Automatic control system and method of tea rolling machine

Abstract

The invention relates to the technical field of tea rolling machines, and discloses an automatic control system of a tea rolling machine, which comprises a hardware control module, a quality monitoring module, an intelligent control module, a process optimization module and a man-machine interaction module, comprising a PLC (Programmable Logic Controller), a sensor network (pressure / torque / temperature and humidity) and an execution mechanism (servo motor and hydraulic system) for real-time data acquisition and instruction issuing, a quality monitoring module is used for evaluating tea rolling quality in real time and detecting key indexes such as a strip forming rate, a breakage rate and a cell breakage rate, and an intelligent control module is used for controlling the intelligent control module. Rolling parameters are dynamically adjusted based on fuzzy control and a PID algorithm, and precise control over pressure and rotating speed is achieved. According to the invention, through multi-modal feature fusion and dynamic weight distribution, the defect detection speed and efficiency are higher, and through the intelligent control module, the system algorithm is more accurate, and rolling parameters are better grasped.

Description

Technical Field

[0001] The invention relates to the technical field of control systems for tea rolling machines, in particular to an automatic control system and method for a tea rolling machine. Background Art

[0002] Tea rolling machine is a kind of tea rolling machine that can keep the fiber structure of tea leaves from being damaged and ensure the uniform quality of tea leaves. It is easy to operate. It mainly consists of a lotus seat, a transmission group, and a cloth rolling rod. The lotus seat has several lotus leaves. The transmission group drives the lotus leaves to open and close, rolling and compressing the tea leaves. Another drive motor drives the lotus seat to rotate, and cooperates with the cloth rolling rod to automatically roll the cloth bag into a knot, so that the storage space of the cloth bag is gradually reduced, achieving the function of double rolling of the tea leaves and reducing the volume.

[0003] After searching, according to the utility model patent disclosed in Chinese Patent No. CN202310240652.1, an automatic control system and method for a tea rolling machine is disclosed. The rolling machine of the invention includes a rolling barrel for holding tea leaves and a pressure cover above the rolling barrel, an electric push rod that drives the pressure cover to move up and down, and a pressure sensor set on the electric push rod. The system includes: an upper computer, a lower computer, a speed control subsystem, and a pressure control subsystem. The upper computer is a touch screen, which is used to display the rolling process and process parameters, process and save data, and set the human-machine interface for parameter setting. The lower computer is a PLC, which is used to perform data acquisition, logical operations, and automatic control of the sensors of the rolling machine. The speed control subsystem includes: a first fuzzy controller, a first A / D conversion module, a frequency converter, a first motor, a first transmission actuator, and a speed measurement feedback module, or a combination thereof. The speed control subsystem is used to control the rolling barrel to perform horizontal circular motion on the rolling disk under the drive of the first motor and the first transmission actuator. The pressure control subsystem includes: a mode control unit and a pressure control unit, which are used to control the rolling machine to perform periodic vertical reciprocating motion. By controlling the speed control subsystem and the pressure control subsystem through PLC, the rolling machine can roll the tea leaves at a precise and constant speed and pressure during operation, thereby improving the quality of the tea leaves.

[0004] However, the system has low pressure control accuracy, poor strip detection effect, and low intelligence during use, which makes the control system less coordinated. Therefore, an automatic control system and method for a tea rolling machine are proposed to solve the above problems. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In view of the deficiencies of the prior art, the present invention provides an automatic control system and method for a tea rolling machine, which has the advantages of high intelligence and more sufficient application of the new generation of electronic information technology, and solves the problems of low pressure control accuracy, poor strip detection effect and low intelligence degree during the use of the system, thus resulting in poor coordination of the control system.

[0007] (2) Technical solution

[0008] To achieve the above-mentioned purpose of high intelligence and more sufficient application of the new generation of electronic information technology, the present invention provides the following technical solution: an automatic control system for a tea rolling machine, including a hardware control module, a quality monitoring module, an intelligent control module, a process optimization module and a human-machine interaction module. The hardware control module is responsible for interacting with physical devices, including real-time data collection and instruction issuance of a PLC controller, a sensor network (pressure / torque / temperature and humidity), and an actuator (servo motor, hydraulic system).

[0009] The quality monitoring module evaluates the quality of tea rolling in real time and detects key indicators such as strip formation rate, breakage rate and cell breakage rate.

[0010] The intelligent control module dynamically adjusts the rolling parameters based on fuzzy control and PID algorithm to achieve precise control of pressure and speed.

[0011] The process optimization module stores and optimizes the rolling process parameters (time / pressure / speed) of different tea varieties, and supports process migration and adaptive adjustment.

[0012] The human-machine interaction module realizes parameter setting, status monitoring and fault diagnosis through a touch screen.

[0013] Preferably, the hardware control module includes a PLC controller and a sensor network module. The PLC controller coordinates the communication of each subsystem and executes logical control instructions, such as pressure regulation and speed switching.

[0014] Preferably, the six-axis force sensor in the sensor network module monitors the rolling pressure (200 - 500N) and torque (accuracy of ±0.5N·m), and the temperature and humidity sensor ensures the stability of environmental parameters.

[0015] Preferably, the quality monitoring module further includes a visual detection unit, a physical property parameter detection unit and an environmental monitoring unit. The visual detection unit includes a depth camera and a dual-channel input network.

[0016] Preferably, in the dual-channel input network, Channel 1: YOLOv5 extracts morphological features including strip formation rate and breakage rate, and Channel 2: CNN extracts fiber distribution and cell breakage degree features.

[0017] Preferably, the physical property parameter detection unit uses a microwave moisture sensor (±0.8% accuracy) and a near-infrared spectrometer to detect the cellulose content of tea leaves, and establishes a mapping relationship between the physical property parameters and the rolling process. The environmental monitoring unit integrates a temperature and humidity sensor (SHT35).

[0018] Preferably, the intelligent control module dynamically adjusts the rolling parameters based on fuzzy control and PID algorithms to achieve precise control of pressure and speed.

[0019] Preferably, the human-computer interaction module, the HMI interface, displays real-time data (pressure / speed / temperature and humidity) and quality scores, supports historical data backtracking, and the remote maintenance interface: supports OTA upgrade (differential compression package < 30MB) and device status log export.

[0020] Preferably, Step 1: Automatically load the pre-trained model at startup, and complete the communication link verification between the edge computing node (NVIDIA Jetson Xavier NX) and the industrial sensors (Basler camera, six-axis force sensor);

[0021] Step 2: Execute the hardware self-check program to confirm that the response delays of the pressure actuator and the motor drive module are < 50ms;

[0022] Step 3: Capture the surface morphology of the rolled tea leaves through the depth camera, calculate the strip rate and breakage rate in real time by YOLOv5, and extract the fiber distribution characteristics by ResNet18;

[0023] Step 4: The six-axis force sensor collects the rolling pressure (200 - 500N) and torque (±0.5N·m accuracy), the temperature and humidity sensor monitors the environmental parameters, and dynamically allocates the YOLO detection results and CNN texture features through the SE attention module;

[0024] Step 5: The CNN-LSTM model inputs 60 seconds of historical data (pressure / speed / temperature and humidity) to predict the optimal parameter combination for the next 5 seconds;

[0025] Step 6: The PLC controller receives the decision instruction, adjusts the speed of the servo motor and the pressure of the hydraulic system through the PID algorithm, and updates the control parameters every 0.5 seconds according to the real-time feedback to ensure the dynamic matching of the rolling force and the physical properties of the tea leaves.

[0026] (III) Beneficial Effects

[0027] Compared with the prior art, the present invention provides an automatic control system and method for a tea rolling machine, having the following beneficial effects:

[0028] 1. The automatic control system and method of the tea rolling machine can achieve higher defect detection speed and efficiency through multi-modal feature fusion and dynamic weight allocation, and make the system algorithm more accurate through the intelligent control module, thus better grasping the rolling parameters.

[0029] 2. The automatic control system and method of the tea rolling machine enhance the maintainability and expandability of the system through human-computer interaction, visual operation and remote management. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the automatic control system of the tea rolling machine of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work shall fall within the protection scope of the present invention.

[0032] Please refer to Figure 1 , an automatic control system of a tea rolling machine, including a hardware control module, a quality monitoring module, an intelligent control module, a process optimization module and a human-computer interaction module. The hardware control module is responsible for interacting with physical devices, including real-time data acquisition and instruction issuance of a PLC controller, a sensor network (pressure / torque / temperature and humidity), and an actuator (servo motor, hydraulic system);

[0033] The quality monitoring module can evaluate the quality of tea rolling in real time and detect key indicators such as strip formation rate, breakage rate and cell breakage rate;

[0034] The intelligent control module dynamically adjusts the rolling parameters based on fuzzy control and PID algorithms to achieve precise control of pressure and speed;

[0035] The process optimization module stores and optimizes the rolling process parameters (time / pressure / speed) of different tea varieties, and supports process migration and adaptive adjustment;

[0036] The human-computer interaction module realizes parameter setting, status monitoring and fault diagnosis through a touch screen.

[0037] In Figure 1 , the hardware control module includes a PLC controller and a sensor network module. The PLC controller coordinates the communication of each subsystem and executes logical control instructions, such as pressure regulation and speed switching.

[0038] Specifically, the PLC controller is PLC Siemens S7-20.

[0039] Edge computing node: NVIDIA Jetson Xavier NX (32 TOPS computing power) runs the CNN model, and industrial camera upgrade: Basler ace2 640-120gm (global shutter, 120 FPS)

[0040] Communicate with the Siemens S7-200 PLC through the Modbus-TCP protocol, retain the pressure closed-loop control function, and the OTA upgrade module supports remote update of the CNN model (incremental update package < 50MB).

[0041] In Figure 1 Among them, the six-axis force sensor in the sensor network module monitors the rolling pressure (200-500N) and torque (±0.5N·m accuracy), and the temperature and humidity sensor ensures the stability of environmental parameters.

[0042] It also includes that the quality monitoring module further includes a visual detection unit, a physical property parameter detection unit, and an environmental monitoring unit. The visual detection unit includes a depth camera and a dual-channel input network. In the dual-channel input network, Channel 1: YOLOv5 extracts morphological features including strip formation rate and breakage rate, and Channel 2: CNN extracts fiber distribution and cell damage degree features.

[0043] Specifically, on the basis of the original YOLOv5 object detection, the ResNet18 convolutional neural network is introduced to perform fine-grained analysis on the texture features of tea leaves, improving the detection accuracy of the strip formation rate (±1.5% → ±0.9%).

[0044] Construct a dual-channel input network:

[0045] Channel 1: YOLOv5 extracts morphological features (strip formation rate, breakage rate)

[0046] Channel 2: CNN extracts fiber distribution and cell damage degree features.

[0047] The number of model parameters is compressed to 1.2M, and the inference speed reaches 25 FPS (Jetson Nano platform) 13

[0048] [[ID=3,0]]Dynamic weight allocation mechanism. Automatically adjust the feature weight ratio of YOLOv5 and CNN through the attention mechanism (SE Block).

[0049] It also includes the construction of a high-precision dataset

[0050] Collect 100,000 images of 5 types of tea (green tea / black tea / oolong tea / white tea / pu-erh tea) and label:

[0051] Strip formation rate grading (Grade A: ≥95%, Grade B: 90-95%)

[0052] Fiber distribution heat map (grayscale gradient map generated by OpenCV)

[0053] Data augmentation: Random rotation, brightness adjustment, simulating environmental fog interference during kneading

[0054] Physical property parameter detection unit, using a microwave moisture sensor (±0.8% accuracy) and a near-infrared spectrometer to detect the cellulose content of tea leaves, establishing a mapping relationship between physical property parameters and the kneading process. The environmental monitoring unit integrates a temperature and humidity sensor (SHT35).

[0055] Three-section inverters are used to achieve stepless speed regulation of the kneading barrel (20 - 35 rpm), adapting to the needs of different tea varieties such as green tea and black tea, with a control accuracy of ±0.5 rpm

[0056] Intelligent control module, dynamically adjusting kneading parameters based on fuzzy control and PID algorithms to achieve precise control of pressure and speed.

[0057] Specifically, the features extracted by CNN (such as the pressure change gradient) are converted into fuzzy rule input variables, and the rule base is extended to 35 rules.

[0058] IF the pressure change rate > 10 N / s AND the moisture content < 55% THEN reduce the speed by 3 rpm

[0059] Human-machine interaction module, HMI interface, displays real-time data (pressure / speed / temperature and humidity) and quality scores, supports historical data backtracking, remote maintenance interface: supports OTA upgrade (differential compression package < 30 MB) and device status log export.

[0060] In this embodiment, the automatic control method of the tea kneading machine includes:

[0061] Step 1: Automatically load the pre-trained model at startup, and complete the communication link verification between the edge computing node (NVIDIA Jetson Xavier NX) and industrial sensors (Basler camera, six-axis force sensor);

[0062] Step 2: Execute the hardware self-check program to confirm that the response delay of the pressure actuator and the motor drive module < 50 ms;[[ID=III]]

[0063] Step 3: Capture the surface morphology of the kneaded tea leaves through a depth camera, calculate the strip rate and breakage rate in real time by YOLOv5, and extract fiber distribution features by ResNet18;

[0064] Step 4: The six-axis force sensor collects the kneading pressure (200 - 500 N), torque (±0.5 N·m accuracy), the temperature and humidity sensor monitors environmental parameters, and dynamically allocates YOLO detection results and CNN texture features through the SE attention module;

[0065] Step 5: The CNN-LSTM model inputs 60 seconds of historical data (pressure / rotation speed / temperature and humidity) and predicts the optimal parameter combination for the next 5 seconds.

[0066] Step 6: The PLC controller receives the decision instruction, adjusts the rotation speed of the servo motor and the pressure of the hydraulic system through the PID algorithm, and updates the control parameters every 0.5 seconds according to the real-time feedback to ensure the dynamic matching of the rolling force and the physical properties of the tea leaves.

[0067] Through the above steps, the problems of low pressure control accuracy, poor strip detection effect, and low intelligence of traditional equipment, resulting in poor coordination of the control system, are solved.

[0068] In summary, the automatic control system and method of the tea rolling machine make the defect detection speed and efficiency higher through multi-modal feature fusion and dynamic weight allocation, make the system algorithm more accurate through the intelligent control module, better grasp the rolling parameters, and enhance the maintainability and scalability of the system through human-computer interaction, visual operation and remote management. Moreover, the problems of low pressure control accuracy, poor strip detection effect, and low intelligence of the system during use, resulting in poor coordination of the control system, are solved.

[0069] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0070] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An automatic control system for a tea rolling machine, comprising a hardware control module, a quality monitoring module, an intelligent control module, a process optimization module and a human-computer interaction module, characterized in that: The hardware control module is responsible for interacting with physical devices, including real-time data acquisition and instruction issuing for PLC controllers, sensor networks (pressure / torque / temperature / humidity), and actuators (servo motors, hydraulic systems). The quality monitoring module evaluates the quality of tea rolling in real time and detects key indicators such as the strip formation rate, breakage rate, and cell breakage rate. The intelligent control module dynamically adjusts the rolling parameters based on fuzzy control and PID algorithms to achieve precise control of pressure and speed. The process optimization module stores and optimizes the rolling process parameters (time / pressure / speed) for different tea varieties, supporting process migration and adaptive adjustment. The human-machine interaction module realizes parameter setting, status monitoring, and fault diagnosis through a touch screen.

2. The automatic control system of a tea rolling machine according to claim 1, characterized in that: The hardware control module includes a PLC controller and a sensor network module. The PLC controller coordinates the communication of each subsystem and executes logic control instructions such as pressure regulation and speed switching.

3. The automatic control system of a tea rolling machine according to claim 1, characterized in that: The six-axis force sensor in the sensor network module monitors the rolling pressure (200 - 500 N) and torque (accuracy of ±0.5 N·m), and the temperature and humidity sensors ensure stable environmental parameters.

4. The automatic control system of a tea rolling machine according to claim 1, characterized in that: The quality monitoring module also includes a visual detection unit, a physical property parameter detection unit, and an environmental monitoring unit. The visual detection unit includes a depth camera and a dual-channel input network.

5. The automatic control system of a tea rolling machine according to claim 4, characterized in that: In the dual-channel input network, Channel 1: YOLOv5 extracts morphological features including the strip formation rate and breakage rate, and Channel 2: CNN extracts fiber distribution and cell breakage degree features.

6. The automatic control system of a tea rolling machine according to claim 1, characterized in that: The physical property parameter detection unit uses a microwave moisture sensor (accuracy of ±0.8%) and a near-infrared spectrometer to detect the cellulose content of tea leaves and establish a mapping relationship between physical property parameters and the rolling process. The environmental monitoring unit integrates temperature and humidity sensors (SHT35).

7. The automatic control system of a tea rolling machine according to claim 1, characterized in that: The intelligent control module dynamically adjusts the rolling parameters based on fuzzy control and PID algorithms to achieve precise control of pressure and speed.

8. The automatic control system of a tea rolling machine according to claim 1, characterized in that: The human-machine interaction module, the HMI interface, displays real-time data (pressure / speed / temperature / humidity) and quality scores, supports historical data retrieval, and the remote maintenance interface: supports OTA upgrade (differential compression package < 30 MB) and device status log export.

9. An automatic control method for a tea rolling machine, characterized in that: A control method for the control system according to any one of claims 1 - 8, the control method comprising the following steps: Step 1: Automatically load the pre-trained model at startup and complete the communication link verification between the edge computing node (NVIDIA Jetson Xavier NX) and industrial sensors (Basler camera, six-axis force sensor). Step 2: Execute the hardware self-check program to confirm that the response delay of the pressure actuator and the motor drive module is < 50 ms. Step 3: Capture the surface morphology of the rolled tea leaves through a depth camera, calculate the strip formation rate and breakage rate in real time by YOLOv5, and extract fiber distribution features by ResNet18. Step 4: The six-axis force sensor collects the rolling pressure (200 - 500 N), torque (accuracy of ±0.5 N·m), the temperature and humidity sensors monitor environmental parameters, and dynamically allocate the YOLO detection results and CNN texture features through the SE attention module. Step 5: The CNN-LSTM model inputs 60 seconds of historical data (pressure / speed / temperature and humidity) to predict the optimal parameter combination for the next 5 seconds. Step 6: The PLC controller receives the decision instruction, adjusts the speed of the servo motor and the pressure of the hydraulic system through the PID algorithm, and updates the control parameters every 0.5 seconds according to the real-time feedback to ensure the dynamic matching of the rolling force and the physical properties of the tea leaves.

Images