On-line detection and regulation system and method for peeling residual quantity of corn harvester

Abstract

The invention belongs to the technical field of harvesting equipment control, and provides a peeling residual quantity on-line detection and regulation system and method for a corn harvester. Calculating the residual skin area, the ear body area and the residual skin coverage rate of the corn target in each frame of image; calculating the residual amount of the residual husks of the corn targets in one detection period according to the residual husk coverage rates of all the corn targets in one detection period; according to the residual amount of the residual husks of the corn target in one detection period, calculating the corn husking rate; calculating a target peeling rate according to the set target residual amount of the residual peels; according to the deviation between the corn husking rate and the target husking rate, a roller set adjusting driving motor is controlled to adjust the gap between the upper-layer husking roller set and the lower-layer husking roller set, and / or a conveying adjusting driving motor is controlled to adjust the position of a feeding point from a conveying plate to the upper-layer husking roller set, and / or the position of a feeding point from the conveying plate to the lower-layer husking roller set is adjusted. And the roller set driving motor is controlled to adjust the rotating speed of the upper-layer peeling roller set and the rotating speed of the lower-layer peeling roller set. According to the invention, continuous self-adaptive regulation and control of the operation process can be realized.

Description

Technical Field

[0001] This invention relates to the field of harvesting equipment control technology, specifically to an online detection and control system and method for the amount of husk residue in a corn harvester. Background Technology

[0002] Corn harvesters are typically equipped with peeling devices to remove the husks from corn ears during field harvesting. Existing peeling devices often employ a structure with multiple peeling rollers and a feeding mechanism, relying on the squeezing, friction, and relative motion between the rollers to peel the corn husks. The peeling effect is closely related to factors such as the roller speed, the gap between the rollers, the feeding state, and the duration (dwell time / contact area) of the corn ear in the peeling zone. To adapt to different varieties, moisture contents, and operating conditions, existing equipment generally offers adjustable roller speeds or roller gaps, but most still rely on manual settings based on experience, requiring intermittent observation or sampling checks by the driver before manual adjustments. In corn seed production plots, to balance the needs of kernel protection, ventilation and drying, storage, and disease control, a certain amount of husk must be retained on the corn ear to meet seed production process specifications. Because there are many varieties of corn used for seed production, small batches, and significant differences in plots and moisture content, the peeling conditions are more prone to change. If the existing peeling method, which is mainly based on experience-based parameter adjustment, is still used, it is often difficult to achieve the target residue control in a long-term and stable manner.

[0003] The existing technology has the following main problems and shortcomings: 1. Most existing peeling devices only improve the peeling effect from the structural level or achieve peeling operation by fixing parameters. During the operation, there is a lack of a real-time detection and quantitative evaluation mechanism for the amount of corn husk residue after peeling, and it is impossible to determine in time whether the peeling has reached the "target number of residual layers" requirement.

[0004] 2. Due to the lack of real-time feedback, drivers usually adjust the speed or clearance by stopping the car to sample and visually assess the amount of corn husks remaining. This method has a significant lag and is greatly affected by the operator's experience, which can easily lead to insufficient or excessive husk removal, resulting in poor consistency in the amount of husks remaining.

[0005] 3. Existing equipment can only adjust the speed of the peeling roller or the gap between the rollers, and lacks a control strategy that coordinates the speed of the peeling roller, the gap between the rollers, and the feeding mechanism (such as the area / time of action affected by the feeding baffle). It is difficult to meet the comprehensive requirements of "residual amount target" and "grain damage control" under different working conditions.

[0006] 4. When operating conditions change, it is difficult to maintain the amount of peeling residue within the target range. Insufficient peeling will result in too much residual peel, which will affect subsequent drying and storage. Excessive peeling may increase mechanical damage and appearance of the grains, reducing seed production quality and consistency. Summary of the Invention

[0007] The purpose of this invention is to solve the above-mentioned technical problems and provide an online detection and control system and method for the amount of husk residue in corn harvesters.

[0008] To achieve the above objectives, some embodiments of the present invention provide the following technical solutions: An online detection and control system for residual husks in corn harvesters: The corn harvester includes a conveying unit and a peeling mechanism unit: The peeling mechanism unit includes an upper peeling roller group and a lower peeling roller group arranged with a gap between them; both the upper and lower peeling roller groups are connected to roller group drive motors to drive the upper and lower peeling roller groups to rotate; the upper peeling roller group is connected to an upper peeling roller group support, and the upper peeling roller group support is connected to a roller group adjustment drive motor via a drive mechanism, and the roller group adjustment drive motor can adjust the gap between the upper and lower peeling roller groups by controlling the drive mechanism; The conveying unit includes a conveying plate that extends to the upper peeling roller group. The conveying plate is connected to an electric push rod, which is connected to a conveying adjustment drive motor. The conveying adjustment drive motor can adjust the position of the conveying plate to the feeding point of the upper peeling roller group by controlling the electric push rod. The online detection and control system includes: Vision unit: Located at the discharge end of the peeling mechanism unit, used to collect images of the peeled corn; Controller: Connected to the vision unit, the roller group drive motor, the roller group adjustment drive motor, and the conveyor adjustment drive motor, and configured as follows: The corn peeling rate was calculated based on the collected images of peeled corn. Based on the deviation between the corn peeling rate and the target peeling rate, the peeling mechanism unit and the conveying unit are controlled, including: The roller group adjustment drive motor adjusts the gap between the upper peeling roller group and the lower peeling roller group, and / or controls the conveying adjustment drive motor to adjust the position of the conveying plate to the feeding point of the upper peeling roller group, and / or controls the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

[0009] This application also proposes an online detection and control method for the peeling residue of corn harvesters, used in an online detection and control system for the peeling residue of corn harvesters, comprising the following steps: Set a detection box and select the corn target in each frame of the peeled corn image acquired by the vision unit; Calculate the first The residual area of ​​each corn target ear area ; According to the The residual area of ​​each corn target ear area Calculate the residual bark coverage rate: ,in, For the first Within the first control cycle, the first The skin coverage rate of each target, It is a positive number; Based on the residual husk coverage of all corn targets within a detection period, calculate the residual husk amount of corn targets within that detection period. ; Based on the residual amount of corn husks within a detection cycle Calculate the corn peeling rate : ; Based on the set target amount of residual skin Calculate the target peeling rate : ; Based on corn peeling rate Deviation from target peeling rate The control of the peeling mechanism unit and the conveying unit includes: controlling the roller group adjustment drive motor to adjust the gap between the upper peeling roller group and the lower peeling roller group, and / or controlling the conveying adjustment drive motor to adjust the position of the conveying plate to the feeding point of the upper peeling roller group, and / or controlling the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

[0010] In some embodiments of this application, the calculation of the first... The ear area of ​​a single corn target ,include: Judge the first Does each corn target have an ear-body mask, which is used to identify the pixel region of the corn ear body after peeling? If so, calculate the ear area: ; If not, the area of ​​the detection frame will be used as the area of ​​the ear:

[0011] in, These are the pixel values ​​for the residual skin mask, where 1 indicates residual skin and 0 indicates no residual skin. Let be the pixel coordinates, where The horizontal index of the pixel This is the vertical index of the pixel; For the first Area of ​​each detection frame For the first Each detection frame width, For the first The detection box height is 1, and the ROI represents the selected sub-region in the image acquired by the visual unit.

[0012] In some embodiments of this application, the residual amount of corn husks within a detection cycle is calculated. ,include:

[0013] in: This represents the total number of all corn targets within a detection period. For the first Confidence level of the corn target It is a positive number.

[0014] In some embodiments of this application, the amount of residual skin is also included. The filtering process includes:

[0015] in: This is the residual amount after filtering. The length of the sliding window. Indicates the order parameter; The calculation of peeling rate It also includes: After filtering Calculate the peeling rate : .

[0016] In some embodiments of this application, based on the corn peeling rate Deviation from target peeling rate The control of the peeling mechanism unit and the conveying unit includes: Error in calculating residual amount : Regarding residual amount error Normalization is performed to obtain the normalization error. ; This is due to the residual amount error from the previous cycle. This is the residual amount error for this cycle; Calculate the rate of change of error: For the rate of change of error Normalization is performed to obtain the rate of change of normalization error. ; Construct membership function rules for the proportional coefficient, integral coefficient, and differential coefficient. Based on these membership function rules, the normalized error is... Normalized error rate of change As input, the proportional parameter adjustment amount is calculated. Integral parameter adjustment amount Differential parameter adjustment amount .

[0017] In some embodiments of this application, the residual amount error is described. Normalization processing is performed, including:

[0018] The rate of change of error Normalization processing is performed, including:

[0019] in: This is the residual error ratio coefficient. This is the proportionality coefficient of the error change rate. To control the cycle.

[0020] In some embodiments of this application, it further includes: Calculate the deviation of the corn target size:

[0021] in: This is due to size deviation; The target size of corn after fusion within a cycle; This serves as a reference size for the corn target.

[0022]

[0023] For the first The target size for each corn cob. For the reference area of ​​the detection frame, For the detection frame width, For the height of the detection frame, For the first Confidence level of the corn target, It is a positive number; Calculate the feedforward term:

[0024] in: For feedforward compensation; Forward coefficients; Corn target supplementary comprehensive output:

[0025] Based on the comprehensive control quantity ultimately used for execution allocation The control roller group adjustment drive motor adjusts the gap between the upper peeling roller group and the lower peeling roller group, and / or controls the conveyor adjustment drive motor to adjust the position of the conveyor plate to the feeding point of the upper peeling roller group, and / or controls the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

[0026] In some embodiments of this application, the detection and control methods include: when To enhance peeling strength, it is necessary to increase the rotation speed, reduce the gap, and increase the extension of the conveyor plate; when Then adjust in the opposite direction.

[0027] In some embodiments of this application, the detection and control method further includes:

[0028] in: For speed increments; This is the speed distribution coefficient; This represents the maximum change in rotational speed over a single cycle.

[0029]

[0030] in: This represents the gap increment; the negative sign ensures... hour This reduces the gap; This is the gap allocation coefficient; This represents the maximum gap change in a single period.

[0031]

[0032] in: This represents the displacement increment of the conveyor plate. Assign coefficients to locations; This represents the maximum change in the conveyor plate size during a single cycle.

[0033] This application addresses the shortcomings of difficulty in online evaluation of peeling residue during the harvesting of seed-producing corn plots, lagging adjustment and reliance on experience, and large fluctuations in peeling quality due to difficulty in coordinated parameter control. It provides a precise control system and method for peeling residue in corn harvesters based on machine vision online detection, which realizes real-time quantitative detection and closed-loop control of peeling residue, so that the amount of corn husk residue on the corn ear after peeling can stably meet the target requirements, while taking into account both operation efficiency and kernel damage control.

[0034] It has the following technical effects: 1. This invention enables online quantitative evaluation of "residual husk content," addressing the lack of real-time detection and evaluation in existing technologies. Current husking devices often rely on fixed parameters or structural modifications, lacking real-time detection and quantitative evaluation of the "residual husk content" of the corn cob after husking, making it impossible to promptly determine whether the target residual layer requirement has been met. This invention deploys cameras and supplementary lighting along the conveying path before the corn is stored. It utilizes a visual model to output detection frames, residual husk masks, and confidence levels, and calculates indicators such as residual content / husking rate online. This transforms the "target residual layer number" into a calculable, recordable, and closed-loop controllable process quantity, fundamentally overcoming the shortcoming of "unmeasurability leading to uncontrollability."

[0035] 2. Upgrading from "spot checks during shutdown + experience-based parameter adjustment" to closed-loop control significantly reduces adjustment lag and reliance on manual intervention. Existing methods typically require stopping the machine for sampling, visual judgment, and then adjusting the speed or clearance, which has significant lag and is heavily influenced by operator experience, easily leading to insufficient or excessive peeling. This invention uses the target residual amount as a reference value. The controller performs closed-loop calculations based on the residual deviation and its rate of change and outputs the adjustment amount, achieving continuous adaptive adjustment during operation. This transforms "post-operation correction" into "online correction," reducing repeated manual adjustments and unnecessary shutdowns.

[0036] 3. Coordinated control of three actuators improves the stability of achieving targets under different working conditions while also controlling grain damage. Existing equipment often only allows for single adjustment of rotational speed or roller gap, lacking a strategy to coordinate the adjustment of rotational speed, gap, and feeding-related mechanisms (such as the area / time of action affected by the conveyor plate). This results in difficulty maintaining the residual amount stably within the target range when working conditions change, and makes it difficult to simultaneously achieve the residual target and grain damage control. This invention coordinates the control quantities to three types of electric actuators: peeling roller rotational speed, peeling roller gap, and the effective length / position of the feeding conveyor plate. This achieves linked control of "peeling intensity (rotational speed / gap) + action process (conveyor plate)," allowing the residual amount to more stably converge to the target requirements and reducing the risk of mechanical damage to the grains caused by over-peeling.

[0037] 4. Robust data links reduce the impact of false detection jitter on control, improving control stability and consistency. This invention fuses multi-target detection results and performs time filtering on residual quantities to form a closed-loop input; simultaneously, it sets up visual validity judgment and gating, freezing / degrading when visual detection is unreliable, avoiding false detections directly driving actuators and causing "over-adjustment-over-adjustment" oscillations, thereby improving the stability and consistency of residual quantity control.

[0038] 5. The fully electric execution and verifiable command conversion mechanism improves adjustment accuracy, response speed, and engineering feasibility. This invention employs a servo-driven closed-loop speed control, a servo + lead screw for gap adjustment, and a push rod for conveyor plate position adjustment. Before issuing the command, it completes the conversion and amplitude constraint of the target quantity to position / angle (or pulse), ensuring that the command can be implemented, repeated, and returned to zero. Compared with coarse mechanical adjustment, it has higher adjustment accuracy and controllability.

[0039] 6. Feedback verification + anomaly handling + safety degradation improve operational safety and system availability. This invention periodically reads the rotational speed, clearance, and conveyor plate position feedback and performs position verification. Simultaneously, it monitors anomalies such as overcurrent, overtorque, stall, limit switches, and communication interruptions. Upon an anomaly, it immediately executes safety actions such as stopping the rollers, opening the clearance, and retrieving the conveyor plate, and records alarms. If visual confidence is consistently insufficient or anomalies persist, it enters a degradation mode, significantly reducing the risks of jamming, overload, and miscontrol, thus improving overall machine safety and continuous operation capability. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic diagram of the corn harvester structure for this application; Figure 2 This is a first-view structural diagram of the peeling structural unit of this application; Figure 3 This is a schematic diagram of the second-view structure of the peeling structural unit in this application; Figure 4 This is a schematic diagram of the logic structure of the online detection and control system for corn harvester peeling residue in this application; Figure 5 This is a schematic diagram of the online detection and control method for corn harvester peeling residue in this application; Figure 6 This is a schematic diagram of the corn image data processing flow of this application; in: 1. Upper peeling roller assembly; 2. Lower peeling roller assembly; 3. Harvester body; 4. Support for the lower peeling roller assembly; 5. Upper peeling roller assembly support; 6. Screw actuator; 7. Guide rail; 8. Roller drive motor; 9. Roller group adjustment drive motor; 10. Conveyor plate; 11. Main conveyor structure; 12. Visual unit; 13. Granary; 14. Bracket; 15. Install rails; 16. Rear conveyor belt; 17. Electric linear actuator. Detailed Implementation

[0042] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0043] The prefixes such as "first" and "second" used in this application embodiment are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this application embodiment does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not constitute unnecessary restrictions due to the use of such prefixes. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0044] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B; the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0045] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0046] This application proposes an online detection and control system for residual husks in a corn harvester. It is used to control the husk mechanism of the corn harvester. By detecting images of the husked corn, the system determines the husk removal effect and controls the actions of the husk mechanism components based on the husk removal effect.

[0047] First, the structure of the corn harvester is explained, with reference to its mechanical structure. Figures 1 to 3 Logic unit composition reference Figure 4 .

[0048] The corn harvester includes a harvester body 3, which serves as the carrying unit for all components. The harvester body 3 is equipped with a peeling module, including a conveying unit and a peeling mechanism unit. The conveying unit is used to convey the harvested corn to be peeled to the peeling mechanism unit and to convey the peeled corn to the grain bin 13, including a conveying unit and a bin-loading unit. The peeling mechanism unit is used to peel the harvested corn.

[0049] In one specific embodiment, the peeling mechanism unit includes an upper peeling roller group 1 and a lower peeling roller group 2 arranged with a gap between them; both the upper peeling roller group 1 and the lower peeling roller group 2 are connected to a roller group drive motor 8 to drive the upper peeling roller group 1 and the lower peeling roller group 2 to rotate. Both the upper peeling roller group 1 and the lower peeling roller group 2 can be mounted on the harvester body 3.

[0050] The upper peeling roller group is connected to the upper peeling roller group support 5. The upper peeling roller group support 5 is connected to the roller group adjustment drive motor 9 via the drive mechanism. The roller group adjustment drive motor 9 can adjust the gap between the upper peeling roller group 4 and the lower peeling roller group 5 by controlling the drive mechanism.

[0051] Specifically, a lower peeling roller assembly support 4 is fixedly installed on the harvester body 3. The lower peeling roller assembly 2 includes three rollers arranged in parallel on the lower peeling roller assembly support 4. An upper peeling roller assembly support 5 is also included, with the upper peeling roller assembly 1 consisting of two rollers arranged in parallel on the upper peeling roller assembly support 5. The two rollers in the upper peeling roller assembly 1 are respectively aligned with the gap between any two rollers in the lower peeling roller assembly 2. The five rollers form an approximately V-shaped structure. Corn can be fed into the lower peeling roller assembly 2 through the gap between the two rollers in the upper peeling roller assembly 1, but will not fall out of the lower peeling roller assembly 2.

[0052] The upper peeling roller assembly support 5 is connected to the drive mechanism. For example, the drive mechanism can be a screw driver 6, with the screw of the screw driver 6 vertically positioned and connected to the upper peeling roller assembly support 5. When the roller assembly adjustment drive motor 9 is started, the position of the upper peeling roller assembly 1 can be adjusted vertically. To ensure the adjustment effect of the upper peeling roller assembly support 5, a guide rail 7 can be provided on the harvester body 3, restricting the upper peeling roller assembly support 5 within the adjustment range defined by the guide rail 7.

[0053] The conveying unit includes a conveyor plate 10 extending to the upper peeling roller assembly. The conveyor plate 10 is connected to an electric push rod 17, which is connected to a conveying adjustment drive motor. The conveying adjustment drive motor can adjust the position of the conveyor plate 10 to the feeding point of the upper peeling roller assembly by controlling the electric push rod 17. The electric push rod 17 has a position feedback output and is equipped with an origin / limit switch for zeroing and stroke protection.

[0054] For example, the conveyor plate 10 is movably mounted on the harvester body 3, and the harvester body 3 is provided with a mounting rail 15 for mounting the conveyor plate 10. The conveyor plate 10 can move along the range defined by the mounting rail 15. Preferably, the direction of the mounting rail 15 is parallel to the direction of the peeling roller assembly, so that the conveyor plate 10 can be adjusted in position parallel to the peeling roller assembly.

[0055] For example, it also includes a main conveyor structure 11, with a conveyor plate 10 serving as a transition structure between the main conveyor structure 11 and the peeling roller assembly. The main conveyor structure 11 may be a conveyor rail.

[0056] The factors affecting the peeling effect mainly include the following three aspects.

[0057] Firstly, the rotational speed of the upper peeling roller group 1 and the lower peeling roller group 2. The faster the rotational speed, the better the peeling effect; the slower the rotational speed, the worse the peeling effect.

[0058] Secondly, the gap between the upper peeling roller group 1 and the lower peeling roller group 2: the smaller the gap, the better the peeling effect; the larger the gap, the worse the peeling effect.

[0059] Thirdly, the position of the feed point from the conveyor plate 10 to the upper peeling roller assembly 1. This position determines the travel distance of the corn between the peeling roller assemblies. The position of the conveyor plate 10 can be adjusted... This can be achieved by changing the contact area or time of the corn on the peeling roller, thereby regulating the amount of peel residue. The further forward the feeding point is, the longer the travel distance and the better the peeling effect; the further backward the feeding point is, the shorter the travel distance and the worse the peeling effect.

[0060] For example, in some embodiments, the corn harvester further includes a grain bin 13 disposed at the rear end of the peeling mechanism unit, the grain bin 13 being used to store peeled corn; a support 14 is disposed at the grain bin 13, and a vision unit 12 is disposed on the support 14, facing the discharge end of the peeling mechanism unit. A rear conveyor belt 16 is also disposed between the grain bin 13 and the peeling mechanism unit.

[0061] The online detection and control system includes: vision unit 12 and controller.

[0062] The vision unit 12, located at the discharge end of the peeling mechanism unit, is used to acquire images of the peeled corn. The vision unit 12 includes a vision acquisition and protection module, and can employ image and video acquisition equipment such as an industrial camera unit. The lens's field of view covers the detection area through which the corn cob passes. To improve video acquisition, light sources and other structures can also be added. A supplementary lighting control unit is arranged around the camera to form stable illumination, reducing the impact of changes in natural field light, shadows, and reflections on the recognition results. Protective covers and transparent windows are installed outside the camera and light source, and dust scrapers or air curtain structures can be added to reduce errors or confidence level reduction caused by dust obstruction. (The purpose of selecting the detection area: Setting the detection area on the conveying path before entering the warehouse directly reflects the actual peeling effect of the peeling mechanism on the corn cob, and there is a clear causal relationship between the detection results and the adjustment of peeling parameters; at the same time, the corn cob posture is relatively controllable and the background is relatively stable at this location, which is conducive to building a robust vision model.) In some embodiments, a protection and cleaning unit is also included for cleaning the camera.

[0063] Controller: Connects to vision unit 12, roller group drive motor 8, roller group adjustment drive motor 9, and conveyor adjustment drive motor, and is configured as follows: Based on the collected images of peeled corn, i.e., the detection object is "corn ears that have been peeled and will enter grain collection bin 13", to avoid false detection of unpeeled samples, the corn peeling rate is calculated. Based on the deviation between the corn peeling rate and the target peeling rate, control instructions are generated, including: Roller gap adjustment command The roller group adjustment drive motor 9 adjusts the gap between the upper peeling roller group 1 and the lower peeling roller group 2, and / or, Target displacement adjustment command Control the conveyor adjustment drive motor to adjust the position of the conveyor plate 10 to the feeding point of the upper peeling roller group 1 (i.e., adjust the position of the conveyor plate 10 along the direction parallel to the peeling roller group), and / or, Speed ​​command The control roller drive motor adjusts the speed of the upper peeling roller group 1 and the lower peeling roller group 2 by 8, and performs closed-loop speed control; the speed encoder outputs the actual speed. Feedback is sent to the driver or controller.

[0064] In some embodiments, the detection and control system further includes an edge computing module (including an edge AI vision computing unit), a central control module (the controller belongs to the central control module), and an execution drive module (including a speed adjustment unit, a gap adjustment unit, and an effective length adjustment unit).

[0065] The edge AI vision computing unit is connected to the central controller via a communication bus using CAN. The AI ​​unit periodically sends visual recognition result data frames, which the controller parses and executes the control algorithm. The controller communicates with the servo drivers and actuator controllers via CAN.

[0066] The power module provides stable power to the camera, light source, AI unit, controller, and driver; the power side and control side are wired separately; the camera signal line and drive power line are routed separately, and EMI filtering and standard grounding are set up to ensure the reliability of visual and control signals.

[0067] After the detection and control system is started, the controller performs self-checks on the camera, light source, edge AI unit, communication link, servo driver (speed adjustment / gap adjustment), push rod control unit, encoder, displacement sensor, limit switch and emergency stop input. If a critical component is offline or an emergency stop is triggered, closed-loop operation is prohibited and an alarm is triggered.

[0068] The gap adjustment servo returns to zero, and the electric push rod 17 controls the position of the conveyor plate 10 to zero; confirm that the gap between the upper adjusting roller group 1 and the lower adjusting roller group 2 and the conveyor plate 10 are within the allowable range, and if necessary, open the gap between the upper adjusting roller group 1 and the lower adjusting roller group 2 to a safe value. Conveyor plate 10 returns to the safe position. .

[0069] This application also provides a method for online detection and control of corn harvester peeling residue, the process of which is as follows: Figure 5 .

[0070] First, initialize the parameters. Set the following parameters: Target residual amount This is used to indicate the number of layers of corn husk that are allowed to remain after the corn cob has been partially peeled. Control cycle This is used to indicate the control cycle of the gap between the upper adjusting roller group 1 and the lower adjusting roller group 2, the rotational speed of the upper adjusting roller group 1 and the lower adjusting roller group 2, and the position of the conveyor plate 10; Sliding filter window ; : Represents the error normalization coefficient (error scaling factor); used to normalize the residual deviation of physical units. Mapping to the normalized universe of discourse used in fuzzy control (usually Within a similar range, we obtain dimensionless values. , This makes it easy to process using triangular membership functions and fuzzy rules; : Represents the error change rate normalization coefficient (error change scaling factor), used to normalize the error change rate in units of "per second" or "per sampling period". Mapping this onto the normalized domain of fuzzy control yields a dimensionless... , ; Initial PID parameters Feedforward coefficient ; : Represents the speed distribution coefficient, which is used to allocate the total control quantity to the control component of speed adjustment. : Represents the gap allocation coefficient, which is used to allocate the total control quantity to the control component of the gap adjustment between the upper peeling roller group and the lower peeling roller group. : Indicates the position allocation coefficient, which represents the proportion of the total control quantity allocated to the control component for adjusting the position of the conveyor plate; used to allocate the comprehensive control quantity output by the fuzzy PID controller. The execution volume is allocated to the three categories according to a predetermined ratio, resulting in corresponding increments. Furthermore, by combining symbol conventions, a synergistic adjustment is achieved where "positive values ​​enhance peeling, and negative values ​​weaken peeling"; the three factors can be selected through empirical tuning or optimization algorithms based on the structural characteristics of the peeling mechanism and calibration test results, satisfying the following: and ; Reference size Feedforward coefficient ; Three execution quantity allocation coefficients ; Membership function parameters, fuzzy rule base, upper and lower limits and rate of change limits for each execution quantity; : Indicates the minimum target rotational speed; : Indicates the target maximum rotational speed; : Indicates the minimum value of the gap target; : Indicates the maximum value of the gap target; : Indicates the minimum value of the target location; : Indicates the maximum value of the target location; : Indicates the maximum speed adjustment step size; This indicates the maximum adjustment step size of the gap; Indicates the maximum adjustment step size of the position; : Represents the visual confidence threshold; degradation strategy parameters. Degradation strategy refers to the rules by which the system automatically switches from "intelligent closed-loop control" to "simplified, safe, and conservative control mode" when visual / execution problems occur. Therefore, "degradation strategy parameters" are the set of thresholds and backup parameters used to control this switching process and degradation state. Degradation strategy parameters are roughly divided into three categories: (1) Condition parameters that trigger degradation Trigger-type degradation strategy parameters are used to limit the situation of "long-term visual unreliability" or "abnormality of actuators persisting". Once the threshold is reached, the system is no longer allowed to continue working in full closed-loop mode, but instead switches to a conservative and safe degradation mode. (2) Backup control parameters adopted after degradation In degradation mode, the controller no longer adjusts in real time according to the visual residual amount of each cycle, but adopts preset fixed parameters or frozen parameters to ensure that the system can still maintain an acceptable peeling level when the vision is unreliable and ensure safety. (3) Condition parameters for recovery from degradation to normal closed loop By setting recovery condition parameters, frequent switching between "normal / degraded" can be avoided, ensuring that the system has enough time to stabilize.

[0071] Set the rotational speed of both the upper peeling roller group 1 and the lower peeling roller group 1 to their initial rotational values. The rotational speed of the upper peeling roller group 1 and the rotational speed of the lower peeling roller group 1, and the gap between the rollers are set to the initial gap value. The position of conveyor plate 10 is set to the initial position value. The camera and light source are activated, and the edge AI unit enters inference standby mode. After the entire machine begins harvesting tasks, it enters an online closed-loop cycle. Among them, with Figure 1 The direction shown is the reference. The distance between the rightmost end of the conveyor plate 10 and the leftmost end of the upper peeling roller group 1.

[0072] This application provides a method for online detection and control of corn harvester peeling residue, which also includes the following steps.

[0073] S1: Image acquisition and preprocessing steps.

[0074] An industrial camera continuously captures image frames at the detection position before the peeled corn is fed into the hopper. And synchronously record frame timestamps / sequence numbers. Preprocessing is performed, including but not limited to distortion correction, ROI cropping, and brightness normalization. The preprocessed image frames are obtained after preprocessing. .

[0075] S2: Detection and analysis of peeling residue steps.

[0076] S21: Set the detection box and select the corn target from each frame of the peeled corn images acquired by the vision unit. The purpose of this step is to select the corn target from the image frames and analyze the peeling effect of the target.

[0077] Specifically, the execution method is as follows.

[0078] Edge AI computing unit processes preprocessed image frames Running the YOLOv11-AgriPeelNet model, the output is: A set of corn ear detection frames , No. Corn husk residue masking , No. Confidence level of corn target and optional ear cover .

[0079] Among them, the corn cob detection box is a result of the machine vision model locating the "corn" target in the image: a rectangle is used to circle the position of a single corn cob in the image to indicate "there is a corn target in this area". The index number indicates "the first" "One detected corn target" The value of is determined by the algorithm. , The total number of corn targets can refer to the total number of all corn targets analyzed within a detection period, that is, the total number of corn targets in all images acquired within a detection period.

[0080] Confidence =Probability of target existence Category Probability , .

[0081] To detect the probability of target presence output by the network at the candidate box location, the value range is... This is used to characterize the confidence level of whether any target exists within the candidate box.

[0082] This represents the probability that the target within the candidate box belongs to the "corn" category, assuming the target exists. The value range is... This is used to characterize the confidence level at which the target is identified as a corn ear. Therefore, The overall confidence level of the detection box is used for threshold filtering and non-maximum suppression ranking.

[0083] Corn husk residue mask This is a pixel-level binary image used to accurately identify residual corn husks (i.e., incompletely removed corn husks) in corn ear images. Its value is either 0 or 1 and is only valid within the Region of Interest (ROI). If the pixel... It is a residue of corn husks. If the pixel is 1, then... Background elements, such as background scenery, corn cobs, etc. It is 0.

[0084] ear cover In corn cob image analysis tasks, a binary image mask is used to accurately identify the pixel region of the corn cob body (i.e., the edible kernel part after removing the husks / corn skin). Its value is 0 or 1, which can more accurately reflect the true shape of the cob.

[0085] For the The detected corn ear target, whose ear mask is denoted as... : If pixel It belongs to the corn cob itself. If the pixel is 1, then... Background elements, such as background surfaces, corn husks, etc. It is 0.

[0086] S22: Calculate the... The residual area of ​​each corn target ear area .

[0087] Among them, the area of ​​residual skin The area of ​​the corn cob remaining on the corn cob after peeling; the area of ​​the corn cob. This refers to the total visible area of ​​the corn ear after peeling.

[0088] In a specific embodiment:

[0089]

[0090] Calculate the first The ear area of ​​a single corn target The methods include: Judge the first Does each corn target have an ear-body mask, which is used to identify the pixel region of the corn ear body after peeling? If so, calculate the ear area: ; If not, the area of ​​the detection frame will be used as the area of ​​the ear: ; in, Pixel values ​​for the residual skin mask, where 1 represents residual skin and 0 represents non-residual skin; ROI represents the sub-region selected in the image acquired by the visual unit. These are the pixel coordinates, i.e., the discrete pixel coordinates of the current image region (or region of interest), where... The horizontal index of the pixel This is the vertical index of the pixel; , Generally, the value is a non-negative integer, and its range is determined by the image resolution or the size of the selected ROI. For example... , , The height of the current image region. The width of the current image region. For the first Area of ​​each detection frame For the first Each detection frame width, For the first The height of each detection frame.

[0091] S23: According to the The residual area of ​​each corn target ear area Calculate the residual bark coverage rate : ,in, For the first Within the first control cycle, the first The skin coverage rate of each target, If the denominator is a positive number, a very small positive number can be used to prevent the denominator from being zero.

[0092] The target peeling rate of a single corn cob is defined as:

[0093] in: This represents the single-target peeling rate; a higher rate indicates more thorough peeling.

[0094] The target size of the corn can be obtained by normalizing the bounding box area:

[0095] in: For the first The target size for each corn cob; The reference area for the detection frame can be set in advance. Used to normalize the size index of corn ears. This can be obtained through factory calibration or deployment calibration: A batch of representative corn ear samples are collected under fixed camera installation location and ROI conditions. The mean / median value is then taken based on the statistical analysis of the mask area segmented by the ear. .

[0096] S24: Based on the residual husk coverage of all corn targets within a detection period, calculate the residual husk amount of corn targets within a detection period. .

[0097] Calculate the residual amount of corn husks within a detection cycle. The residual husk coverage rate for each maize target is weighted and summed according to its confidence level, including:

[0098] in: This represents the total number of all corn targets within a detection period. For the first Confidence level of the corn target It is a positive number, and a very small positive number to prevent the denominator from being zero.

[0099] Assuming there are 3 corn targets within a detection period, It is 0.9. It is 0.6. It is 0.6. It is 0.2. It is 0.5. It is 0.3. Pick Then it can be calculated. .

[0100] The same applies to both size indicators:

[0101] in: This refers to the target size indicator for corn after the fusion of this cycle.

[0102] To reduce vibration, Perform moving average filtering (window size) ):

[0103] in: The filtered residual value serves as the control input. The length of the window; This is a residual amount from historical integration.

[0104] If Perform smoothing filtering to obtain The overall peeling rate output is:

[0105] Based on the set target amount of residual skin Calculate the target peeling rate :

[0106] in: For the overall confidence level of this period, when Time can be specified ; The target residue level (a pre-set calibration value, for example, the target is two layers of corn husks remaining).

[0107] The overall confidence level for this period is calculated by averaging or weighted averaging the confidence levels of all targets. For example, this embodiment uses the following method to calculate .

[0108]

[0109] The above-mentioned calculation results can provide feedback on whether the visual detection is effective. If it is ineffective, the calculation results are frozen and the re-acquired image is used for calculation.

[0110] S25: Based on corn peeling rate Deviation from target peeling rate The control of the peeling mechanism unit and the conveying unit includes: controlling the roller group adjustment drive motor to adjust the gap between the upper peeling roller group and the lower peeling roller group, and / or controlling the conveying adjustment drive motor to adjust the position of the conveying plate to the feeding point of the upper peeling roller group, and / or controlling the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

[0111] In some embodiments of this application, based on the corn peeling rate Deviation from target peeling rate Controlling the peeling mechanism unit and the conveying unit includes the following steps.

[0112] Error in calculating residual amount : Regarding residual amount error Normalization is performed to obtain the normalization error. ; This is due to the residual amount error from the previous cycle. This is the residual amount error for this cycle.

[0113] For residual error Normalization is performed to obtain the normalized residual error. ,include:

[0114] Calculate the rate of change of error: For the rate of change of error Normalization is performed to obtain the rate of change of normalization error. .

[0115] For the rate of change of error Normalization is performed to obtain the rate of change of normalization error. ,include:

[0116] in: This is the residual error ratio coefficient. This is the proportionality coefficient of the error change rate. To control the cycle.

[0117] The input to the online detection and control system for residual husk content in corn harvesters is the residual husk content error. (Normalized residual error) Error change rate (Normalized error rate of change) ); the output is These refer to the control values ​​of the proportional, integral, and derivative parameters. A modular PID control method is used for control.

[0118] Define the membership function.

[0119] For any input , No. The triangular membership function is defined as follows:

[0120] in: Membership degree; These are the parameters for the left endpoint, vertex, and right endpoint of the triangle. Corresponding language variable set .

[0121] In this embodiment, the triangular membership function parameters By first determining the normalized domain of discourse, and then evenly distributing the central points of the language set on the domain of discourse... Then, the left and right endpoints are determined by the adjacent center points. The selection method is as follows: the boundary language terms are processed using truncated triangles; the triangle parameters at the output end are constructed in the same way after the output domain is given, and fine-tuned through simulation and field calibration to ensure the coverage and control performance of fuzzy inference.

[0122] In fuzzy inference, this is a utility function used to transform numerical variables into "linguistic variable membership degrees." It's used to calculate the membership degrees of inputs / outputs, which are the activation strengths of rules. The source is determined by output synthesis and defuzzification. The output value of the PID controller affects the PID parameters and actuator commands.

[0123] For output variables Set the same number of membership functions for the universe of discourse, and use This indicates the output membership degree.

[0124] If input belong ,and belong ,but belong , belong , belong ; in, Enter a language label for the residual error. Enter a language label for the rate of change of error. Output language labels for the scaling parameters. Output language labels for the integral parameters. Output language labels for the differential parameters.

[0125] No. The activation strength of the fuzzy control rule is:

[0126] in: For rule activation degree; For the error in the language set Membership degree; For the rate of change in the language set The degree of membership. and It is calculated based on the input quantity and the triangular membership function rule.

[0127] for The output fuzzy set is synthesized using maximization.

[0128] in: The membership degree of the synthesized output; To output language variables The corresponding membership function.

[0129] for The output fuzzy set is synthesized using maximization.

[0130] in: The membership degree of the synthesized output; To output language variables The corresponding membership function.

[0131] for The output fuzzy set is synthesized using maximization.

[0132] in: The membership degree of the synthesized output; To output language variables The corresponding membership function.

[0133] According to the obtained , , ,right , , Defuzzification is performed to convert the parameters into precise adjustment values, ensuring that the PID parameters are within a reasonable physical range.

[0134] The synthesized fuzzy output set 、 、 This is converted into a definite real number, which is then used to update the parameters.

[0135]

[0136]

[0137]

[0138] in, Output the numerical value of the scale parameter after defuzzification; Output the numerical values ​​of the integral parameters after defuzzification; Output the numerical values ​​of the differential parameters after defuzzification; The output universe of discourse is the variable; integration is performed on the output universe of discourse.

[0139] PID parameters updated online and limited:

[0140] in: This is the updated proportional coefficient for this period; This is the proportional coefficient for the previous period; This refers to the allowable range of the proportionality coefficient; This is the amplitude limiting function.

[0141]

[0142] in: The integral coefficients are the ones updated for this period; This is the proportional coefficient of the previous period. Its scope.

[0143]

[0144] in: These are the differential coefficients updated in this period; This is the proportional coefficient of the previous period. Its scope.

[0145] PID output is defined as the overall regulation quantity:

[0146] in: It is a comprehensive control quantity (which can be dimensionless or normalized); its positive or negative sign corresponds to the direction of "strengthening / weakening peeling"; This is an approximation of the rate of change of error.

[0147] In some embodiments of this application, a corn target size feedforward compensation step is also included.

[0148] Corn target size deviation:

[0149] in: This is due to size deviation; This refers to the target size indicator for corn after the fusion of this cycle; This is a reference size for the corn target.

[0150] Calculate the feedforward term:

[0151] in: For feedforward compensation; is the feedforward coefficient.

[0152] Corn target compensation comprehensive output:

[0153] in: This is the final comprehensive control quantity used for allocation of execution quantities.

[0154] In some embodiments of this application, the following step is also included: based on the comprehensive control quantity ultimately used for execution quantity allocation. The control roller group adjustment drive motor adjusts the gap between the upper peeling roller group and the lower peeling roller group, and / or controls the conveyor adjustment drive motor to adjust the position of the conveyor plate to the feeding point of the upper peeling roller group, and / or controls the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

[0155] Specifically, it includes the following steps.

[0156] when When there is a tendency for excessive residue (insufficient peeling), the peeling intensity needs to be increased: increase the rotation speed, reduce the gap, and increase the extension of the conveyor plate; when Then adjust in the opposite direction.

[0157] The methods for distributing and adjusting the control components of speed, clearance, and displacement are as follows:

[0158] in: For speed increments; This is the speed distribution coefficient; This represents the maximum change in rotational speed over a single cycle.

[0159]

[0160] in: This represents the gap increment; the negative sign ensures... hour This reduces the gap; This is the gap allocation coefficient; This represents the maximum gap change in a single period.

[0161]

[0162] in: This represents the displacement / effective length increment of the conveyor plate. Assign coefficients to locations; This represents the maximum change in the conveyor plate size during a single cycle.

[0163] In some embodiments of this application, the detection and control method further includes the following steps.

[0164] Define the gating factor:

[0165] in: The gating factor; The confidence threshold; when This indicates that the vision is unreliable in this cycle, and the control output is frozen or a degradation strategy is entered.

[0166] Execute gating (freeze when visual invalidation fails):

[0167] in: The gating coefficient; when The increment of the execution volume at each of the three times is 0.

[0168] Update and limit to physical boundaries:

[0169] in: Set the rotational speed for the next cycle; The upper and lower speed limits are set for the upper peeling roller group 1 and the lower peeling roller group 2.

[0170]

[0171] in: Set the interval for the next cycle; The upper and lower limits of the gap between the upper peeling roller group 1 and the lower peeling roller group 2.

[0172]

[0173] in: Set the conveyor plate displacement for the next cycle; These are the upper and lower limits of the conveyor plate's travel.

[0174] The central controller will calculate the setpoint for the next cycle. The speed is respectively distributed to roller group drive motor 8, roller group adjustment drive motor 9, and conveyor adjustment drive motor. The speed channel directly... As the target speed command; the gap channel first determines the calibration relationship. The target clearance is converted into the target displacement of the bearing housing, and the target angle / pulse number of the motor is further calculated based on the lead screw and reduction ratio. The position command is then issued; the conveyor channel... The electric actuator extends or retracts as a target displacement command. Each actuator executes locally in a closed loop and sends the data back. Used for on-time verification and fault diagnosis.

[0175] The above instructions are sent to the controller, which periodically reads the feedback values ​​of rotational speed, clearance, and conveyor plate position. The system compares the result with the set value for on-time verification and provides feedback. Simultaneously, it monitors for abnormal signals such as overcurrent, overtorque, stall, limit triggering, and communication interruption in the servo / push rod. If incomplete execution or timeout is detected, or any abnormality is detected, a safety strategy is immediately triggered: the peeling roller speed is reduced to 0, and the gap is opened to a safe value. The conveyor plate is retracted to a safe position. It will also output alarms / record faults. If the visual confidence level remains insufficient for an extended period or the anomaly persists, the system will enter a degraded mode (freezing current parameters or reverting to empirical parameters). (This process continues until the system returns to normal or manual intervention is initiated. Once normal operation is restored, the image acquisition and preprocessing steps can be repeated.)

[0176] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

[0177] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. It should be noted that any modifications, equivalent substitutions, and improvements made by those skilled in the art within the spirit and principles of the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of this patent application should be determined by the scope of the appended claims.

Claims

1. An online detection and control system for residual husk content in a corn harvester, characterized in that: The corn harvester includes a conveying unit and a peeling mechanism unit: The peeling mechanism unit includes an upper peeling roller group and a lower peeling roller group arranged with a gap between them; both the upper and lower peeling roller groups are connected to roller group drive motors to drive the upper and lower peeling roller groups to rotate; the upper peeling roller group is connected to an upper peeling roller group support, and the upper peeling roller group support is connected to a roller group adjustment drive motor via a drive mechanism, the roller group adjustment drive motor being able to adjust the gap between the upper and lower peeling roller groups by controlling the drive mechanism; The conveying unit includes a conveying plate that extends to an upper peeling roller assembly. The conveying plate is connected to an electric push rod, which is connected to a conveying adjustment drive motor. The conveying adjustment drive motor can adjust the position of the conveying plate to the feeding point of the upper peeling roller assembly by controlling the electric push rod. The online detection and control system includes: Vision unit: Located at the discharge end of the peeling mechanism unit, used to collect images of the peeled corn; Controller: Connected to the vision unit, the roller group drive motor, the roller group adjustment drive motor, and the conveyor adjustment drive motor, and configured as follows: The corn peeling rate was calculated based on the collected images of peeled corn. Based on the deviation between the corn peeling rate and the target peeling rate, the peeling mechanism unit and the conveying unit are controlled, including: The roller group adjustment drive motor adjusts the gap between the upper peeling roller group and the lower peeling roller group, and / or controls the conveying adjustment drive motor to adjust the position of the conveying plate to the feeding point of the upper peeling roller group, and / or controls the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

2. A method for online detection and control of residual husk content in a corn harvester, characterized in that, The online detection and control system for residual husk content in a corn harvester as described in claim 1 includes the following steps: Set a detection box and select the corn target in each frame of the peeled corn image acquired by the vision unit; Calculate the first The residual area of ​​each corn target ear body area ; According to the The residual area of ​​each corn target ear body area Calculate the residual bark coverage rate: ,in, For the first Within the first control cycle, the first The skin coverage rate of each target, It is a positive number; Based on the residual husk coverage of all corn targets within a detection period, calculate the residual husk amount of corn targets within that detection period. ; Based on the residual amount of corn husks within a detection cycle Calculate the corn peeling rate : ; Based on the set target amount of residual skin Calculate the target peeling rate : ; Based on corn peeling rate Deviation from target peeling rate The control of the peeling mechanism unit and the conveying unit includes: controlling the roller group adjustment drive motor to adjust the gap between the upper peeling roller group and the lower peeling roller group, and / or controlling the conveying adjustment drive motor to adjust the position of the conveying plate to the feeding point of the upper peeling roller group, and / or controlling the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

3. The method for online detection and control of residual husk content in corn harvesters according to claim 2, characterized in that, Calculate the first The ear area of ​​a single corn target ,include: Judge the first Does each corn target have an ear-body mask, which is used to identify the pixel region of the corn ear body after peeling? If so, calculate the ear area: ; If not, the area of ​​the detection frame is used as the area of ​​the ear: in, These are the pixel values ​​for the residual skin mask, where 1 indicates residual skin and 0 indicates no residual skin. Let be the pixel coordinates, where This is the horizontal index of the pixel. This is the vertical index of the pixel; For the first Area of ​​each detection frame For the first Each detection frame width, For the first The detection box height is 1, and the ROI represents the selected sub-region in the image acquired by the visual unit.

4. The method for online detection and control of residual husk content in corn harvesters according to claim 2, characterized in that, Calculate the residual amount of corn husks within a detection cycle. ,include: in: This represents the total number of all corn targets within a detection period. For the first Confidence level of the corn target, It is a positive number.

5. The online detection and control system for residual husk content in a corn harvester according to claim 2, characterized in that, It also includes the amount of residual skin. The filtering process includes: in: This is the residual amount after filtering. The length of the sliding window. Indicates the order parameter; The calculation of peeling rate It also includes: After filtering Calculate the peeling rate : .

6. The method for online detection and control of residual husk content in a corn harvester according to claim 2, characterized in that, Based on corn peeling rate Deviation from target peeling rate The control of the peeling mechanism unit and the conveying unit includes: Error in calculating residual amount : Regarding residual amount error Normalization is performed to obtain the normalization error. ; This is due to the residual amount error from the previous cycle. This is the residual amount error for this cycle; Calculate the rate of change of error: For the rate of change of error Normalization is performed to obtain the rate of change of normalization error. ; Construct membership function rules for the proportional coefficient, integral coefficient, and differential coefficient. Based on these membership function rules, the normalized error is... Normalized error rate of change As input, the proportional parameter adjustment amount is calculated. Integral parameter adjustment amount Differential parameter adjustment amount .

7. The online detection and control system for residual husk content in a corn harvester according to claim 6, characterized in that, The error in residual amount Normalization processing is performed, including: The rate of change of error Normalization processing is performed, including: in: This is the residual error ratio coefficient. This is the proportionality coefficient of the error change rate. To control the cycle.

8. The method for online detection and control of corn harvester peeling residue as described in claim 2 or 6, characterized in that, Also includes: Calculate the deviation of the corn target size: in: This is due to size deviation; The target size of corn after fusion within a cycle; This serves as a reference size for the corn target. For the first The target size for each corn cob. For the reference area of ​​the detection frame, For the detection frame width, For the height of the detection frame, For the first Confidence level of the corn target, It is a positive number; Calculate the feedforward term: in: For feedforward compensation; Forward coefficients; Corn target supplementary comprehensive output: Based on the comprehensive control quantity ultimately used for execution allocation The control roller group adjustment drive motor adjusts the gap between the upper peeling roller group and the lower peeling roller group, and / or controls the conveyor adjustment drive motor to adjust the position of the conveyor plate to the feeding point of the upper peeling roller group, and / or controls the roller group drive motor to adjust the rotation speed of the upper peeling roller group and the lower peeling roller group.

9. The method for online detection and control of residual husk content in a corn harvester according to claim 8, characterized in that, include: when To enhance peeling strength, it is necessary to increase the rotation speed, reduce the gap, and increase the extension of the conveyor plate; when Then adjust in the opposite direction.

10. The method for online detection and control of residual husk content in a corn harvester according to claim 8, characterized in that, include: in: For speed increments; This is the speed distribution coefficient; This represents the maximum change in rotational speed over a single cycle. in: This represents the gap increment; the negative sign ensures... hour This reduces the gap; This is the gap allocation coefficient; This represents the maximum gap change in a single cycle. in: This represents the displacement increment of the conveyor plate. Assign coefficients to locations; This represents the maximum change in the conveyor plate size during a single cycle.

Images