A device and method for peeling panax ginseng

Abstract

The application relates to the American ginseng processing technical field and provides a peeling device and method for American ginseng, which are applied to a control device of the peeling device for American ginseng. The peeling device for American ginseng further comprises a support, a driving motor, a roller and a blower. The method controls the driving motor to start, drives the roller to rotate at a predetermined rotating speed, drives American ginseng located in the roller to rub against the iron wire mesh in the inner circumferential wall of the roller, realizes preliminary peeling of the American ginseng, controls the blower to start in the process of rotating the roller, transports airflow at a predetermined wind speed to the air inlet through the air supply hole, makes the airflow enter the roller through the air inlet, blows off and carries out the American ginseng from the roller the skin scraps generated in the preliminary peeling process, separates the skin scraps from the American ginseng, and controls the driving motor and the blower to stop after running for a predetermined working time.

Description

Technical Field

[0001] This application relates to the field of American ginseng processing technology, and in particular to a peeling device and method for American ginseng. Background Technology

[0002] American ginseng (Panax quinquefolius L.) is a valuable medicinal and economic crop. Its roots, after processing, possess properties that replenish qi and yin, clear heat and promote fluid production, and are widely used in the pharmaceutical and health product industries. Peeling is a crucial step in American ginseng processing, aiming to remove the outer skin from the roots, ensuring the finished product's appearance while enhancing its specific medicinal effects.

[0003] The existing peeling technology for American ginseng has the following problems: (1) Traditional manual peeling: It relies on manual rubbing or brushing, which has three major drawbacks: ① High labor intensity and extremely low efficiency (only 5-10 kg can be processed per hour), which cannot meet the processing needs after large-scale planting; ② Uneven manual force, which can easily lead to scratches and damage to the ginseng body, destroying the medicinal components and appearance integrity, and reducing the commodity value; ③ High labor cost, with the processing cost increasing year by year as labor prices rise.

[0004] (2) Existing mechanical peeling devices: In order to solve the problem of manual labor, two types of mechanical devices have appeared on the market, but both have fatal defects: Single friction peeling device: The roller rotates to drive the American ginseng to rub against the inner wall, but the inner wall of the roller is mostly a smooth metal surface or ordinary mesh, the friction effect is poor, and the skin flakes are easy to adhere to the surface of the American ginseng during the peeling process, requiring manual secondary cleaning, which increases the processing steps and costs; Single blowing cleaning device: The blower blows air to assist in the separation of skin flakes, but it is not coordinated with the roller friction action, the air supply direction is not aligned with the air inlet of the roller, the airflow cannot effectively enter the inside of the roller, the skin flakes are not completely separated, and the peeling function cannot be realized.

[0005] In addition, the existing mechanical equipment has a very low degree of automation, requiring manual monitoring of the peeling process, making it difficult to ensure the consistency of each batch and failing to meet the requirements of standardized production. Summary of the Invention

[0006] This application provides a peeling device and method for American ginseng, aiming to solve the problems of the extremely low degree of automation of existing mechanical devices, the need for manual monitoring of the peeling process, the difficulty in ensuring the consistency of each batch of processing, and the inability to meet the requirements of standardized production.

[0007] In a first aspect, this application provides a method for peeling American ginseng, and a control device for an American ginseng peeling device. The American ginseng peeling device further includes a support, a drive motor, a roller, and a blower. The support has a hollow portion. The drive motor is fixedly mounted on the support. One end of the roller is connected to the output shaft of the drive motor, and the other end of the roller is rotatably connected to the support. The roller is located within the hollow portion, and the rotation center axis of the roller is parallel to the horizontal direction. The inner circumferential wall of the roller is covered with a wire mesh. An air inlet is provided at the end of the roller facing away from the drive motor. The blower is fixedly mounted on the support, and the air outlet of the blower faces the air inlet. The control device is fixedly mounted on the support and electrically connected to the drive motor and the blower. The method includes: The drive motor is started to make the drum rotate at a predetermined speed, which causes the American ginseng located inside the drum to rub against the wire mesh on the inner wall of the drum, thus achieving the initial peeling of the American ginseng. During the rotation of the drum, the blower is started and airflow at a predetermined speed is delivered to the air inlet through the air outlet. The airflow enters the drum through the air inlet and blows the skin flakes generated during the initial peeling process off the surface of the American ginseng and carries them out of the drum, thus completing the separation of skin flakes from the American ginseng. The drive motor and the blower are controlled to run for a predetermined working time and then stop.

[0008] In some embodiments, controlling the start of the drive motor to rotate the drum at a predetermined speed, causing the American ginseng located inside the drum to rub against the wire mesh on the inner circumferential wall of the drum, thereby achieving preliminary peeling of the American ginseng, includes: acquiring the variety information and individual weight information of the American ginseng to be peeled; determining a predetermined speed matching the variety information and individual weight information of the current American ginseng to be peeled based on the pre-stored correspondence between the American ginseng variety, individual weight, and predetermined speed; and controlling the drive motor to rotate at the matched predetermined speed.

[0009] In some embodiments, the step of delivering an airflow at a predetermined speed to the air inlet through the air outlet, the airflow entering the drum through the air inlet, and blowing away the skin flakes generated during the initial peeling process from the surface of the American ginseng and carrying them out of the drum, thereby completing the separation of skin flakes from the American ginseng, includes: real-time detection of the skin flake concentration inside the drum by a dust sensor installed inside the drum; adjusting the airflow speed of the blower according to the real-time skin flake concentration, wherein when the skin flake concentration is higher than a preset concentration threshold, the blower speed is increased to a first adjusted speed, and when the skin flake concentration is lower than or equal to the preset concentration threshold, the blower speed is decreased to a second adjusted speed, wherein the first adjusted speed is greater than the second adjusted speed.

[0010] In some embodiments, the step of supplying airflow at a predetermined speed to the air inlet through the air outlet, and the airflow entering the drum through the air inlet to blow away the skin flakes generated during the initial peeling process from the surface of the American ginseng and carry them out of the drum, thus completing the separation of skin flakes from the American ginseng, includes: adjusting the airflow direction of the blower so that the airflow enters from the air inlet along the axial direction of the drum and then obliquely passes through the gaps between the American ginseng inside the drum; the airflow carrying the skin flakes passes through the gaps in the wire mesh on the inner peripheral wall of the drum and is discharged from the outer peripheral wall of the drum to the hollow part of the support; the skin flake carry-out rate is detected by a skin flake sensor installed on the outer peripheral wall of the drum; when the skin flake carry-out rate is lower than a preset carry-out rate threshold, the angle between the airflow direction and the axial direction of the drum is increased; when the skin flake carry-out rate is higher than or equal to the preset carry-out rate threshold, the angle is decreased.

[0011] In some embodiments, the method further includes: collecting historical peeling operation data, the historical data including variety information of the ginseng to be peeled, individual ginseng weight information, initial humidity information, actual rotation speed of the drive motor, actual wind speed of the blower, and corresponding peeling effect score; the peeling effect score is comprehensively evaluated based on the ginseng damage rate and skin residue rate; training a machine learning model using the historical data, the machine learning model taking the variety of the ginseng to be peeled, individual ginseng weight, and initial humidity as input, and outputting a recommended predetermined rotation speed and predetermined wind speed; when the current ginseng to be peeled has its variety, individual ginseng weight, and initial humidity information obtained, inputting the current ginseng to be peeled variety, individual ginseng weight, and initial humidity information into the machine learning model to obtain the recommended predetermined rotation speed and wind speed; controlling the drive motor to rotate at the recommended rotation speed, and controlling the blower to deliver air at the recommended wind speed.

[0012] In some embodiments, the method further includes: capturing a surface image of the American ginseng in real time using an image sensor installed inside the drum; performing image recognition processing on the surface image to extract features of the peeled areas, calculating the proportion of the peeled areas to the total surface area, and obtaining the peeling rate; the features include color and texture changes; when the peeling rate is lower than a first preset threshold, increasing the speed of the drive motor to a first adjustment speed higher than a predetermined speed, and increasing the speed of the blower to a first adjustment speed higher than a predetermined speed; when the peeling rate is higher than a second preset threshold, decreasing the speed of the drive motor to a second adjustment speed lower than a predetermined speed, and decreasing the speed of the blower to a second adjustment speed lower than a predetermined speed; wherein the first preset threshold is less than the second preset threshold.

[0013] In some embodiments, the method further includes: detecting the vibration value of the drum in real time using a vibration sensor on the support, and detecting the current value in real time using a current sensor on the drive motor; inputting the vibration value and current value into a pre-stored anomaly judgment model; the anomaly judgment model is pre-trained using vibration or current characteristics when the drum is stuck or the motor is overloaded; if the anomaly judgment model determines that the vibration value exceeds a preset vibration threshold and the current value exceeds a preset current threshold, then it is determined that there is a jam or overload; controlling the drive motor to reduce its speed to a safe speed, stopping the blower from blowing air, and sending a fault warning signal to the user terminal.

[0014] In some embodiments, the method further includes: acquiring real-time temperature and humidity inside the drum; adjusting the blower's air supply temperature and humidity according to pre-stored adjustment rules for temperature, humidity, and air supply parameters; the adjustment rules include reducing the air supply temperature when the temperature is higher than 30°C or the humidity is higher than 70%; and reducing the air supply humidity when the humidity is higher than 75%.

[0015] In some embodiments, the method further includes: recording complete parameters for each peeling operation, including the quantity and variety of American ginseng to be peeled, the average weight of a single American ginseng, the rotational speed curve of the drive motor, the wind speed curve of the blower, the peeling time, and the final peeling effect score; performing statistical analysis on the parameters to calculate the correlation between each parameter and the peeling effect score; the correlation includes the correlation between rotational speed and damage rate and the correlation between wind speed and dander removal rate; identifying parameters with correlations higher than a preset threshold as key influencing parameters; the key influencing parameters include average rotational speed, average wind speed, and peeling time; optimizing pre-stored predetermined rotational speed and wind speed setting rules based on the statistical results of the key parameters; and determining the predetermined parameters according to the optimized rules when performing a peeling operation again.

[0016] Secondly, this application also provides a peeling device for American ginseng, comprising: The support has a hollow section; The drive motor is fixedly mounted on the bracket. A roller, one end of which is connected to the output shaft of the drive motor, and the other end of which is rotatably connected to the bracket. The roller is located inside the hollow part. The rotation center axis of the roller is arranged parallel to the horizontal direction. The inner peripheral wall of the roller is covered with wire mesh. An air inlet is provided at the end of the roller facing away from the drive motor. A blower is fixedly mounted on the bracket, with the blower's air outlet facing the air inlet; A control device is fixedly mounted on the bracket and electrically connected to the drive motor and the blower, used to implement the steps of the ginseng peeling method described in the first aspect above.

[0017] This application utilizes automated control to drive the motor and blower, eliminating the need for manual monitoring and increasing processing efficiency compared to manual methods, thus meeting the processing needs of large-scale cultivation. The wire mesh on the inner wall of the drum provides uniform friction, preventing damage to the ginseng due to uneven manual force. Furthermore, the blower's directional airflow (with air holes aligned with the drum's air inlet) thoroughly removes the peeling debris, ensuring the cleanliness of the American ginseng surface and preserving its medicinal components. Automation reduces manual labor requirements, and the complete separation of peeling debris eliminates the need for secondary cleaning, lowering labor and processing costs. By setting parameters such as predetermined rotation speed, wind speed, and time, the peeling effect of each batch of American ginseng is ensured to be consistent, solving the consistency problem of manual or existing mechanical processing and meeting standardized production requirements.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the structure of the ginseng peeling device provided in this application from one perspective; Figure 2 This is a schematic diagram of the structure of an embodiment of the ginseng peeling device provided in this application from another perspective; Figure 3 yes Figure 1 Schematic diagram of the middle roller; Figure 4 yes Figure 1 Schematic diagram of the auxiliary rotating support assembly; Figure 5 yes Figure 4 Exploded view of the auxiliary rotation support assembly; Figure 6 This is a schematic flowchart illustrating the steps of a method for peeling American ginseng according to an embodiment of this application; Figure 7 This is a schematic block diagram of the structure of a control device provided in an embodiment of this application.

[0021] Figure label: 100, Bracket; 110, Hollow section; 200, Drive motor; 300, Roller; 310, Wire mesh; 320, Air inlet; 330, Door hole; 340, Cover plate; 341, Locking tongue; 342, Grip handle; 350, Locking buckle; 400, Blower; 410, Air outlet; 500, Control device; 600, Auxiliary rotation support assembly; 610, Fixed shaft; 611, Fixed plate; 611a, Through hole; 620, Bearing; 630, Roller sleeve; 631, Insertion hole; 632, Annular rib; 700, Drawer; 710, Receiving slot.

[0022] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0025] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0026] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0027] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0028] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0029] American ginseng (Panax quinquefolius L.) is a valuable medicinal and economic crop. Its roots, after processing, possess properties that replenish qi and yin, clear heat and promote fluid production, and are widely used in the pharmaceutical and health product industries. Peeling is a crucial step in American ginseng processing, aiming to remove the outer skin from the roots, ensuring the finished product's appearance while enhancing its specific medicinal effects.

[0030] The existing peeling technology for American ginseng has the following problems: (1) Traditional manual peeling: It relies on manual rubbing or brushing, which has three major drawbacks: ① High labor intensity and extremely low efficiency (only 5-10 kg can be processed per hour), which cannot meet the processing needs after large-scale planting; ② Uneven manual force, which can easily lead to scratches and damage to the ginseng body, destroying the medicinal components and appearance integrity, and reducing the commodity value; ③ High labor cost, with the processing cost increasing year by year as labor prices rise.

[0031] (2) Existing mechanical peeling devices: In order to solve the problem of manual labor, two types of mechanical devices have appeared on the market, but both have fatal defects: Single friction peeling device: The roller rotates to drive the American ginseng to rub against the inner wall, but the inner wall of the roller is mostly a smooth metal surface or ordinary mesh, the friction effect is poor, and the skin flakes are easy to adhere to the surface of the American ginseng during the peeling process, requiring manual secondary cleaning, which increases the processing steps and costs; Single blowing cleaning device: The blower blows air to assist in the separation of skin flakes, but it is not coordinated with the roller friction action, the air supply direction is not aligned with the air inlet of the roller, the airflow cannot effectively enter the inside of the roller, the skin flakes are not completely separated, and the peeling function cannot be realized.

[0032] In addition, the existing mechanical equipment has a very low degree of automation, requiring manual monitoring of the peeling process, making it difficult to ensure the consistency of each batch and failing to meet the requirements of standardized production.

[0033] To resolve the above issues, please refer to [link / reference]. Figures 1 to 5 The ginseng peeling device provided by this invention includes: a support frame 100, a drive motor 200, a roller 300, a blower 400, and a control device 500. The support frame 100, as the main load-bearing structure of the device, can be welded from angle steel, channel steel, or other profiles to ensure sufficient structural strength and stability. A hollow section 110 is formed in the middle of the support frame 100 to accommodate the roller 300.

[0034] The drive motor 200 is securely mounted on one end of the bracket 100 via a motor mount or flange plate. Preferably, the drive motor 200 is a geared motor with speed adjustment function to allow for adjustment of the output speed according to different peeling process requirements. The output shaft of the drive motor 200 is coaxially connected to one end of the roller 300 via a coupling or directly, providing rotational power to the roller 300.

[0035] The roller 300 is horizontally positioned, with its rotation axis parallel to the horizontal direction. The right end of the roller 300 is connected to the output shaft of the drive motor 200, while the left end is rotatably connected to the support 100 via bearings and other structures. The main body of the roller 300 is located within the hollow portion 110 of the support 100. A layer of wire mesh 310 is laid on the inner circumferential wall of the roller 300. The mesh size of this wire mesh 310 needs to be designed to allow sufficient friction against the surface of the ginseng to remove the outer skin, while also preventing the ginseng from getting stuck in the mesh or leaking out. The wire mesh 310 can be fixed to the inner wall of the roller 300 by welding, riveting, or high-strength bonding.

[0036] Additionally, an air inlet 320 is provided on the upper part of the end cap or side wall of the end of the roller 300 facing away from the drive motor 200. The blower 400 is fixedly mounted on the bracket 100, and its air outlet 410 is directly opposite the air inlet 320 of the roller 300. The blower 400 is preferably a centrifugal fan or a high-pressure vortex fan to provide a drying airflow with sufficient air pressure and volume.

[0037] The control device 500 is typically an electrical control box, fixedly mounted on the bracket 100 for easy operation and protection from moisture contamination. Internally, the control device 500 integrates a circuit breaker, contactor, frequency converter (if the motor is speed-adjustable), PLC or microcontroller controller, as well as operation buttons, indicator lights, and a display screen for convenient operation. The control device 500 is electrically connected to the drive motor 200 and the blower 400 via cables, enabling control and setting of parameters such as start / stop, speed (wind speed), and operating time for both.

[0038] When the equipment is in operation, the operator starts the equipment via control device 500. Drive motor 200 drives drum 300 to rotate at a constant speed. The American ginseng placed inside the drum rotates with the drum under the influence of gravity, continuously tumbling and falling. Its surface continuously rubs and scrapes against the wire mesh 310 on the inner wall of the drum, thus achieving peeling. Simultaneously, blower 400 starts, generating dry, clean airflow that blows into the drum 300 through air inlet 320. The airflow dries the freshly peeled, still damp ginseng skin, reducing its stickiness; simultaneously, the airflow carries the dried skin flakes, which, with the assistance of the rotating airflow inside the drum, pass through the mesh of the wire mesh 310 and are discharged from the drum. Once peeling is complete and the skin flakes have been mostly blown away, the equipment is stopped, and the American ginseng is removed.

[0039] Furthermore, to improve the smoothness of the roller 300's operation, especially under heavy loads, the present invention also includes two sets of auxiliary rotating support assemblies 600. One end of each of these two sets of auxiliary rotating support assemblies 600 is fixedly installed at a certain interval on the crossbeam or vertical beam of the bracket 100 by welding or bolting. The other ends of the two assemblies 600 are respectively connected to the outer peripheral wall of the roller 300 from below or the side to form rolling support connections. In this way, the outer edge of the roller 300 obtains two auxiliary support points, which, together with the support at the drive end, constitute a stable multi-point support system, effectively preventing roller deflection and vibration during operation.

[0040] Please see Figure 4 and Figure 5 The auxiliary rotation support assembly 600 includes a fixed shaft 610, a bearing 620, and a rolling sleeve 630. The fixed shaft 610 is a rigid short shaft, one end of which is fixedly connected to the bracket 100 by welding or bolting. The other end of the fixed shaft 610, extending into the hollow portion 110, has a shoulder for mounting the bearing 620. The bearing 620 is preferably a deep groove ball bearing or a self-aligning roller bearing, with its inner ring interference-fitted with the fixed shaft 610 or fixed by a set screw. The rolling sleeve 630 is generally a cylindrical sleeve with a central insertion hole 631. The diameter of the insertion hole 631 is slightly larger than the outer diameter of the bearing 620, forming a clearance fit, allowing the rolling sleeve 630 to easily fit onto the outer ring of the bearing 620 and rotate freely with it. The outer peripheral wall of the rolling sleeve 630 is polished or inlaid with wear-resistant material, and at least a portion of its outer wall remains in contact with the outer peripheral wall of the left end of the roller 300, forming rolling friction.

[0041] Furthermore, to enhance the connection strength between the fixed shaft 610 and the bracket 100, two fixing plates 611 are welded or integrally formed on the outer peripheral wall of the fixed shaft 610, with the two fixing plates 611 located on the left and right sides of the fixed shaft 610, respectively. Each fixing plate 611 has at least one through hole 611a. During installation, high-strength bolts are used to pass through the through holes 611a and lock them with the corresponding screw holes on the bracket 100. By increasing the connection area and providing two fixing points, the connection is extremely stable and can resist forces in various directions.

[0042] In some embodiments, an annular rib is provided at the end of the roller sleeve facing the fixed shaft. The annular rib extends circumferentially along the roller sleeve, and the end face of the annular rib is in clearance fit with the end face of the roller facing away from the drive motor. An annular rib 632 extends circumferentially from the end of the roller sleeve 630 facing the fixed shaft 610. After installation, a gap is maintained between the end face of the annular rib 632 and the end face of the left end of the roller 300. This clearance fit structure achieves axial positioning of the roller 300. When the roller 300 generates a small axial force due to materials or other reasons during rotation, the annular rib 632 abuts against the end face of the roller, preventing it from continuing to move axially, thereby precisely limiting the axial position of the roller within the hollow portion 110, while avoiding the frictional resistance and wear that may result from direct rigid contact.

[0043] Please see Figures 1 to 3 To facilitate the insertion and removal of ginseng, a door opening 330 is provided on the outer peripheral wall of the drum 300. The size of the door opening 330 should be sufficient to allow operators to easily place and remove materials. A cover plate 340 is installed on the drum 300 corresponding to the door opening 330. The shape of the cover plate 340 matches the door opening 330 and can be hinged to the drum, or designed as a completely removable flat plate. To reliably close the door opening 330 during equipment operation, a sliding latch 341 is installed on the cover plate 340, and a latch 350 is installed at the corresponding position on the outer peripheral wall of the drum 300. By pushing or pulling the latch 341 to engage or disengage the latch 350, the cover plate 340 can be quickly locked and unlocked. A raised gripping handle 342 is also installed on the side of the latch 341 for easy finger application, making operation more effortless and faster.

[0044] Please see Figure 3 In a preferred embodiment, the entire sidewall of the roller 300 is enclosed by a wire mesh 310, which not only reduces the overall weight of the roller 300 but also allows dander to be smoothly discharged from all sides of the roller. To collect the discharged dander, a drawer 700 is slidably installed inside the hollow portion 110 of the support 100, directly below the roller 300. The drawer 700 is connected to the support 100 via a slide rail and can be easily pulled out or pushed in. The upward-facing side of the drawer 700 (i.e., the side facing the roller 300) is designed with a recessed receiving groove 710 to catch the falling dander. When a processing operation is completed or the receiving groove 710 is nearly full, simply pulling out the drawer 700 allows for the centralized cleaning of accumulated dander, maintaining the cleanliness of the equipment's interior and greatly simplifying maintenance.

[0045] like Figure 6 As shown, one embodiment of this application provides a method for peeling American ginseng. This method for peeling American ginseng can be performed by... Figures 1 to 5 The control device for the peeling equipment of American ginseng shown is implemented.

[0046] Specifically, such as Figure 6 As shown, the peeling method for American ginseng provided includes steps S101 to S103, which are detailed below: Step S101. Control the drive motor to start, so that the drum rotates at a predetermined speed, causing the American ginseng located inside the drum to rub against the wire mesh on the inner wall of the drum, thereby achieving the initial peeling of the American ginseng.

[0047] Specifically, the core of this step is to drive the drum to rotate using a drive motor, causing the American ginseng to tumble and fall inside the drum, and to form continuous physical friction with the wire mesh on the inner wall of the drum, thus completing the initial peeling of the outer skin of the American ginseng.

[0048] The power transmission and speed control system is a drive motor (preferably a speed-regulating and geared motor) coaxially connected to the drum, outputting stable rotational power. The speed can be flexibly adjusted according to the processing requirements such as the variety, size, and humidity of the American ginseng, ensuring that the friction force is appropriate and does not damage the ginseng body. The material movement mechanism involves the rotation of a horizontal drum. Under the influence of gravity, the American ginseng rises along the drum wall and falls after reaching a certain height, repeatedly forming a cycle of "rolling-friction-falling". This ensures that all surfaces of the ginseng can fully contact the wire mesh. The friction medium design utilizes a wire mesh laid on the inner wall of the drum with optimized mesh openings. This design not only scrapes and rubs the ginseng skin through the mesh edges for efficient peeling but also prevents the ginseng from getting stuck in or leaking out of the mesh, ensuring the integrity of the material. Furthermore, it caters to different user needs: those who want to replenish their vital energy and improve their immunity should use ginseng with the skin on; while those who are sensitive to heat and simply want to nourish yin and generate fluids should use peeled ginseng.

[0049] In the early stages, the operator sets the predetermined speed of the drive motor through the control device (electric control box) (adjusted according to the specifications of the American ginseng, usually suitable for the low to medium speed range of small and medium-sized ginseng bodies to avoid damage to the ginseng bodies due to high speed), and checks whether the drum cover is locked by the locking tongue and the latch to ensure that there is no material leakage during operation; The start command is sent through the operation button of the control device. The control device triggers the drive motor to start via cable. The motor output shaft drives the drum to rotate at a set speed. The initial peeling process involves the rotating drum causing the ginseng inside to tumble and fall continuously. The surface of the ginseng is repeatedly rubbed and scraped against the wire mesh, gradually loosening and peeling off the outer skin, thus completing the initial peeling. During this process, the auxiliary rotation support components of the drum (fixed shaft, bearing, roller sleeve) provide rolling support from below / side to prevent the drum from bending or vibrating due to the weight of the material, ensuring stable rotation and further preventing damage to the ginseng due to equipment shaking.

[0050] Step S102. During the rotation of the drum, the blower is started and an airflow of a predetermined speed is delivered to the air inlet through the air outlet. The airflow enters the drum through the air inlet and blows the skin flakes generated during the initial peeling process off the surface of the American ginseng and carries them out of the drum, thus completing the separation of skin flakes from the American ginseng.

[0051] Specifically, this step involves using a blower to deliver air while the drum rotates to peel the ginseng, achieving real-time separation of the dander from the ginseng and solving the problem of dander adhesion and the need for secondary cleaning after traditional peeling.

[0052] The airflow is precisely delivered by aligning the air outlet of the blower with the air inlet of the drum, forming a "directional air delivery channel" to ensure that the dry airflow enters the drum efficiently and avoids the reduction in separation efficiency caused by airflow leakage. The airflow has a dual function: first, it blows and dries the freshly removed wet ginseng skin, reducing the stickiness of the skin flakes and preventing them from re-adhering to the surface of the American ginseng; second, it carries the dried skin flakes through the mesh of the wire mesh inside the drum with the assistance of the rotating airflow, thus achieving complete separation of the skin flakes from the ginseng body. The coordinated operation is achieved by synchronizing the blower and the rotation of the drum, allowing peeling and dander removal to proceed simultaneously without additional steps, thus improving processing efficiency.

[0053] The wind speed setting is achieved by the operator presetting the blower's speed via the control device (using a centrifugal fan or a high-pressure vortex fan; the wind speed must be adapted to the peeling rhythm to remove dander without damaging the sample). Synchronous start-up is achieved by the control device simultaneously triggering the drive motor (or after a very short delay) and simultaneously triggering the blower's start-up. The dry, high-pressure airflow generated by the blower is precisely blown into the drum's air inlet through the air outlet, reaching directly into the drum. During the dander separation process, after the airflow enters the drum, the initially removed wet dander is first dried to reduce its stickiness. Then, carrying the dried dander, it is diffused by the rotating airflow inside the drum, passing through the wire mesh and exiting the drum. The discharged dander falls into the drawer receiving slot directly below the drum for centralized collection, preventing dander from scattering and contaminating the equipment or environment. This process requires no manual intervention; the continuous airflow ensures no dander residue adheres.

[0054] Step S103. Control the drive motor and the blower to run for a predetermined working time and then stop.

[0055] Specifically, this step uses a control device to preset the working time, enabling the drive motor and blower to automatically stop at set times, ensuring the consistency of each batch of American ginseng peeling and processing, and meeting the requirements of standardized production.

[0056] Precise time control is achieved through the integrated PLC / microcontroller controller in the control device, which can accurately set the total processing time. Once the time is up, the power to the motor and blower will be automatically cut off to avoid excessive peeling that could damage the sample or insufficient peeling that could affect the quality. The automated closed-loop system automates the entire process from startup, peeling, chip removal to shutdown, eliminating the need for real-time manual monitoring, reducing labor costs, and ensuring consistent processing parameters across different batches, thereby improving the finished product qualification rate. After the machine stops, the roller stops rotating and the airflow is interrupted. The operator can safely open the roller cover and take out the American ginseng that has been peeled and cleaned of skin flakes.

[0057] The time preset allows operators to set a predetermined working time based on the peeling difficulty of the ginseng (e.g., old ginseng, young ginseng) and the processing volume via the control device's display screen or operating buttons (usually determined based on experimental data to ensure thorough peeling without damage to the ginseng). Automatic operation and shutdown are achieved by the equipment running according to preset parameters. The drive motor continuously peels the ginseng while the blower continuously blows air to remove debris. The control device keeps a real-time timer. When the set working time is reached, an automatic shutdown command is issued, sequentially cutting off the power to the drive motor and blower, causing both to stop simultaneously. After shutdown, the operator unlocks the latch on the drum cover, opens the cover, and removes the finished ginseng. Simultaneously, the drawer under the drum can be pulled out to clean the collected dander from the receiving trough, completing the entire single processing cycle and preparing for the next batch.

[0058] In some embodiments, controlling the start of the drive motor to rotate the drum at a predetermined speed, causing the American ginseng located inside the drum to rub against the wire mesh on the inner circumferential wall of the drum, thereby achieving preliminary peeling of the American ginseng, includes: acquiring the variety information and individual weight information of the American ginseng to be peeled; determining a predetermined speed matching the variety information and individual weight information of the current American ginseng to be peeled based on the pre-stored correspondence between the American ginseng variety, individual weight, and predetermined speed; and controlling the drive motor to rotate at the matched predetermined speed.

[0059] The core of this embodiment is to achieve adaptive speed matching based on the material characteristics of American ginseng, which solves the problem that traditional fixed speed cannot be adapted to different varieties and sizes of American ginseng, and avoids damage to the ginseng body or incomplete peeling due to improper speed.

[0060] Material information sensing obtains the variety information of the American ginseng to be peeled and the weight information of a single American ginseng through manual input or automatic identification (such as weighing sensors, barcode identification of varieties), thus clarifying the basic characteristics of the material; The preset parameter association is established through a pre-stored database of the correspondence between "variety - weight of a single American ginseng - predetermined rotation speed". This relationship is based on a large amount of experimental data to ensure that American ginseng with different characteristics can be matched with the optimal friction speed. Precise speed control uses a control device to drive a speed-regulating and reducing motor to output the corresponding speed according to the matched speed command, so that the rotation force of the drum is precisely matched with the characteristics of the material, taking into account both peeling efficiency and material integrity.

[0061] Information acquisition involves operators inputting the variety information (e.g., soft or hard ginseng) of the ginseng to be peeled via the input module of the control device (e.g., touchscreen, barcode scanner), or automatically collecting the weight information of a single ginseng through a weighing sensor inside the drum, and transmitting this information to the processor of the control device. Speed ​​matching involves the processor retrieving a pre-stored database of corresponding relationships, comparing the current variety and weight information of a single ginseng with the database entries, and selecting a matching predetermined speed (e.g., low speed for lighter ginseng, medium to high speed for heavier ginseng). Execution control involves the control device sending speed commands to the drive motor, which adjusts its output speed via a frequency converter or speed control module, driving the drum to rotate at the matching speed. The ginseng rubs against the wire mesh at the appropriate speed, achieving efficient and non-destructive initial peeling.

[0062] In some embodiments, the step of delivering an airflow at a predetermined speed to the air inlet through the air outlet, the airflow entering the drum through the air inlet, and blowing away the skin flakes generated during the initial peeling process from the surface of the American ginseng and carrying them out of the drum, thereby completing the separation of skin flakes from the American ginseng, includes: real-time detection of the skin flake concentration inside the drum by a dust sensor installed inside the drum; adjusting the airflow speed of the blower according to the real-time skin flake concentration, wherein when the skin flake concentration is higher than a preset concentration threshold, the blower speed is increased to a first adjusted speed, and when the skin flake concentration is lower than or equal to the preset concentration threshold, the blower speed is decreased to a second adjusted speed, wherein the first adjusted speed is greater than the second adjusted speed.

[0063] This embodiment addresses the issues of incomplete separation due to dander accumulation or excessive energy consumption caused by high airflow by real-time monitoring of dander concentration inside the drum and closed-loop adjustment of the blower speed, thus achieving a balance between dander separation efficiency and energy consumption. Real-time concentration detection: A dust sensor is installed inside the drum to collect dander concentration data in real time and feed it back to the control device, forming a "detection-feedback" closed loop. The wind speed dynamic response system determines the separation status based on the concentration threshold through the control device. When the concentration is too high, the wind speed is increased to enhance the cleaning ability. When the concentration meets the standard, the wind speed is reduced to save energy and avoid ineffective high wind speed operation. Separation efficiency is optimized by dynamically adjusting the wind speed according to the amount of dander produced, ensuring that the dander generated during the peeling process can be blown away in time without residue or accumulation, thus improving the separation effect.

[0064] The sensor deployment involves installing dust sensors on the inner wall of the roller (in an area away from the air inlet to avoid interference from direct airflow). The sensors are connected to the control device wirelessly or via wired connection to transmit dander concentration data in real time. Threshold setting is achieved by the operator preseting the dandruff concentration threshold (e.g., 50 mg / m³) via the control device. 3 The dust sensor detects a real-time dander concentration higher than the preset threshold, and the control device sends a command to the blower to increase the wind speed to the first adjustment wind speed (e.g., 20% higher than the base wind speed) to enhance airflow carrying capacity and quickly clear accumulated dander. If the real-time dander concentration is lower than or equal to the preset threshold, the control device reduces the blower speed to the second adjustment wind speed (e.g., the base wind speed or 10% lower than the base wind speed) to reduce energy consumption while ensuring separation effect. Throughout the entire dander removal process, the dust sensor continuously detects and the wind speed is dynamically adjusted in real time until the dander concentration stabilizes and meets the standard.

[0065] In some embodiments, the step of supplying airflow at a predetermined speed to the air inlet through the air outlet, and the airflow entering the drum through the air inlet to blow away the skin flakes generated during the initial peeling process from the surface of the American ginseng and carry them out of the drum, thus completing the separation of skin flakes from the American ginseng, includes: adjusting the airflow direction of the blower so that the airflow enters from the air inlet along the axial direction of the drum and then obliquely passes through the gaps between the American ginseng inside the drum; the airflow carrying the skin flakes passes through the gaps in the wire mesh on the inner peripheral wall of the drum and is discharged from the outer peripheral wall of the drum to the hollow part of the support; the skin flake carry-out rate is detected by a skin flake sensor installed on the outer peripheral wall of the drum; when the skin flake carry-out rate is lower than a preset carry-out rate threshold, the angle between the airflow direction and the axial direction of the drum is increased; when the skin flake carry-out rate is higher than or equal to the preset carry-out rate threshold, the angle is decreased.

[0066] This embodiment solves the problems of poor penetration and low dander carry-out rate of traditional axial airflow by optimizing the airflow path (oblique air supply) and adjusting the air supply angle in a closed loop, thereby improving the separation efficiency of dander and ginseng.

[0067] The oblique airflow design adjusts the direction of the blower's airflow so that the airflow enters along the roller axis and then obliquely penetrates the gaps between the American ginseng, increasing the contact area between the airflow and the ginseng body and skin flakes, and improving the skin flake carrying capacity. Real-time monitoring of carry-out rate involves installing a dander sensor on the outer peripheral wall of the drum (outside the wire mesh) to detect the amount of dander carried out in real time, calculate the carry-out rate (amount of dander carried out / total amount of dander generated), and quantify the separation effect. The adaptive angle adjustment dynamically adjusts the angle between the air supply direction and the roller axis based on the comparison between the carry-out rate and the preset threshold, ensuring that the airflow is always at the optimal penetration angle and maximizing the carry-out rate.

[0068] The air supply direction can be adjusted by using the rotatable nozzle or guide plate of the blower to make the air supply direction at a certain angle with the axis of the drum (the initial angle can be set to 30°-45°), so that the airflow enters the drum obliquely and passes through the gaps of the American ginseng; the path of skin dander discharge: the airflow carrying skin dander passes through the gaps in the wire mesh, is discharged from the outer peripheral wall of the drum to the hollow part of the support, and finally falls into the drawer below; The carry-out rate detection and adjustment system uses a dander sensor to collect the amount of dander discharged in real time. The control device calculates the dander carry-out rate and compares it with a preset carry-out rate threshold (e.g., 90%). If the carry-out rate is lower than the preset threshold, it indicates insufficient airflow penetration. The control device increases the angle between the airflow direction and the roller axis (e.g., to 50°) to enhance the oblique penetration capability of the airflow. If the carry-out rate is higher than or equal to the preset threshold, it indicates that the airflow angle is suitable. The control device decreases the angle (e.g., to 30°) to avoid excessive angle causing airflow dispersion and maintain efficient separation. Throughout the peeling process, the carry-out rate is monitored in real time, and the airflow angle is dynamically fine-tuned to ensure that the dander carry-out rate always meets the standard.

[0069] In some embodiments, the method further includes: collecting historical peeling operation data, the historical data including variety information of the ginseng to be peeled, individual ginseng weight information, initial humidity information, actual rotation speed of the drive motor, actual wind speed of the blower, and corresponding peeling effect score; the peeling effect score is comprehensively evaluated based on the ginseng damage rate and skin residue rate; training a machine learning model using the historical data, the machine learning model taking the variety of the ginseng to be peeled, individual ginseng weight, and initial humidity as input, and outputting a recommended predetermined rotation speed and predetermined wind speed; when the current ginseng to be peeled has its variety, individual ginseng weight, and initial humidity information obtained, inputting the current ginseng to be peeled variety, individual ginseng weight, and initial humidity information into the machine learning model to obtain the recommended predetermined rotation speed and wind speed; controlling the drive motor to rotate at the recommended rotation speed, and controlling the blower to deliver air at the recommended wind speed.

[0070] This embodiment trains a machine learning model using historical data to intelligently recommend peeling parameters (rotation speed, wind speed), replacing manual experience-based settings and improving parameter adaptability and processing standardization.

[0071] A database was built by collecting comprehensive data on historical peeling operations, including material characteristics (variety, weight of a single American ginseng, initial humidity), equipment parameters (actual rotation speed, wind speed), and processing results (damage rate, skin residue rate, peeling effect score). The model is trained by taking material properties as input and optimal rotation speed and wind speed as output, and using historical data to train a machine learning model (such as neural networks and decision trees) so that the model learns the intrinsic relationship between material properties and processing parameters. During the current operation, the model directly outputs recommended rotation speed and wind speed based on the input material information. The control device then controls the equipment according to the recommended parameters, achieving "one-click intelligent processing".

[0072] Historical data collection involves the control device automatically recording operation data (material information, equipment parameters, and effect scores) after each peeling operation is completed, storing the data in the database, and continuously accumulating historical data. Model training involves periodically (e.g., every 100 assignments) using historical data from the database to train the machine learning model, optimize model parameters, and improve recommendation accuracy. Before the operation, the recommended parameters are entered by the operator or automatically collected by the equipment, including the variety of the American ginseng to be peeled, the weight of a single ginseng, and the initial humidity information. These are then input into a trained machine learning model. The model quickly calculates and outputs the recommended preset rotation speed and preset wind speed, which are then transmitted to the control device. The equipment operates by controlling the drive motor to rotate at the recommended speed and the blower to deliver air at the recommended wind speed according to the recommended parameters, thus completing the peeling and separation of skin flakes without the need for repeated manual parameter adjustments.

[0073] In some embodiments, the method further includes: capturing a surface image of the American ginseng in real time using an image sensor installed inside the drum; performing image recognition processing on the surface image to extract features of the peeled areas, calculating the proportion of the peeled areas to the total surface area, and obtaining the peeling rate; the features include color and texture changes; when the peeling rate is lower than a first preset threshold, increasing the speed of the drive motor to a first adjustment speed higher than a predetermined speed, and increasing the speed of the blower to a first adjustment speed higher than a predetermined speed; when the peeling rate is higher than a second preset threshold, decreasing the speed of the drive motor to a second adjustment speed lower than a predetermined speed, and decreasing the speed of the blower to a second adjustment speed lower than a predetermined speed; wherein the first preset threshold is less than the second preset threshold.

[0074] This embodiment uses an image sensor to monitor the peeling progress in real time and dynamically adjust the equipment parameters, solving the problem that traditional timed processing cannot accurately control the degree of peeling, and avoiding excessive peeling that damages the ginseng or incomplete peeling that affects the quality.

[0075] Visual perception uses image sensors (such as high-definition cameras) installed inside the drum to capture real-time images of the ginseng surface and detect changes in the color and texture of the ginseng during the peeling process. Peeling rate calculation uses image recognition algorithms to extract features of peeled areas (such as color changing from dark brown to light yellow, texture changing from rough to smooth), calculates the proportion of peeled areas to the total surface area, obtains the peeling rate, and quantifies the peeling progress; parameter dynamic adaptation adjusts the rotation speed and wind speed in real time based on the comparison between the peeling rate and the preset threshold. When peeling is insufficient, the parameters are increased to accelerate peeling, and when peeling is excessive, the parameters are decreased to protect the system.

[0076] Image acquisition is achieved by installing an image sensor on the inner wall of the drum (away from the friction area to avoid material obstruction). The sensor rotates synchronously with the drum or is fixed on a bracket to align with the inside of the drum, capturing real-time images of the ginseng surface and transmitting them to the image processing module of the control device. Peeling rate calculation involves preprocessing the image (denoising and enhancement) using the image processing module to extract the color and texture features of the ginseng surface; through feature comparison (comparing with the feature libraries of unpeeled and fully peeled ginseng), the peeled areas are identified, and the peeling rate (peeled area / total surface area) is calculated. The parameter adjustment is performed by setting a first preset threshold (e.g., 80%, the critical value for insufficient peeling) and a second preset threshold (e.g., 95%, the critical value for excessive peeling), with the first preset threshold < the second preset threshold. If the peeling rate is lower than the first preset threshold, it indicates that the peeling progress is slow. The control device increases the drive motor speed to the first adjusted speed (15% higher than the predetermined speed) and increases the blower speed to the first adjusted wind speed (15% higher than the predetermined wind speed) to accelerate peeling and separation of skin flakes. If the peeling rate is higher than the second preset threshold, it indicates that the peeling is excessive. The control device decreases the drive motor speed to the second adjusted speed (10% lower than the predetermined speed) and decreases the blower speed to the second adjusted wind speed (10% lower than the predetermined wind speed) to stop excessive friction and protect the ginseng. Precise shutdown is achieved when the peeling rate reaches the target value (e.g., 90%-95%), the control device automatically stops to ensure that the peeling degree of each batch of American ginseng is consistent.

[0077] In some embodiments, the method further includes: detecting the vibration value of the drum in real time using a vibration sensor on the support, and detecting the current value in real time using a current sensor on the drive motor; inputting the vibration value and current value into a pre-stored anomaly judgment model; the anomaly judgment model is pre-trained using vibration or current characteristics when the drum is stuck or the motor is overloaded; if the anomaly judgment model determines that the vibration value exceeds a preset vibration threshold and the current value exceeds a preset current threshold, then it is determined that there is a jam or overload; controlling the drive motor to reduce its speed to a safe speed, stopping the blower from blowing air, and sending a fault warning signal to the user terminal.

[0078] This embodiment uses a vibration sensor, a current sensor, and an anomaly detection model to achieve real-time prediction and automatic protection against faults such as roller jamming and motor overload, thereby avoiding equipment damage and material loss and improving equipment operation safety.

[0079] Multi-dimensional monitoring comprehensively collects equipment operating status data by installing vibration sensors on the support (to detect the vibration of the drum during operation) and current sensors on the drive motor (to detect the motor load current). Intelligent anomaly judgment uses an anomaly judgment model trained with pre-stored fault data. The model learns the vibration characteristics (sudden increase in vibration value) and current characteristics (excessive current value) during jamming and overload, achieving accurate fault identification. Active safety protection immediately executes operations such as speed reduction, shutdown, and warning once an anomaly is detected to prevent the fault from escalating, while simultaneously notifying operators to handle the situation promptly.

[0080] The sensor deployment involves installing vibration sensors on the crossbeams / vertical beams of the support frame to collect the vibration values ​​of the rollers in real time; and installing current sensors on the power supply line of the drive motor to collect the operating current values ​​of the motor in real time; both types of sensors are connected to the control device to transmit data in real time. The anomaly detection is achieved by inputting real-time vibration and current values ​​into a pre-stored anomaly detection model through the control device. The model compares the real-time data with preset vibration and current thresholds (set based on normal operation data). If the vibration value exceeds the preset vibration threshold and the current value exceeds the preset current threshold, it is determined that the roller is stuck or the motor is overloaded. The safety protection system immediately sends a command to the drive motor via the control device to reduce the speed to a safe speed (e.g., 30% of the rated speed) and simultaneously stops the blower to prevent the fault from worsening. The control device sends a fault warning signal to the user terminal (e.g., mobile phone, computer, on-site display screen) via wireless or wired means, indicating "Drum jamming / Motor overload, please check". Troubleshooting: After receiving the warning, the operator should stop the machine in time to check, and after eliminating the problem of jammed material or overload, restart the equipment.

[0081] In some embodiments, the method further includes: acquiring real-time temperature and humidity inside the drum; adjusting the blower's air supply temperature and humidity according to pre-stored adjustment rules for temperature, humidity, and air supply parameters; the adjustment rules include reducing the air supply temperature when the temperature is higher than 30°C or the humidity is higher than 70%; and reducing the air supply humidity when the humidity is higher than 75%.

[0082] This embodiment optimizes the airflow drying effect by monitoring the temperature and humidity inside the drum in real time and dynamically adjusting the temperature and humidity of the supplied air. This prevents the skin flakes from sticking to the ginseng body in high humidity conditions and the active ingredients of the ginseng body from being damaged in high temperature conditions, thereby improving the processing quality.

[0083] Environmental sensing involves installing temperature and humidity sensors inside the drum to collect real-time temperature and humidity data, thus understanding the processing environment. Regularized adjustment uses pre-stored "temperature and humidity - air supply parameters" adjustment rules to precisely adjust the blower's air supply temperature and humidity according to changes in ambient temperature and humidity, ensuring the airflow adapts to the processing environment. Quality assurance involves optimizing air supply temperature and humidity to ensure rapid drying of dander without sticking, while preventing high-temperature, high-humidity airflow from damaging the medicinal components of the ginseng, thus guaranteeing the quality of the finished product.

[0084] Temperature and humidity data are collected in real time by installing temperature and humidity sensors on the inner wall of the drum (away from the air inlet and friction area), and transmitted to the control device. Adjustment rules are preset by the operator via the control device, such as: reducing the air supply temperature (e.g., below 25℃) when the temperature is above 30℃ or the humidity is above 70%; reducing the air supply humidity (e.g., below 60% via a dehumidification module) when the humidity is above 75%. Air supply parameter adjustment involves comparing the real-time temperature and humidity with the preset rules. If the adjustment condition is triggered, a command is immediately sent to the blower's temperature and humidity control module (e.g., heating / cooling module, dehumidification module). The blower adjusts the temperature and humidity of the output airflow according to the command, ensuring that the airflow entering the drum is always in a state suitable for the processing environment, quickly drying the skin flakes, preventing adhesion, and protecting the material. Throughout the peeling process, the temperature and humidity sensors continuously monitor and adjust the air supply temperature and humidity in real time to ensure optimal environmental and airflow coordination.

[0085] In some embodiments, the method further includes: recording complete parameters for each peeling operation, including the quantity and variety of American ginseng to be peeled, the average weight of a single American ginseng, the rotational speed curve of the drive motor, the wind speed curve of the blower, the peeling time, and the final peeling effect score; performing statistical analysis on the parameters to calculate the correlation between each parameter and the peeling effect score; the correlation includes the correlation between rotational speed and damage rate and the correlation between wind speed and dander removal rate; identifying parameters with correlations higher than a preset threshold as key influencing parameters; the key influencing parameters include average rotational speed, average wind speed, and peeling time; optimizing pre-stored predetermined rotational speed and wind speed setting rules based on the statistical results of the key parameters; and determining the predetermined parameters according to the optimized rules when performing a peeling operation again.

[0086] This embodiment achieves continuous upgrading of the peeling process by recording all operation parameters, statistically analyzing correlations, and iteratively optimizing rules, shifting from "experience-driven" to "data-driven" and improving processing standardization and finished product qualification rate.

[0087] Full parameter recording automatically records all dimensions of parameters related to materials, equipment, and effects for each operation, building a complete operation database to provide a data foundation for analysis and optimization; Correlation analysis uses statistical analysis (such as Pearson correlation coefficient) to identify key parameters affecting peeling effect (such as the correlation between rotation speed and damage rate, and wind speed and carry-out rate), clarifying the direction of process optimization; Rule iterative optimization optimizes the original preset rules for rotation speed and wind speed based on the statistical results of key parameters, making subsequent operation parameters more accurate and continuously improving processing effect.

[0088] The operation parameter recording system automatically records complete parameters during each peeling operation, including: material parameters: quantity, variety, and average weight of a single American ginseng; equipment parameters: drive motor speed curve, blower wind speed curve, and total peeling time; and effect parameters: final peeling effect score (comprehensive ginseng damage rate and skin residue rate). The system periodically (e.g., monthly) statistically analyzes the parameters in the database, calculating the correlation between each parameter and the peeling effect score (e.g., higher speed corresponds to a higher damage rate, showing a positive correlation; higher wind speed corresponds to a higher carry-out rate, also showing a positive correlation). Parameters with correlations higher than a preset threshold (e.g., 0.7) are identified as key influencing parameters (e.g., average speed, average wind speed, peeling time). Rule optimization is based on the statistical results of the key parameters, optimizing the original "variety-weight-speed" and "concentration-wind speed" rules. The system uses preset rules (such as reducing the base rotation speed of larger ginseng by 5% to reduce damage rate; increasing the skin concentration threshold by 10% to improve carry-out rate); the optimized rules are updated to the control device's database, replacing the original rules; during the next peeling operation, the control device determines the preset rotation speed, wind speed, and other parameters according to the optimized rules, realizing continuous iterative upgrades of the process and gradually improving the peeling effect and finished product quality.

[0089] In some embodiments, by introducing reinforcement learning algorithms, a closed-loop control system of "agent-environment-reward" is constructed to replace traditional fixed rules or static models, achieving dynamic adaptive optimization of rotation speed, wind speed, and working time throughout the entire peeling process (S101-S103). Its innovation lies in the fact that it does not rely on a large amount of historical data for initialization; through real-time interactive learning, it balances the three major objectives of "peeling efficiency, ginseng integrity, and energy consumption," adapting to fluctuations in ginseng materials from different batches and in different states, thus solving the problem of "poor adaptability and inability to cope with sudden changes in materials" in traditional algorithms.

[0090] The intelligent agent deployment uses the control device as a reinforcement learning agent, and the material state inside the drum and the equipment operating parameters as the environmental state. The action space is defined by the actions that the agent can output, including adjusting the drive motor speed, adjusting the blower speed, and dynamically correcting the working time. The reward function includes setting reward values ​​for comprehensive peeling efficiency (peeling amount per unit time), material damage rate, energy consumption (motor + blower power), and dander residue rate, maximizing the reward function to achieve multi-objective balance. Real-time learning involves updating the strategy based on environmental feedback (sensor data) during each batch of peeling, gradually optimizing the action output to achieve "more and more accurate with use".

[0091] The reinforcement learning system is built by deploying reinforcement learning agents (using Q-learning or DQN algorithms) in the control device and setting core parameters: environmental state space (initial humidity of American ginseng, weight of a single American ginseng, peeling rate in the drum, dander concentration, motor speed, blower speed, real-time energy consumption), action space (speed adjustment step size ±5%, wind speed adjustment step size ±5%, working time correction range ±10%), and reward function (reward value = 100 - 20 × damage rate - 15 × dander residue rate - 5 × energy consumption ratio + 10 × peeling efficiency, reward value range 0-100).

[0092] In the S101 stage, upon initial operation, the agent outputs an initial rotation speed (based on preset basic parameters, such as medium speed) to control the drive motor to start and rotate the drum. Simultaneously, the image sensor, weighing sensor, and humidity sensor inside the drum collect environmental conditions in real time and feed them back to the agent. Based on the initial conditions (such as the material being too wet or too heavy), the agent gradually adjusts the rotation speed (such as appropriately increasing the rotation speed) and records the corresponding environmental feedback (changes in peeling rate, damage status) and updates the Q-value table.

[0093] In phase S102, execution is synchronized with S101. The agent outputs the initial wind speed to control the blower to start. Dust sensors and dander sensors collect dander concentration and carry-out rate in real time and feed them back to the agent. If the dander residue rate is too high and the reward value decreases, the agent increases the wind speed. If the energy consumption is too high and the reward value decreases, the agent appropriately decreases the wind speed and adjusts the matching relationship between the rotation speed and the wind speed (e.g., when the rotation speed increases, the wind speed is increased simultaneously to ensure timely dander separation), continuously optimizing the action strategy.

[0094] In the S103 stage, the intelligent agent dynamically adjusts the predetermined working time based on real-time changes in peeling rate and reward value. If the peeling rate meets the standard and the reward value reaches its peak, the equipment can be stopped in advance (shortening the working time and reducing energy consumption). If the peeling rate is insufficient and the reward value is too low, the working time can be extended (not exceeding the preset upper limit), while adjusting the rotation speed and wind speed to the optimal combination. When the drive motor and blower run to the adjusted working time, the intelligent agent controls both to stop synchronously.

[0095] After each batch of work is completed, the agent summarizes the environmental state, action output, and reward value of that batch, updates the parameters of the reinforcement learning model, and optimizes the action strategy for the next batch. After learning through multiple batches, the agent can quickly adapt to different material states and output the optimal parameter combination without human intervention.

[0096] In some embodiments, to address the problem of fuzzy uncertainty in the characteristics of American ginseng materials (variety, humidity, size) (such as the inability to accurately quantify "high humidity" or "uneven size"), a fuzzy control algorithm is introduced to replace traditional threshold judgment and linear adjustment, achieving flexible adaptive adjustment of rotation speed (S101) and wind speed (S102). Its innovation lies in its ability to handle fuzzy and nonlinear material characteristic parameters, achieving smooth parameter adjustment without the need for precise threshold quantification, avoiding damage to the ginseng body or incomplete peeling due to parameter mutations, and adapting to the peeling needs of various varieties and states of American ginseng.

[0097] The fuzzy controller design uses "initial humidity of American ginseng, weight deviation of individual American ginseng, and peeling rate deviation" as inputs and "rotation speed adjustment and wind speed adjustment" as outputs. Fuzzification processing transforms the inputs and outputs into fuzzy linguistic variables (e.g., humidity is categorized as "low, medium, and high," and rotation speed adjustment is categorized as "slight decrease, unchanged, slight increase, and significant increase"). The fuzzy rule base establishes fuzzy rules based on peeling experience and experimental data (e.g., "high humidity and low peeling rate → significantly increase rotation speed + significantly increase wind speed," "large weight deviation and high damage rate → slightly increase rotation speed + slightly decrease wind speed"). Defuzzification transforms the fuzzy outputs into precise control quantities, which are then output to the drive motor and blower to achieve smooth adjustment.

[0098] Fuzzy controller deployment: Integrate a fuzzy controller in the control device and set the fuzzy range of input and output quantities, including: (1) Input quantities: initial humidity (fuzzy language: low, medium, high, corresponding to quantization range 0-30%, 30%-60%, 60%-100%); weight deviation of a single American ginseng (fuzzy language: small, moderate, large, corresponding to quantization range -50%, -50%~+50%, +50% and above); peeling rate deviation (fuzzy language: large deviation, medium deviation, small deviation, corresponding to quantization range <-20%, -20%~+10%, >+10%); (2) Output quantities: speed adjustment quantity (fuzzy language: significantly reduced, slightly reduced, unchanged, slightly increased, significantly increased, corresponding to quantization range -20%, -10%, 0, +10%, +20%); wind speed adjustment quantity (consistent with the fuzzy range of speed adjustment quantity).

[0099] The establishment of the fuzzy rule base includes: based on experimental data and peeling experience, 27 core fuzzy rules are established, such as: ① Low humidity + moderate weight + small peeling rate deviation → constant rotation speed and constant wind speed; ② High humidity + large weight + large peeling rate deviation → significantly increased rotation speed and significantly increased wind speed; ③ Medium humidity + small weight + medium peeling rate deviation → slightly increased rotation speed and slightly increased wind speed; ④ High humidity + small weight + small peeling rate deviation → slightly decreased rotation speed and constant wind speed. S101 stage execution: Before operation, the initial humidity and weight of each American ginseng to be peeled are collected, and the weight deviation of each American ginseng is calculated (compared to the standard weight); these two parameters are input into the fuzzy controller, which outputs the rotation speed adjustment amount according to the fuzzy rule base. Combined with the base rotation speed, the actual rotation speed for S101 stage is obtained, and the drive motor is started; during operation, the peeling rate is collected in real time, the peeling rate deviation is calculated, and the rotation speed is dynamically adjusted to achieve flexible adjustment (e.g., as the peeling rate deviation decreases, the rotation speed adjustment amount is gradually reduced). S102 Stage Execution: Synchronized with S101, the fuzzy controller outputs a wind speed adjustment based on initial humidity, peeling rate deviation, and weight deviation. Combined with the base wind speed, this controls the blower's startup. When the dust sensor detects a change in dander concentration, no additional threshold judgment is needed; the fuzzy controller automatically adjusts the wind speed based on the humidity change corresponding to the dander concentration (e.g., if the dander concentration is high, the corresponding humidity is high, and the wind speed adjustment is automatically increased), ensuring thorough dander separation while avoiding damage to the material due to sudden wind speed changes. The fuzzy controller uses the center-of-gravity method to convert the fuzzy output into precise rotational speed and wind speed adjustment values, outputting them to the control modules of the drive motor and blower. This achieves smooth and continuous parameter adjustment, adapting to the fuzzy fluctuations in material characteristics.

[0100] In some embodiments, by combining digital twin technology with deep learning algorithms, a digital twin model of the peeling equipment and materials is constructed to achieve real-time mapping, effect prediction, and advance control of the peeling process. This solves the drawbacks of traditional "post-processing adjustments" and improves processing accuracy and consistency. Its innovation lies in digitizing the physical peeling process (drum rotation, material friction, and dander separation), predicting the peeling effect over a future period through deep learning models, adjusting parameters in advance to avoid insufficient or excessive peeling, and simulating the processing effects of different parameter combinations to achieve parameter pre-optimization.

[0101] The digital twin model is constructed based on the three-dimensional structure of the peeling equipment (support, roller, blower) and the characteristics of American ginseng material, establishing a one-to-one mapping between the physical entity and the digital model, and synchronizing equipment operating parameters and material status in real time. The deep learning prediction model uses the LSTM deep learning algorithm, taking historical operating data (rotation speed, wind speed, temperature and humidity, peeling rate) as input, to predict the peeling rate, damage rate, and dander residue rate in the next 5-10 minutes. The prediction-control closed loop uses the digital twin model to provide real-time feedback on the material status, and the deep learning model outputs the prediction results. If the prediction results are not up to standard (such as excessively high damage rate or insufficient peeling rate), the control device adjusts the parameters in advance to achieve proactive control. Parameter pre-optimization uses the digital twin model to simulate the processing effects of different combinations of rotation speed and wind speed, selects the optimal parameter combination, and applies it to the actual peeling process.

[0102] Digital twin model construction: Using 3D modeling software (such as SolidWorks) to build a digital twin model of the peeling equipment, and integrating sensor data interfaces to achieve: (1) Equipment status synchronization: Real-time collection of parameters such as drive motor speed, blower wind speed, roller vibration value, temperature and humidity, etc., synchronized to the digital twin model to realize the digital mapping of the equipment operation status; (2) Material status synchronization: Real-time collection of the status of American ginseng (peeling rate, damage, and skin residue) through image sensors and weighing sensors, synchronized to the digital twin model to build a dynamic mapping of the material peeling process.

[0103] The deep learning prediction model was trained by collecting historical peeling operation data (no less than 500 batches), including parameters such as rotation speed, wind speed, temperature and humidity, time, and corresponding peeling rate, damage rate, and dander residue rate. The first 3 / 4 of the data was used as the training set to train the LSTM deep learning model and optimize the model parameters to keep the prediction error within 5%. The remaining 1 / 4 of the data was used as the test set to verify the accuracy of the model.

[0104] The S101-S102 stage prediction and control includes: At the beginning of the operation, the processing effect of different combinations of rotation speed and wind speed is simulated through a digital twin model to select the optimal initial parameters, control the start of the drive motor and blower, and execute steps S101 and S102; During operation, the digital twin model synchronizes the equipment parameters and material status in the physical world in real time, inputs the current parameters (rotation speed, wind speed, temperature and humidity, current peeling rate) into the LSTM prediction model, and predicts the peeling rate, damage rate, and dander residue rate for the next 5 minutes; If the predicted damage rate is higher than the preset threshold (e.g., 5%), the control device reduces the rotation speed in advance (e.g., by 10%), and at the same time fine-tunes the wind speed to avoid further damage; If the predicted peeling rate is lower than the preset threshold (e.g., 85%), the rotation speed and wind speed are increased in advance to ensure timely achievement of the target; If the predicted dander residue rate is too high, the air supply direction and wind speed are adjusted in advance to enhance the separation effect. In the S103 stage, the LSTM model predicts the peeling effect in real time. When the predicted peeling rate reaches the target value (e.g., 90%-95%) and the damage rate and dander residue rate are below the threshold, the control device preemptively stops the drive motor and blower, eliminating the need to wait for the preset working time and achieving precise shutdown. If the prediction fails to meet the target, the working time is appropriately extended, and parameters are adjusted to ensure the processing effect. After each batch of work is completed, the actual processing data is fed back to the deep learning model and digital twin model to optimize model parameters and mapping accuracy, thereby improving the accuracy of prediction and control.

[0105] In some embodiments, to address the issues of multiple peeling machines operating simultaneously, limited data volume per machine, and insufficient parameter optimization accuracy, a federated learning algorithm is introduced. This enables collaborative training and parameter sharing among multiple machines, improving the optimization accuracy of peeling parameters for a single machine while protecting the privacy data of each machine (such as material type and processing technology). Its innovation lies in the fact that it eliminates the need to aggregate the raw data from each machine (avoiding privacy leaks). Through federated learning, it achieves "data without leaving the factory, model co-training," allowing a single machine to learn from the processing experience of multiple machines, adapting to the peeling needs of more varieties and scenarios of American ginseng, and solving the problems of limited data volume and poor model generalization ability of a single machine.

[0106] The federated node deployment uses the control device of each peeling device as a local node for federated learning, and sets up a central server as a federated aggregation node. Local model training involves each local node training a local machine learning model (such as a decision tree model) based on its own historical job data and outputting local model parameters. Global model aggregation involves the central server collecting the model parameters of each local node, aggregating them using a weighted average method to obtain the globally optimal model parameters, and feeding them back to each local node. Collaborative optimization involves each local node updating its own parameter recommendation rules based on the global model parameters, thereby improving the parameter optimization accuracy of a single device. Privacy protection involves using differential privacy technology to encrypt the local model parameters to prevent the leakage of raw data.

[0107] The deployment of the federated learning system includes: (1) Node setup: Assuming there are N peeling machines, the control device of each machine is used as a local node to deploy a local machine learning model (input: material type, weight of a single ginseng, initial humidity; output: optimal rotation speed, wind speed, working time); and a central server is set up as an aggregation node to be responsible for parameter aggregation and distribution. (2) Privacy protection configuration: Differential privacy modules are deployed in each local node to add small noise to the parameters of the trained local model, so as to avoid inferring the original data (such as material type, processing technology) through the model parameters and protect data privacy.

[0108] The global model training includes: (1) Local training: Each local node trains its local model based on its own historical operation data (such as the processing data of ginseng variety A in equipment 1 and the processing data of ginseng variety B in equipment 2), obtains local model parameters, and uploads them to the central server; (2) Global aggregation: The central server collects the model parameters of all local nodes, and aggregates them according to the data volume weight of each node (the more data, the greater the weight) using the weighted average method to obtain the global model parameters, ensuring that the global model can take into account the processing experience of each node; (3) Model update: The central server feeds back the global model parameters to each local node, and each node updates its own local model according to the global parameters to complete one collaborative training; Global aggregation is performed once a week to continuously optimize the model accuracy.

[0109] The following steps are performed in stages S101-S103: (1) Before the operation, each local node collects information on the variety of the American ginseng to be peeled, the weight of a single American ginseng, and the initial humidity, and inputs it into the updated local model (integrating global experience) to obtain the recommended rotation speed (S101), wind speed (S102), and working time (S103) parameters; (2) In stage S101: the control device controls the drive motor to start according to the recommended rotation speed, drives the roller to rotate, and achieves preliminary peeling; if the local equipment detects that the peeling effect is not good (such as the damage rate is too high), the rotation speed is adjusted in real time, and the adjusted parameters and effect data are used as local data for the next local model training. (3) In stage S102: the blower is controlled to start according to the recommended wind speed to achieve peeling; if the peeling residue rate is too high, the wind speed is dynamically adjusted in combination with the wind speed adjustment rules recommended by the global model, and the adjustment data is recorded to supplement the local dataset. (4) S103 stage: Control the equipment to stop according to the recommended working time; if the actual peeling effect meets the standard, record the parameter combination as valid data; if it does not meet the standard, adjust the working time and feed the data back to the local model to optimize the next parameter recommendation.

[0110] As each node continues to accumulate data and participate in global aggregation, the generalization ability of the global model is constantly improved, and the parameter recommendation accuracy of each local node is also gradually improved. Even if a single device encounters an unprocessed American ginseng variety, it can output the optimal parameters based on global experience, thereby achieving multi-device collaborative optimization.

[0111] In some embodiments, to address the problem that traditional peeling rate detection can only obtain overall progress and cannot identify local unpeeled areas (such as ginseng folds or roots), a deep learning semantic segmentation algorithm is introduced to accurately identify local peeled areas on the surface of American ginseng. This allows for dynamic adjustment of the drum speed and blower speed, specifically addressing the problem of insufficient local peeling. Its innovation lies in overcoming the limitations of "overall peeling rate" and achieving "precise detection of local areas + targeted parameter control," avoiding uneven product quality caused by overall peeling rate meeting standards but local unpeeled areas, while also reducing ginseng damage and energy waste due to excessive peeling.

[0112] The semantic segmentation model employs the U-Net deep learning semantic segmentation model to accurately segment the "unpeeled area, peeled area, and damaged area" on the surface of American ginseng, quantifying the area proportion of each region. Local feature analysis uses the segmentation results output by the model to analyze the location (such as wrinkles, roots) and area of ​​local unpeeled areas, determining the reasons for not peeling (such as insufficient friction or lack of airflow penetration). Targeted control strategies adjust the rotation speed (to enhance friction) or wind speed (to enhance airflow penetration), or adjust the matching relationship between the two, based on the reasons for local unpeeling, to achieve precise optimization of local peeling. Real-time feedback involves performing semantic segmentation every 10-15 seconds to update the local peeling status in real time and dynamically adjust parameters.

[0113] The semantic segmentation model training includes: (1) Dataset construction: collect surface images of different ginseng varieties at different peeling stages (no less than 1,000 images), manually label "unpeeled areas, peeled areas, and damaged areas" to construct a semantic segmentation dataset; (2) Model training: use the U-Net semantic segmentation model, with the labeled images as the training set, optimize the model parameters to achieve a segmentation accuracy of over 95%, and accurately identify local unpeeled areas with an area ratio of ≥1%; (3) Model deployment: integrate the trained semantic segmentation model into the image processing module of the control device, link it with the high-definition image sensor in the drum, and process the collected ginseng surface images in real time.

[0114] S101 stage local control: (1) Image acquisition and segmentation: The image sensor takes an image of the surface of the American ginseng every 10 seconds and transmits it to the image processing module. The semantic segmentation model is used to segment the unpeeled area, peeled area and damaged area, and calculate the area ratio of each area; (2) Judgment of the cause of unpeeled skin: If the unpeeled area is mainly concentrated in the folds of the ginseng body and the root (caused by insufficient friction), and the proportion of the damaged area is lower than the threshold (such as 3%), it is judged as "insufficient friction"; if the unpeeled area is evenly distributed and the proportion of the damaged area is high, it is judged as "excessive speed leading to local excessive friction and uneven overall friction"; (3) Targeted adjustment: ① Insufficient friction: Control the drive motor to increase the speed (such as increasing by 10%-15%) to enhance the tumbling and friction of the American ginseng in the drum, and focus on solving the problem of local unpeeled skin; ② Uneven friction: Appropriately reduce the speed (such as reducing by 5%-10%), and at the same time extend the tumbling time to make the friction of each part of the ginseng body uniform and avoid local excessive damage.

[0115] S102 stage local control: (1) Combined segmentation results analysis: If semantic segmentation shows that there are many danders attached to the surface of the unpeeled area (caused by airflow not penetrating), it is judged as "dander adhesion, affecting local peeling"; (2) Targeted adjustment: ① Increase the blower speed (e.g., increase by 10%), and at the same time adjust the air supply direction so that the airflow penetrates the folds and roots of the body obliquely, blowing away the attached danders, creating conditions for local peeling; ② If the dander carry-out rate is low, simultaneously increase the angle between the air supply direction and the roller axis to enhance airflow penetration, ensure timely separation of local danders, and promote local peeling.

[0116] After each parameter adjustment, the semantic segmentation model detects the local peeling status in real time. If the proportion of the area of ​​the unpeeled area drops below the threshold (e.g., 1%), the rotation speed and wind speed are restored to the base speed. If there are still areas that have not peeled, the parameters are fine-tuned until the overall peeling is uniform and there are no local residues, ensuring consistent product quality.

[0117] Please see Figure 7 , Figure 7 This is a schematic block diagram of the control device provided in an embodiment of this application. The control device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0118] The storage medium may store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any method of peeling ginseng.

[0119] The processor provides computing and control capabilities to support the operation of the entire control unit.

[0120] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When executed by a processor, the computer program can enable the processor to perform any method of peeling ginseng.

[0121] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. The specific control device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0122] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0123] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: The drive motor is started to make the drum rotate at a predetermined speed, which causes the American ginseng located inside the drum to rub against the wire mesh on the inner wall of the drum, thus achieving the initial peeling of the American ginseng. During the rotation of the drum, the blower is started and airflow at a predetermined speed is delivered to the air inlet through the air outlet. The airflow enters the drum through the air inlet and blows the skin flakes generated during the initial peeling process off the surface of the American ginseng and carries them out of the drum, thus completing the separation of skin flakes from the American ginseng. The drive motor and the blower are controlled to run for a predetermined working time and then stop.

[0124] In some embodiments, controlling the start of the drive motor to rotate the drum at a predetermined speed, causing the American ginseng located inside the drum to rub against the wire mesh on the inner circumferential wall of the drum, thereby achieving preliminary peeling of the American ginseng, includes: acquiring the variety information and individual weight information of the American ginseng to be peeled; determining a predetermined speed matching the variety information and individual weight information of the current American ginseng to be peeled based on the pre-stored correspondence between the American ginseng variety, individual weight, and predetermined speed; and controlling the drive motor to rotate at the matched predetermined speed.

[0125] In some embodiments, the step of delivering an airflow at a predetermined speed to the air inlet through the air outlet, the airflow entering the drum through the air inlet, and blowing away the skin flakes generated during the initial peeling process from the surface of the American ginseng and carrying them out of the drum, thereby completing the separation of skin flakes from the American ginseng, includes: real-time detection of the skin flake concentration inside the drum by a dust sensor installed inside the drum; adjusting the airflow speed of the blower according to the real-time skin flake concentration, wherein when the skin flake concentration is higher than a preset concentration threshold, the blower speed is increased to a first adjusted speed, and when the skin flake concentration is lower than or equal to the preset concentration threshold, the blower speed is decreased to a second adjusted speed, wherein the first adjusted speed is greater than the second adjusted speed.

[0126] In some embodiments, the step of supplying airflow at a predetermined speed to the air inlet through the air outlet, and the airflow entering the drum through the air inlet to blow away the skin flakes generated during the initial peeling process from the surface of the American ginseng and carry them out of the drum, thus completing the separation of skin flakes from the American ginseng, includes: adjusting the airflow direction of the blower so that the airflow enters from the air inlet along the axial direction of the drum and then obliquely passes through the gaps between the American ginseng inside the drum; the airflow carrying the skin flakes passes through the gaps in the wire mesh on the inner peripheral wall of the drum and is discharged from the outer peripheral wall of the drum to the hollow part of the support; the skin flake carry-out rate is detected by a skin flake sensor installed on the outer peripheral wall of the drum; when the skin flake carry-out rate is lower than a preset carry-out rate threshold, the angle between the airflow direction and the axial direction of the drum is increased; when the skin flake carry-out rate is higher than or equal to the preset carry-out rate threshold, the angle is decreased.

[0127] In some embodiments, the method further includes: collecting historical peeling operation data, the historical data including variety information of the ginseng to be peeled, individual ginseng weight information, initial humidity information, actual rotation speed of the drive motor, actual wind speed of the blower, and corresponding peeling effect score; the peeling effect score is comprehensively evaluated based on the ginseng damage rate and skin residue rate; training a machine learning model using the historical data, the machine learning model taking the variety of the ginseng to be peeled, individual ginseng weight, and initial humidity as input, and outputting a recommended predetermined rotation speed and predetermined wind speed; when the current ginseng to be peeled has its variety, individual ginseng weight, and initial humidity information obtained, inputting the current ginseng to be peeled variety, individual ginseng weight, and initial humidity information into the machine learning model to obtain the recommended predetermined rotation speed and wind speed; controlling the drive motor to rotate at the recommended rotation speed, and controlling the blower to deliver air at the recommended wind speed.

[0128] In some embodiments, the method further includes: capturing a surface image of the American ginseng in real time using an image sensor installed inside the drum; performing image recognition processing on the surface image to extract features of the peeled areas, calculating the proportion of the peeled areas to the total surface area, and obtaining the peeling rate; the features include color and texture changes; when the peeling rate is lower than a first preset threshold, increasing the speed of the drive motor to a first adjustment speed higher than a predetermined speed, and increasing the speed of the blower to a first adjustment speed higher than a predetermined speed; when the peeling rate is higher than a second preset threshold, decreasing the speed of the drive motor to a second adjustment speed lower than a predetermined speed, and decreasing the speed of the blower to a second adjustment speed lower than a predetermined speed; wherein the first preset threshold is less than the second preset threshold.

[0129] In some embodiments, the method further includes: detecting the vibration value of the drum in real time using a vibration sensor on the support, and detecting the current value in real time using a current sensor on the drive motor; inputting the vibration value and current value into a pre-stored anomaly judgment model; the anomaly judgment model is pre-trained using vibration or current characteristics when the drum is stuck or the motor is overloaded; if the anomaly judgment model determines that the vibration value exceeds a preset vibration threshold and the current value exceeds a preset current threshold, then it is determined that there is a jam or overload; controlling the drive motor to reduce its speed to a safe speed, stopping the blower from blowing air, and sending a fault warning signal to the user terminal.

[0130] In some embodiments, the method further includes: acquiring real-time temperature and humidity inside the drum; adjusting the blower's air supply temperature and humidity according to pre-stored adjustment rules for temperature, humidity, and air supply parameters; the adjustment rules include reducing the air supply temperature when the temperature is higher than 30°C or the humidity is higher than 70%; and reducing the air supply humidity when the humidity is higher than 75%.

[0131] In some embodiments, the method further includes: recording complete parameters for each peeling operation, including the quantity and variety of American ginseng to be peeled, the average weight of a single American ginseng, the rotational speed curve of the drive motor, the wind speed curve of the blower, the peeling time, and the final peeling effect score; performing statistical analysis on the parameters to calculate the correlation between each parameter and the peeling effect score; the correlation includes the correlation between rotational speed and damage rate and the correlation between wind speed and dander removal rate; identifying parameters with correlations higher than a preset threshold as key influencing parameters; the key influencing parameters include average rotational speed, average wind speed, and peeling time; optimizing pre-stored predetermined rotational speed and wind speed setting rules based on the statistical results of the key parameters; and determining the predetermined parameters according to the optimized rules when performing a peeling operation again.

[0132] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement the steps of the ginseng peeling method provided in the above embodiments of this application.

[0133] The computer-readable storage medium can be an internal storage unit of the control device described in the foregoing embodiments, such as the hard disk or memory of the control device. Alternatively, the computer-readable storage medium can be an external storage device of the control device, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the control device.

[0134] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for peeling American ginseng, characterized in that, A control device for peeling equipment used in American ginseng. The peeling equipment also includes a support frame, a drive motor, a roller, and a blower. The support frame has a hollow section. The drive motor is fixedly mounted on the support frame. One end of the roller is connected to the output shaft of the drive motor, and the other end of the roller is rotatably connected to the support frame. The roller is located inside the hollow section. The rotation center axis of the roller is parallel to the horizontal direction. The inner circumferential wall of the roller is covered with a wire mesh. An air inlet is provided at the end of the roller facing away from the drive motor. The blower is fixedly mounted on the support frame, and the air outlet of the blower faces the air inlet. The control device is fixedly mounted on the bracket and electrically connected to the drive motor and the blower; the method includes: The drive motor is started to make the drum rotate at a predetermined speed, which causes the American ginseng located inside the drum to rub against the wire mesh on the inner wall of the drum, thus achieving the initial peeling of the American ginseng. During the rotation of the drum, the blower is started and airflow at a predetermined speed is delivered to the air inlet through the air outlet. The airflow enters the drum through the air inlet and blows the skin flakes generated during the initial peeling process off the surface of the American ginseng and carries them out of the drum, thus completing the separation of skin flakes from the American ginseng. The drive motor and the blower are controlled to run for a predetermined working time and then stop.

2. The method according to claim 1, characterized in that, The control of starting the drive motor causes the drum to rotate at a predetermined speed, which in turn causes the American ginseng located inside the drum to rub against the wire mesh on the inner wall of the drum, thus achieving the initial peeling of the American ginseng, including: Obtain the variety information and individual weight information of the American ginseng to be peeled; Based on the pre-stored correspondence between American ginseng varieties, individual American ginseng weights and predetermined rotation speeds, a predetermined rotation speed matching the variety information and individual weight information of the current American ginseng to be peeled is determined. Control the drive motor to rotate at the predetermined speed.

3. The method according to claim 1, characterized in that, The process involves supplying airflow at a predetermined speed through the air inlet, which then enters the drum, blowing away and carrying away the dander generated during the initial peeling process from the surface of the ginseng, thus separating the dander from the ginseng. This process includes: The concentration of dander inside the drum is detected in real time by a dust sensor installed inside the drum. The blower speed is adjusted according to the real-time dandruff concentration. When the dandruff concentration is higher than a preset concentration threshold, the blower speed is increased to a first adjusted speed. When the dandruff concentration is lower than or equal to the preset concentration threshold, the blower speed is decreased to a second adjusted speed. The first adjusted speed is greater than the second adjusted speed.

4. The method according to claim 1, characterized in that, The process involves supplying airflow at a predetermined speed through the air inlet, which then enters the drum, blowing away and carrying away the dander generated during the initial peeling process from the surface of the ginseng, thus separating the dander from the ginseng. This process includes: Adjust the airflow direction of the blower so that the airflow enters from the air inlet along the axial direction of the drum and then passes obliquely through the gaps between the American ginseng roots inside the drum. The airflow carries the dander through the gaps in the wire mesh on the inner peripheral wall of the roller and is discharged from the outer peripheral wall of the roller to the hollow part of the support. The dander carry-out rate is detected by a dander sensor installed on the outer peripheral wall of the roller. When the dander carry-out rate is lower than a preset carry-out rate threshold, the angle between the air supply direction and the roller axis is increased. When the dander carry-out rate is higher than or equal to the preset carry-out rate threshold, the angle is decreased.

5. The method according to claim 1, characterized in that, The method further includes: Collect historical peeling operation data, including information on the variety of American ginseng to be peeled, the weight of a single ginseng, the initial humidity, the actual rotation speed of the drive motor, the actual wind speed of the blower, and the corresponding peeling effect score; the peeling effect score is comprehensively evaluated based on the ginseng damage rate and the skin residue rate; The historical data is used to train a machine learning model. The input of the machine learning model is the variety of the American ginseng to be peeled, the weight of a single American ginseng, and the initial humidity. The output is the recommended predetermined rotation speed and predetermined wind speed. When the variety, weight of a single ginseng, and initial humidity information of the American ginseng to be peeled are obtained, the variety, weight of a single ginseng, and initial humidity information of the American ginseng to be peeled are input into the machine learning model to obtain the recommended predetermined rotation speed and wind speed. The drive motor is controlled to rotate at the recommended speed, and the blower is controlled to deliver air at the recommended wind speed.

6. The method according to claim 1, characterized in that, The method further includes: The surface image of the American ginseng is captured in real time by an image sensor installed inside the drum; The surface image is subjected to image recognition processing to extract features of the peeled areas, and the proportion of the peeled areas to the total surface area is calculated to obtain the peeling rate; the features include color and texture changes; When the peeling rate is lower than the first preset threshold, the drive motor speed is increased to a first adjusted speed higher than the predetermined speed, and the blower speed is increased to a first adjusted speed higher than the predetermined speed. When the peeling rate is higher than the second preset threshold, the drive motor speed is reduced to a second adjustment speed lower than the predetermined speed, and the blower speed is reduced to a second adjustment speed lower than the predetermined speed; wherein the first preset threshold is less than the second preset threshold.

7. The method according to claim 1, characterized in that, The method further includes: The vibration value of the roller is detected in real time by a vibration sensor on the bracket, and the current value is detected in real time by a current sensor on the drive motor. The vibration and current values ​​are input into a pre-stored anomaly detection model; the anomaly detection model is pre-trained using vibration or current characteristics under conditions of drum jamming and motor overload. If the anomaly detection model determines that the vibration value exceeds the preset vibration threshold and the current value exceeds the preset current threshold, then it is determined that there is jamming or overload. The system controls the drive motor to reduce its speed to a safe speed, stops the blower from blowing air, and sends a fault warning signal to the user terminal.

8. The method according to claim 1, characterized in that, The method further includes: To obtain real-time temperature and humidity inside the drum; According to the pre-stored adjustment rules for temperature, humidity and air supply parameters, adjust the air supply temperature and humidity of the blower; the adjustment rules include reducing the air supply temperature when the temperature is higher than 30℃ or the humidity is higher than 70%; and reducing the air supply humidity when the humidity is higher than 75%.

9. The method according to claim 1, characterized in that, The method further includes: Record complete parameters for each peeling operation, including the number and variety of American ginseng to be peeled, the average weight of a single American ginseng, the speed curve of the drive motor, the wind speed curve of the blower, the peeling time, and the final peeling effect score. Statistical analysis was performed on the parameters to calculate the correlation between each parameter and the peeling effect score; the correlation included the correlation between rotation speed and damage rate and the correlation between wind speed and dander removal rate. Parameters with a correlation higher than a preset threshold are identified as key influencing parameters; key influencing parameters include average rotational speed, average wind speed, and peeling time. Based on the statistical results of key parameters, optimize the pre-stored preset rotation speed and wind speed setting rules; When performing the peeling operation again, the predetermined parameters are determined according to the optimized rules.

10. A peeling device for American ginseng, characterized in that, include: The support has a hollow section; The drive motor is fixedly mounted on the bracket. A roller, one end of which is connected to the output shaft of the drive motor, and the other end of which is rotatably connected to the bracket. The roller is located inside the hollow part. The rotation center axis of the roller is arranged parallel to the horizontal direction. The inner peripheral wall of the roller is covered with wire mesh. An air inlet is provided at the end of the roller facing away from the drive motor. A blower is fixedly mounted on the bracket, with the blower's air outlet facing the air inlet; A control device is fixedly mounted on the bracket and electrically connected to the drive motor and the blower, for implementing the method as described in any one of claims 1 to 9.

Images