A three-motor cooperative smoothie maker based on food material recognition and closed-loop control

Abstract

The application provides a three-motor collaborative smoothie maker based on food material recognition and closed-loop control, and belongs to the technical field of food processing equipment. The maker is composed of a hardware device system and a software control system. The hardware end integrates a food material active recognition and state detection module, a three-motor power drive module, a multi-dimensional feedback detection module, and an intelligent control and storage module. The software end is equipped with data acquisition preprocessing, adaptive parameter adjustment, three-motor collaborative control, commercial batch production, full life cycle life management, and hierarchical fault diagnosis and early warning units. Through dynamic collaboration of stirring and lifting nonlinear torque coupling and cleaning and lifting PI water pressure linkage, combined with food material active recognition and local iterative learning, the precise adaptation of food materials with different characteristics is realized. At the same time, single-cup customization and commercial batch double-mode switching are supported, effectively improving stirring uniformity, production efficiency and equipment running stability, and meeting the production needs in family and commercial scenarios.

Description

Technical Field

[0001] This invention relates to the field of food processing equipment technology, and in particular to a three-motor collaborative drive intelligent slush-type thick beverage making machine that integrates active ingredient recognition, multi-dimensional collaborative control, and full life cycle fault management. Background Technology

[0002] In the food processing industry, smoothies and similar thick beverages are highly favored for their rich texture. Equipment for making these smoothies needs to meet diverse requirements, including processing ingredients of varying thicknesses, customizing personalized flavors, ensuring efficient and stable operation, and facilitating mass production in commercial settings. Existing smoothie makers mostly employ a three-motor collaborative solution. However, this traditional collaborative model has significant shortcomings in hardware detection accuracy, algorithmic adaptability, scenario compatibility, and fault handling mechanisms. This leads to problems such as insufficient mixing uniformity, low efficiency, poor commercial adaptability, and high equipment failure rates, failing to meet the diverse needs of smart home appliances. Specific defects are as follows: 1. Significant limitations of adaptive and collaborative algorithms: Existing equipment mostly adopts a single time-triggered collaborative or simple closed-loop feedback mode. The former cannot adapt to ingredients with different consistency, while the latter only relies on a small amount of local historical data for simple iteration. It lacks self-learning adaptive logic for multi-dimensional data fusion and has not achieved dynamic coupling and collaboration of three motors. It is easy to have uneven stirring, over-stirring, or energy waste. There is no intelligent collaborative mechanism for the whole process of "real-time feedback - parameter adjustment - multi-source data learning - personalized optimization".

[0003] 2. Low intelligence and scenario adaptability: Existing systems mostly rely on preset parameters to operate, and cannot automatically identify the type of ingredients and their initial state and match the optimal process. Parameters need to be manually adjusted, which is cumbersome and lacks accuracy. At the same time, there is no batch production function designed for commercial scenarios, and only single-cup production is supported, which cannot meet the high-frequency and continuous production needs of commercial scenarios such as milk tea shops and coffee shops.

[0004] 3. Weak fault tolerance and management capabilities: Existing technologies have not solved the problem of linkage and adaptation between feedback signals and adaptive algorithms, resulting in insufficient coordination accuracy. When faced with faults such as motor jamming, stalling, and abnormal water pressure, they only have simple passive repair and alarm mechanisms, lacking the ability to predict faults in equipment operation status and failing to classify faults for handling. Minor faults can easily lead to equipment shutdowns, and serious faults lack targeted emergency protection, which can easily lead to equipment damage, affecting user experience and service life. Summary of the Invention

[0005] To address the problems of poor adaptability, insufficient intelligence, limited application scenarios, and weak fault tolerance in existing technologies, this invention provides a three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control. Through precise configuration of the hardware system and intelligent collaboration of the software control system, dynamic optimization of the entire smoothie making process is achieved. Furthermore, by actively recognizing ingredients through hardware modules, core detection and drive components, iterating adaptive logic, and adding multi-dimensional collaborative algorithms, the commercial batch production function is expanded, and an equipment operation statistics module is added. An innovative fault prediction algorithm is upgraded to a hierarchical self-healing logic, achieving deep integration of the hardware and software control systems. This simultaneously meets the dual needs of personalized home production and commercial batch production, improving the equipment's intelligence level, operational stability, and service life.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: This invention provides a three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control, including a hardware device system and a software control system. The hardware device system and the software control system achieve data interaction and collaborative control through a main control unit. The hardware device system includes: The active food ingredient identification and status detection module is deployed above the mixing station and the tray. It integrates a high-definition industrial vision recognition module, a material density sensor, an integrated probe for food ingredient temperature and hardness, and a liquid level sensor. It is used for food ingredient type identification, food ingredient solid-liquid ratio and ice particle size detection, food ingredient temperature and hardness quantitative detection, and food ingredient liquid level detection. The power drive module includes a stirring motor, a cleaning motor, and a lifting motor. Both the stirring motor and the cleaning motor support FOC field orientation control, and the lifting motor is equipped with a running position encoder. The feedback detection module includes a thin-film pressure sensor attached to the bottom of the cup holder, a photoelectric limit switch for lifting limit, a temperature sensor for monitoring the temperature of the motor and the beverage, and current sensors for each of the three motors, used to collect feedback data on the operating status of the equipment and the status of the ingredients. The intelligent control and storage module includes a main control unit, a storage unit, and a touch screen display, which is used for data processing, control command output, process parameter and operating data storage, and human-machine interaction. The software control system is stored in the intelligent control and storage module and executed by the main control unit. It includes a data acquisition and preprocessing module, an adaptive adjustment module, and an intelligent control strategy module. The intelligent control strategy module is configured to execute coordinated stirring-lifting control and coordinated cleaning-lifting control: In the coordinated control of stirring and lifting, the real-time torque of the stirring motor is used as the input, and the running speed and micro-lifting amplitude of the lifting motor are adjusted nonlinearly to form a torque-lifting coupling closed loop. In the cleaning-lifting coordinated control, the real-time water pressure of the cleaning motor is used as the input, and the cleaning movement speed of the lifting motor is adjusted through PI control linkage. Combined with the segmented cleaning path, it achieves cleaning without dead angles.

[0007] As a further embodiment of the present invention, the integrated probe for measuring the temperature and hardness of the food is linked to a thin-film pressure sensor at the bottom of the cup. The method for quantifying the hardness of the ice is as follows: by applying downward pressure, the probe is inserted into the ice to a preset depth, and the pressure value required to reach that depth is used as the basis for quantifying the hardness of the ice.

[0008] As a further aspect of the present invention, the stirring motor has a speed adjustment range of 200-5000 rpm and is equipped with an incremental encoder for real-time acquisition of phase current signals and obtaining real-time torque values ​​through a current-torque conversion model; wherein, the current-torque conversion model is: In the formula, For the motor output torque, The torque constant of the motor. This is the quadrature-axis current of the motor.

[0009] As a further embodiment of the present invention, the cleaning motor is a DC brushless motor, equipped with an analog water pressure sensor and an incremental encoder, and the output water pressure is controlled by speed adjustment; the lifting motor is a stepper motor equipped with a dedicated driver, which is responsible for the lifting and lowering drive of the cup-holding platform and provides positioning support for cup replacement.

[0010] As a further embodiment of the present invention, the storage unit is a dual storage architecture of EEPROM and SD card, storing no less than 500 sets of manufacturing process parameters, and simultaneously storing cutter head life data, equipment operation data, fault data, and production ledger data; the touch screen is a 4.3-inch touch screen, supporting switching between commercial mode and single-cup mode, cutter head life display, batch production quantity setting, process parameter locking, and production ledger query and export.

[0011] As a further aspect of the present invention, the adaptive adjustment module is configured to perform intelligent matching of pre-parameters, dynamic optimization of real-time parameters, and iterative learning of local multi-source data: Intelligent matching of pre-parameters: Based on the detection results of the active ingredient recognition and status detection module, the optimal initial process parameters of the corresponding ingredients stored locally are automatically called. Real-time parameter dynamic optimization: Based on real-time feedback of torque, speed, beverage temperature, and ingredient characteristics, the FOC drive parameters and lifting fine-tuning actions are adjusted; the lifting motor operating speed is dynamically determined by the stirring motor torque value. When the real-time torque exceeds 120% of the preset torque value, the lifting speed is reduced; when the real-time torque is lower than the preset torque value, the lifting speed is gradually increased but does not exceed the lifting speed limit; when the torque fluctuation is ≤5% and lasts for 2 stirring cycles, and the solid-liquid ratio and particle size of the ingredients reach the preset standards, the stirring is considered complete. Local multi-source data iterative learning: The entire process data of each production is synchronized to local storage, and the parameters of the same type of ingredients and the same mode are continuously iterated and optimized.

[0012] As a further aspect of the present invention, in the real-time parameter dynamic optimization, if the hardness of the current food is detected to be greater than 20% of the preset value and there is no corresponding process parameter locally, the closest process parameter of the same type of food is automatically matched as the initial operating parameter; during the stirring process, for the case of high hardness, the torque of the stirring motor is increased first to maintain the preset speed stable.

[0013] As a further aspect of the present invention, in the stirring-lifting coordinated control, the real-time torque of the stirring motor is defined as follows: The preset reference torque is The real-time operating speed of the lifting motor is The real-time micro-lift amplitude is The food hardness correction factor is ,in, The baseline slight rise is ; The rate-torque coupling uses a nonlinear piecewise rule as follows: when When this occurs, the stall protection is triggered, stopping the mixing motor; when At that time, the operating speed of the lifting motor decreases non-linearly as the torque increases; when At this time, the lifting motor maintains the base operating speed; when At the same time, the operating speed of the lifting motor gradually increases but does not exceed the maximum lifting speed; The formula for the lift amplitude-torque coupling is: The maximum slight lifting amplitude shall not exceed 10mm; when And if the duration exceeds 1 second, a slight lifting action is triggered.

[0014] As a further aspect of the present invention, the execution steps of the stirring-lifting coordinated control are as follows: After the stirring motor starts, the phase current is sampled every 100ms to calculate the real-time torque. ; Substitute the coupling rules and formulas to calculate the current acceleration / deceleration rate. With slight rise ; The lifting motor performs adjustment actions, and the encoder feeds back the actual position and speed to form a closed-loop correction. When the torque returns to the normal range, gradually restore the baseline parameters and stop the slight lifting. The process repeats until the mixing completion condition is met, then exits the mixing-lifting coordination and enters the cleaning phase.

[0015] As a further aspect of the present invention, in the cleaning-lifting coordinated control, the real-time water pressure of the cleaning motor is defined as follows: The preset reference water pressure is The lifting motor moves at a speed of [missing information - likely a speed measurement]. The standard cleaning lifting speed is Water pressure deviation ,set up This is the proportionality coefficient. The integral coefficient is used; therefore, the PI control formula for water pressure-lift / lower speed is: .

[0016] As a further aspect of the present invention, the cleaning path is divided into a standard cleaning mode and a rapid cleaning mode: The standard cleaning mode path is divided into 5 segments: bottom stop → uniform speed rise → top stop → uniform speed descent → bottom second stop. The stop time and stroke ratio of each segment are fixed, which is used for comprehensive cleaning of different ingredients after preparation. The quick cleaning mode has a three-stage process: bottom pause → short-range reciprocating rise and fall → bottom reset pause, with a total duration of 40% of the standard mode. It is used for quick cleaning after the same ingredients have been cooked continuously.

[0017] As a further aspect of the present invention, the execution steps of the cleaning-lifting coordinated control are as follows: (1) The lifting motor sends the cup to the cleaning station with a positioning accuracy of ±0.1mm; (2) Start the cleaning motor and pre-pressurize to the reference water pressure. The water pressure sensor collects real-time water pressure data every 200ms. ; (3) Substitute into the PI formula to calculate the cleaning lifting speed. Control the lifting motor to move according to the planned segmented path; (4) Synchronously adjust the FOC parameters of the cleaning motor to compensate for the water pressure. Within the range; (5) After the preset cleaning time is reached, the cleaning motor stops and the lifting motor is reset to the standby position.

[0018] As a further embodiment of the present invention, the intelligent control strategy module is also configured to perform batch production collaborative control. In commercial mode, the stirring, lifting, and cleaning motors are linked with the external commercial cup rack to achieve full-process automated collaboration of stirring completion, cup lowering, cleaning, automatic cup replacement, new cup rising, and stirring, with cleaning and feeding operations running in parallel.

[0019] As a further aspect of the present invention, the automatic cup-changing process in the batch production collaborative control is as follows: (1) Initial preparation stage: Identify the label information of the first raw cup, set the initial cleaning mode to standard cleaning, and move the raw cup to the cup holding station; (2) Parallel mixing and pre-decision stage: While mixing the current cup, the label of the next cup of raw food is identified in parallel, the types of ingredients are compared and the cleaning mode is preset: the same type of ingredients are preset to the quick cleaning mode, and different types of ingredients are preset to the standard cleaning mode; (3) Finished product transfer stage: After the current cup is stirred, the finished product is transferred to the finished product area; (4) Pre-planned cleaning stage: Cleaning is immediately performed according to the preset cleaning mode; (5) Cup placement stage: After cleaning, the next raw cup is moved to the cup holder station and the next production cycle begins.

[0020] As a further aspect of the present invention, the software control system also includes a commercial batch production function module, which supports custom configuration of the number of batch productions, automatic cup replacement, local commercial production ledger generation and export, and single-cup / batch mode switching; the production ledger includes the total batch production time, the average production time per cup, the total number of productions, and energy consumption data.

[0021] As a further aspect of the present invention, the software control system further includes a tool life and equipment operation statistics module, used for: Real-time display of remaining cutter head life and cumulative working time; graded reminders when life threshold is reached; maximum torque is limited before forced replacement. The system provides real-time statistics on the overall operating time of the equipment and the individual operating time of each motor. When a preset maintenance node is reached, it automatically prompts for maintenance and displays the parts that need maintenance. It supports querying cutter head life and equipment operation data by time and production batch, and generates equipment maintenance ledgers.

[0022] As a further aspect of the present invention, the software control system further includes a fault diagnosis and early warning module, comprising a fault prediction unit and a hierarchical self-repair unit: Fault prediction unit: Based on multi-dimensional data such as motor current, torque, speed, temperature rise, water pressure, and cutter head wear, a health model of the equipment is built. Potential faults are predicted by data anomalies. When a fault is predicted, a pop-up reminder is given and the data is recorded. Hierarchical self-healing unit: Classified into three levels of handling based on the severity of the fault. Minor fault: Automatically initiates a basic repair procedure, and runs normally after repair; Moderate fault: Initiate the intermediate repair and derating operation procedure, and simultaneously trigger a pop-up alarm. The equipment will continue to operate at a derating rate. Severe fault: Immediately activate the advanced emergency protection program, cut off the power to the faulty motor, trigger the light alarm, and display the fault code and solution.

[0023] As a further aspect of the present invention, the minor faults include slight jamming of the lifting motor, slight fluctuation of the cleaning water pressure, and slight drift of the sensor data. The corresponding primary repair procedures include the lifting motor outputting 50 pulses in reverse to unlock, the cleaning motor fine-tuning the FOC parameters to compensate for the water pressure, and automatic calibration of the sensor data. The moderate faults include slight stalling of the stirring motor, blade wear approaching the threshold, and excessive motor temperature rise. The corresponding intermediate repair and derating procedures include emergency speed reduction of the stirring motor and starting a light flushing of the cleaning motor, limiting the maximum torque of the motor, and reducing the stirring efficiency. The severe malfunctions include complete stalling of the stirring motor, jamming of the lifting motor that cannot be released, sensor failure, and completely abnormal water pressure. The corresponding advanced emergency protection procedures include cutting off the power to the faulty motor and starting the cleaning motor for 30 seconds to prevent food from solidifying when the stirring motor stalls.

[0024] As a further aspect of the present invention, the data acquisition and preprocessing module has an acquisition frequency of not less than 10Hz, and the acquisition dimensions include food feature data, equipment operating status data, and commercial production process parameter data. After acquisition, the data is processed by signal conditioning and filtering to extract food feature vectors, equipment status vectors, process parameter vectors, and production efficiency vectors, providing data support for collaborative control.

[0025] As a further aspect of the present invention, the device performs a collaborative link self-test after powering on, including: Lifting and positioning and safety interlock self-test: Verify the positional accuracy of the lifting motor and the safety interlock that prevents the stirring motor from starting if the lifting position is not reached; Stirring-torque-lifting speed coupling self-test: Verify the effectiveness of the linkage between stirring torque changes and lifting motor speed adjustments; Cleaning-lifting speed-water pressure coupling self-test: Verify the correspondence between changes in lifting speed and water pressure fluctuations, and confirm the effectiveness of the PI regulation link.

[0026] Compared with existing technologies, the three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control provided in this application has the following advantages: This invention achieves dynamic closed-loop coordination of three motors through nonlinear torque-rate coupling of stirring and lifting, and PI water pressure-speed linkage of cleaning and lifting. Combined with real-time adaptive adjustment logic, it can adapt to ingredients of different hardness and consistency, effectively improving mixing uniformity, avoiding over-mixing, reducing blade wear, and extending the service life of core components. This smoothie maker integrates multiple sensors for active ingredient identification and status detection, automatically matching optimal process parameters without manual adjustment. It supports dual modes of single-cup customization and commercial batch processing. In batch mode, cleaning and cup changing are performed in parallel, significantly improving commercial continuous production efficiency and catering to diverse needs in both home and commercial scenarios. A fault prediction model based on multi-source data can identify potential faults in advance. A three-level self-repair mechanism is employed: automatic repair for minor faults, reduced operation for moderate faults, and emergency protection for severe faults, minimizing downtime, reducing production losses in commercial scenarios, and effectively extending the overall service life of the equipment. Through local data storage and iterative learning mechanisms, the equipment can continuously optimize process parameters for similar ingredients based on historical production data, achieving continuous standardization and intelligent upgrading of the product's taste. Attached Figure Description

[0027] Figure 1 This invention provides a flowchart of the stirring-lifting coordinated control process in a three-motor coordinated smoothie maker based on ingredient recognition and closed-loop control.

[0028] Figure 2 This invention provides a flowchart of the cleaning-lifting coordinated control process in a three-motor coordinated smoothie maker based on food ingredient recognition and closed-loop control.

[0029] Figure 3 This invention provides a complete flowchart of the commercial batch production process of a three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control.

[0030] Figure 4 This is a hardware system architecture block diagram of a three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control provided by the present invention.

[0031] Figure 5 This is a flowchart of the equipment preparation stage in a three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control, provided by the present invention. Detailed Implementation

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0033] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0034] To address the issues of poor adaptability, insufficient intelligence, limited application scenarios, and weak fault tolerance in existing technologies, this invention provides a three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control. Through precise hardware system configuration and intelligent collaboration with the software control system, it achieves dynamic optimization of the entire smoothie making process. Furthermore, by incorporating an active ingredient recognition hardware module, core detection and drive components, it iterates adaptive logic and adds multi-dimensional collaborative algorithms, expanding commercial batch production capabilities and adding a device operation statistics module. It also innovates fault prediction algorithms and upgrades them to hierarchical self-healing logic, achieving deep integration of the hardware and software control systems. This simultaneously meets the dual needs of personalized home production and commercial batch production, improving the device's intelligence level, operational stability, and lifespan.

[0035] This invention provides a three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control, comprising a hardware device system and a software control system. The hardware device system and the software control system achieve data interaction and collaborative control through a main control unit. The hardware device system comprises four main components: an active ingredient recognition and status detection module, a power drive module, a feedback detection module, and an intelligent control and storage module.

[0036] In this embodiment, the active food identification and status detection module is deployed above the stirring station and the tray to achieve accurate pre-positioning identification and real-time status monitoring of the food. This active food identification and status detection module integrates a high-definition industrial vision recognition module, a material density sensor, an integrated probe for food temperature and hardness, and a liquid level sensor. It is used for food type identification, detection of food solid-liquid ratio and ice particle size, quantitative detection of food temperature and hardness, and detection of food liquid level.

[0037] The high-definition industrial vision recognition module is used to perform image recognition of food types (ice, fruit, milk, etc.) and can also obtain food type information by recognizing the label on the cup. The material density sensor is used to directly detect the solid-liquid ratio, ice particle size, moisture content and other core characteristic parameters of the food. The integrated food temperature and hardness probe is used in conjunction with the thin film pressure sensor at the bottom of the cup to accurately detect the initial temperature and hardness of the food and quantify the initial consistency of the food. The ice hardness quantification rule is: by applying downward pressure to make the probe penetrate the ice to a preset depth, the pressure value required to reach that depth is used as the quantitative basis for judging the ice hardness. The liquid level sensor is used to detect the remaining food and liquid level in the cup to prevent dry stirring and overflow, and at the same time provide data support for the quantitative preparation of food in commercial batches.

[0038] The integrated probe for measuring the temperature and hardness of the food is linked to a thin-film pressure sensor at the bottom of the cup. The method for quantifying the hardness of the ice is as follows: by applying downward pressure, the probe is inserted into the ice to a preset depth, and the pressure value required to reach that depth is used as the basis for quantifying the hardness of the ice.

[0039] In this embodiment, the power drive module includes a stirring motor, a cleaning motor, and a lifting motor, corresponding to stirring, cleaning, cup lifting, and cup changing operations, respectively. Both the stirring and cleaning motors support FOC (Field-Oriented Control) magnetic field control, and the lifting motor is equipped with a running position encoder. The stirring motor has an adjustable speed range of 200-5000 rpm, supports FOC magnetic field control, and can acquire phase current signals in real time and obtain torque values ​​through a current-torque conversion model. An incremental encoder is used to assist in speed measurement. The torque conversion formula is: In the formula, For the motor output torque, The torque constant of the motor. The current is the quadrature-axis current of the motor. The cleaning motor is a DC brushless motor, equipped with an analog water pressure sensor and an incremental encoder, and controls the output water pressure through speed regulation; the lifting motor is a stepper motor with a dedicated driver, responsible for the lifting and lowering of the cup-holding platform, providing positioning support for cup replacement.

[0040] In this embodiment, the feedback detection module includes a thin-film pressure sensor attached to the bottom of the cup holder, a photoelectric limit switch for lifting and lowering limits, a temperature sensor for monitoring the temperature of the motor and the beverage, and current sensors for each of the three motors. These sensors are used to collect feedback data on the equipment's operating status and the state of the ingredients. By deploying multiple types of high-precision sensors at key locations on the equipment, a data foundation is provided for closed-loop control and fault diagnosis. Specifically, the thin-film pressure sensor, attached to the bottom of the cup holder, detects the hardness of the beverage and, in conjunction with an integrated temperature and hardness probe, quantifies the state of the ingredients. The photoelectric limit switch locates the extreme positions of the lifting motor and triggers hard limit protection. The temperature sensors are deployed inside the motor and at the stirring station, monitoring the motor temperature for overheat protection and simultaneously monitoring the beverage temperature, providing initial beverage state data in conjunction with the pressure sensor. The current sensors, each for one of the three motors, collect motor phase current signals, providing data support for closed-loop control.

[0041] In this embodiment, the intelligent control and storage module includes a main control unit, a storage unit, and a touch screen display. It is used for data processing, control command output, process parameter and operating data storage, and human-machine interaction. The main control unit uses an MCU with integrated fuzzy control algorithms to process feedback signals and output control commands. The storage unit employs an EEPROM + SD card dual-storage architecture, capable of storing over 500 sets of manufacturing process parameters (speed, torque, water pressure, duration, etc.), as well as cutting head life data, equipment operating data, fault data, and production ledger data, enabling local iteration and management of massive amounts of data. The touch screen display is a 4.3-inch screen that supports switching between commercial and single-cup modes, cutting head life display, equipment operating time statistics, batch production quantity setting, process parameter locking, and production ledger query / export functions.

[0042] The storage unit is a dual-storage architecture of EEPROM and SD card, storing no less than 500 sets of manufacturing process parameters, as well as cutting head life data, equipment operation data, fault data, and production ledger data; the touch screen is a 4.3-inch touch screen, supporting switching between commercial mode and single-cup mode, cutting head life display, batch production quantity setting, process parameter locking, and production ledger query and export.

[0043] The software control system is stored in the intelligent control and storage module and executed by the main control unit. It includes a data acquisition and preprocessing module, an adaptive adjustment module, and an intelligent control strategy module. The intelligent control strategy module is configured to execute coordinated stirring-lifting control and coordinated cleaning-lifting control. In the coordinated control of stirring and lifting, the real-time torque of the stirring motor is used as the input, and the running speed and micro-lifting amplitude of the lifting motor are adjusted nonlinearly to form a torque-lifting coupling closed loop. In the cleaning-lifting coordinated control, the real-time water pressure of the cleaning motor is used as the input, and the cleaning movement speed of the lifting motor is adjusted through PI control linkage. Combined with the segmented cleaning path, it achieves cleaning without dead angles.

[0044] The data acquisition and preprocessing module is used to collect multi-dimensional data in real time, including food characteristics (type, solid-liquid ratio, particle size), equipment operating status (torque, speed, motor temperature rise, blade wear, water pressure), and commercial production process parameters (stirring time, lifting fine adjustment, cleaning water pressure, single cup / batch production data). The acquisition frequency is ≥10Hz. After signal conditioning and filtering to remove noise interference, the raw data is converted into a standardized format, and key features such as food feature vectors, equipment status vectors, process parameter vectors, and production efficiency vectors are extracted to provide accurate data support for multi-dimensional collaborative control.

[0045] The adaptive adjustment module is configured to perform intelligent matching of pre-parameters, dynamic optimization of real-time parameters, and iterative learning of local multi-source data, thereby achieving full-process adaptive control from initial parameter matching to dynamic process optimization and local data iteration. Intelligent matching of pre-set parameters: Based on the detection results of the active ingredient recognition and status detection module, the system automatically calls the optimal initial process parameters of the corresponding ingredients stored locally; no manual adjustment is required, and the standardized parameters are matched as soon as the ingredients are put into the cup, ensuring that the taste of different batches of ingredients is consistent.

[0046] Real-time parameter dynamic optimization: The MCU combines real-time feedback of torque, speed, beverage temperature, and ingredient characteristics to adjust the FOC drive parameters and lifting fine-tuning actions; the lifting motor operating speed is dynamically determined by the stirring motor torque value. When the real-time torque exceeds 120% of the preset torque value, the lifting speed is reduced; when the real-time torque is lower than the preset torque value, the lifting speed is gradually increased but does not exceed the lifting speed limit; when the torque fluctuation is ≤5% and lasts for 2 stirring cycles, and the solid-liquid ratio and particle size of the ingredients reach the preset standards, the stirring is considered complete. Local multi-source data iterative learning: The entire process data of each production (core characteristics of ingredients, initial matching process parameters, dynamic parameters that are adaptively adjusted during the stirring process, stirring completion time, and final ingredient status data) are synchronized to local storage, and the parameters of the same type of ingredients and the same mode are continuously iteratively optimized.

[0047] In the real-time parameter dynamic optimization, the core logic for adjusting the FOC drive parameters and the rise and fall fine-tuning actions is as follows: (1) If the hardness of the current food is detected to be greater than 20% of the preset value, and there is no corresponding process parameter for this type of food in the local storage, the MCU will automatically match the closest process parameter of the same type of food in the local storage as the initial running parameter and start the stirring operation. (2) During the mixing process, the MCU monitors the speed, torque and food status of the mixing motor in real time through torque and encoder; for the case of food with high hardness, the torque of the mixing motor is increased first to ensure the stability of the preset speed and avoid the mixing effect caused by the high hardness of the food. (3) The operating speed of the lifting motor and the torque of the stirring motor are linked and adaptively adjusted: if the real-time torque exceeds 120% of the preset torque value, the operating speed of the lifting motor is immediately reduced until the real-time torque falls back to within 120% of the preset torque value; if the real-time torque is lower than the preset torque value, the operating speed of the lifting motor is gradually increased, and the increased operating speed does not exceed the lifting speed limit set by the equipment, so as to minimize the wear of the blade while ensuring the stirring effect. (4) Judgment of completion of stirring: The torque sensor continuously monitors the torque fluctuation and the food identification module detects the changes in the solid-liquid ratio and ice particle size of the food in real time, and comprehensively judges the stirring status of the food. When the torque fluctuation is ≤5% and lasts for 2 stirring cycles, and the solid-liquid ratio and particle size of the food reach the preset stirring standard, the stirring is judged to be completed and the stirring action is stopped immediately.

[0048] In this embodiment, during real-time parameter dynamic optimization, if the hardness of the current food ingredient is detected to be greater than 20% of the preset value and there is no corresponding process parameter locally, the closest process parameter of the same type of food ingredient is automatically matched as the initial operating parameter; during the stirring process, for the case of high hardness, the torque of the stirring motor is increased first to maintain the preset speed stable.

[0049] In this embodiment, the real-time torque of the stirring motor is defined as follows in the stirring-lifting coordinated control: The preset reference torque is The real-time operating speed of the lifting motor is The real-time micro-lift amplitude is The food hardness correction factor is ,in, The baseline slight rise is (Default value 5mm), maximum slight lifting amplitude not exceeding 10mm.

[0050] The rate-torque coupling uses a nonlinear piecewise rule as follows: when When the stall occurs (during the stall interval), the stall protection is triggered, and the mixing motor stops. when During periods of high load, the operating speed of the lifting motor decreases non-linearly as torque increases. when During normal load periods, the lifting motor maintains the baseline operating speed. when During periods of low load, the operating speed of the lifting motor gradually increases but does not exceed the maximum lifting speed. The formula for the lift amplitude-torque coupling is: The maximum slight lifting amplitude shall not exceed 10mm; when If the duration exceeds 1 second, a micro-lift action is triggered; if the calculated result exceeds 10 mm, it is executed as 10 mm.

[0051] In this embodiment, see Figure 1 As shown, the execution steps of the stirring-lifting coordinated control are as follows: 1. After the stirring motor starts, the phase current is sampled every 100ms to calculate the real-time torque. ; 2. Substitute the coupling rules and formulas to calculate the current acceleration / deceleration rate. With slight rise ; 3. The lifting motor performs adjustment actions, and the encoder feeds back the actual position and speed to form a closed-loop correction; 4. When the torque returns to the normal range, gradually restore the baseline parameters and stop the slight lifting; 5. Repeat the process until the mixing is completed, then exit the mixing-lifting coordination and enter the cleaning phase.

[0052] In this embodiment, in the cleaning-lifting coordinated control, the real-time water pressure of the cleaning motor is defined as follows: The preset reference water pressure is The lifting motor moves at a speed of [missing information - likely a speed measurement]. The standard cleaning lifting speed is Water pressure deviation ,set up This is the proportionality coefficient. The integral coefficient is used; therefore, the PI control formula for water pressure-lift / lower speed is: .

[0053] In this embodiment, the cleaning path is divided into a standard cleaning mode and a quick cleaning mode: The standard cleaning path consists of 5 segments: bottom pause → constant speed ascent → top pause → constant speed descent → second bottom pause. The pause time and percentage of the travel in each segment are fixed, and this is used for thorough cleaning of different ingredients after preparation. See Table 1 for details. Table 1 Standard Cleaning Mode Parameters

[0054] The quick cleaning mode has a three-stage cleaning path: bottom pause → short-range reciprocating rise and fall → bottom reset pause, with a total duration of 40% of the standard mode. It is used for quick cleaning after continuous preparation of the same ingredients; see Table 2 for details. Table 2 Parameters for Quick Cleaning Mode

[0055] In this embodiment, see Figure 2 As shown, the execution steps of the cleaning-lifting coordinated control are as follows: (1) The lifting motor sends the cup to the cleaning station with a positioning accuracy of ±0.1mm; (2) Start the cleaning motor and pre-pressurize to the reference water pressure. The water pressure sensor collects real-time water pressure data every 200ms. ; (3) Substitute into the PI formula to calculate the cleaning lifting speed. Control the lifting motor to move according to the planned segmented path; (4) Synchronously adjust the FOC parameters of the cleaning motor to compensate for the water pressure. Within the range; (5) After the preset cleaning time is reached, the cleaning motor stops and the lifting motor is reset to the standby position.

[0056] In this embodiment, the intelligent control strategy module is also configured to perform batch production collaborative control. In commercial mode, the stirring, lifting, and cleaning motors are linked with the external commercial cup rack to achieve fully automated collaboration of the entire process, including stirring completion, cup lowering, cleaning, automatic cup replacement, new cup rising, and stirring. Cleaning and feeding are carried out in parallel.

[0057] In this embodiment, the automatic cup-changing process in the batch production collaborative control is as follows: (1) Initial preparation stage: Place raw cups with food labels in the preparation area. After the equipment completes its initial self-test, identify the label information of the first raw cup, set the initial cleaning mode to standard cleaning, and move the raw cup to the middle cup holding station. (2) Parallel mixing and pre-decision stage: While executing the mixing process of the current cup, the system simultaneously identifies the label information of the next raw cup in the preparation area, compares the types of ingredients in the current cup and the next cup, and presets the cleaning mode in advance: the same type of ingredients are preset to the fast cleaning mode, and different types of ingredients are preset to the standard cleaning mode. (3) Finished product transfer stage: After the current cup is stirred, immediately transfer the finished product from the cup holder to the finished product area on the right to avoid cross-contamination between the finished product and the raw cup; (4) Pre-planned cleaning stage: Based on the pre-set cleaning mode, the cleaning process of the blade and cup is immediately executed without any additional waiting time; (5) Cup placement stage: After cleaning, move the next raw cup from the preparation area to the cup holder station, update the current cup's ingredient information, and enter the next production cycle.

[0058] In this embodiment, the software control system further includes a commercial batch production function module, which supports custom configuration of the number of batch productions, automatic cup replacement, local commercial production ledger generation and export, and single cup / batch mode switching; the production ledger includes the total batch production time, average production time per cup, total production quantity, and energy consumption data. Among its features, the batch parameter customization configuration allows users to set the number of servings on the touchscreen. The system automatically matches optimized process parameters for commercial batch production, shortening the average production and cleaning time per cup and improving overall production efficiency. During automatic cup replacement, the lifting motor and external commercial cup rack work together to achieve automated and precise cup replacement without manual operation. After completing the production and quick cleaning of a single cup, it automatically switches to the next cup, adapting to the high-frequency feeding rhythm of commercial production. The local commercial production ledger automatically calculates the total production time, average production time per cup, total production, energy consumption data, etc., generating a commercial production ledger and storing it in local FLASH memory. It can be exported to a USB drive at any time, facilitating merchants to perform production efficiency statistics and cost accounting. The free switching between single cup / batch modes allows for quick switching between commercial batch production and single cup customization, meeting the dual needs of mass production during peak hours and personalized customization during off-peak hours.

[0059] In this embodiment, the software control system further includes a tool life and equipment operation statistics module, used for: Cutter head life management: Real-time display of remaining cutter head life and cumulative working time, graded reminders when life threshold is reached, and maximum torque is limited before forced replacement; Equipment operation statistics: Real-time statistics of the overall running time of the equipment and the individual working time of each motor. When the preset maintenance node is reached (such as 2000 hours of cumulative operation), the system will automatically prompt for maintenance and display the parts to be maintained (such as spray heads and blades). Local data management: Supports querying cutter head life and equipment operation data by time and production batch, and generates equipment maintenance ledgers.

[0060] In this embodiment, the software control system further includes a fault diagnosis and early warning module, comprising a fault prediction unit and a hierarchical self-repair unit: Fault prediction unit: Based on multi-dimensional equipment operation data (motor current, torque, speed, temperature rise, water pressure, blade wear), a commercial-grade equipment health model is constructed to analyze the operating status in real time and predict potential faults: If the lifting motor current fluctuation is consistently higher than the initial processing parameters for the corresponding food ingredient, ball screw wear is predicted; if the stirring motor torque curve is consistently higher than the processing parameters for the corresponding food ingredient, blade dulling or stirring paddle jamming is predicted; slight fluctuations in sensor data are predicted as sensor malfunctions. When a potential fault is predicted, a pop-up reminder is displayed on the touchscreen, and the predicted data is recorded locally, allowing merchants to arrange maintenance in advance and avoid sudden equipment downtime. Tiered self-healing unit: Based on the severity of the fault and combined with commercial production needs, it is divided into three levels of handling to minimize equipment downtime. ① Minor faults (such as slight jamming of the lifting motor, slight fluctuation of cleaning water pressure, and slight drift of sensor data): The system automatically starts the primary repair procedure, including the lifting motor outputting 50 pulses in reverse to unlock, fine-tuning the FOC parameters of the cleaning motor to compensate for water pressure, and automatic calibration of sensor data; after the repair is completed, it will run normally without downtime, and is suitable for commercial peak production. ② Moderate faults (such as slight stalling of the mixing motor, blade wear approaching the threshold, and excessive motor temperature rise): The system initiates a medium-level repair and derated operation procedure, including emergency speed reduction of the mixing motor and initiation of a light flushing of the cleaning motor, limiting the maximum torque of the motor, and reducing the mixing efficiency; at the same time, a pop-up alarm on the touch screen prompts the operator to check in time during off-peak periods. The equipment can continue to operate at derated speed to meet the needs of emergency commercial production. ③ Severe malfunctions (such as complete stalling of the stirring motor, jamming of the lifting motor, sensor failure, or completely abnormal water pressure): The system immediately activates the advanced emergency protection program, including cutting off the power to the faulty motor, activating the cleaning motor for 30 seconds to prevent food from solidifying when the stirring motor stalls, and shutting off the motor power; at the same time, it triggers an audible and visual alarm, displays the commercial fault code, fault location, and quick solution on the touch screen, and records the fault data locally to facilitate quick troubleshooting and repair by maintenance personnel, reducing commercial downtime losses.

[0061] In this embodiment, the minor faults include slight jamming of the lifting motor, slight fluctuation of the cleaning water pressure, and slight drift of the sensor data. The corresponding primary repair procedures include the lifting motor outputting 50 pulses in reverse to unlock, fine-tuning the FOC parameters of the cleaning motor to compensate for the water pressure, and automatic calibration of the sensor data. The moderate faults include slight stalling of the stirring motor, blade wear approaching the threshold, and excessive motor temperature rise. The corresponding intermediate repair and derating procedures include emergency speed reduction of the stirring motor and starting a light flushing of the cleaning motor, limiting the maximum torque of the motor, and reducing the stirring efficiency. The severe malfunctions include complete stalling of the stirring motor, jamming of the lifting motor that cannot be released, sensor failure, and completely abnormal water pressure. The corresponding advanced emergency protection procedures include cutting off the power to the faulty motor and starting the cleaning motor for 30 seconds to prevent food from solidifying when the stirring motor stalls.

[0062] In this embodiment, the data acquisition and preprocessing module acquires data at a frequency of no less than 10Hz, and the acquisition dimensions include food feature data, equipment operating status data, and commercial production process parameter data. After acquisition, the data is processed by signal conditioning and filtering to extract food feature vectors, equipment status vectors, process parameter vectors, and production efficiency vectors, providing data support for collaborative control.

[0063] In this embodiment, the device performs a collaborative link self-test after powering on, including: Lifting and positioning and safety interlock self-test: Verify the positional accuracy of the lifting motor and the safety interlock that prevents the stirring motor from starting if the lifting position is not reached; Stirring-torque-lifting speed coupling self-test: Verify the effectiveness of the linkage between stirring torque changes and lifting motor speed adjustments; Cleaning-lifting speed-water pressure coupling self-test: Verify the correspondence between changes in lifting speed and water pressure fluctuations, and confirm the effectiveness of the PI regulation link.

[0064] Compared with existing technologies, the three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control provided in this application has the following advantages: This invention achieves dynamic closed-loop coordination of three motors through nonlinear torque-rate coupling of stirring and lifting, and PI water pressure-speed linkage of cleaning and lifting. Combined with real-time adaptive adjustment logic, it can adapt to ingredients of different hardness and consistency, effectively improving mixing uniformity, avoiding over-mixing, reducing blade wear, and extending the service life of core components. This smoothie maker integrates multiple sensors for active ingredient identification and status detection, automatically matching optimal process parameters without manual adjustment. It supports dual modes of single-cup customization and commercial batch processing. In batch mode, cleaning and cup changing are performed in parallel, significantly improving commercial continuous production efficiency and catering to diverse needs in both home and commercial scenarios. A fault prediction model based on multi-source data can identify potential faults in advance. A three-level self-repair mechanism is adopted: automatic repair for minor faults, reduced operation for moderate faults, and emergency protection for severe faults, minimizing downtime, reducing production losses in commercial scenarios, and effectively extending the overall service life of the equipment. Through local data storage and iterative learning mechanisms, the equipment can continuously optimize process parameters for similar ingredients based on historical production data, achieving continuous standardization and intelligent upgrading of the product's taste.

[0065] The present invention will now be described in further detail with reference to the accompanying drawings and typical embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0066] The hardware architecture of the three-motor collaborative intelligent smoothie / thick beverage maker described in this embodiment is as follows: Figure 4 As shown, the complete workflow is divided into the preparation stage, the mixing and production stage, the cleaning and cup changing stage, and the batch finishing stage, with the fault diagnosis module embedded throughout the process.

[0067] I. Preparation Stage The preparation phase process is as follows: Figure 5 As shown, the specific steps are as follows: Routine initialization and self-test: After the device is powered on, the MCU completes the initialization of each module, performs routine hardware self-tests, and verifies the status of sensors, three motors, and drivers; at the same time, it uses a local fault prediction algorithm to evaluate the health of the device in real time, and if a potential fault is predicted, a pop-up reminder will be displayed on the touch screen.

[0068] Special self-test for the collaborative link: Based on the routine self-test, perform verification of the three-motor collaborative link to ensure the effectiveness of the core control logic hardware. Lifting and positioning and safety interlock self-test: Control the lifting motor to run from the lowest point to the mixing station and the cleaning station, and verify the position accuracy (±0.1mm) through encoder feedback; at the same time, verify the safety interlock logic that "the mixing motor cannot start when the lifting position is not in place"; Stirring-torque-lifting speed coupling self-test: Control the stirring motor to run under no-load, collect the no-load torque and compare it with the factory benchmark torque; increase the stirring motor speed to increase torque, verify whether the lifting motor running speed is adjusted synchronously with the torque change, and confirm that the coupling control link hardware is normal; Cleaning-lifting speed-water pressure coupling self-test: Start the cleaning motor and adjust it to the reference speed. Control the lifting motor to run at low speed and high speed respectively, collect real-time water pressure data, verify the corresponding fluctuation of water pressure with lifting speed, and confirm the effectiveness of PI regulation link.

[0069] Ingredient identification and parameter matching: When the operator puts the ingredients into the cup (in commercial mode, they are put into the commercial cup rack in batches), the active ingredient identification module is automatically activated: the visual recognition module identifies the type of ingredients, the density sensor detects the solid-liquid ratio and ice particle size of the ingredients, and the liquid level sensor detects the remaining amount of ingredients. All data is transmitted to the main control MCU in real time. The MCU calls the corresponding optimal initial process parameters of the ingredients stored locally according to the detection results to complete the parameter initialization.

[0070] Taking smoothies, milkshakes, and slushies as typical examples, the initial production parameters are shown in Table 3: Table 3 Initialization parameters for smoothies, milkshakes, and slushies

[0071] See Figure 1 As shown, the core of the stirring stage is the coordinated stirring and lifting control, and the specific steps are as follows: 1. After the operator clicks the start button, the lifting motor drives the cup to be accurately delivered to the mixing station with a positioning accuracy of ±0.1mm; 2. The stirring motor starts according to the initial parameters, the FOC driver collects the phase current signal in real time and converts it into real-time torque, and the encoder provides accurate speed feedback; the MCU monitors the food status and motor operation status simultaneously through multi-sensor data fusion calibration. 3. Real-time dynamic adjustment: If the detected real-time torque is greater than 120% of the preset value, the MCU immediately adjusts the FOC drive parameters of the stirring motor to maintain stable speed, and at the same time triggers the lifting motor to reduce speed and perform a slight lifting action; if the motor temperature rises too high or the cutter head is close to the end of its life, the MCU automatically slows down the speed increase rate or reduces the maximum torque to ensure continuous operation of the equipment. 4. The closed-loop control of torque acquisition, parameter calculation, and lifting adjustment is executed cyclically until the torque fluctuation is ≤5% and continues for 2 cycles. At the same time, the solid-liquid ratio and particle size of the ingredients reach the preset standards, the stirring is judged to be completed, and the stirring motor is stopped.

[0072] See Figure 2 and Figure 3 As shown, after stirring, the cleaning stage begins. In single-cup mode, standard cleaning is performed; in commercial batch mode, cleaning and cup changing are performed in parallel and collaboratively. The overall batch process is as follows: Figure 3 As shown, the cleaning-lifting coordinated process is as follows: Figure 2 As shown.

[0073] The cleaning and cup-changing process in the commercial batch mode is as follows: 1. After the first cup is stirred, the lifting motor drives the cup rack to descend, sending the finished product to the cup retrieval port, and at the same time sending the cup tray to the cleaning station; 2. The cleaning motor starts and executes the corresponding cleaning-lifting coordinated control according to the pre-determined cleaning mode (quick cleaning of the same type of food, standard cleaning of different food); 3. While the cleaning process is underway, the commercial cup holder automatically replaces the cups, delivering the next raw cup to the stirring position; 4. After cleaning and cup replacement, the lifting motor will send the new cup to the mixing station to start making the next cup; 5. Repeat the above process until the preset batch production quantity is completed.

[0074] After all production is completed, the equipment will perform a finishing process, which includes the following steps: 1. The lifting motor and commercial cup holder are reset; 2. The cleaning motor starts the deep cleaning program for the whole machine, which thoroughly cleans the stirring head, spray pipe, and cup holder contact areas to meet the hygiene standards for commercial food processing. 3. The cutter head life and equipment operation statistics module automatically updates the cumulative working time of the cutter head, the remaining life, and the equipment running time, generating the production ledger for this production (total time, average time per cup, total production, energy consumption, etc.). 4. Operators can check production records via touch screen, or export data such as cutter head life, equipment operation, and production records via USB flash drive; 5. The touch screen displays a completion notification. If the lifespan of the cutting head or the operating time of the equipment reaches the maintenance threshold, a pop-up window will simultaneously remind you to perform maintenance.

[0075] The fault diagnosis and early warning module runs through the entire equipment operation process, monitors the equipment status in real time, and executes corresponding self-repair and protection logic according to the fault level to ensure stable equipment operation.

[0076] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can still combine, add, delete, or otherwise adjust the features of the various embodiments of the present invention according to the circumstances without conflict or creative effort, thereby obtaining different technical solutions that do not fundamentally depart from the concept of the present invention. These technical solutions also fall within the scope of protection of the present invention.

Claims

1. A three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control, characterized in that, It includes a hardware device system and a software control system, which achieve data interaction and collaborative control through a main control unit; The hardware system includes: The active food ingredient identification and status detection module is deployed above the mixing station and the tray. It integrates a high-definition industrial vision recognition module, a material density sensor, an integrated probe for food ingredient temperature and hardness, and a liquid level sensor. It is used for food ingredient type identification, food ingredient solid-liquid ratio and ice particle size detection, food ingredient temperature and hardness quantitative detection, and food ingredient liquid level detection. The power drive module includes a stirring motor, a cleaning motor, and a lifting motor. Both the stirring motor and the cleaning motor support FOC field orientation control, and the lifting motor is equipped with a running position encoder. The feedback detection module includes a thin-film pressure sensor attached to the bottom of the cup holder, a photoelectric limit switch for lifting limit, a temperature sensor for monitoring the temperature of the motor and the beverage, and current sensors for each of the three motors, used to collect feedback data on the operating status of the equipment and the status of the ingredients. The intelligent control and storage module includes a main control unit, a storage unit, and a touch screen display, which is used for data processing, control command output, process parameter and operating data storage, and human-machine interaction. The software control system is stored in the intelligent control and storage module and executed by the main control unit. It includes a data acquisition and preprocessing module, an adaptive adjustment module, and an intelligent control strategy module. The intelligent control strategy module is configured to execute coordinated stirring-lifting control and coordinated cleaning-lifting control: In the coordinated control of stirring and lifting, the real-time torque of the stirring motor is used as the input, and the running speed and micro-lifting amplitude of the lifting motor are adjusted nonlinearly to form a torque-lifting coupling closed loop. In the cleaning-lifting coordinated control, the real-time water pressure of the cleaning motor is used as the input, and the cleaning movement speed of the lifting motor is adjusted through PI control linkage. Combined with the segmented cleaning path, it achieves cleaning without dead angles.

2. The three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 1, characterized in that, The integrated probe for measuring the temperature and hardness of the food is linked to a thin-film pressure sensor at the bottom of the cup. The method for quantifying the hardness of the ice is as follows: by applying downward pressure, the probe is inserted into the ice to a preset depth, and the pressure value required to reach that depth is used as the basis for quantifying the hardness of the ice.

3. The three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 1, characterized in that, The stirring motor has a speed adjustment range of 200-5000 rpm and is equipped with an incremental encoder for real-time acquisition of phase current signals and obtaining real-time torque values ​​through a current-torque conversion model; wherein, the current-torque conversion model is as follows: In the formula, For the motor output torque, The torque constant of the motor. This is the quadrature-axis current of the motor.

4. The three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 1, characterized in that, The cleaning motor is a DC brushless motor, equipped with an analog water pressure sensor and an incremental encoder, and controls the output water pressure through speed adjustment; the lifting motor is a stepper motor with a dedicated driver, responsible for the lifting and lowering of the cup-holding platform, providing positioning support for cup replacement.

5. The three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 1, characterized in that, The storage unit adopts a dual storage architecture of EEPROM and SD card, storing no less than 500 sets of manufacturing process parameters, as well as cutting head life data, equipment operation data, fault data, and production ledger data; the touch screen is a 4.3-inch touch screen, supporting switching between commercial mode and single-cup mode, cutting head life display, batch production quantity setting, process parameter locking, and production ledger query and export.

6. The three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 1, characterized in that, The adaptive adjustment module is configured to perform intelligent matching of pre-parameters, dynamic optimization of real-time parameters, and iterative learning of local multi-source data. Intelligent matching of pre-parameters: Based on the detection results of the active ingredient recognition and status detection module, the optimal initial process parameters of the corresponding ingredients stored locally are automatically called. Real-time parameter dynamic optimization: Based on real-time feedback of torque, speed, beverage temperature, and ingredient characteristics, adjust FOC drive parameters and lifting fine-tuning actions; The operating speed of the lifting motor is dynamically determined by the torque value of the mixing motor. When the real-time torque exceeds 120% of the preset torque value, the lifting speed is reduced. When the real-time torque is lower than the preset torque value, the lifting speed is gradually increased but does not exceed the lifting speed limit. When the torque fluctuation is ≤5% and continues for 2 stirring cycles, and the solid-liquid ratio and particle size of the ingredients reach the preset standards, the stirring is considered complete. Local multi-source data iterative learning: The entire process data of each production is synchronized to local storage, and the parameters of the same type of ingredients and the same mode are continuously iterated and optimized.

7. The three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 1, characterized in that, In the aforementioned stirring-lifting coordinated control, the real-time torque of the stirring motor is defined as follows: The preset reference torque is The real-time operating speed of the lifting motor is The real-time micro-lift amplitude is The food hardness correction factor is ,in, The baseline slight rise is ; The rate-torque coupling uses a nonlinear piecewise rule as follows: when When this occurs, the stall protection is triggered, stopping the mixing motor; when At that time, the operating speed of the lifting motor decreases non-linearly as the torque increases; when At this time, the lifting motor maintains the base operating speed; when At the same time, the operating speed of the lifting motor gradually increases but does not exceed the maximum lifting speed; The formula for the lift amplitude-torque coupling is: The maximum slight lifting amplitude shall not exceed 10mm; when And if the duration exceeds 1 second, a slight lifting action is triggered.

8. The three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 7, characterized in that, The execution steps of the stirring-lifting coordinated control are as follows: After the stirring motor starts, the phase current is sampled every 100ms to calculate the real-time torque. ; Substitute the coupling rules and formulas to calculate the current acceleration / deceleration rate. With slight rise ; The lifting motor performs adjustment actions, and the encoder feeds back the actual position and speed to form a closed-loop correction. When the torque returns to the normal range, gradually restore the baseline parameters and stop the slight lifting. The process repeats until the mixing completion condition is met, then exits the mixing-lifting coordination and enters the cleaning phase.

9. A three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 1, characterized in that, In the aforementioned cleaning-lifting coordinated control, the real-time water pressure of the cleaning motor is defined as follows: The preset reference water pressure is The lifting motor moves at a speed of [missing information - likely a speed measurement]. The standard cleaning lifting speed is Water pressure deviation ,set up This is the proportionality coefficient. The integral coefficient is used; therefore, the PI control formula for water pressure-lift / lower speed is: 。 10. A three-motor collaborative smoothie maker based on ingredient recognition and closed-loop control as described in claim 9, characterized in that, The execution steps of the cleaning-lifting coordinated control are as follows: The lifting motor delivers the cup to the cleaning station with a positioning accuracy of ±0.1mm; The cleaning motor starts and pre-pressurizes to the reference water pressure. The water pressure sensor collects real-time water pressure data every 200ms. ; Substitute into the PI formula to calculate the cleaning lifting speed. Control the lifting motor to move according to the planned segmented path; Synchronously adjust the FOC parameters of the cleaning motor to compensate for water pressure. Within the range; After the preset cleaning time is reached, the cleaning motor stops and the lifting motor returns to the standby position.

Images