Self-adaptive cooperative control system and method for food processing stirring process

Abstract

The invention belongs to the technical field of process automatic control, and discloses a self-adaptive cooperative control system and method for a food processing stirring process. The system comprises a multi-dimensional process analysis unit used for collecting a multi-dimensional parameter data stream in real time; and a cooperative control module. A process stage characteristic model database and a cooperative control strategy library are stored in the cooperative control module, and the cooperative control module is configured to process the multi-dimensional parameter data flow in real time and match with the model database in combination with the change rate trend of the multi-dimensional parameter data flow so as to diagnose the current process stage of the material; then, a preset cooperative control strategy corresponding to the stage is called to cooperatively control the operation states of a low-shear main stirring mechanism, a high-shear dispersing and emulsifying mechanism, a self-adaptive wall scraping mechanism and a temperature control unit. Through active process stage diagnosis and cooperative control, adaptive optimization of the process is realized, and the product quality, the production efficiency and the process stability are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of process automatic control, and in particular relates to an adaptive collaborative control system and method for a food processing and stirring process. Background Art

[0002] Mixing is an essential core unit operation in the food, chemical, and pharmaceutical industries. Its purpose is to achieve uniform mixing of materials, promote mass and heat transfer, or achieve specific physical and chemical changes such as dissolution, emulsification, and crystallization. Traditional mixing equipment typically operates with preset fixed parameters (such as fixed stirring speed and stirring time) or relies entirely on manual adjustment based on operator experience. This approach suffers from low control precision and difficulty adapting to the complex nonlinear changes that occur during material processing, resulting in inconsistent product quality between batches and high energy consumption.

[0003] To improve control accuracy, some stirring systems with sensors have emerged. For example, Chinese patent CN110935337A discloses a stirring system capable of monitoring parameters such as viscosity, temperature, and stirring speed during the stirring process. Its core control logic determines whether the system is experiencing an anomaly by monitoring whether the viscosity change rate (dη / dt) deviates from a learned "normal range." However, this existing technology is essentially a process stability monitoring and abnormal state detection mechanism. Its goal is to maintain the entire stirring process within a preset, single "normal" trajectory.

[0004] The limitation of this approach is that it fails to distinguish between the multiple process stages inherent in complex food processing, each with distinct physicochemical properties. For example, the production process of a typical mayonnaise or salad dressing involves several stages, including wetting and dissolving the powdered raw materials, shear dispersion of the oil phase, oil-water emulsification, and finally cooling and thickening. The optimal processing conditions for each stage (such as macro-mixing intensity, micro-shear intensity, and heat transfer efficiency requirements) are completely different. Judging the entire process using a unified set of "normal / abnormal" criteria fails to provide optimized, differentiated control strategies tailored to the specific requirements of different stages, such as dissolution, dispersion, emulsification, and heat transfer. Therefore, when dealing with such complex, multi-stage, nonlinear processes, it still suffers from low processing efficiency, high energy consumption, and difficulty achieving optimal microstructure and quality uniformity in the final product. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the defect that the existing technology can only perform anomaly detection but cannot achieve stage-by-stage adaptive optimization, and provide an adaptive collaborative control system and method that can actively identify the different physical and chemical stages of complex processes and call the most optimized collaborative control strategy for each stage, thereby significantly improving processing efficiency, product quality and process stability, and reducing dependence on manual experience.

[0006] The technical solution adopted in the present invention is: In order to solve the above technical problems, the present invention provides an adaptive collaborative control system for a food processing stirring process, which is applied to a stirring equipment including a tank body, a reconfigurable low-shear main stirring mechanism located in the tank body, a high-shear dispersion and emulsification mechanism, and an adaptive scraping mechanism. The system includes: a multidimensional process analysis unit and a collaborative control module.

[0007] The multidimensional process analysis unit is configured to collect in real time multidimensional parameters of the material in the tank, including at least viscosity, temperature, motor power of the low-shear main stirring mechanism, motor power of the high-shear dispersion and emulsification mechanism, and particle size, to form a real-time data stream.

[0008] The collaborative control module internally stores a process stage characteristic model database and a preset collaborative control strategy library corresponding to each process stage. The collaborative control module is configured to: receive and process the real-time data stream from the multidimensional process analysis unit in real time; match the real-time data stream and its change rate trend with the process stage characteristic model database to diagnose the process stage of the current material from multiple preset process stages; and call the corresponding collaborative control strategy from the preset collaborative control strategy library for the diagnosed current process stage to collaboratively and automatically control at least two of the configuration and speed of the low-shear main stirring mechanism, the operating parameters of the high-shear dispersion and emulsification mechanism, and the scraping pressure of the adaptive scraping mechanism.

[0009] Preferably, the collaborative control module is also configured to: after at least one process batch is completed, receive the final quality data of the batch product, and associate and analyze the quality data with the real-time data stream recorded for the batch, so as to automatically update or optimize the process stage feature model database or the preset collaborative control strategy library.

[0010] Preferably, the collaborative control module is also configured to perform predictive judgment: by analyzing the time derivatives of one or more parameters in the real-time data stream, to predict the upcoming process stage transition or potential process anomaly, and call the corresponding collaborative control strategy in advance.

[0011] Preferably, the multidimensional process analysis unit further includes at least one of an acoustic sensor or an online pH meter, and the collaborative control module uses the acoustic characteristics collected by the acoustic sensor or the pH value collected by the online pH meter as additional dimensional parameters for matching diagnosis.

[0012] Preferably, the process stages preset in the process stage characteristic model database include at least: powder wetting and dissolving stage, high viscosity shear dispersion stage, oil-water emulsification stage, and cooling and thickening stage; The collaborative control strategy corresponding to the oil-water emulsification stage stored in the collaborative control strategy library is a closed-loop collaborative control strategy, wherein the collaborative control module instructs the low-shear main stirring mechanism to switch to an asymmetric configuration for generating a directional pumping flow and operate at a low speed. At the same time, based on the real-time particle size feedback of the online particle size analyzer in the multidimensional process analysis unit, the operating parameters of the high-shear dispersion emulsification mechanism are closed-loop adjusted to make the material particle size reach a preset target.

[0013] Preferably, the preset collaborative control strategy also includes temperature control of the tank jacket, and the collaborative control module adjusts the flow rate or temperature of the medium entering the jacket according to the feedback from the multi-point temperature sensor in the multi-dimensional process analysis unit to collaboratively achieve the heat transfer target of a specific process stage.

[0014] The present invention also provides a method for using the above system, comprising the steps of: a) Through the multi-dimensional process analysis unit, the multi-dimensional parameters of the material in the mixing equipment are collected in real time to form a real-time data stream; b) The collaborative control module receives and processes the real-time data stream, matches it and its rate of change trend with the internally stored process stage characteristic model database to diagnose the process stage of the current material; c) For the diagnosed current process stage, the collaborative control module calls the corresponding collaborative control strategy to collaboratively and automatically control the various actuators of the stirring equipment until the process is completed.

[0015] Preferably, the method of use further comprises the steps of: d) After at least one process batch is completed, the collaborative control module receives the final quality data of the batch product and associates and analyzes the quality data with the real-time data stream recorded for the batch to automatically update or optimize the process stage feature model database or the preset collaborative control strategy library.

[0016] Preferably, in step b), when the current process stage is diagnosed as the powder wetting and dissolution stage, the collaborative control strategy called in step c) is: instructing the low-shear main stirring mechanism to operate at a high speed in a symmetrical configuration to produce strong macroscopic mixing, and instructing the high-shear dispersion and emulsification mechanism to remain closed.

[0017] Preferably, in step b), when the current process stage is diagnosed as the cooling and thickening stage, the collaborative control strategy called in step c) is: instructing the high-shear dispersion and emulsification mechanism to shut down; instructing the low-shear main stirring mechanism to operate at a low speed in a full sweep configuration to enhance heat transfer; and instructing the adaptive scraping mechanism to increase the scraping pressure to keep the heat exchange surface clean.

[0018] The beneficial effects of the present invention are: 1. A leap from "passive reaction" to "active cognition and prediction": This invention no longer passively corrects process deviations. Instead, it actively and intelligently identifies the physical and chemical stage of the current process through multi-dimensional data fusion and model matching. It even predicts stage transitions by analyzing data change rate trends, achieving a deep understanding and proactive control of complex processes.

[0019] 2. Achieved system-level collaborative optimization and energy efficiency improvement: By calling exclusive collaborative control strategies for different stages, the present invention can enable different hardware units (such as main stirring, high shear unit) to play the optimal role (such as macro mixer, high-efficiency delivery pump) at different stages. While ensuring product quality, it avoids the energy waste caused by "one-size-fits-all" control and maximizes the overall energy efficiency and processing efficiency of the system.

[0020] 3. Introducing self-learning and evolutionary capabilities to achieve continuous process improvement: The system of the present invention can associate final product quality with process data, and self-optimize control models and strategies through machine learning algorithms, giving the equipment the ability to "accumulate experience" and "iterate processes", breaking away from long-term dependence on senior operators.

[0021] 4. Significantly improved product quality and process stability: This invention enables previously unattainable refined control. For example, by performing closed-loop feedback control of particle size during the emulsification stage, it is possible to achieve product microstructure, taste, and high batch-to-batch consistency that was difficult to achieve with traditional control methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the overall structure of the food processing and stirring equipment of the present invention.

[0023] Figure 2 It is a partial structural diagram of the multidimensional process analysis unit of the present invention.

[0024] Figure 3 It is a schematic diagram of an embodiment of the variable geometry stirring blade of the present invention.

[0025] Figure 4 It is a working principle diagram of the food processing and stirring equipment of the present invention.

[0026] In the figure: 1-tank body; 21-drive unit; 22-main stirring shaft; 23-actuator; 24-inner control rod; 25-first support arm; 26-connecting rod; 27-telescopic rod; 28-blade body; 29-linkage structure; 3-high shear dispersion and emulsification mechanism; 41-second support arm; 42-adaptive unit; 43-scraper; 51-online viscosity sensor; 52-multi-point temperature sensor; 53-online particle size analyzer. DETAILED DESCRIPTION

[0027] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present invention and are not to be construed as limiting the present invention. Generally, the components of the embodiments of the present invention described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations.

[0028] The present invention provides an adaptive collaborative control system and method, which is generally applied to Figure 1 The food processing and mixing equipment shown in the figure comprises a jacketed tank 1, a low-shear main mixing mechanism located within the tank 1, a high-shear dispersing and emulsifying mechanism 3 integrated into the bottom of the tank 1, and an adaptive scraping mechanism mounted on the main mixing shaft 22. The tank 1 is secured by a support structure and is used to hold material (i.e., food slurry). The interior surface of the tank 1 is mirror-polished, and the bottom is arc-shaped for easy discharge and cleaning. The jacket is used to pass media such as steam, hot water, or cold water.

[0029] The low shear main stirring mechanism is located in the center of the tank body 1, and its core is reconfigurable. The low shear main stirring mechanism is configured to switch between at least two configurations, which include at least: a symmetrical configuration for generating strong macroscopic mixing, and an asymmetric configuration for generating a directional pumping flow.

[0030] Specifically, the low shear main stirring mechanism includes a driving unit 21, a main stirring shaft 22 and at least one layer of variable geometry stirring blades installed thereon. Figure 3As shown, the angle and extension length of the blade body 28 can be adjusted via a transmission mechanism (composed of an actuator 23, an inner control rod 24, a first support arm 25, a connecting rod 26, a telescopic rod 27, and a linkage structure 29). In this embodiment, the main agitator shaft 22 is a hollow shaft. Multiple actuators 23 are connected to the main agitator shaft 22. Each actuator 23 is connected to an inner control rod 24, which is used to drive the inner control rod 24 to move up and down. The connecting rod 26 is connected to the first support arm 25 fixed to the outer wall of the main agitator shaft 22 via one or more hinges. One end of the connecting rod 26 is connected to the inner control rod 24 via a linkage structure 29. The other end of the connecting rod 26 is connected to the telescopic rod 27 via a rotating member. The telescopic rod 27 is connected to the blade body 28. The rotating member can drive the telescopic rod 27 to rotate circumferentially, allowing the telescopic rod 27 to extend along its own axial length, thereby driving the blade body 28 to extend or retract. The rotating member and the telescopic rod 27 are both driven by independent power units.

[0031] The high-shear dispersing and emulsifying mechanism 3 features a multi-stage structure, integrating at least two rotor-stator systems with different tooth profiles. The first stage utilizes coarser teeth for initial crushing and dispersion, while the second stage utilizes finer teeth for homogenization and emulsification. The discharge port features a switchable return path, controlled by a valve controlled by a coordinated control module. This allows for return flow to the upper liquid surface of the tank 1 to promote overall circulation, or to the bottom of the main agitation mechanism for rapid, localized reprocessing.

[0032] The adaptive scraping mechanism includes a second support arm 41 connected to the main stirring shaft 22, an adaptive unit 42 (e.g., a cylinder) connected to the second support arm 41, and a scraper 43 connected to the adaptive unit 42. The scraper 43 is made of a flexible food-grade material. The adaptive unit 42 provides scraping pressure to ensure that the scraper 43 adheres to the inner wall and bottom of the tank body 1.

[0033] The adaptive collaborative control system provided by the present invention includes a multi-dimensional process analysis unit and a collaborative control module, such as Figure 1 、 Figure 2 and Figure 4 As shown, the multi-dimensional process analysis unit is the perception center of the present invention, which captures material status information in real time and continuously through a series of online sensors. In this embodiment, it includes at least: Online viscosity sensor 51: Real-time monitoring of the core macrorheological indicators of materials - viscosity (η) and its rate of change (dη / dt).

[0034] Multi-point temperature sensors 52: located at the upper, middle and lower parts of the tank and at the inlet and outlet of the jacket, used to monitor temperature uniformity and heat transfer efficiency.

[0035] Dual motor power sensor (not shown): monitors the motor power consumption (Pmain ,P hs ), which directly reflects their respective operating loads.

[0036] Online particle size analyzer 53: connected to the bottom of the tank 1 or the outlet of the high shear dispersion emulsification mechanism 3 through a bypass, it measures the particle size distribution of the slurry (such as D50, D90) in real time, and provides direct microscopic quality feedback for the emulsification, homogenization and other processes.

[0037] In a preferred embodiment, in order to further improve the dimension and accuracy of perception, this unit can also integrate: Acoustic sensor (not shown): Installed on the exterior of tank 1, it collects sound or vibration signals during the mixing process. Different process stages (such as the rustling sound of powder wetting, the dull thud of emulsification, and the cavitation sound of foam generation) have distinct acoustic signatures, providing a powerful tool for identifying phases and anomalies.

[0038] Online pH or conductivity meter (not shown): For processes where pH or ion concentration are critical parameters (such as yogurt fermentation and salt dissolution), it provides real-time data on key chemical parameters.

[0039] like Figure 4 As shown, the collaborative control module is the decision-making core of the present invention. It is an industrial computer or PLC system integrated with advanced control algorithms. Its core functions are based on two key internal databases: a process stage feature model database and a preset collaborative control strategy library.

[0040] The process stage characteristic model database is established by analyzing, summarizing, and modeling a large amount of experimental data from successful process batches (for example, through training with machine learning algorithms such as support vector machines, decision trees, or neural networks). It maps abstract "process stages" to specific multi-dimensional parameter feature vectors.

[0041] For example, in the process of making salad dressing, the model library contains: Model 1: Powder wetting and dissolution stage Characteristic vector: {low viscosity η & low dη / dt, low P main ,P hs =0, particle size D90>100μm and decreasing, specific low-frequency acoustic characteristics} Model 2: Oil-water emulsification and microstructure formation stage Eigenvector: {high viscosity η, high P main &High P hs , particle size D90 < 10 μm and approaching the target value, specific high-frequency acoustic characteristics} Model 3: Cooling and thickening stage Characteristic vector: {viscosity η increases sharply as temperature T decreases, Phs =0,P main The viscosity increases and the temperature T decreases significantly. The preset collaborative control strategy library corresponds one-to-one to each of the above process stage models, and predefines how the various hardware units (actuators) work together in this stage to achieve the specific process goals of this stage.

[0042] Strategy 1 (corresponding to Model 1): Target: Rapid macro-mixing. Coordinated actions: The low-shear main stirring mechanism switches to a symmetrical configuration and operates at high speed; the high-shear dispersing and emulsifying mechanism 3 is shut down; and the adaptive scraping mechanism operates at low pressure.

[0043] Strategy 2 (corresponding to Model 2): ​​Target: Precise microemulsification. Coordinated Action: The low-shear main agitator switches to an asymmetric pumping configuration, operating at low speed to steadily transport the material to the high-shear zone. The speed of the high-shear dispersing and emulsifying mechanism 3 is regulated by the coordinated control module using PID or fuzzy logic closed-loop control based on the real-time D90 readings from the online particle size analyzer 53. Cooling water is introduced into the jacket for precise temperature control.

[0044] Strategy 3 (corresponding to Model 3): Goal: Efficient and uniform heat transfer. Coordinated actions: High-shear dispersing and emulsifying mechanism 3 is deactivated; the low-shear main stirring mechanism switches to a full sweep configuration, operating at very low speeds; and the adaptive scraping mechanism increases scraping pressure to keep the heat exchange surface clean.

[0045] A key innovation of this system is that it not only performs post-matching but also pre-emptive prediction. The collaborative control module continuously calculates the time derivatives (first- and second-order rates of change) of key parameters. For example, if the viscosity rate of change (dη / dt) changes from positive to negative and reaches an inflection point, the system can predict that emulsification is nearing completion and is about to enter the shear-thinning or stabilization phase. This allows for proactive control adjustments to prevent excessive shearing.

[0046] Another core innovation of this system is its self-evolutionary capability. After a process batch is completed, the final offline quality test data for that batch (such as laboratory-measured final viscosity, shelf stability score, and sensory score) is fed into the collaborative control module. Machine learning algorithms within the module (such as reinforcement learning or regression analysis models) correlate these final results with the complete process data stream for that batch, fine-tuning the boundaries of the "process stage characteristic model" or specific parameters in the "collaborative control strategy" (such as the optimal target particle size in the emulsification stage). In this way, the equipment continuously accumulates experience during production, enabling continuous and autonomous process optimization.

[0047] Example of Adaptive Collaborative Control Method: Fully Automatic Production of High-Quality Salad Dressing 1. Startup and Dissolution: The operator adds aqueous ingredients such as water, egg yolk, and vinegar, and powders such as starch and spices, to Tank 1. On the touchscreen, they select the "Salad Dressing - High Quality" recipe and start the process. The system begins collecting multidimensional parameters and immediately matches them to the model library. It then identifies the current stage as "Powder Wetting and Dissolution" and automatically invokes the corresponding Strategy 1.

[0048] 2. Emulsification: After about 5 minutes, the system determines that the powder wetting and dissolution stage is complete based on changes in parameters such as viscosity and particle size. At this time, the system prompts the operator to start adding the oil phase through the metering pump. As the oil is added, the system monitors that the parameter characteristic vector drifts toward the model of "oil-water emulsification and microstructure formation stage" (i.e., Model 2). When the matching degree exceeds the preset threshold, the system automatically switches seamlessly to the corresponding strategy 2. The low-shear main stirring mechanism changes to a low-speed asymmetric pumping mode, and the high-shear dispersion emulsification mechanism 3 is started. Its speed is dynamically adjusted under the closed-loop control of the collaborative control module. The goal is to stabilize the reading of the online particle size analyzer 53 at D90=4.5±0.5μm while optimizing the motor energy consumption.

[0049] 3. Cooling: After the oil phase is added and emulsification continues for a period of time, the system determines that the oil-water emulsification and microstructure formation phases are complete based on the stable particle size and no significant change in energy consumption, and automatically enters the cooling process. The system recognizes that the parameter characteristics match Model 3 of the "Cooling and Thickening Phase" and immediately invokes the corresponding Strategy 3.

[0050] 4. Completion and Learning: When the average reading of the multi-point temperature sensor 52 reaches the preset 20°C, the system issues a process completion notification and saves all process data for this batch for quality traceability. If quality inspection data for this batch is subsequently input, the system will automatically execute the learning algorithm in the background to optimize the model for the next batch of production.

[0051] The present invention is not limited to the above-mentioned optional implementation modes. Anyone can derive other forms of products under the inspiration of the present invention. However, no matter what changes are made in the shape or structure, any technical solution that falls within the scope defined by the claims of the present invention falls within the scope of protection of the present invention.

Claims

1. An adaptive collaborative control system for a food processing stirring process, which is applied to a stirring device comprising a tank, a reconfigurable low-shear main stirring mechanism located within the tank, a high-shear dispersing and emulsifying mechanism, and an adaptive scraping mechanism, characterized in that: The system comprises: a multidimensional process analysis unit configured to collect, in real time, multidimensional parameters of the material in the tank, including at least viscosity, temperature, motor power of the low-shear main stirring mechanism, motor power of the high-shear dispersing and emulsifying mechanism, and particle size, to form a real-time data stream; The collaborative control module stores a database of process stage feature models and a preset collaborative control strategy library corresponding to each process stage. The collaborative control module is configured as follows: receiving and processing the real-time data stream from the multi-dimensional process analysis unit in real time; Matching the real-time data stream and its rate of change trend with the process stage characteristic model database to diagnose the process stage of the current material from a plurality of preset process stages; For the diagnosed current process stage, the corresponding collaborative control strategy is called from the preset collaborative control strategy library to collaboratively and automatically control at least two of the configuration and speed of the low-shear main stirring mechanism, the operating parameters of the high-shear dispersion and emulsification mechanism, and the scraping pressure of the adaptive scraping mechanism.

2. The system according to claim 1, wherein: The collaborative control module is also configured to: after at least one process batch is completed, receive the final quality data of the batch product, and associate and analyze the quality data with the real-time data stream recorded for the batch to automatically update or optimize the process stage feature model database or the preset collaborative control strategy library.

3. The system according to claim 2, characterized in that: The collaborative control module is further configured to perform predictive judgments by analyzing the time derivatives of one or more parameters in the real-time data stream to predict upcoming process stage transitions or potential process anomalies and to call corresponding collaborative control strategies in advance.

4. The system according to claim 1, wherein: The multidimensional process analysis unit further includes at least one of an acoustic sensor or an online pH meter, and the collaborative control module uses the acoustic characteristics collected by the acoustic sensor or the pH value collected by the online pH meter as additional dimensional parameters for matching diagnosis.

5. The system according to claim 1, wherein: The process stages preset in the process stage characteristic model database include at least: powder wetting and dissolving stage, high viscosity shear dispersion stage, oil-water emulsification stage, and cooling and thickening stage; The collaborative control strategy corresponding to the oil-water emulsification stage stored in the collaborative control strategy library is a closed-loop collaborative control strategy, wherein the collaborative control module instructs the low-shear main stirring mechanism to switch to an asymmetric configuration for generating a directional pumping flow and operate at a low speed. At the same time, based on the real-time particle size feedback of the online particle size analyzer in the multidimensional process analysis unit, the operating parameters of the high-shear dispersion emulsification mechanism are closed-loop adjusted to make the material particle size reach a preset target.

6. The system according to claim 1, wherein: The preset collaborative control strategy also includes temperature control of the tank jacket. The collaborative control module adjusts the flow rate or temperature of the medium entering the jacket based on feedback from multi-point temperature sensors in the multi-dimensional process analysis unit to collaboratively achieve the heat transfer target of a specific process stage.

7. An adaptive collaborative control method for a food processing and mixing process, characterized in that: The following steps are involved: a) using a multidimensional process analysis unit to collect, in real time, multidimensional parameters of the material in the stirring device, including at least viscosity, temperature, motor power of the low-shear main stirring mechanism, motor power of the high-shear dispersing and emulsifying mechanism, and particle size, to form a real-time data stream; b) The collaborative control module receives and processes the real-time data stream, matches it and its rate of change trend with an internally stored process stage feature model database, and diagnoses the process stage of the current material from a plurality of preset process stages; c) For the current process stage diagnosed in step b), the collaborative control module calls the corresponding collaborative control strategy from the preset collaborative control strategy library to collaboratively and automatically control at least two of the configuration and speed of the low-shear main stirring mechanism in the stirring equipment, the operating parameters of the high-shear dispersion and emulsification mechanism, and the scraping pressure of the adaptive scraping mechanism until the process is completed.

8. The method according to claim 7, characterized in that Also includes the steps: d) After at least one process batch is completed, the collaborative control module receives the final quality data of the batch product and associates and analyzes the quality data with the real-time data stream recorded for the batch to automatically update or optimize the process stage feature model database or the preset collaborative control strategy library.

9. The method according to claim 7 or 8, characterized in that In step b), when the current process stage is diagnosed as the powder wetting and dissolution stage, the collaborative control strategy called in step c) is: instructing the low-shear main stirring mechanism to operate at a high speed in a symmetrical configuration to produce strong macroscopic mixing, and instructing the high-shear dispersion and emulsification mechanism to remain closed.

10. The method according to claim 7 or 8, characterized in that In step b), when the current process stage is diagnosed as the cooling and thickening stage, the collaborative control strategy called in step c) is: instructing the high-shear dispersion and emulsification mechanism to shut down; instructing the low-shear main stirring mechanism to operate at a low speed in a full sweep configuration to enhance heat transfer; and instructing the adaptive scraping mechanism to increase the scraping pressure to keep the heat exchange surface clean.

Images