-learning end of the self-learning control module. The drive end of the control processing module is electrically connected to the control end of the drive control module. The drive end of the drive control module is electrically connected to the control end of the cylinder (04). The acquisition end of the control processing module is electrically connected to the acquisition signal end of the signal acquisition module. The self-learning control module is equipped with a convolutional neural network model, a recurrent neural network model, a deep reinforcement learning convolutional network model, a memory-enhanced network model, and a graph neural network model. The signal acquisition module transmits the acquired signals to the convolutional neural network model for feature vector processing through the control processing module. The convolutional neural network model then transmits the signals to the recurrent neural network model for context feature vector processing. Finally, the signals are processed by the deep reinforcement learning convolutional network model, which has a memory-enhanced network model and a graph neural network model working together, before being sent to the control processing module.

Images