Polyethylene wax forming device and control method thereof
By designing automated feeding, demolding, and cooling systems, combined with intelligent temperature control, the problems of inconvenient operation and low demolding efficiency of polyethylene wax molding equipment in high-temperature environments have been solved, realizing a safe and efficient polyethylene wax molding process.
Patent Information
- Application Number
- CN202411018937.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-27
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing polyethylene wax molding equipment is inconvenient to operate in high-temperature environments and poses safety hazards. Furthermore, traditional demolding methods are inefficient and cannot meet the usage requirements.
A polyethylene wax molding device was designed, comprising a feeding mechanism, a pushing mechanism, a demolding mechanism, and a cooling and collection component. Combined with an intelligent temperature control system, automated feeding, demolding, and cooling are achieved through drive motors, servo motors, and cooling fans. Reinforcement learning and adaptive fuzzy control algorithms are introduced to optimize temperature control.
It improves ease of operation and production efficiency, reduces the labor intensity of operators, ensures safety and molding quality, and achieves efficient temperature control.
Smart Images

Figure CN118906355A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of polyethylene wax processing and molding equipment, in particular to a polyethylene wax molding device and a control method thereof. Background Art
[0002] Polyethylene wax is widely used due to its excellent cold resistance, heat resistance, chemical resistance and wear resistance. It can increase the gloss and processing performance of the product. A polyethylene wax molding device is a commonly used polyethylene wax processing method. First, polyethylene wax particles are placed in a melting device and heated to above the melting point to melt it into a liquid state. Then, the molten polyethylene wax is made into the required shape and size of a container through processes such as extrusion, injection molding and calendering. Finally, the obtained product is cooled and solidified to obtain a final polyethylene wax product. When the existing polyethylene wax molding device is used, due to the high temperature outside the high-temperature reactor, the operator needs to face a high temperature environment during the loading process, which not only increases the difficulty of operation, but also may pose a threat to the safety of the operator. Since the polyethylene wax product still has a certain temperature after molding, the operator needs to be extra careful during the collection process to avoid scalding or other safety accidents. In addition, the traditional demolding method is often inefficient and cannot meet the use requirements. Therefore, a polyethylene wax molding device and a control method thereof are proposed to solve the above-mentioned problems.
[0003] The above information disclosed in the background technology is only used to increase the understanding of the background technology of the present application, and therefore, it may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention
[0004] 1. Technical issues to be solved
[0005] In view of the deficiencies in the prior art, the present invention provides a polyethylene wax molding device and a control method thereof, which have the advantages of being able to effectively facilitate operators to load materials and effectively improve the production efficiency of the device, and solve the problems raised in the above-mentioned background technology.
[0006] (II) Technical solution
[0007] To achieve the above objectives, the present invention provides the following technical solutions: a polyethylene wax molding device, comprising a mounting frame, support frames are arranged on the left and right sides of the bottom of the mounting frame, a melting reaction kettle extending to the inner side thereof is arranged on the top of the mounting frame, a stirring motor extending to the inside thereof is arranged on the top of the melting reaction kettle, a discharge pipe communicating with the inside thereof is arranged at the bottom of the melting reaction kettle, a control valve is arranged at the bottom of the discharge pipe, a fixed mold base communicating with the control valve is arranged on the inner bottom wall of the mounting frame, a fixed mold is arranged on the right side of the fixed mold base, a liquid injection hole communicating with the control valve is arranged inside the fixed mold, a feeding mechanism communicating with the inside of the melting reaction kettle and extending to the left side of the mounting frame is arranged on the left side of the melting reaction kettle, a pushing mechanism penetrating into the inside of the fixed mold is arranged on the left side of the mounting frame, a demoulding mechanism penetrating into the inside of the fixed mold is arranged on the inner bottom wall of the mounting frame, and a cooling and collecting assembly is arranged at the bottom of the mounting frame;
[0008] The feeding mechanism includes a feeding barrel, which is fixedly installed on the left side of the melting reactor and connected with it, a driving motor is fixedly installed on the top of the feeding barrel, a connecting frame is fixedly installed on the outside of the feeding barrel, a feeding hopper that penetrates to the inside of the connecting frame is fixedly installed on the top of the connecting frame, and a material guide pipe connected with the inside of the feeding barrel is fixedly installed on the bottom of the feeding hopper.
[0009] Preferably, the pushing mechanism includes an electric push rod, which is fixedly installed on the left side of the mounting frame, and the output end of the electric push rod passes through a connecting plate inside the mounting frame, and a pushing rod that passes through the inside of the fixed mold is fixedly installed on the right side of the connecting plate.
[0010] Preferably, the demolding mechanism includes a fixed seat, a fixed seat located on the right side of the fixed mold seat is fixedly installed on the inner bottom wall of the mounting frame, a servo motor is fixedly installed on the right side of the fixed seat, the output shaft of the servo motor passes through the fixed seat and is fixedly connected to a threaded rod, the outer side of the threaded rod is threadedly installed with a threaded sleeve extending to its left side, a movable frame is movably installed on the inner bottom wall of the mounting frame and is sleeved on the outside of the threaded sleeve, a return spring located on the outside of the threaded sleeve is fixedly installed between the movable frame and the fixed seat, a movable mold seat fixedly connected to the threaded sleeve is slidably installed on the inner top wall of the movable frame, a molding mold extending to the inside of the fixed mold and fitted therewith is fixedly installed on the left side of the movable mold seat, and a demolding push rod passing through the inside of the molding mold is fixedly installed on the right side of the inner wall of the movable frame.
[0011] Preferably, the cooling collection assembly includes a connecting frame, the connecting frame is fixedly mounted on the bottom of the mounting frame, cooling fans connected to the outside of the connecting frame are fixedly mounted on the front and rear sides of the inner wall of the connecting frame, an electric telescopic rod penetrating into the interior of the connecting frame is fixedly mounted on the left side of the connecting frame, a supporting frame is fixedly mounted on the bottom of the connecting frame, a receiving plate located below the electric telescopic rod is movably mounted inside the connecting frame, a buffer spring is fixedly mounted between the bottom of the receiving plate and the inner bottom wall of the supporting frame, a collection frame located on the right side of the support frame is fixedly mounted on the bottom of the connecting frame, and a collection box penetrating into the outside of the collection frame is movably mounted inside the collection frame.
[0012] Preferably, the output shaft of the stirring motor passes through the interior of the molten reactor and is fixedly connected to a stirrer, the loading barrel is inclined in design, and the output shaft of the drive motor passes through the interior of the loading barrel and is fixedly connected to a spiral auger rotatably connected thereto.
[0013] Preferably, the number of the pushing rods is several and they are distributed in a rectangular array, and the interiors of the fixed mold base and the fixed mold are provided with through holes that are compatible with the pushing rods.
[0014] Preferably, a circular hole adapted to the threaded sleeve is provided inside the movable frame, the number of the reset springs is several and distributed in a rectangular array, and the fixed mold and the fixed mold base as well as the forming mold and the movable mold base are all installed by several bolt threads.
[0015] Preferably, the mounting frame and the connecting frame are both provided with interconnected material receiving holes, the outer side of the connecting frame is fixedly provided with a dust screen located on the outer side of the heat dissipation fan, the output end of the electric telescopic rod is fixedly provided with a pushing plate located above the material receiving plate, the left side of the pushing plate is fixedly provided with an auxiliary rod that passes through the left side of the connecting frame and is located on the front and rear sides of the electric telescopic rod, the top of the material receiving plate is fixedly provided with a heat insulating rubber pad, and the number of the buffer springs is several and distributed in a rectangular array.
[0016] Preferably, the inner bottom wall of the connecting frame is provided with a material drop hole adapted to the collecting frame, and a handle is fixedly mounted on the front side of the collecting box.
[0017] The present invention also discloses a method for using the polyethylene wax molding device, which specifically comprises the following steps:
[0018] Step 1: The operator pours the polyethylene wax material into the loading hopper, and drives the spiral auger to rotate by controlling the driving motor. The polyethylene wax material is introduced into the loading barrel from the guide pipe and transported to the melting reactor by the spiral auger for melting treatment, thereby effectively facilitating the operator to perform the loading operation and improving the operation convenience of the device;
[0019] Step 2: After the polyethylene wax material liquid is molded by the fixed mold and the molding mold, the operator can drive the threaded rod to rotate through the servo motor, and the threaded rod drives the movable mold base and the molding mold to move away from the fixed mold base and the fixed mold through the threaded sleeve. At the same time, when the movable mold base and the molding mold move to the right, the demoulding push rod demoulds the molded polyethylene wax product from the outside of the molding mold, and at the same time, the connecting plate is driven to move by controlling the electric push rod, and the connecting plate drives the push rod to push the polyethylene wax product out of the fixed mold, thereby effectively helping the user to perform the demoulding operation on the molding device and effectively improving the molding efficiency of the device;
[0020] Step 3. When the polyethylene wax product falls from the receiving hole to the receiving plate after demolding, the operator can control the heat dissipation fan to cool the polyethylene wax product, and then use the electric telescopic rod to drive the push plate to move, so that the push plate pushes the cooled polyethylene wax product into the collection box for collection, thereby helping the operator to cool and collect the molded polyethylene wax product and reducing the operator's labor intensity.
[0021] In step 1, temperature sensors and fuzzy control algorithms are introduced. Temperature sensors are installed inside the melting reactor. The temperature is monitored and adjusted in real time through the fuzzy control algorithm to ensure the stability of the melting temperature of the polyethylene wax and improve the molding quality. Then, adaptive control is implemented, and the adaptive control algorithm is used to automatically adjust the heating power according to the melting point and viscosity of the polyethylene wax, optimize the melting process, and form an intelligent temperature control system for the melting reactor.
[0022] (III) Beneficial effects
[0023] Compared with the prior art, the present invention provides a polyethylene wax molding device and a control method thereof, which have the following beneficial effects:
[0024] 1. The polyethylene wax molding device and the control method thereof are provided with a loading hopper. The operator pours the polyethylene wax material into the loading hopper and controls the driving motor to drive the spiral auger to rotate. The polyethylene wax material is introduced into the loading barrel from the material guide pipe and transported to the melting reactor by the spiral auger for melting treatment, thereby effectively facilitating the operator to perform the loading operation and improving the operation convenience of the device.
[0025] 2. The polyethylene wax molding device and its control method are as follows: a servo motor is set to drive the threaded rod to rotate, and the threaded rod drives the movable mold base and the molding mold to move away from the fixed mold base and the fixed mold through the threaded sleeve. At the same time, when the movable mold base and the molding mold move to the right, the demoulding push rod demoulds the molded polyethylene wax product from the outer side of the molding mold, and at the same time, the electric push rod is controlled to drive the connecting plate to move, and the connecting plate drives the push rod to push the polyethylene wax product out of the fixed mold, thereby effectively helping the user to perform the demoulding operation on the molding device, and effectively improving the molding efficiency of the device. After demoulding, the polyethylene wax product falls from the receiving hole to the receiving plate, and the operator can cool the polyethylene wax product by controlling the heat dissipation fan, and then drive the push plate to move by the electric telescopic rod, so that the push plate pushes the cooled polyethylene wax product into the collection box for collection, thereby helping the operator to cool and collect the molded polyethylene wax product, reducing the labor intensity of the operator and facilitating the use of the user.
[0026] 3. The polyethylene wax molding device and its control method, by introducing an intelligent temperature control system based on reinforcement learning and adaptive fuzzy control, realizes efficient and accurate control of the polyethylene wax molding device, and greatly improves the overall performance of the device. The combination of the value iteration network VIN in reinforcement learning and the proximal policy optimization PPO enables the control strategy to self-optimize in a complex dynamic environment. VIN provides the basis for the system to take the best action under different states by estimating the state-action value function Q value. PPO ensures the stability and convergence of the training process by limiting the strategy update amplitude, effectively avoiding the problems of excessive strategy update and performance fluctuation. The introduction of the adaptive fuzzy control algorithm enables the system to have higher flexibility and adaptability when adjusting the heating power in real time. At the same time, the adaptive adjustment mechanism dynamically adjusts the fuzzy rule parameters according to the temperature change rate, further improving the system's responsiveness to environmental changes. Through the optimization of the intelligent algorithm, the temperature control accuracy, system stability and production efficiency of the polyethylene wax molding device are improved, providing a strong guarantee for high-quality and efficient production. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a three-dimensional schematic diagram of the overall structure of the present invention;
[0028] Figure 2 It is a side perspective schematic diagram of the overall structure of the present invention;
[0029] Figure 3 It is a three-dimensional schematic diagram of the connection between the structural mounting frame and the fixed mold base in the present invention;
[0030] Figure 4 It is a three-dimensional schematic diagram of the structural feeding mechanism in the present invention;
[0031] Figure 5 It is a three-dimensional schematic diagram of the connection between the structural mounting frame and the demoulding mechanism in the present invention;
[0032] Figure 6 It is a three-dimensional schematic diagram of the structural driving mechanism in the present invention;
[0033] Figure 7 It is a three-dimensional schematic diagram of the structural demoulding mechanism in the present invention;
[0034] Figure 8 It is a three-dimensional schematic diagram of the structural cooling collection assembly in the present invention;
[0035] Fig. 9 It is a three-dimensional schematic diagram of the connection between the structural material receiving plate and the buffer spring in the present invention.
[0036] Fig.10 This is the intelligent temperature control process based on reinforcement learning and adaptive fuzzy control in the present invention.
[0037] In the figure: 1. mounting frame; 2. supporting frame; 3. melting reactor; 4. stirring motor; 5. discharging pipe; 6. feeding mechanism; 61. feeding barrel; 62. driving motor; 63. connecting frame; 64. feeding hopper; 65. material guide pipe; 7. pushing mechanism; 71. electric push rod; 72. connecting plate; 73. pushing rod; 8. demoulding mechanism; 81. fixing seat; 82. servo motor; 83. threaded rod; 84. threaded sleeve; 85. moving frame; 86. reset spring; 87. moving mold base; 88. forming mold; 89. demoulding push rod; 9. cooling and collecting assembly; 91. connecting frame; 92. cooling fan; 93. electric telescopic rod; 94. supporting frame; 95. receiving plate; 96. buffer spring; 97. collecting frame; 98. collecting box; 10. control valve; 11. fixed mold base; 12. fixed mold; 13. injection hole. DETAILED DESCRIPTION
[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0039] See also Figure 1-9, a polyethylene wax molding device and a control method thereof, comprising a mounting frame 1, a support frame 2 is fixedly installed on both the left and right sides of the bottom of the mounting frame 1, a molten reactor 3 extending to the inside thereof is fixedly installed on the top of the mounting frame 1, a stirring motor 4 extending to the inside thereof is fixedly installed on the top of the molten reactor 3, an output shaft of the stirring motor 4 penetrates into the inside of the molten reactor 3 and is fixedly connected with a stirrer, a discharge pipe 5 connected to the inside of the molten reactor 3 is fixedly connected to the bottom, a control valve 10 is fixedly connected to the bottom of the discharge pipe 5, a fixed mold base 11 connected to the control valve 10 is fixedly installed on the inner bottom wall of the mounting frame 1, a fixed mold 12 is fixedly installed on the right side of the fixed mold base 11, an injection hole 13 connected to the control valve 10 is opened in the fixed mold 12, a feeding mechanism 6 connected to the inside of the molten reactor 3 and extending to the left side of the mounting frame 1 is fixedly connected to the left side of the molten reactor 3, the feeding mechanism 6 comprises a feeding cylinder 61, the feeding cylinder 61 is fixedly installed on the left side of the molten reactor 3 and connected with it, and the feeding cylinder 61 is The inclined design has a driving motor 62 fixedly installed on the top of the loading barrel 61, and the output shaft of the driving motor 62 passes through the interior of the loading barrel 61 and is fixedly connected to a spiral auger rotatably connected thereto. A connecting frame 63 is fixedly installed on the outside of the loading barrel 61, and a loading hopper 64 passing through the inside thereof is fixedly installed on the top of the connecting frame 63, and a material guide pipe 65 connected to the interior of the loading barrel 61 is fixedly installed on the bottom of the loading hopper 64. It should be noted that the specific structures and working principles of the melting reactor 3, the stirring motor 4, the control valve 10 and the driving motor 62 are all referred to the prior art. The polyethylene wax molding device and the control method thereof, by setting the loading hopper 64, the operator pours the polyethylene wax material into the loading hopper 64 and controls the driving motor 62 to drive the spiral auger to rotate, and the polyethylene wax material is introduced from the material guide pipe 65 into the loading barrel 61 and is transported to the melting reactor 3 by the spiral auger for melting treatment, thereby effectively facilitating the operator to perform the loading operation and improving the convenience of operation of the device.
[0040] A pushing mechanism 7 penetrating into the interior of the fixed mold 12 is fixedly installed on the left side of the mounting frame 1. The pushing mechanism 7 includes an electric push rod 71. The electric push rod 71 is fixedly installed on the left side of the mounting frame 1. The output end of the electric push rod 71 penetrates through a connecting plate 72 inside the mounting frame 1. A pushing rod 73 penetrating into the interior of the fixed mold 12 is fixedly installed on the right side of the connecting plate 72. The number of the pushing rods 73 is several and they are distributed in a rectangular array. The interiors of the fixed mold base 11 and the fixed mold 12 are provided with through holes adapted to the pushing rods 73. A demoulding mechanism 8 penetrating into the interior of the fixed mold 12 is fixedly installed on the inner bottom wall of the mounting frame 1. The demoulding mechanism 8 includes a fixed seat 81. The inner bottom wall of the mounting frame 1 is fixedly installed with a fixed seat 81 located on the right side of the fixed mold base 11. A servo motor 82 is fixedly installed on the right side of the fixed seat 81, and the output shaft of the servo motor 82 passes through the fixed seat 81 and is fixedly connected to a threaded rod 83. The outer side of the threaded rod 83 is threadedly installed with a threaded sleeve 84 extending to its left side. A moving frame 85 sleeved on the outer side of the threaded sleeve 84 is movably installed on the inner bottom wall of the mounting frame 1. A circular hole matching the threaded sleeve 84 is opened inside the moving frame 85. A reset spring 86 located on the outer side of the threaded sleeve 84 is fixedly installed between the moving frame 85 and the fixed seat 81. The number of the reset springs 86 is several and distributed in a rectangular array. The setting of the reset spring 86 can support and reset the moving frame 85. A moving mold base 87 fixedly connected to the threaded sleeve 84 is slidably installed on the inner top wall of the moving frame 85. The moving mold base 8 The left side of the mounting frame 7 is fixedly installed with a molding die 88 extending into the interior of the fixed mold 12 and fitting therewith. The fixed mold 12 and the fixed mold base 11 as well as the molding die 88 and the movable mold base 87 are all installed by a plurality of bolt threads. The bolt installation can facilitate the user to disassemble and assemble the mold. The right side of the inner wall of the movable frame 85 is fixedly installed with a demoulding push rod 89 penetrating into the interior of the molding die 88. The arrangement of the demoulding push rod 89 can prevent the polyethylene wax product from adhering to the molding die 88. The bottom of the mounting frame 1 is fixedly installed with a cooling collection component 9. The cooling collection component 9 includes a connecting frame 91, which is fixedly installed at the bottom of the mounting frame 1. The mounting frame 1 and the connecting frame 91 are both provided with mutually connected material receiving holes. The inner wall of the connecting frame 91 has two inner walls, front and rear. A heat dissipation fan 92 connected to the outside is fixedly installed on both sides of the connecting frame 91, a dust-proof net located on the outside of the heat dissipation fan 92 is fixedly installed on the outside of the connecting frame 91, and the dust-proof net can reduce the impurities entering the device. An electric telescopic rod 93 that runs through the inside of the connecting frame 91 is fixedly installed on the left side of the connecting frame 91, and a supporting frame 94 is fixedly installed on the bottom of the connecting frame 91. A receiving plate 95 located below the electric telescopic rod 93 is movably installed inside the connecting frame 91, and a pushing plate located above the receiving plate 95 is fixedly installed on the output end of the electric telescopic rod 93. An auxiliary rod that runs through the left side of the connecting frame 91 and is located on the front and rear sides of the electric telescopic rod 93 is fixedly installed on the left side of the pushing plate. The auxiliary rod can improve the stability of the pushing plate when it moves, and a heat-insulating rubber pad is fixedly installed on the top of the receiving plate 95.A buffer spring 96 is fixedly installed between the bottom of the receiving plate 95 and the inner bottom wall of the support frame 94. The number of buffer springs 96 is several and distributed in a rectangular array. The matching arrangement of the buffer spring 96 and the thermal insulation pad can protect the polyethylene wax products. A collecting frame 97 located on the right side of the support frame 94 is fixedly installed at the bottom of the connecting frame 91. The inner bottom wall of the connecting frame 91 is provided with a blanking hole adapted to the collecting frame 97. A collecting box 98 is movably installed inside the collecting frame 97 and penetrates to the outside thereof. A handle is fixedly installed on the front side of the collecting box 98. The setting of the handle can facilitate the operator to take out the collecting box 98. It should be noted that the specific structure and working principle of the fixed mold 12, the fixed mold base 11, the electric push rod 71, the servo motor 82, the molding mold 88, the movable mold base 87, the heat dissipation fan 92 and the electric telescopic rod 93 can all be referred to the prior art. The polyethylene wax molding device and its control method drive the threaded rod 83 to rotate by setting the servo motor 82. The rod 83 drives the movable die base 87 and the molding die 88 away from the fixed die base 11 and the fixed die 12 through the threaded sleeve 84. At the same time, when the movable die base 87 and the molding die 88 move to the right, the demoulding push rod 89 demoulds the molded polyethylene wax product from the outer side of the molding die 88. At the same time, the electric push rod 71 is controlled to drive the connecting plate 72 to move, and the connecting plate 72 drives the push rod 73 to push the polyethylene wax product out of the fixed die 12, thereby effectively helping the user to perform the demoulding operation on the molding device and effectively improving the molding efficiency of the device. After demoulding, the polyethylene wax product falls from the receiving hole to the receiving plate 95. The operator can cool the polyethylene wax product by controlling the heat dissipation fan 92, and then drive the push plate to move through the electric telescopic rod 93, so that the push plate pushes the cooled polyethylene wax product into the collection box 98 for collection, thereby helping the operator to cool and collect the molded polyethylene wax product and reducing the labor intensity of the operator.
[0041] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.
[0042] The present invention also discloses a method for using the polyethylene wax molding device, which specifically comprises the following steps:
[0043] Step 1: The operator pours the polyethylene wax material into the loading hopper 64, and drives the spiral auger to rotate by controlling the driving motor 62. The polyethylene wax material is introduced into the loading barrel 61 from the guide pipe 65 and transported to the melting reactor 3 by the spiral auger for melting treatment, thereby effectively facilitating the operator to perform the loading operation and improving the convenience of operation of the device;
[0044] Step 2, when the polyethylene wax material liquid is molded by the fixed mold 12 and the molding mold 88, the operator can drive the threaded rod 83 to rotate through the servo motor 82, and the threaded rod 83 drives the movable mold base 87 and the molding mold 88 to move away from the fixed mold base 11 and the fixed mold 12 through the threaded sleeve 84. At the same time, when the movable mold base 87 and the molding mold 88 move to the right, the demoulding push rod 89 demoulds the molded polyethylene wax product from the outer side of the molding mold 88, and at the same time, the electric push rod 71 is controlled to drive the connecting plate 72 to move, and the connecting plate 72 drives the push rod 73 to push the polyethylene wax product out of the fixed mold 12, thereby effectively helping the user to perform the demoulding operation on the molding device and effectively improving the molding efficiency of the device;
[0045] Step three, when the polyethylene wax product falls from the receiving hole onto the receiving plate 95 after demolding, the operator can control the heat dissipation fan 92 to cool the polyethylene wax product, and then drive the push plate to move through the electric telescopic rod 93, so that the push plate pushes the cooled polyethylene wax product into the collection box 98 for collection, thereby helping the operator to cool and collect the molded polyethylene wax product, reducing the labor intensity of the operator, facilitating the use of the user, and solving the problems raised in the above-mentioned background technology.
[0046] like Fig.10 In step 1, the temperature sensor and fuzzy control algorithm are introduced. The temperature sensor is installed inside the melting reactor. The fuzzy control algorithm is used to monitor and adjust the temperature in real time to ensure the stability of the melting temperature of the polyethylene wax and improve the molding quality. Then, adaptive control is implemented. The adaptive control algorithm is used to automatically adjust the heating power according to the melting point and viscosity of the polyethylene wax, optimize the melting process, and form an intelligent temperature control system for the melting reactor.
[0047] Introducing temperature sensor and adaptive fuzzy control algorithm
[0048] Temperature sensor installation
[0049] Several high-precision temperature sensors are installed at different locations inside the melting reactor: bottom, middle and top, to monitor the temperature distribution inside the reactor in real time. The specific locations are as follows:
[0050] Bottom temperature sensor T b :Installed at the bottom of the melting reactor to monitor the bottom temperature.
[0051] Middle temperature sensor T m :Installed in the middle of the melting reactor to monitor the middle temperature.
[0052] Top temperature sensor T t :Installed on the top of the melting reactor to monitor the top temperature.
[0053] Type of temperature sensor: Use high-precision thermocouple temperature sensors, which have the characteristics of fast response, high measurement accuracy and high temperature resistance. Each sensor must ensure long-term stable operation in a high temperature environment.
[0054] Data transmission and processing: The temperature sensor sends real-time temperature data to the central control system via data line or wireless transmission. Data line transmission is suitable for short distances and fixed installations, while wireless transmission is suitable for situations where wiring is inconvenient or flexible movement is required.
[0055] Data collection and processing
[0056] The central control system receives real-time temperature data from the temperature sensor. b , T m , T t ) Collect temperature data at different locations inside the melting reactor in real time, and transmit the data to the central control system CCS through the wireless transmission module.
[0057] Preprocess, filter, denoise and calibrate the temperature data to ensure the accuracy and stability of the data.
[0058] The temperature data is filtered using a discrete low-pass filter:
[0059] T filtered (t) = α·T raw (t)+(1-α)·T filtered (t-1)
[0060] Among them, T filtered (t) is the temperature data after filtering; T filtered (t-1) is the temperature data before filtering; T raw (t) is the original temperature data; α is the filter coefficient;
[0061] Use the moving average method to denoise the filtered data to further improve the stability of the data:
[0062]
[0063] Among them, T denoised (t) is the temperature data after denoising; N is the size of the moving average window; T filtered (ti) represents the filtered temperature data at time ti; t represents the current time; i represents the time offset relative to the current time step t;
[0064] The sensor data is then calibrated according to a known calibration curve or standard temperature data to ensure the accuracy of the temperature data.
[0065] T calibrated (t) = a·T denoised (t)+b
[0066] Among them, T calibrated (t) is the temperature data after calibration; T denoised (t) is the temperature after denoising; data a and b are calibration coefficients, determined by calibration experiments;
[0067] Implementation of Adaptive Fuzzy Control Algorithm
[0068] Establishment of fuzzy rule base: Based on experience and experimental data, a fuzzy rule base for the melting reactor is established. The rule base includes the fuzzy relationship between temperature and heating power.
[0069] Fuzzy variable definition:
[0070] Input variable: Temperature error E(t) = T set -T current And the temperature change rate ΔE(t) = E(t)-E(t-1). Where, T set is the target temperature; T current is the current temperature; E(t) is the current temperature error; E(t-1) is the temperature error at the previous moment;
[0071] Output variable: heating power adjustment ΔP.
[0072] Fuzzy sets:
[0073] Temperature error E(t): divided into {large negative, medium negative, small negative, zero, small positive, medium positive, large positive}; that is (BN, MN, SN, Z, SP, MP, BP).
[0074] Temperature change rate ΔE(t): divided into {large negative, medium negative, small negative, zero, small positive, medium positive, large positive}; that is (BN, MN, SN, Z, SP, MP, BP).
[0075] Heating power adjustment amount AP: divided into {large decrease, medium decrease, small decrease, unchanged, small increase, medium increase, large increase}; that is (BD, MD, SD, NC, SI, MI, BI).
[0076] Fuzzy rules:
[0077] If E(t) is strongly negative and ΔΔE(t) is strongly negative, then AP is strongly increasing.
[0078] If E(t) is zero and ΔE(t) is zero, then ΔP is unchanged.
[0079] If E(t) is strongly positive and ΔE(t) is strongly positive, then ΔP is strongly negative.
[0080] Fuzzy reasoning: Based on the real-time temperature data, the fuzzy reasoning algorithm is used to calculate the currently required heating power adjustment value. The fuzzy reasoning process includes fuzzification, rule reasoning and defuzzification.
[0081] Fuzzification: Membership function of temperature error E(t):
[0082]
[0083] Where, E(t) is the temperature error; μ BN (E) is the membership of the temperature error E(t) in the fuzzy set “big negative” BN; To define the upper limit of the “large negative” BN fuzzy set; Define the upper limit of the "medium negative" MN fuzzy set;
[0084] Rule reasoning: Based on the fuzzy rule base, the fuzzy reasoning algorithm is used to calculate the activation degree of each rule.
[0085] α rule1 =min(μ BN (E(t)), μ BN (ΔE(t)))
[0086] Among them, α rule1 is the activation degree of rule 1, indicating the degree of compliance with the rule under the current conditions; μ BN (E) is the membership degree of temperature error E(t) in the fuzzy set “big negative” BN; μ BN (ΔE(t)) is the membership of the temperature error change rate ΔE(t) in the fuzzy set “large negative” BN; min means taking the minimum value of the two memberships;
[0087] Defuzzification: Take the weighted average of the output results of all rules to obtain the exact value of the heating power adjustment AP.
[0088]
[0089] Among them, α i is the activation degree of the ith rule; P i is the output value corresponding to the i-th rule; ΔP is the heating power adjustment;
[0090] Adaptive adjustment: Introduce an adaptive control mechanism to dynamically adjust the parameters of the fuzzy rules according to changes in real-time temperature data to improve the adaptability of the system.
[0091] Introduction of adaptive factor: The adaptive factor λ(t) is based on the temperature change rate Dynamically adjust the parameters of fuzzy rules.
[0092]
[0093] Among them, λ(t) is the adaptive factor; k is the adjustment factor; is the temperature change rate;
[0094] Fuzzy rule parameter adjustment: According to the adaptive factor λ(t), adjust the membership function of the fuzzy rule. Adjust the membership function of the temperature error E(t):
[0095] μ′ BN (E) = λ(t)·μ BN (E)
[0096] Among them, μ′ BN (E) is the membership degree of the adjusted temperature error E(t); λ(t) is the adaptive factor; μ BN (E) is the membership of the temperature error E(t) in the fuzzy set “big negative” BN;
[0097] Update heating power: Update heating power P(t+1)=P(t)+ΔP according to ΔP;
[0098] Wherein, P(t) is the current heating power, P(t+1) is the heating power at the next moment; ΔP is the heating power adjustment amount;
[0099] The above steps are repeated in a loop to carry out the next round of fuzzy adaptive control.
[0100] Introducing reinforcement learning algorithms: Construction of reinforcement learning environment
[0101] State space: defines the state space of the system, including current temperature, target temperature, and heating power. The state vector is expressed as:
[0102]
[0103] Among them, s(t) is the state vector; T(t) is the current temperature, which means the real-time temperature of the system at time t; T set is the target temperature; P(t) is the current heating power; ΔT(t) is the temperature change rate;
[0104] Action space: Define the action space of the system, including increasing heating power, decreasing heating power, and keeping heating power unchanged. The specific actions can be quantified into different heating power adjustment steps. The action set is expressed as:
[0105] A={a1,a2,a3}
[0106] Among them, A is the action set; a1 means increasing the heating power, a2 means reducing the heating power, and a3 means keeping the heating power unchanged;
[0107] Reward function: Design a reward function to encourage the system to stabilize the temperature at the target value in the shortest time while reducing energy consumption. It can be defined as:
[0108] R(t)=-|T(t)-T set |-λP(t)
[0109] Among them, R(t) is the reward function; T(t) is the current temperature, which means the real-time temperature of the system at time t; T set is the target temperature; P(t) is the current heating power; λ is the weight coefficient for weighing temperature error and energy consumption;
[0110] Training of reinforcement learning algorithms
[0111] Algorithm selection: Adopt the latest model-based reinforcement learning algorithm, based on the value iteration network VIN method. Train the reinforcement learning model in a simulation environment, use the collected temperature data and heating power data, and continuously adjust the model parameters so that it can optimize temperature control in actual operation.
[0112] The state value function V(s) is defined as:
[0113] V(s)=max a Q(s,a)
[0114] The state-action value function Q(s, a) is defined as:
[0115]
[0116] Where V(s) is the state value function; Q(s, a) is the state-action value function, which represents the expected cumulative reward after taking action a in state s; s is the current state; max a is the maximization operator, which means selecting action a that maximizes Q(s, a); R(s, a) is the immediate reward, which means the direct reward obtained by taking action a in state s; Y is the discount factor, which ranges from 0 to 1; P(s′|s, a) is the state transition probability, which means the probability of transitioning to the next state s′ after taking action a in state s; V(s′) is the state value function of the next state s′;
[0117] During the training process, the following steps are repeated to update the Q value:
[0118] Initialize the Q value:
[0119] Q value iterative update formula:
[0120]
[0121] Policy Update:
[0122] Where Q(s, a) is the state-action value function; s is the current state; a is the current action; a is the current action; The assignment operation is performed for all states s and all actions a; R(s, a) is the immediate reward; γ is the discount factor; P(s′|s, a) is the state transition probability, which indicates the probability of transitioning to the next state s′ after taking action a in state s; π(s) is the strategy, which indicates the optimal action selected in state s; Q(s′, a′) indicates the expected cumulative reward after taking action a′ in state s′; represents the maximum Q value of taking all possible actions a′ in state s′; It means to select the action a that can maximize Q(s, a);
[0123] PPO optimization is used in policy update. The PPO algorithm avoids excessive policy updates by limiting the change range between the new and old policies. Then, the policy changes are controlled by cutting the policy ratio and introducing the KL divergence penalty term. The steps are as follows:
[0124] Collect samples: Run the current new strategy πθ in the environment, θ is the strategy parameter, and collect the state, action, reward, and next state.
[0125] Calculate the advantage function
[0126] The advantage function is calculated at each time step using the time difference method The advantage function is defined as:
[0127]
[0128] in, is the advantage function; δ t is the time difference error, δ t =R(t)+γV(s t+1 )-V(s t ); R(t) is the immediate reward; γ is the discount factor; λ is the discount factor of the advantage function; the combination of γλ represents the reward for the future time step; V(s t+1 ) is the next state s t+1 The state value function of V(s t ) is the current state s t The state value function of
[0129] The strategy parameter θ is updated using the clipping strategy ratio method. The specific update formula is as follows:
[0130] The ratio is r t (θ):
[0131]
[0132] Among them, r t (θ) is the clipping strategy ratio; θ is the strategy parameter, π θ For new strategies, For the old strategy; θ (a t |s t ) means in state s t Next, take action a according to the new policy parameter θ t The probability of Indicates that in state s t Next, according to the old policy parameters θ old Take action a t The probability of
[0133] Update the policy parameters using the clipping policy ratio method:
[0134]
[0135] Among them, L CLIP (θ) is the objective function after clipping; r t (θ) is the clipping strategy ratio; E t The expected value at time t; is the advantage function estimate; clip(r t (θ), 1-∈, 1+∈) represents the strategy ratio r t (θ) is clipped to be in the range of [1-∈, 1+∈]; ∈ is a hyperparameter used to limit the variation of the strategy ratio;
[0136] The objective function of the KL divergence penalty method is:
[0137]
[0138] Among them, L KL (θ) is the objective function of the KL divergence penalty method; E t The expected value at time t; r t (θ) is the strategy ratio; is the advantage function estimate; β is a hyperparameter that controls the weight of the KL divergence penalty term; is the KL divergence, measuring the old strategy With the new strategy π θ In status t The distribution differences under
[0139] Update the value function: minimize the mean square error and update the parameters of the value function V(s):
[0140] L VF (φ)=E t [(R(t)+γV(st+1 )-V(s t )) 2 ]
[0141] Among them, L VF (φ) is the loss function of the value function, which is used to minimize the mean square error to update the parameter φ of the value function V(s); E t The expected value at time t; R(t) is the immediate reward; γ is the discount factor; V(s t+1 ) is the next state at t+1 The state value function of V(s t ) is the current state s t The state value function of
[0142] By introducing PPO to optimize VIN and limiting the strategy update range, the stability and convergence of the training process are ensured, effectively avoiding the problems of excessive strategy updates and performance fluctuations.
[0143] Online learning and adjustment
[0144] Online learning: During the actual operation process, the system continuously collects operation data and uses online learning algorithms to continuously update the reinforcement learning model to maintain the optimal performance of the model.
[0145] Dynamic adjustment: Combined with the adaptive fuzzy control algorithm, the reinforcement learning model dynamically adjusts the heating power according to real-time data to ensure the stable operation of the system under various working conditions.
[0146] Through the above specific steps, combined with online learning and dynamic adjustment, the reinforcement learning model can be continuously optimized and adjusted in actual operation to ensure the temperature control accuracy of the polyethylene wax molding device and the adaptability of the system, thereby improving production efficiency and product quality.
[0147] Numerical case analysis:
[0148] There is a polyethylene wax molding device, the goal is to stabilize the temperature inside the reactor at 200℃ to ensure uniform melting and high-quality molding of the polyethylene wax. An intelligent temperature control system based on value iteration network VIN and proximal policy optimization (PPO) combined with adaptive fuzzy control will be introduced to achieve this goal.
[0149] Target temperature: T set =200; initial temperature: T(0)=150;
[0150] Initial heating power: P(0) = 50; discount factor: γ = 0.9;
[0151] Weight coefficient: λ = 0.01; Clipping parameter: ∈ = 0.2;
[0152] KL divergence penalty weight: β = 0.01; filter coefficient: a = 0.2;
[0153] Moving average window size: N = 5; Adaptive factor adjustment parameter: k = 0.1;
[0154] Real-time temperature data collection:
[0155] Bottom temperature sensor T b (t): collected value 150℃
[0156] Middle temperature sensor T m (t): collected value 152℃
[0157] Top temperature sensor T t (t): collected value 153℃
[0158] Filtered temperature value T filtered (t):
[0159] T filtered (t)=0.2·153+0.8·152=152.2℃
[0160] Moving average denoising temperature value T denoised (t):
[0161]
[0162] Temperature Calibration:
[0163] T calibrated (t)=1.02·151.6+0.5=154.132℃
[0164] Reinforcement Learning Model Update
[0165] Instant Reward Calculation:
[0166] R(t)=-|T(t)-T set |-λP(t)=-|154.132-200|-0.01·50
[0167] =-45.868-0.5=-46.368
[0168] The Q value of the current state-action pair is updated to:
[0169]
[0170] Strategy Optimization PPO
[0171] Strategy ratio calculation: r(θ) = 1.1
[0172] Objective function after clipping: L CLIP(θ)=10
[0173] Adaptive Fuzzy Control
[0174] Calculate the temperature error and rate of change:
[0175] E(t)=T sct -T calibrated (t) = 200 - 154.132 = 45.868
[0176] ΔE(t)=E(t)-E(t-1)=45.868-46.5=-0.632 Fuzzy control calculation:
[0177] Fuzzy membership function:
[0178] μ BN (E) = 0, μ MN (E) = 0.3, μ SN (E) = 0.7 Activation calculation:
[0179] α rulel =min(μ BN (E), μ BN (ΔE))=0
[0180] Heating power adjustment AP:
[0181] AP = -6.5
[0182] Dynamic adjustment, new heating power:
[0183] P(t+1)=P(t)+ΔP=50+(-6.5)=43.5%
[0184] Adaptive Adjustment:
[0185] Adaptive factor λ(t):
[0186]
[0187] Adjusted membership function:
[0188] μ′ BN (E) = λ(t)·μ BN (E) = 0.975 0.3 = 0.2925
[0189] The above specific numerical examples show that the system can dynamically adjust the heating power based on real-time data, so that the temperature gradually approaches the target value, and can effectively reduce energy consumption, improve production efficiency and product quality. Through online learning and dynamic adjustment, the system can maintain stable operation under different working conditions, giving full play to the advantages of intelligent control algorithms.
[0190] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0191] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A polyethylene wax molding device, comprising a mounting frame (1), characterized in that: Support frames (2) are arranged on both left and right sides of the bottom of the mounting frame (1); a molten reaction kettle (3) extending to the inside of the mounting frame (1) is arranged on the top of the mounting frame (1); a stirring motor (4) extending to the inside of the molten reaction kettle (3) is arranged on the top of the molten reaction kettle (3); a discharge pipe (5) communicating with the inside of the molten reaction kettle (3) is arranged on the bottom of the molten reaction kettle (3); a control valve (10) is arranged on the bottom of the discharge pipe (5); a fixed mold base (11) communicating with the control valve (10) is arranged on the inner bottom wall of the mounting frame (1); the right side of the fixed mold base (11) is provided with a plurality of control valves (10); and a plurality of control valves (10) are arranged on the bottom of the mounting frame (1). A fixed mold (12) is provided, and a liquid injection hole (13) connected to a control valve (10) is provided inside the fixed mold (12); a loading mechanism (6) connected to the inside of the molten reactor (3) and extending to the left side of the mounting frame (1) is provided on the left side of the molten reactor (3); a pushing mechanism (7) penetrating into the inside of the fixed mold (12) is provided on the left side of the mounting frame (1); a demoulding mechanism (8) penetrating into the inside of the fixed mold (12) is provided on the inner bottom wall of the mounting frame (1); and a cooling collection assembly (9) is provided at the bottom of the mounting frame (1); The feeding mechanism (6) comprises a feeding barrel (61), the feeding barrel (61) is fixedly mounted on the left side of the molten reactor (3) and is connected thereto, a driving motor (62) is fixedly mounted on the top of the feeding barrel (61), a connecting frame (63) is fixedly mounted on the outside of the feeding barrel (61), a feeding hopper (64) penetrating to the inside of the connecting frame (63) is fixedly mounted on the top of the connecting frame (63), and a material guide pipe (65) connected to the inside of the feeding barrel (61) is fixedly mounted on the bottom of the feeding hopper (64).
2. A polyethylene wax molding device according to claim 1, characterized in that: The pushing mechanism (7) comprises an electric push rod (71), the electric push rod (71) is fixedly mounted on the left side of the mounting frame (1), the output end of the electric push rod (71) passes through a connecting plate (72) inside the mounting frame (1), and a pushing rod (73) is fixedly mounted on the right side of the connecting plate (72) and passes through the interior of the fixed mold (12).
3. A polyethylene wax molding device according to claim 1, characterized in that: The demoulding mechanism (8) comprises a fixed seat (81), the inner bottom wall of the mounting frame (1) is fixedly mounted with the fixed seat (81) located on the right side of the fixed mold seat (11), the right side of the fixed seat (81) is fixedly mounted with a servo motor (82), the output shaft of the servo motor (82) passes through the fixed seat (81) and is fixedly connected with a threaded rod (83), the outer side of the threaded rod (83) is threadedly mounted with a threaded sleeve (84) extending to the left side thereof, and the inner bottom wall of the mounting frame (1) is movably mounted with a threaded sleeve (84) sleeved on the outer side of the threaded sleeve (84). A movable frame (85) is fixedly installed between the movable frame (85) and the fixed seat (81) with a return spring (86) located outside the threaded sleeve (84); a movable die seat (87) fixedly connected to the threaded sleeve (84) is slidably installed on the inner top wall of the movable frame (85); a molding die (88) extending into the interior of the fixed mold (12) and fitting therewith is fixedly installed on the left side of the movable die seat (87); and a demoulding push rod (89) penetrating into the interior of the molding die (88) is fixedly installed on the right side of the inner wall of the movable frame (85).
4. A polyethylene wax molding device according to claim 1, characterized in that: The cooling collection assembly (9) comprises a connection frame (91), the connection frame (91) being fixedly mounted on the bottom of the mounting frame (1), a heat dissipation fan (92) in communication with the outside of the connection frame (91) being fixedly mounted on both the front and rear sides of the inner wall of the connection frame (91), an electric telescopic rod (93) penetrating into the interior of the connection frame (91) being fixedly mounted on the left side of the connection frame (91), a support frame (94) being fixedly mounted on the bottom of the connection frame (91), a material receiving plate (95) located below the electric telescopic rod (93) being movably mounted inside the connection frame (91), a buffer spring (96) being fixedly mounted between the bottom of the material receiving plate (95) and the inner bottom wall of the support frame (94), a collection frame (97) located on the right side of the support frame (94) being fixedly mounted on the bottom of the connection frame (91), and a collection box (98) penetrating into the outside of the collection frame (97) being movably mounted inside the collection frame (97).
5. A polyethylene wax molding device according to claim 1, characterized in that: The output shaft of the stirring motor (4) passes through the interior of the molten reactor (3) and is fixedly connected to a stirrer. The loading barrel (61) is designed to be inclined. The output shaft of the driving motor (62) passes through the interior of the loading barrel (61) and is fixedly connected to a spiral auger rotatably connected thereto.
6. A polyethylene wax molding device according to claim 2, characterized in that: The number of the pushing rods (73) is several and they are distributed in a rectangular array, and the interiors of the fixed mold base (11) and the fixed mold (12) are provided with through holes that are compatible with the pushing rods (73).
7. A polyethylene wax molding device according to claim 3, characterized in that: The movable frame (85) has a circular hole inside that matches the threaded sleeve (84). The number of the return springs (86) is several and they are distributed in a rectangular array. The fixed mold (12) and the fixed mold base (11) as well as the molding mold (88) and the movable mold base (87) are all installed by means of several bolt threads.
8. A polyethylene wax molding device according to claim 4, characterized in that: The mounting frame (1) and the connecting frame (91) are provided with interconnected material receiving holes inside, a dust screen located outside the heat dissipation fan (92) is fixedly installed on the outside of the connecting frame (91), a push plate located above the material receiving plate (95) is fixedly installed on the output end of the electric telescopic rod (93), an auxiliary rod penetrating the left side of the connecting frame (91) and located on the front and rear sides of the electric telescopic rod (93) is fixedly installed on the left side of the push plate, a heat insulating rubber pad is fixedly installed on the top of the material receiving plate (95), and the buffer springs (96) are multiple and distributed in a rectangular array.
9. A polyethylene wax molding device according to claim 4, characterized in that: The inner bottom wall of the connection frame (91) is provided with a material drop hole matched with the collection frame (97), and a handle is fixedly mounted on the front side of the collection box (98).
10. A control method for a polyethylene wax molding device according to any one of claims 1 to 9, characterized in that: The specific steps include: Step 1: The operator pours the polyethylene wax material into the loading hopper (64), and drives the spiral auger to rotate by controlling the driving motor (62). The polyethylene wax material is introduced into the loading barrel (61) from the material guide pipe (65) and transported by the spiral auger to the melting reactor (3) for melting treatment, thereby effectively facilitating the operator to perform the loading operation and improving the convenience of operation of the device; Step 2: After the polyethylene wax material liquid is molded by the fixed mold (12) and the molding mold (88), the operator can drive the threaded rod (83) to rotate through the servo motor (82), and the threaded rod (83) drives the movable mold base (87) and the molding mold (88) to move away from the fixed mold base (11) and the fixed mold (12) through the threaded sleeve (84). At the same time, when the movable mold base (87) and the molding mold (88) move to the right, the demoulding push rod (89) demoulds the molded polyethylene wax product from the outer side of the molding mold (88), and at the same time, the electric push rod (71) is controlled to drive the connecting plate (72) to move, and the connecting plate (72) drives the push rod (73) to push the polyethylene wax product out of the fixed mold (12), thereby effectively helping the user to perform the demoulding operation on the molding device and effectively improving the molding efficiency of the device; Step 3: When the polyethylene wax product is demoulded and falls from the receiving hole onto the receiving plate (95), the operator can control the heat dissipation fan (92) to cool the polyethylene wax product, and then drive the push plate to move through the electric telescopic rod (93), so that the push plate pushes the cooled polyethylene wax product into the collection box (98) for collection, thereby helping the operator to cool and collect the molded polyethylene wax product, reducing the operator's labor intensity.
11. The control method of a polyethylene wax molding device according to claim 10, characterized in that: In step 1, a temperature sensor and a fuzzy control algorithm are introduced. A temperature sensor is installed inside the melting reactor. The temperature is monitored and adjusted in real time through the fuzzy control algorithm to ensure the stability of the melting temperature of the polyethylene wax and improve the molding quality. Then, adaptive control is implemented and an adaptive control algorithm is used to automatically adjust the heating power according to the melting point and viscosity of the polyethylene wax to optimize the melting process, thus forming an intelligent temperature control system for the melting reactor. Introducing temperature sensor and fuzzy control algorithm Temperature sensor installation: Several high-precision temperature sensors are installed at different locations inside the melting reactor: bottom, middle and top, to monitor the temperature distribution inside the reactor in real time; the specific locations are as follows: Bottom temperature sensor T b :Installed at the bottom of the melting reactor to monitor the bottom temperature; Middle temperature sensor T m : Installed in the middle of the melting reactor to monitor the temperature in the middle; Top temperature sensor T t :Installed on the top of the melting reactor to monitor the top temperature; Type of temperature sensor: Use high-precision thermocouple temperature sensors, which have the characteristics of fast response, high measurement accuracy and high temperature resistance; each sensor must ensure long-term stable operation in a high temperature environment; Data transmission and processing: The temperature sensor sends real-time temperature data to the central control system via data line or wireless transmission; data line transmission is suitable for short distance and fixed installation, while wireless transmission is suitable for situations where wiring is inconvenient or flexible movement is required; Data collection and processing: The central control system receives real-time temperature data from the temperature sensor. b , T m , T t ) Real-time collection of temperature data at different locations inside the melting reactor, and transmission of the data to the central control system CCS via a wireless transmission module; Preprocess, filter, denoise and calibrate temperature data to ensure data accuracy and stability; The temperature data is filtered using a discrete low-pass filter: T filtered (t)=α·T raw (t)+(1-α)·T filtered (t-1) Among them, T filtered (t) is the temperature data after filtering; T filtered (t-1) is the temperature data before filtering; T raw (t) is the original temperature data; α is the filter coefficient; Use the moving average method to denoise the filtered data to further improve the stability of the data: Among them, T denoised (t) is the temperature data after denoising; N is the size of the moving average window; T filtered (ti) represents the filtered temperature data at time ti; t represents the current time; i represents the time offset relative to the current time step t; Then calibrate the sensor data according to the known calibration curve or standard temperature data to ensure the accuracy of the temperature data; T calibrated (t)=a·T denoised (t)+b Among them, T calibrated (t) is the temperature data after calibration; T denoised (t) is the temperature after denoising; data a and b are calibration coefficients, determined by calibration experiments; Implementation of adaptive fuzzy control algorithm: Establishment of fuzzy rule base: Based on experience and experimental data, a fuzzy rule base for the melting reactor is established; the rule base includes the fuzzy relationship between temperature and heating power; Fuzzy variable definition: Input variable: Temperature error E(t) = T set -T current and temperature change rate ΔE(t)=E(t)-E(t-1); where T set is the target temperature; T current is the current temperature; E(t) is the current temperature error; E(t-1) is the temperature error at the previous moment; Output variable: heating power adjustment ΔP; Fuzzy Sets: Temperature error E(t): divided into {large negative, medium negative, small negative, zero, small positive, medium positive, large positive}; that is (BN, MN, SN, Z, SP, MP, BP); Temperature change rate ΔΔE(t): divided into {large negative, medium negative, small negative, zero, small positive, medium positive, large positive}; that is (BN, MN, SN, Z, SP, MP, BP); Heating power adjustment ΔP: divided into {large decrease, medium decrease, small decrease, unchanged, small increase, medium increase, large increase}; i.e. (BD, MD, SD, NC, SI, MI, BI); Fuzzy rules: If E(t) is strongly negative and ΔE(t) is strongly negative, then ΔP is strongly increasing; If E(t) is zero and ΔΔE(t) is zero, then ΔP is unchanged; If E(t) is strongly positive and ΔE(t) is strongly positive, then ΔP is strongly negative; Fuzzy reasoning: Based on real-time temperature data, the fuzzy reasoning algorithm is used to calculate the currently required heating power adjustment value; the fuzzy reasoning process includes fuzzification, rule reasoning and defuzzification; Fuzzification: Membership function of temperature error E(t): Where, E(t) is the temperature error; μ BN (E) is the membership degree of the temperature error E(t) in the fuzzy set "big negative" BN; is the upper limit value for defining the "large negative" BN fuzzy set; Define the upper limit of the "medium negative" MN fuzzy set; Rule reasoning: Based on the fuzzy rule base, the activation degree of each rule is calculated using the fuzzy reasoning algorithm; a rule1 =min(μ BN (E(t)), μ BN (ΔE(t))) Among them, α rule1 is the activation degree of rule 1, indicating the degree of compliance with the rule under the current conditions; μ BN (E) is the membership degree of temperature error E(t) in the fuzzy set "big negative" BN; μ BN (ΔE(t)) is the membership of the temperature error change rate ΔE(t) in the fuzzy set "large negative" BN; min means taking the minimum value of the two memberships; Defuzzification: Take the weighted average of the output results of all rules to obtain the exact value of the heating power adjustment ΔP: Among them, α i is the activation degree of the ith rule; P i is the output value corresponding to the i-th rule; ΔP is the heating power adjustment amount; Adaptive adjustment: Introduce adaptive control mechanism to dynamically adjust the parameters of fuzzy rules according to the changes of real-time temperature data to improve the adaptability of the system; Introduction of adaptive factor: The adaptive factor λ(t) is based on the temperature change rate Dynamically adjust the parameters of fuzzy rules; Among them, λ(t) is the adaptive factor; k is the adjustment factor; is the temperature change rate; Fuzzy rule parameter adjustment: According to the adaptive factor λ(t), adjust the membership function of the fuzzy rule; adjust the membership function of the temperature error E(t): m′ BN (E)=λ(t)·μ BN (E) Among them, μ′ BN (E) is the membership degree of the adjusted temperature error E(t); λ(t) is the adaptive factor; μ BN (E) is the membership degree of the temperature error E(t) in the fuzzy set "big negative" BN; Update heating power: Update heating power P(t+1)=P(t)+ΔP according to ΔP; Wherein, P(t) is the current heating power, P(t+1) is the heating power at the next moment; ΔP is the heating power adjustment amount; The above steps are repeated in a loop to carry out the next round of fuzzy adaptive control; Introducing reinforcement learning algorithms: Construction of reinforcement learning environment; State space: Define the state space of the system, including current temperature, target temperature, and heating power; the state vector is expressed as: Among them, s(t) is the state vector; T(t) is the current temperature, which means the real-time temperature of the system at time t; T set is the target temperature; P(t) is the current heating power; ΔT(t) is the temperature change rate; Action space: Define the action space of the system, including increasing heating power, decreasing heating power, and keeping heating power unchanged; the specific actions can be quantified into different heating power adjustment steps; the action set is expressed as: A={a1,a2,a3} Among them, A is the action set; a1 means increasing the heating power, a2 means reducing the heating power, and a3 means keeping the heating power unchanged; Reward function: Design a reward function to encourage the system to stabilize the temperature at the target value in the shortest time while reducing energy consumption; it can be defined as: R(t)=-|T(t)-T set |-λP(t) Among them, R(t) is the reward function; T(t) is the current temperature, which means the real-time temperature of the system at time t; T set is the target temperature; P(t) is the current heating power; λ is the weight coefficient for weighing temperature error and energy consumption; Training of reinforcement learning algorithms: Algorithm selection: Adopt the latest model-based reinforcement learning algorithm, based on the value iteration network (VIN) method; train the reinforcement learning model in a simulation environment, and use the collected temperature data and heating power data to continuously adjust the model parameters so that it can optimize temperature control in actual operation; The state value function V(s) is defined as: V(s)=max a Q(s,a) The state-action value function Q(s, a) is defined as: Where V(s) is the state value function; Q(s, a) is the state-action value function, which represents the expected cumulative reward after taking action a in state s; s is the current state; max a is the maximization operator, which means selecting action a that maximizes Q(s,a); R(s,a) is the immediate reward, which means the direct reward obtained by taking action a in state s; γ is the discount factor, ranging from 0 to 1; P(s′|s,a) is the state transition probability, which means the probability of transitioning to the next state s′ after taking action a in state s; V(s′) is the state value function of the next state s′; During the training process, the following steps are repeated to update the Q value: Initialize the Q value: Q value iterative update formula: Policy Update: Among them, Q(s, a) is the state-action value function; s is the current state; a is the current action; a performs assignment operations on all states s and all actions a; R(s, a) is the immediate reward; γ is the discount factor; P(s′|s, a) is the state transition probability, which indicates the probability of transitioning to the next state s′ after taking action a in state s; π(s) is the strategy, which indicates the optimal action selected in state s; Q(s′, a′) indicates the expected cumulative reward after taking action a′ in state s′; represents the maximum Q value of all possible actions a′ taken in state s′; It means to select the action a that can maximize Q(s,a); PPO optimization is used in strategy update. The PPO algorithm avoids excessive strategy update by limiting the change range between the new and old strategies. It also controls the change of strategy by cutting the strategy ratio and introducing the KL divergence penalty term. The steps are as follows: Collect samples: Run the current new policy π in the environment θ , θ is the policy parameter, collecting the state, action, reward and next state; Calculate the advantage function The advantage function is calculated at each time step using the time difference method The advantage function is defined as: in, is the advantage function; δ t is the time difference error, δ t =R(t)+γV(s t+1 )-V(s t ); R(t) is the immediate reward; γ is the discount factor; λ is the discount factor of the advantage function; the combination of γλ represents the reward for the future time step; V(s t+1 ) is the next state S t+1 The state value function of V(s t ) is the current state s t The state value function of The strategy parameter θ is updated using the clipping strategy ratio method. The specific update formula is as follows: The ratio is r t (θ): Among them, r t (θ) is the clipping strategy ratio; θ is the strategy parameter, π θ For new strategies, For the old strategy; θ (a t |s t ) means in state s t Next, take action a according to the new policy parameter θ t probability; Indicates that in state s t Next, according to the old policy parameters θ old Take action a t probability; Update the policy parameters using the clipping policy ratio method: Among them, L CLIP (θ) is the objective function after clipping; r t (θ) is the clipping strategy ratio; E t The expected value at time t; is the advantage function estimate; clip(r t (θ), 1-∈, 1+∈) represents the strategy ratio r t (θ) is clipped to be in the range of [1-∈, 1+∈]; ∈ is a hyperparameter used to limit the variation of the strategy ratio; The objective function of the KL divergence penalty method is: Among them, L KL (θ) is the objective function of the KL divergence penalty method; E t The expected value at time t; r t (θ) is the strategy ratio; is the advantage function estimate; β is a hyperparameter that controls the weight of the KL divergence penalty term; is the KL divergence, measuring the old strategy With the new strategy π θ In status t The distribution differences under Update the value function: minimize the mean square error and update the parameters of the value function V(s): 50 VF (φ)=E t [(R(t)+γV(s t+1 )-V(s t )) 2 ] Among them, L VF (φ) is the loss function of the value function, which is used to minimize the mean square error to update the parameter φ of the value function V(s); V(s t+1 ) is the next state a t+1 The state value function of V(s t ) is the current state s t state value function; through the above specific steps, combined with online learning and dynamic adjustment, the reinforcement learning model can continuously optimize and adjust in actual operation to ensure the temperature control accuracy of the polyethylene wax molding device and the adaptability of the system, thereby improving production efficiency and product quality.
Citation Information
Patent Citations
Plastic modification device and method
CN114618387A
Friction stir welding fuzzy control algorithm based on neural network
CN115903491A
Air conditioner temperature control method based on deep reinforcement learning
CN116358114A
Reaction kettle temperature control method based on fuzzy control
CN116700393A
Silane intermittent furnace and heating control method thereof
CN118149606A
Cited By
Injection molding mechanism for wax piece injection molding machining
CN119704520A
An injection molding mechanism for wax part injection molding
CN119704520B
