Intelligent control type expansion coating preparation reaction kettle device

Abstract

The invention discloses an intelligent control type expansion coating preparation reaction kettle device which comprises a reaction kettle body, the interior of the reaction kettle body is divided into a plurality of independent stirring cabins through partition plates, each stirring cabin is internally provided with an independent stirring assembly, a feeding assembly and an observation assembly, and each stirring assembly comprises a stirring shaft, stirring blades and a driver; the stirring blade is provided with a semiconductor chilling plate and a heating element, and the feeding assembly comprises a plurality of feeding channels connected with the stirring cabin, is provided with a flow meter, an electromagnetic valve and a dynamic adjusting module, and is used for independently controlling the feeding amount and the feeding speed of each cabin; the control system comprises a multi-channel data acquisition module, a learning module and a coordination control module, the multi-channel data acquisition module is provided with a sensor network for monitoring the temperature, the pressure and the material state of each stirring cabin in real time for each stirring cabin, and the control system performs coordination control on the stirring cabins according to preset parameters or autonomous learning results. And the stirring speed, temperature and feeding strategy of each cabin are dynamically adjusted.

Description

Technical Field

[0001] The present invention relates to the field of reactor devices, and particularly to an intelligent control type expansion coating preparation reactor device. Background Art

[0002] Reactors are widely used in the fields of petroleum, chemical industry, rubber, pesticides, dyes, medicine, and food, etc. They are pressure vessels used to complete processes such as vulcanization, nitrification, hydrogenation, alkylation, polymerization, and condensation. A stirring device is provided inside the reactor to stir and react the mixed components. Some reactors are equipped with temperature facilities to control the temperature inside the reactor.

[0003] During the preparation of existing coatings, it is necessary to precisely control the reaction temperature, pressure, material mixing uniformity, and feeding sequence. When it is necessary to experiment to obtain the optimal temperature, optimal pressure, and optimal feeding sequence, since traditional reactors mostly adopt a single chamber structure, it is necessary to conduct experiments in multiple reactors to obtain the experimental results, which reduces the overall efficiency. It is difficult to simultaneously control different reaction temperatures, pressures, and feeding sequences under the single production state of a single stirring chamber, resulting in low efficiency. Summary of the Invention

[0004] The present invention aims at the disadvantages of the prior art that it requires relatively many experiments to obtain the best stirring effect in a single reactor and has low efficiency, and provides an intelligent control type expansion coating preparation reactor device.

[0005] In order to solve the above technical problems, the present invention is solved by the following technical solutions:

[0006] An intelligent control type expansion coating preparation reactor device, characterized in that it includes:

[0007] A reactor main body, inside which there are several independent stirring chambers separated by partition plates. Each stirring chamber is provided with an independent stirring component, a feeding component, and an observation component. The stirring component includes a stirring shaft, stirring blades, and a driver for driving the stirring shaft to rotate. The stirring blades are provided with a semiconductor refrigerating sheet and a heating element for adjusting the temperature of the corresponding stirring chamber. The feeding component includes several feeding channels, each feeding channel is connected to the corresponding stirring chamber, and is configured with a flow meter, a solenoid valve, and a dynamic adjustment module for independently controlling the feeding amount and feeding speed of each chamber;

[0008] A control system, including a multi-channel data acquisition module, a learning module, and a coordinated control module. The multi-channel data acquisition module is provided with a sensor network for each stirring chamber to monitor the temperature, pressure, and material state of each stirring chamber in real time. The control system dynamically adjusts the stirring speed, temperature, and feeding strategy of each chamber according to preset parameters or the results of autonomous learning.

[0009] By adopting the above technical solutions, multiple independent stirring chambers are separated by partition plates to achieve physical isolation of different reaction stages, avoid experimental data contamination, improve reaction controllability, and the stirring blades integrated with temperature regulation function, semiconductor refrigeration chips cooperate with heating elements, which can directly regulate the local temperature during the stirring process, shorten the heat conduction path, improve the temperature control accuracy and response speed. The multi-channel feeding component cooperates with the dynamic adjustment module, which supports independent control of the feeding parameters of each chamber, such as gradient feeding, to meet the precise feeding requirements of complex formulas. The sensor network is combined with the control system, and through real-time data acquisition and coordinated control, the dynamic optimization of multi-chamber parameters is realized, and the overall reaction efficiency and product quality consistency are improved.

[0010] The present invention is further configured as: the observation component includes a high-definition camera and a spectral analyzer, which are used to collect the mixing uniformity, color change and reaction process data of the materials in each stirring chamber in real time, and generate a fusion degree score through an image processing algorithm.

[0011] By adopting the above technical solutions, the high-definition camera combined with the spectral analyzer can capture the color change and mixing uniformity of the materials in real time, replace the traditional manual sampling detection, reduce the risk of reaction interruption, and the fusion degree score generated by the image processing algorithm provides a quantitative index for the control system, which is convenient for quickly judging the reaction process and triggering adjustment strategies, such as feeding supplementary materials or terminating the reaction.

[0012] The present invention is further configured as: the dynamic adjustment module sets different feeding reference temperatures, feeding speeds and feeding timings for different stirring chambers, and corrects the deviation in real time through the feedback data.

[0013] By adopting the above technical solutions, the gradient feeding strategy controls the differential reference temperature, timing and speed to match the physical and chemical properties of each raw material, such as viscosity and reaction activity, and reduces the phenomena of agglomeration or stratification caused by feeding conflicts.

[0014] The function of real-time correction of feedback data can dynamically compensate for environmental interference, such as temperature fluctuation, to ensure the fitting degree of the feeding process with the preset theoretical model.

[0015] The present invention is further configured as: the learning module is based on the reinforcement learning algorithm. By comparing the observed data of each stirring chamber with the target performance index, it generates an optimized combination of stirring parameters and automatically updates them to the coordinated control module;

[0016] The reinforcement learning algorithm adopts the deep Q-network (DQN). The input parameters include the temperature gradient of the stirring chamber, the pressure fluctuation range, the change of material viscosity and the fusion degree score, and the output parameters are the adjustment amounts of the stirring speed, heating / cooling power and feeding speed.

[0017] By adopting the above technical solution, based on the DQN-based reinforcement learning algorithm, through multi-parameter coupling analysis of temperature gradient, pressure fluctuation, etc., the optimal stirring parameter combination is automatically generated, reducing the cost of manual trial and error. The learning module continuously accumulates data and updates the control strategy, gradually enhancing the system's adaptability to complex working conditions, such as raw material batch differences, and realizing long-term process optimization.

[0018] The present invention is further configured such that: the control system further includes a remote interaction module, which supports receiving external instructions through the industrial Internet of Things protocol and uploading the process data of each stirring chamber to the cloud database in real time for cross-device process synchronization;

[0019] The cloud database stores multiple groups of historical process templates. The coordination control module automatically recommends initial parameters and starts the adaptive optimization process by matching the similarity between the current material characteristics and the templates.

[0020] By adopting the above technical solution, the best process data is saved in the cloud database, which is convenient for synchronization to other reactor devices, thus achieving both optimization of production and reducing the experimental time for stirring materials with high similarity through the historical process templates, and increasing work efficiency.

[0021] The present invention is further configured such that: the sensor network includes distributed temperature sensors, pressure sensors, and viscosity sensors. The distributed temperature sensors, pressure sensors, and viscosity sensors are all embedded in the inner wall of the stirring chamber in a grid layout and encapsulated by an anti-corrosion coating.

[0022] By adopting the above technical solution, the grid-layout embedded sensors cover the entire area of the stirring chamber, avoiding monitoring blind spots and ensuring the comprehensiveness of data collection, such as the detection of local overheating or pressure concentration points. The anti-corrosion coating encapsulation extends the service life of the sensors, adapts to the highly corrosive environment in the preparation of expansion coatings, and reduces the maintenance frequency.

[0023] The present invention is further configured such that: the reactor is provided with a compartmentalized safety pressure relief system. Each stirring chamber is equipped with a first pressure relief valve. When abnormal pressure is detected, only the first pressure relief valve of the corresponding chamber is triggered.

[0024] By adopting the above technical solution, the compartmentalized pressure relief design realizes precise control of abnormal pressure, avoids global pressure relief caused by a single chamber failure, and reduces raw material waste and equipment downtime.

[0025] The independent first pressure relief valve reduces the risk of misoperation and ensures the continuity of the reaction in other chambers, especially suitable for the parallel control scenario of multi-stage reactions.

[0026] The present invention is further configured such that: the compartmentalized safety pressure relief system is equipped with a buffer gas tank. When it is detected that the pressure in the mixing chamber is higher than the preset pressure, the pressure relief valve one of the corresponding chamber is triggered, and the gas discharged by the pressure relief valve one enters the buffer gas cavity. When it is detected that the pressure in the mixing chamber is lower than the preset pressure, the pressure relief valve one of the corresponding chamber is triggered, and the gas in the buffer gas tank enters the corresponding mixing chamber through the pressure relief valve one. The buffer gas cavity is provided with a pneumatic pump and a pressure relief valve two.

[0027] By adopting the above technical solution, the buffer gas tank maintains the dynamic balance of the pressure in the mixing chamber through bidirectional pressure regulation, pressure relief and pressure compensation, avoiding mechanical wear caused by frequent opening and closing of the pressure relief valve one.

[0028] The pneumatic pump is linked with the pressure relief valve one to realize the recycling of gas, such as inert gas circulation, reducing energy consumption and emissions, and meeting the requirements of green production.

[0029] The present invention is further configured such that: the mixing chamber is separated by a partition, the partition is detachably connected to the reaction kettle, and the partition is provided with an openable and closable material communication valve. The control system controls the opening and closing of the material communication valve according to the material state of each chamber to realize multi-chamber collaborative or independent reactions.

[0030] By adopting the above technical solution, the openable and closable material communication valve supports flexible switching of the multi-chamber collaborative mode, such as mixing after step-by-step reactions, expanding the applicable scenarios of the equipment, such as the composite preparation of multi-component coatings.

[0031] Based on the material state to control the opening and closing of the valve, realizing the sequential management of the reaction process, such as starting mixing after the pre-reaction is completed, improving the degree of freedom of process design.

[0032] The present invention is further configured such that: the driver of the mixing component is a servo motor that supports forward and reverse rotation switching and speed adjustment. The surface of the mixing shaft is provided with spiral diversion grooves, and the mixing blades cooperate with the diversion grooves to form a mixing mode of vortex and laminar flow.

[0033] By adopting the above technical solution, the functions of forward and reverse rotation switching and speed adjustment of the servo motor meet the mixing requirements of materials with different viscosities, such as low speed to prevent splashing in the high-viscosity stage and high speed to improve efficiency in the low-viscosity stage. The spiral diversion grooves cooperate with the mixing blades to optimize the shear force distribution through the vortex-laminar flow mixing mode, reducing the deterioration of material properties caused by local overmixing, such as bubble generation or particle breakage. Description of the Drawings

[0034] Figure 1 is a system architecture diagram of an intelligent control type expansion coating preparation reaction kettle device;

[0035] Figure 2 is a structural cross-sectional view of the reaction kettle device.

[0036] The names of the parts referred to by the respective numerical labels in the above drawings are as follows: 1. Reactor main body; 2. Partition board; 3. Stirring chamber; 4. Stirring assembly; 5. Stirring shaft; 6. Stirring blade; 7. Driver; 8. Feeding channel; 9. Material connection valve; 10. Erection ring; 11. Pressure relief valve I. Specific implementation mode

[0037] The present invention will be further described in detail below in conjunction with the drawings and embodiments.

[0038] Embodiment:

[0039] This embodiment provides an intelligent control type expansion coating preparation reactor device, including a reactor main body 1 and a control system. The inside of the reactor main body 1 is divided into four independent stirring chambers 3 by a partition board 2. The four independent stirring chambers 3 are arranged in sequence from high to low along the height direction of the reactor. However, it is not limited to four stirring chambers 3, and it can also be two or more. Each stirring chamber 3 is provided with an independent stirring assembly 4, a feeding assembly and an observation assembly. Erection rings are evenly spaced along the height direction inside the reactor main body 1. The partition board 2 is erected on the erection rings and connected by bolts. The partition board 2 connected by bolts is convenient for disassembly and cleaning or adjusting the number of chambers. A material connection valve 9 that can be opened and closed is provided on the partition board 2 for realizing multi-chamber collaborative reaction or independent reaction. An outlet for discharging materials is arranged at the bottom of the reactor main body 1, through the material.

[0040] The stirring assembly 4 in each stirring chamber 3 includes a stirring shaft 5, a stirring blade 6 and a driver 7. The surface of the stirring shaft 5 is provided with spiral diversion grooves. The stirring blade 6 cooperates with the diversion grooves to form a mixing mode of vortex flow and laminar flow, which is suitable for the uniform mixing of materials with different viscosities. The driver 7 adopts a servo motor, which supports forward and reverse rotation switching and speed adjustment. The stirring blade 6 is integrated with a semiconductor refrigerating sheet and a heating element, and can accurately adjust the temperature within the range of -10°C to 150°C. A fixed sleeve for fixing the stirring shaft 5 is arranged outside the stirring shaft 5. The fixed sleeve is connected and arranged in cooperation with the reactor main body 1. The stirring blades 6 of each stirring chamber 3 are aligned and arranged along the central position of the reactor main body 1. When the driver 7 corresponding to the stirring chamber 3 is located on the side of the reactor main body 1, a 90° commutator is arranged on the stirring shaft 5 cooperating with the driver 7 for driving the stirring blade 6 to be arranged along the central position of the reactor main body 1.

[0041] The feeding assembly includes at least eight feeding channels 8. Each stirring chamber 3 is correspondingly provided with at least two feeding channels 8. The two feeding channels 8 in each stirring chamber 3 are used to circulate materials of different media. Each feeding channel 8 is configured with a high-precision flowmeter, a solenoid valve and a dynamic adjustment module. The dynamic adjustment module supports a gradient feeding strategy and can set different feeding reference temperatures, feeding speeds and feeding sequences according to different stirring chambers 3. For example, the feeding reference temperature of chamber A is 25°C and the feeding speed is 10 mL / min. The feeding reference temperature of chamber B is 40°C and the feeding speed is 5 mL / min. Deviations are corrected in real time through feedback data to ensure feeding accuracy.

[0042] An observation window is configured on the reactor body 1 facing each stirring chamber 3. The observation assembly is arranged opposite to the observation window. The observation assembly includes a high-definition camera and a spectral analyzer. The high-definition camera uses a high-resolution and high-frame-rate imaging device to collect data on the mixing uniformity and color change of the materials in real time. The spectral analyzer covers the visible light to near-infrared band and is used to monitor chemical changes during the reaction process. The observation data generates a fusion degree score through an image processing algorithm, such as 0-100 points. When the score is lower than the preset score, for example, lower than the preset 80 points, the system adjustment strategy is triggered.

[0043] The control system includes a multi-channel data acquisition module, a learning module, a coordinated control module and a remote interaction module.

[0044] The multi-channel data acquisition module monitors the temperature, pressure and material state of each stirring chamber 3 in real time through a sensor network. The sensor network includes distributed temperature sensors, pressure sensors and viscosity sensors, which are embedded in the inner wall of the stirring chamber 3 in a grid layout and encapsulated with an anti-corrosion coating.

[0045] The learning module, based on the deep Q-network (DQN) reinforcement learning algorithm, has input parameters including temperature gradient, pressure fluctuation range, material viscosity change and fusion degree score, and output parameters are the adjustment amounts of stirring speed, heating / cooling power and feeding speed. For example, when it is detected that the material viscosity increases, the system automatically increases the stirring speed and reduces the heating power.

[0046] The coordinated control module dynamically adjusts the stirring speed, temperature and feeding strategy of each chamber according to the preset parameters or the optimization results of the learning module.

[0047] The remote interaction module supports receiving external instructions through an industrial Internet of Things protocol (such as MQTT) and uploading process data to the cloud database in real time. The cloud database stores multiple groups of historical process templates. The coordinated control module automatically recommends initial parameters and starts an adaptive optimization process by matching the similarity between the current material characteristics and the templates.

[0048] The compartmented safety pressure relief system is equipped with an independent pressure relief valve - 11 in each mixing compartment 3. When abnormal pressure is detected, only the pressure relief valve - 11 of the corresponding compartment is triggered. The pressure relief system is also equipped with a buffer gas tank. When the pressure in the compartment is higher than the preset value, such as 0.5 MPa, the pressure relief valve - 11 opens, and the gas enters the buffer gas tank; when the pressure is lower than the preset value, such as 0.3 MPa, the gas in the buffer gas tank returns to the compartment through the pressure relief valve - 11. The buffer gas tank is provided with a pneumatic pump and a pressure relief valve - 2 to ensure pressure balance. Another purpose of setting the buffer gas tank is to control the gas entering the compartment to avoid the reaction between the external gas and the components in the tank, and it can also uniformly process the discharged gas to avoid polluting the environment with toxic and harmful gases.

[0049] Taking the preparation of two - component expanding paint as an example:

[0050] Step 1: Add equal amounts of component A and equal amounts of component B into compartments 1 to 4 respectively.

[0051] Step 2: Set different stirring temperatures or stirring speeds for compartments 1 to 4. For example, the temperature of compartment 1 is 30 °C and the stirring speed is 500 rpm; the temperature of compartment 2 is 30 °C and the stirring speed is 300 rpm; the temperature of compartment 3 is 40 °C and the stirring speed is 500 rpm; the temperature of compartment 4 is 40 °C and the stirring speed is 300 rpm.

[0052] Step 3: Real - time monitor the material state through the observation component. When the integration degree score of compartment 1 reaches 90 points, open the material connection valve 9 on the partition 2 to transfer component A to compartment 2 for mixing.

[0053] Step 4: The learning module optimizes the stirring parameters according to the real - time data, such as increasing the stirring speed to 800 rpm and reducing the temperature of compartment 2 to 35 °C, to ensure the mixing uniformity.

[0054] Step 5: When abnormal pressure is detected, the compartmented safety pressure relief system automatically adjusts to ensure the reaction safety.

[0055] Step 6: After the reaction is completed, the remote interaction module uploads the process data to the cloud database for subsequent production reference.

[0056] Among them, Step 1 can also be transformed. For example, adjust the feeding order of component A and component B, or control the feeding speed when continuously pouring component A or component B.

[0057] Through the compartmentalized design and independent control, precise preparation of multi-component coatings is achieved, an experimental control group is added, the stirring blade 6 with integrated temperature regulation function and the dynamic feeding strategy are adopted to improve the reaction efficiency and product quality consistency. The intelligent control system based on reinforcement learning significantly reduces the need for manual intervention, can adapt to complex working conditions, and can also save the best process data through the cloud database for easy synchronization to other reactor devices. The parameters of the best reaction efficiency obtained from the experiment can be replicated in other reactors. Through the historical process template, the experimental time for stirring materials with high similarity can be reduced, and the work efficiency can be increased.

Claims

1. An intelligent control type intumescent coating preparation reactor device, characterized in that: include: A reactor body (1) is provided with a plurality of independent stirring chambers (3) separated by partitions (2). Each stirring chamber (3) is provided with an independent stirring assembly (4), a feeding assembly and an observation assembly. The stirring assembly (4) comprises a stirring shaft (5), a stirring blade (6) and a driver (7) for driving the stirring shaft (5) to rotate. The stirring blade (6) is provided with a semiconductor cooling plate and a heating element for adjusting the temperature of the corresponding stirring chamber (3). The feeding assembly comprises a plurality of feeding channels (8). Each feeding channel (8) is connected to a corresponding stirring chamber (3) and is provided with a flow meter, a solenoid valve and a dynamic adjustment module for independently controlling the feeding amount and feeding speed of each chamber. The control system comprises a multi-channel data acquisition module, a learning module and a coordination control module, wherein the multi-channel data acquisition module is provided with a sensor network for real-time monitoring of the temperature, pressure and material state of each stirring chamber (3), and the control system dynamically adjusts the stirring speed, temperature and feeding strategy of each chamber according to preset parameters or autonomous learning results.

2. The intelligent control type intumescent coating preparation reactor device according to claim 1, characterized in that: The observation component comprises a high-definition camera and a spectrum analyzer, which are used to collect the mixing uniformity, color change and reaction progress data of the materials in each mixing chamber (3) in real time, and generate a fusion degree score through an image processing algorithm.

3. The intelligent control type intumescent coating preparation reactor device according to claim 1, characterized in that: The dynamic adjustment module sets differentiated feed reference temperatures, feed speeds and feed timings for different mixing chambers (3), and corrects deviations in real time through feedback data.

4. The intelligent control type intumescent coating preparation reactor device according to claims 1 to 3, characterized in that: The learning module is based on a reinforcement learning algorithm, and generates an optimized stirring parameter combination by comparing the observed data of each stirring chamber (3) with the target performance index, and automatically updates it to the coordination control module; The reinforcement learning algorithm adopts a deep Q network, the input parameters include the temperature gradient of the stirring chamber (3), the pressure fluctuation range, the material viscosity change and the fusion degree score, and the output parameters are the stirring speed, the heating / cooling power and the adjustment amount of the feeding speed.

5. The intelligent control type intumescent coating preparation reactor device according to claim 4, characterized in that: The control system also includes a remote interaction module, which supports receiving external instructions through the industrial Internet of Things protocol and uploading the process data of each mixing chamber (3) to the cloud database in real time for cross-device process synchronization; The cloud database stores multiple groups of historical process templates. The coordination control module automatically recommends initial parameters and starts an adaptive optimization process by matching current material characteristics with template similarities.

6. The intelligent control type intumescent coating preparation reactor device according to claim 1, characterized in that: The sensor network comprises distributed temperature sensors, pressure sensors and viscosity sensors, which are all embedded in the inner wall of the stirring chamber (3) in a grid-like layout, and are all encapsulated by an anti-corrosion coating.

7. The intelligent control type intumescent coating preparation reactor device according to claim 1, characterized in that: The reactor is provided with a compartment-type safety pressure relief system, and each of the mixing chambers (3) is provided with a pressure relief valve (11). When abnormal pressure is detected, only the pressure relief valve (11) of the corresponding chamber is triggered.

8. The intelligent control type intumescent coating preparation reactor device according to claim 7, characterized in that: The compartment-type safety pressure relief system is equipped with a buffer gas tank. When it is detected that the pressure in the mixing chamber (3) is higher than a preset pressure, the pressure relief valve (11) of the corresponding chamber is triggered, and the gas discharged from the pressure relief valve (11) enters the buffer gas cavity. When it is detected that the pressure in the mixing chamber (3) is lower than the preset pressure, the pressure relief valve (11) of the corresponding chamber is triggered, and the gas in the buffer gas tank enters the corresponding mixing chamber (3) through the pressure relief valve (1). The buffer gas cavity is provided with an air pressure pump and a pressure relief valve (2).

9. The intelligent control type intumescent coating preparation reactor device according to claim 1, characterized in that: The mixing chamber (3) is separated by a partition (2), and the partition (2) is detachably connected to the reaction kettle. An openable and closable material connecting valve (9) is provided on the partition (2). The control system controls the opening and closing of the material connecting valve (9) according to the material state of each chamber, thereby realizing coordinated or independent reaction of multiple chambers.

10. The intelligent control type intumescent coating preparation reactor device according to claim 1, characterized in that: The driver (7) of the stirring assembly (4) is a servo motor that supports forward and reverse switching and speed adjustment. The surface of the stirring shaft (5) is provided with a spiral guide groove, and the stirring blade (6) cooperates with the guide groove to form a mixed mode of laminar flow and vortex flow.

Images