Threshing and separating device for rice seed harvesting and harvester and control method

Abstract

This invention provides a threshing and separating device, a harvester, and a control method for rice seed harvesting. The threshing and separating device includes a threshing drum, a threshing drum top cover, a stepped concave plate, and a guide strip angle adjustment device. A spiral feeding head, a threshing transition section, a threshing section, and a straw removal section of the threshing drum sequentially pass through and are fixed to the threshing drum shaft. These components are connected end-to-end. This invention's threshing and separating device ensures stable and uniform flow of the plant within the threshing space, guaranteeing threshing and separating quality. It uses a feeding rate prediction model to predict the feed rate of the combine harvester at the next moment, overcoming the problem of feed rate lag. Based on the predicted feed rate, grain breakage rate, and grain entrainment loss rate at the next moment, the harvester's operating parameters are adjusted to improve threshing and separating performance, prevent clogging, and reduce seed damage and entrainment loss.

Description

Technical Field

[0001] This invention belongs to the field of structural design and adaptive control of working parameters of combine harvesters, and particularly relates to a threshing and separating device for rice seed harvesting, a harvester, and a control method. Background Technology

[0002] Grain is the foundation of the nation, and seeds are the basis of grain production. Seeds are considered the "chips" of grain crops; seed security is crucial to food security, and seed quality determines grain yield. Rice is my country's largest grain crop, and rice yield largely depends on the quality of rice seeds. Currently, the technology of tracked combine harvesters used for harvesting paddy rice in my country is becoming increasingly mature. However, most of the machinery used for harvesting rice seed production fields is ordinary combine harvesters. When harvesting the parent plants using ordinary rice combine harvesters, seed breakage and loss rates remain high, drum blockages are frequent, and operational efficiency is low. There are no mature combine harvesters specifically designed for rice seed production on the domestic market; most are still in the research and testing stage. Foreign-developed seed-producing combine harvesters are highly adaptable. By changing key components such as the header, threshing elements, and cleaning mechanisms, they can meet the needs of breeding and harvesting processes for corn, wheat, rapeseed, sesame, and various small-grain crops, achieving multi-purpose functionality. The machines generally have a high degree of intelligence, incorporating mature technologies such as grain self-cleaning, yield measurement, grain moisture content monitoring, automatic operating speed control, sensor fault diagnosis, GPS navigation automatic driving, electro-hydraulic control, and multi-functional lever control. They offer strong human-machine interaction, enabling real-time display, recording, and analysis of operating and performance parameters, as well as adaptive adjustment of relevant operating parameters. They boast advantages such as good self-cleaning effect, high operating efficiency, low grain loss rate, and low grain breakage rate, and are developing towards serialization, industrialization, and precision.

[0003] As a crucial component of combine harvesters, the threshing and separating device directly impacts the harvester's performance when harvesting hybrid rice seeds. The space comprised of the threshing drum, threshing concave plate, and threshing drum top cover is called the threshing and separating space. The flow state of the plants within this space directly affects the threshing and separating effect. Currently, the size of the threshing and separating space in existing domestic combine harvesters is consistent across each cross-section along the length of the threshing drum, failing to consider the variable quality process of threshing and separating. Existing research and product development primarily aim to improve the threshing and separating quality of hybrid rice female parent seeds by optimizing the structural parameters of the threshing drum. However, experimental results show that this method has limited improvement in threshing and separating effect. Furthermore, the angle of the guide strips on the top cover of existing threshing drums is fixed, as are the gap between the fixed and rotating blades of the stalk crushing device and the rotation speed of the moving blades. This prevents adaptive adjustments based on stalk characteristics during operation, resulting in poor adaptability for harvesting different seed varieties. Specifically, the angle of the top cover guide strip directly affects the flow of the plants within the threshing drum and the number of times they are repeatedly struck. If the top cover guide strip angle is not set properly, the threshed plants cannot be discharged from the machine in a timely manner, increasing the probability of drum blockage and the number of times the plants are repeatedly struck, leading to increased seed breakage. Severely struck plants produce a large amount of fine, short stems with high moisture content and impurities. These wet, sticky impurities adhere to the concave sieve holes, hindering seed passage and increasing seed loss. Furthermore, if the stems discharged from the threshing drum through the discharge port are not processed by the stem crushing device in a timely manner, the stems will accumulate at the crushing device, affecting their flow within the threshing drum and causing blockages in both the threshing drum and the crushing device. Therefore, setting appropriate structural parameters for the threshing separation device and the stem crushing device based on changes in the feed rate, guiding the plants to flow evenly and stably within the threshing space and discharge them from the machine in a timely manner, is crucial to ensuring the effective operation of the threshing separation device.

[0004] Regarding the operational parameter control model, existing research has only preliminarily established simple linear models or fuzzy control rules based on the monitored drum speed changes to adjust the machine's forward speed. This results in unstable overall machine operating load and an inability to guarantee the overall machine operation quality, especially the threshing and separation quality, which affects the harvesting quality of hybrid rice seed production mother lines. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the aforementioned technical problems. To this end, the present invention proposes a threshing and separating device for rice seed harvesting, a combine harvester, and a control method to prevent blockage of the combine harvester, reduce seed breakage rate and entrainment losses, reduce threshing and separating power consumption, and improve threshing and separating performance.

[0006] This invention has wide applicability and is particularly suitable for harvesting hybrid rice seed production female parent seeds, improving the quality of threshing and separation, thereby improving the harvest quality of hybrid rice seed production female parent seeds.

[0007] The technical solution of this invention is:

[0008] A rice seed threshing and separating device includes a threshing and separating device; the threshing and separating device includes a threshing drum, a threshing drum top cover, a stepped concave plate, and a guide strip angle adjustment device;

[0009] The top cover of the threshing drum is located on the upper part of the threshing drum, and the stepped concave plate is located on the lower part of the threshing drum. The top cover of the threshing drum is connected to the stepped concave plate. The threshing drum is connected to the frame of the combine harvester. The space enclosed between the threshing drum, the top cover of the threshing drum, and the stepped concave plate is the threshing separation space. The threshing separation space is a gradual threshing separation space.

[0010] The top cover of the threshing drum is provided with a guide strip on the side facing the threshing drum, and the guide strip angle adjustment device is connected to the guide strip.

[0011] In the above scheme, the threshing drum includes a threshing drum shaft, a threshing drum spiral feed head, a threshing transition section of the threshing drum, a threshing section of the threshing drum, short-textured bar teeth, nail teeth, a straw discharge section of the threshing drum, and straw discharge blades of the threshing drum;

[0012] The threshing drum spiral feed head, threshing drum transition section, threshing drum threshing section and threshing drum straw discharge section pass through the threshing drum shaft in sequence and are fixed. The threshing drum spiral feed head, threshing drum transition section, threshing drum threshing section and threshing drum straw discharge section are connected end to end.

[0013] The threshing drum spiral feed head includes a trapezoidal frustum and spiral blades, with the spiral blades arranged spirally along the surface of the trapezoidal frustum;

[0014] The short-toothed rod plate teeth are spirally arranged and fixed in the threshing transition section and the front section of the threshing section of the threshing drum, while the nail teeth are spirally arranged in the middle and rear sections of the threshing section of the threshing drum.

[0015] The straw-discharging blades of the threshing drum are spirally arranged along the surface of the straw-discharging section of the threshing drum.

[0016] The cross-sectional area of ​​the threshing drum spiral feed head and the threshing transition section of the threshing drum gradually increases along the threshing direction, making the cross-sections of the threshing drum spiral feed head and the threshing transition section of the threshing drum trapezoidal. The cross-sectional area of ​​the straw discharge section of the threshing drum gradually decreases along the threshing direction, making the cross-section of the straw discharge section of the threshing drum trapezoidal. This results in a gradually changing threshing separation space between the threshing drum spiral feed head, the threshing transition section of the threshing drum, the straw discharge section of the threshing drum, the top cover of the threshing drum, and the stepped concave plate.

[0017] In the above scheme, the top cover of the threshing drum facing the threshing drum is provided with a feeding section, a first threshing separation section, a transition section, a second threshing separation section and a straw discharge section along the length of the threshing drum; the cross-sectional radii of the feeding section, the first threshing separation section and the second threshing separation section increase in size sequentially, the cross-sectional area of ​​the transition section gradually increases along the threshing separation direction, and the cross-sectional area of ​​the straw discharge section gradually decreases along the threshing separation direction;

[0018] The feeding section is equipped with a feeding section guide bar, the first threshing and separation section is equipped with a first threshing and separation section guide bar, and the second threshing and separation section is equipped with a second threshing and separation section guide bar. The second threshing and separation section guide bar is connected to the guide bar angle adjustment device.

[0019] Furthermore, the guide bar angle adjustment device includes a telescopic motor, a connecting plate, a push rod, a rotating shaft for the guide bar of the second threshing and separation section, and a connecting rod for the movable end of the guide bar of the second threshing and separation section;

[0020] One end of the guide bar of the second threshing and separation section has a through hole. The rotating shaft passes through the through hole to install the guide bar of the second threshing and separation section on the top cover of the threshing drum, so that the guide bar of the second threshing and separation section can rotate around the rotating shaft. The other end of the guide bar of the second threshing and separation section is installed on the connecting plate through the connecting rod, so that each guide bar of the second threshing and separation section is linked together through the connecting plate. The telescopic motor is installed on the top cover of the threshing drum. The electric push rod of the telescopic motor is connected to the connecting rod through the push rod, so that when the electric push rod of the telescopic motor is working, it drives the connecting plate to move through the push rod, thereby driving the guide bar of the second threshing and separation section to rotate around the rotating shaft.

[0021] In the above scheme, the stepped concave plate includes a first concave plate and a second concave plate;

[0022] The first concave plate and the second concave plate are arranged along the length of the threshing drum, and the volume of the second concave plate is larger than that of the first concave plate.

[0023] A harvester includes the aforementioned rice seed threshing and separating device.

[0024] The above solution also includes a stalk crushing device, a forward speed monitoring device, a feed amount prediction device, a conveying trough, a forward speed adjustment device, a grain entrainment loss monitoring sensor, a grain breakage rate monitoring device, a guide bar angle monitoring sensor, a blade gap monitoring sensor, a torque measurement sensor, a rotating blade fixed shaft speed monitoring sensor, and a control device.

[0025] The threshing and separating device is connected to the conveying trough at its feed end and to the stalk crushing device at its output end. The control device is connected to the stalk crushing device, the forward speed monitoring device, the threshing and separating device, the feed amount prediction device, the forward speed adjustment device, the grain entrainment loss monitoring sensor, and the grain breakage rate monitoring device, respectively.

[0026] The forward speed monitoring device monitors the rotational speed of the combine harvester's track drive wheels; the feed rate prediction device has a feed rate prediction model to predict the feed rate at the next moment; the forward speed adjustment device adjusts the forward speed of the combine harvester; the grain entrainment loss monitoring sensor monitors the grain entrainment loss rate; the grain breakage rate monitoring device monitors the grain breakage rate; the guide bar angle monitoring sensor monitors the angle of the guide bar in the second threshing and separating section; the blade gap monitoring sensor monitors the gap between the rotating blade and the fixed blade in the stalk crushing device; the stalk crushing device rotating blade fixed shaft speed monitoring sensor monitors the speed of the rotating blade fixed shaft in the stalk crushing device; the control device uses the real-time feed rate prediction value for the next moment, the grain breakage rate monitored by the grain breakage rate monitoring device, and the grain entrainment loss rate monitored by the grain entrainment loss monitoring sensor, as well as the current forward speed of the combine harvester and the second threshing and separating section... The angle of the guide bar in the second threshing section of the threshing device, the gap between the rotating blades and the fixed blades in the stalk crushing device, and the rotational speed of the fixed shaft of the rotating blades in the stalk crushing device are input into the model prediction controller for processing. This yields predicted values ​​for the combine harvester's forward speed, the angle of the guide bar in the second threshing section of the threshing device, the gap between the rotating blades and the fixed blades in the stalk crushing device, and the rotational speed of the fixed shaft of the rotating blades in the stalk crushing device at the next moment. The controller then controls the forward speed adjustment device to adjust the forward speed of the combine harvester, the guide bar angle adjustment device to adjust the angle of the guide bar in the second threshing section, the linear telescopic electric cylinder of the stalk crushing device to adjust the gap between the rotating blades and the fixed blades, and the hydraulic motor of the stalk crushing device to adjust the rotational speed of the fixed shaft of the rotating blades. This adjusts the forward speed adjustment device, the angle of the guide bar in the second threshing section, the gap between the rotating blades and the fixed blades in the stalk crushing device, and the rotational speed of the fixed shaft of the rotating blades in the stalk crushing device to the predicted values ​​for the next moment.

[0027] A control method for the harvester includes the following steps:

[0028] The control device processes the real-time predicted feed rate for the next moment, the grain breakage rate monitored by the grain breakage rate monitoring device, the grain entrainment loss rate monitored by the grain entrainment loss monitoring sensor, and the current combine harvester forward speed, the angle of the guide bar in the second threshing and separating section, the gap between the rotating blade and the fixed blade in the stalk crushing device, and the rotational speed of the fixed shaft of the rotating blade in the stalk crushing device. This results in the predicted values ​​for the next moment's combine harvester forward speed, the angle of the guide bar in the second threshing and separating section, the gap between the rotating blade and the fixed blade in the stalk crushing device, and the rotational speed of the fixed shaft of the rotating blade in the stalk crushing device. This allows the forward speed adjustment device, the angle of the guide bar in the second threshing and separating section, the gap between the rotating blade and the fixed blade in the stalk crushing device, and the rotational speed of the fixed shaft of the rotating blade in the stalk crushing device to be adjusted to the predicted values ​​for the next moment, thereby improving threshing and separating performance and preventing blockages.

[0029] In the above scheme, the feed volume prediction model is:

[0030]

[0031] Where Q is the predicted feed amount, v is the forward speed, and h is the forward speed. c h is the height of the cutting platform, w is the plant height, and ρ is the cutting width.

[0032] ρ=p·M

[0033] Where p is the plant density and M is the rice plant mass equation;

[0034]

[0035] Where, m e denoted as h, where h is the average mass of the rice panicle; and denoted as m, where m is the mass per unit length of the stem.

[0036] In step S1, the plant height h in the harvesting area is:

[0037] h = h1 - l x sinθ

[0038] Where h1 is the installation height of the image acquisition unit, l x Let θ be the straight-line distance from the spikelet of the plant within the region of interest to the camera, and l be the pitch angle of the image acquisition unit. x sinθ is the height from the camera to the spike layer of the plant.

[0039] In the above scheme, the model prediction controller is an MPC controller, and the following prediction model is provided:

[0040] x(k+1)=Ax(k)+Bu(k)

[0041] in, Let x1(k) be the system state variable at time k, x2(k) be the grain entrainment loss, x2(k) be the grain breakage rate, and u(k) be the system input vector. u1(k) is the 2-3 angle of the guide bar in the second threshing and separation section, u2(k) is the forward speed v, u3(k) is the gap between the rotating blade and the fixed blade in the stalk crushing device, u4(k) is the rotational speed of the fixed shaft of the rotating blade in the stalk crushing device, A is the system state matrix, and B is the input matrix.

[0042] The cost function J of the MPC controller is:

[0043] J = x(k) T Gx(k)+U(k) T HU(k)+2x(k) T EU(k)

[0044] in,

[0045]

[0046] T represents the matrix transpose.

[0047] Furthermore, x(k) and u(k) should satisfy the following constraints:

[0048]

[0049]

[0050] Where, x min (k) is the minimum value of the system state variables, x max (k) is the maximum value of the system state variable, u min (k) is the minimum value of the system's input vector, u max (k) is the maximum value of the system's input vector.

[0051] In the above scheme, x1(k) and x2(k) satisfy the following constraints:

[0052] 0≤x1(k)≤Q·P·1%

[0053] 0≤x²(k)≤Q·P·2%

[0054] Among them, the grain entrainment loss x1(k) and the breakage rate x2(k) are system state variables, Q is the predicted value of the feed amount of the combine harvester in kg / s, and P is the proportion of grain in the feed amount.

[0055] Let u1(k), u2(k), u3(k), and u4(k) satisfy the following constraints:

[0056] 20°≤u1(k)≤50°

[0057] 0 m / s ≤ u2(k) ≤ 1.5 m / s

[0058] 5mm≤u3(k)≤20mm

[0059] 2500rpm≤u4(k)≤3000rpm.

[0060] Compared with the prior art, the beneficial effects of the present invention are:

[0061] 1. In this invention, the cross-sections of the trapezoidal frustum, the threshing transition section, and the straw discharge section of the threshing drum are trapezoidal, so that the threshing space between the threshing drum, the threshing concave plate, and the top cover of the threshing drum forms a gradual threshing space, which allows the plants to flow stably and evenly within the threshing space, ensuring the quality of threshing and separation.

[0062] 2. This invention constructs a feeding amount prediction model through a feeding amount prediction device, which can predict the feeding amount of the combine harvester at the next moment, thereby adjusting the relevant working parameters of the threshing and separating device in real time, reducing control lag, overcoming the problem of lag in traditional feeding amount measurement, and significantly improving the threshing and separating effect.

[0063] 3. The control device of this invention, based on the predicted value of the feed amount of the combine harvester at the next moment, the grain breakage rate monitored by the grain breakage rate monitoring device, and the grain entrainment loss rate monitored by the grain entrainment loss monitoring sensor, processes the predicted values ​​of the forward speed of the combine harvester, the angle of the guide strip in the threshing and separating device, and the gap between the rotating blade and the fixed blade in the stalk crushing device at the next moment. This allows the forward speed adjustment device, the angle of the guide strip in the threshing and separating device, the gap between the rotating blade and the fixed blade in the stalk crushing device, and the rotational speed of the fixed shaft of the rotating blade in the stalk crushing device to the predicted values ​​for the next moment, thereby improving threshing and separating performance, preventing blockage, and reducing seed damage and entrainment loss rate. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of the threshing and separation device according to one embodiment of the present invention.

[0065] Figure 2 This is a schematic front view of a threshing and separation device according to an embodiment of the present invention.

[0066] Figure 3 This is a schematic diagram of a combine harvester according to one embodiment of the present invention.

[0067] Figure 4 This is a schematic diagram of a threshing drum according to an embodiment of the present invention.

[0068] Figure 5This is a schematic diagram of a threshing drum spiral feed head according to an embodiment of the present invention.

[0069] Figure 6 This is a schematic diagram of the force analysis of the threshing drum screw feed head according to an embodiment of the present invention.

[0070] Figure 7 This is a schematic diagram of a short-grooved bar plate tooth structure according to an embodiment of the present invention.

[0071] Figure 8 This is a schematic diagram of a gradual threshing and separation spatial structure according to an embodiment of the present invention.

[0072] Figure 9 This is a schematic diagram of a guide strip inside the top cover of the threshing drum according to one embodiment of the present invention.

[0073] Figure 10 This is a schematic diagram of the dimensions of the guide strip structure in the feeding section according to an embodiment of the present invention.

[0074] Figure 11 This is a schematic diagram of the structural dimensions of the guide strip in the first threshing and separation section according to an embodiment of the present invention.

[0075] Figure 12 This is a schematic diagram of the guide bar angle adjustment device according to one embodiment of the present invention.

[0076] Figure 13 This is a schematic diagram of the front view of the first concave plate according to an embodiment of the present invention.

[0077] Figure 14 This is a schematic diagram of the left side view of the first concave plate according to an embodiment of the present invention.

[0078] Figure 15 This is a schematic diagram of the front view of the second concave plate according to an embodiment of the present invention.

[0079] Figure 16 This is a schematic diagram of the left side view of the second concave plate according to an embodiment of the present invention.

[0080] Figure 17 This is a front view showing the positional relationship between the first and second concave plates according to an embodiment of the present invention.

[0081] Figure 18 This is a left view showing the positional relationship between the first and second concave plates according to an embodiment of the present invention.

[0082] Figure 19 This is a schematic diagram of the stalk crushing device according to one embodiment of the present invention.

[0083] Figure 20 This is a schematic diagram of the principle of the cutting platform height measuring mechanism according to one embodiment of the present invention.

[0084] Figure 21 This is a flowchart of the plant density acquisition process according to one embodiment of the present invention.

[0085] Figure 22 A flowchart illustrating the process of obtaining plant height according to one embodiment of the present invention.

[0086] Figure 23 A flowchart illustrating the process of obtaining the cutting width according to one embodiment of the present invention.

[0087] Figure 24 A schematic diagram of the cutting width detection according to an embodiment of the present invention.

[0088] Figure 25 A diagram of the RBF-MLP neural network structure according to an embodiment of the present invention.

[0089] Figure 26 A training loss curve of an RBF neural network according to an embodiment of the present invention.

[0090] Figure 27 A training loss curve of an MLP neural network according to an embodiment of the present invention.

[0091] Figure 28 The training loss curve of the RBF-MLP neural network according to one embodiment of the present invention.

[0092] Figure 29 This is a schematic diagram of the forward speed control device according to one embodiment of the present invention.

[0093] Figure 30 This is a flowchart of a fuzzy PID control according to an embodiment of the present invention.

[0094] In the diagram: 1. Threshing drum; 1-1. Threshing drum shaft; 1-2. Threshing drum screw feed head; 1-3. Threshing transition section of threshing drum; 1-4. Threshing section of threshing drum; 1-5. Short grooved bar teeth; 1-5-1. Short grooved bar; 1-5-2. Bar teeth; 1-6. Spike teeth; 1-7. Straw discharge section of threshing drum; 1-8. Straw discharge blades of threshing drum; 2. Threshing drum top cover; 2-1. Feeding section guide bar; 2-2. First threshing separation section guide bar; 2-3. Second threshing separation section guide bar. 2-4. Guide bar for the second threshing and separation section; 2-5. Telescopic motor; 2-6. Connecting plate; 2-7. Push rod; 2-8. Rotating shaft of the guide bar for the second threshing and separation section; 2-9. Connecting rod at the movable end of the guide bar for the second threshing and separation section; 2-10. Feeding section; 2-11. First threshing and separation section; 2-12. Transition section; 2-13. Second threshing and separation section; 2-14. Straw discharge section; 3. Feeding section grid plate; 4. First section concave plate; 4-1. Vertical grid bars of the first section concave plate; 4-2. Horizontal grid bars of the first section concave plate; 5. 5-1. Second concave plate; 5-2. Second concave plate vertical grid bars; 5-3. Second concave plate horizontal grid bars; 6. Guide bar angle adjustment device; 7. Stalk crushing device; 7-1. Rotating blade; 7-2. Hydraulic continuously variable speed motor; 7-3. Linear telescopic electric cylinder; 7-4. Moving chute; 7-5. Fixed blade fixing rod; 7-6. Fixed blade; 7-7. Torque measurement sensor; 8. Forward speed monitoring device; 10. Threshing and separation device; 11. RealSense depth camera ; 12. Header height monitoring device; 13. Conveying trough; 14. Forward speed adjustment device; 15. Grain entrainment loss monitoring sensor; 13-1. Connecting piece; 13-2. Fisheye bearing; 13-3. Metal rod; 13-4. Fisheye bearing; 13-5. Connecting plate; 13-6. Angle sensor; 13-7. Connecting piece; 14-1. HST push rod; 14-2. Electric cylinder connecting piece; 14-3. Electric cylinder; 14-4. Displacement sensor; 14-5. Support structure. Detailed Implementation

[0095] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0096] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "front," "rear," "left," "right," "upper," "lower," "axial," "radial," "vertical," "horizontal," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0097] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0098] like Figure 1 , 2 3 is a preferred embodiment of the rice seed harvesting threshing and separating device of the present invention, wherein the threshing and separating device includes a threshing and separating device 10; the threshing and separating device 10 includes a threshing drum 1, a threshing drum top cover 2, a stepped concave plate and a guide bar angle adjustment device 6;

[0099] The threshing drum top cover 2 is located on the upper part of the threshing drum 1, and the stepped concave plate is located on the lower part of the threshing drum 1. The threshing drum top cover 2 is connected to the stepped concave plate. The threshing drum 1 is connected to the combine harvester frame. The space enclosed between the threshing drum 1, the threshing drum top cover 2, and the stepped concave plate is the threshing separation space, which is a gradual threshing separation space.

[0100] The top cover 2 of the threshing drum is provided with a guide strip on one side facing the threshing drum 1, and the guide strip angle adjustment device 6 is connected to the guide strip.

[0101] like Figure 4 , 5As shown in Figures 6 and 7, preferably, the threshing drum 1 includes a threshing drum shaft 1-1, a threshing drum spiral feed head 1-2, a threshing drum transition section 1-3, a threshing drum threshing section 1-4, short-patterned rod teeth 1-5, nail teeth 1-6, a threshing drum straw discharge section 1-7, and a threshing drum straw discharge blade 1-8; the threshing drum spiral feed head 1-2, the threshing drum transition section 1-3, the threshing drum threshing section 1-4, and the threshing drum straw discharge section 1-7 sequentially pass through the threshing drum shaft 1-1 and are fixed, and the threshing drum spiral feed head 1-2, the threshing drum transition section 1-3, the threshing drum threshing section 1-4, and the threshing drum straw discharge section 1-7 are connected end to end; the threshing drum spiral feed head 1-2 includes a trapezoidal frustum 1-2-1 and spiral blades 1-2-2, the spiral blades 1-2-2 being spirally arranged and fixed along the surface of the trapezoidal frustum 1-2-1; Short-toothed rods 1-5 are spirally arranged and fixed in the front section of the threshing transition section 1-3 and the threshing section 1-4 of the threshing drum; nails 1-6 are spirally arranged and fixed in the middle and rear section of the threshing section 1-4 of the threshing drum; the straw-discharging blades 1-8 of the threshing drum are spirally arranged and fixed along the surface of the straw-discharging section 1-7 of the threshing drum; the cross-sectional area of ​​the spiral feed head 1-2 and the threshing transition section 1-3 of the threshing drum gradually increases along the threshing direction, so that the cross-sections of the spiral feed head 1-2 and the threshing transition section 1-3 of the threshing drum form a trapezoid, and the cross-sectional area of ​​the straw-discharging section 1-7 of the threshing drum gradually decreases along the threshing direction, so that the cross-section of the straw-discharging section 1-7 of the threshing drum forms a trapezoid, so that the threshing separation space between the spiral feed head 1-2, the threshing transition section 1-3, the straw-discharging section 1-7 of the threshing drum and the top cover 2 and the stepped concave plate of the threshing drum is gradually changing.

[0102] like Figure 7 As shown, preferably, the short grooved bar plate teeth 1-5 includes short grooved bar 1-5-1 and plate teeth 1-5-2; the short grooved bar 1-5-1 and plate teeth 1-5-2 are connected, and the plate teeth 1-5-2 are installed on the threshing transition section 1-3 and the threshing section 1-4 of the threshing drum.

[0103] like Figure 8 , 9 As shown in Figures 10, 11, and 12, preferably, the top cover 2 of the threshing drum facing the threshing drum 1 is provided with a feeding section 2-9, a first threshing separation section 2-10, a transition section 2-11, a second threshing separation section 2-12, and a straw discharge section 2-13 along the length of the threshing drum 1. The cross-sectional radii of the feeding section 2-9, the first threshing separation section 2-10, and the second threshing separation section 2-12 increase sequentially, the cross-sectional area of ​​the transition section 2-11 along the threshing separation direction gradually increases, and the cross-sectional area of ​​the straw discharge section 2-13 along the threshing separation direction gradually decreases.

[0104] like Figure 9As shown, a feeding section guide bar 2-1 is installed on the feeding section 2-9, a first threshing and separation section guide bar 2-2 is installed on the first threshing and separation section 2-10, and a second threshing and separation section guide bar 2-3 is installed on the second threshing and separation section 2-11. The second threshing and separation section guide bar 2-3 is connected to the guide bar angle adjustment device 6. There are multiple feeding section guide bars 2-1, first threshing and separation section guide bars 2-2, and second threshing and separation section guide bars 2-3.

[0105] like Figure 12 As shown, preferably, the guide bar angle adjustment device 6 includes a telescopic motor 2-4, a connecting plate 2-5, a push rod 2-6, a rotating shaft 2-7 for the second threshing and separation section guide bar, and a connecting rod 2-8 for the movable end of the second threshing and separation section guide bar. A through hole is provided at one end of the second threshing and separation section guide bar 2-3, and the rotating shaft 2-7 installs the second threshing and separation section guide bar 2-3 into the second threshing and separation section of the threshing and separation drum top cover 2 through the through hole, so that the second threshing and separation section guide bar 2-3 can rotate around the rotating shaft 2-7. The other end of the guide bar 2-3 of the second threshing and separation section is mounted on the connecting plate 2-5 via the connecting rod 2-8, so that the guide bar 2-3 of the second threshing and separation section is linked through the connecting plate 2-5. The telescopic motor 2-4 is mounted on the top cover 2 of the threshing drum. The electric push rod of the telescopic motor 2-4 is connected to the connecting rod 2-8 via the push rod 2-6, so that when the electric push rod of the telescopic motor 2-4 is working, it drives the connecting plate 2-5 to move through the push rod 2-6, thereby driving the guide bar 2-3 of the second threshing and separation section to rotate around the rotating shaft 2-7.

[0106] The movable end connecting rods 2-8 of each second threshing and separation section guide bar are linked through the connecting plate 2-5. During operation, the electric push rod 2-4 drives the connecting handle push rod 2-6 to transmit power. The push rod 2-6 is connected to the connecting plate 2-5 by bolts. Under the drive of the push rod 2-6, the second threshing and separation section guide bar 2-3 rotates around the second threshing and separation section guide bar rotation axis 2-7, realizing stepless adjustment of the angle of the second threshing and separation section guide bar 2-3.

[0107] The guide bar angle adjustment device 6 affects the residence time of plants in the threshing and separating device. Based on the monitored changes in operating performance data such as grain entrainment loss rate and grain breakage rate, as well as the feeding amount, the angle is adjusted to control the residence time of plants in the threshing and separating device. Increasing the angle of the guide bar allows the plants to stay longer, which can improve the threshing and separating performance. Decreasing the angle allows the plants to be expelled quickly, avoiding blockage.

[0108] Preferably, it also includes a feeding section grid plate 3, located below the spiral feed head 1-2 of the threshing drum. Its function is to allow some of the mature and easily detached grains to fall through the holes on the feeding section grid plate 3 onto the vibrating screen below when the plants pass through the spiral feed head 1-2 of the threshing drum, thereby reducing the subsequent cleaning load.

[0109] like Figure 13 , 14 As shown in 15, 16, 17, and 18, preferably, the stepped concave plate includes a first concave plate 4 and a second concave plate 5; the first concave plate 4 and the second concave plate 5 are arranged along the length direction of the threshing drum 1, and the volume of the second concave plate 5 is larger than the volume of the first concave plate 4.

[0110] like Figure 19 As shown, preferably, the stalk crushing device 7 includes a rotating blade 7-1, a fixed shaft for the rotating blade 7-1, a hydraulic continuously variable speed motor 7-2, a linear telescopic electric cylinder 7-3, a moving slide 7-4, a fixed blade fixing rod 7-5, a fixed blade 7-6, and a torque measuring sensor 7-7.

[0111] The rotating blades 7-1 are evenly arranged on the mounting shaft, which is connected to the power output shaft of the hydraulic continuously variable motor 7-2 via a torque measuring sensor 7-7. The torque measuring sensor 7-7 is used to collect the torque of the fixed shaft of the rotating blades 7-1 and transmit it to the controller. The fixed blade 7-6 is mounted on the fixed blade fixing rod 7-5, which is slidably connected to the moving slide 7-4. The fixed blade fixing rod 7-5 is connected to the linear telescopic electric cylinder 7-3, which is fixed on the frame of the combine harvester. This allows the linear telescopic electric cylinder 7-3 to drive the fixed blade fixing rod 7-5 to move on the moving slide 7-4, thereby moving the fixed blade 7-6 and changing the gap between the rotating blades 7-1 and the fixed blades 7-6. The stalks discharged from the threshing drum 1 enter the stalk crushing device 7. The control device can adaptively adjust the gap between the rotating blades 7-1 and the fixed blades 7-6, and the rotational speed of the rotating blades 7-1 fixed shaft, based on the angle of the guide strips on the top cover of the threshing drum, the torque of the fixed shaft of the rotating blades 7-1 / grain entrainment loss rate, and the grain breakage rate, to prevent clogging of the stalk crushing device 7 and the threshing and separating device 10. In one embodiment of the present invention, preferably, the threshing drum 1 is a closed threshing drum, with the diameter of the trapezoidal frustum 1-2-1 gradually increasing from 296 mm to 377 mm and a taper of 12°.

[0112] In one embodiment of the present invention, preferably, the diameter of the threshing transition section 1-3 of the threshing drum gradually increases from 377 mm to 440 mm, with a taper of 12°, and the diameter of the threshing section 1-4 of the threshing drum is 440 mm. Short grooved rod teeth 1-5 are installed in a spiral distribution on the first 1 / 3 of the threshing transition section 1-3 and the first 1 / 3 of the threshing section 1-4 of the threshing drum, and nail teeth 1-6 are installed in a spiral distribution on the last 2 / 3 of the threshing section 1-4 of the threshing drum.

[0113] In one embodiment of the present invention, preferably, the diameter of the straw discharge section 1-7 of the threshing drum gradually decreases from 440mm to 380mm, with a taper of 12°. The number of blades in the straw discharge blades 1-8 of the threshing drum is 2. The spiral lead of the feeding head is 1000mm, and the pitch is 600mm for rapid straw discharge. The material friction angle β on the spiral feeding head 1-2 of the threshing drum is 28.8°, and the spiral helix angle θ is 32°. Figure 6 As shown, the number of spiral blades 1-2 of the threshing drum spiral feed head 1-2 is 3, the spiral lead of the feed head is 1200mm, the pitch is 400mm, and the taper of the threshing drum feed head is 12°.

[0114] In one embodiment of the present invention, preferably, on the threshing transition section 1-3 and the threshing section 1-4 of the threshing drum, the number of rows of short grooved bar teeth 1-5 is 6, the number of spiral wire heads is 3, the short grooved bar 1-5-1 is a specially made D-shaped grooved bar with a diameter of 140mm and a height of 50mm, the tooth 1-5-2 is 10mm higher than the short grooved bar by h, the thickness of the tooth 1-5-2 is 6mm, and the short grooved bar and tooth are rolled from 65 manganese steel.

[0115] In one embodiment of the present invention, preferably, the diameter of the rod teeth of the nail teeth 1-6 is 14mm and the height is 90mm. The nail teeth are made of 45 steel, and the surface hardness after heat treatment is HRC50. In addition, a 10° backward tilt angle is provided 30mm below the top of the nail teeth 1-6 to prevent straw from getting stuck during the threshing and separation process. The working range of the drum speed is 635rpm-725rpm.

[0116] After being cut, the stalks first enter the threshing and separation space through the feeding section of the threshing drum 1. In the feeding section of the threshing drum 1, the distance from the mounting plane of the threshing drum top cover 6 to the threshing drum top cover is the smallest. The trapezoidal frustum 1-2-1 of the spiral feeding head with gradually increasing diameter can accommodate a large number of complete stalk groups in the feeding section. At this point, a certain pressure can be applied to the stalks, which can improve the gripping ability of the spiral blades on the stalks and prevent blockage in the feeding section of the threshing drum. As the threshing drum 1 rotates, the stalks pass through the threshing transition section 1-3 and enter the first threshing separation section 1-4. The first 1 / 3 of the threshing transition section 1-3 and the first threshing separation section 1-4 mainly serve as the working space for the short-grooved stalk teeth 1-5. The threshing separation space is slightly increased here, which can form a uniform and stable thin layer of stalks within the threshing separation space, reducing the threshing separation load. The short-grooved stalk teeth 1-5 can reduce grain damage and ensure the threshing separation effect. The grains with high maturity and easy threshing are first separated from the first concave plate 4, reducing the amount of grains carried into the second threshing separation section 1-4 along with the stalks. Next, the stalks pass through the action zone of the nail teeth 1-6 in the second threshing and separation section 1-4. Crops with poor maturity and difficult threshing are threshed under the strong impact of the nail teeth 1-6. Here, the distance from the mounting plane of the threshing drum top cover to the top cover of the threshing drum is the largest, and the distance from the center of the drum shaft 1-1 to the bottom of the concave plate 5 is the largest. The separation area is also the largest here. The stalk group begins to become relatively loose, and most of the grains can easily pass through the sieve, which can reduce the loss of entrainment. By adjusting the angle of the guide strips 2-3 on the top cover of the threshing drum, the number of rotations of the stalks in the threshing and separation space can be adjusted, and the speed at which the stalks are discharged from the machine can be controlled. Finally, under the action of the spiral blades in the straw discharge section of the threshing drum, the stalks are discharged from the threshing drum.

[0117] In one embodiment of the present invention, preferably, in the feeding section, the distance from the mounting plane of the threshing drum top cover to the threshing drum top cover 2 is 215 mm, and the length of the threshing drum top cover 2 in the feeding section is 420 mm; in the first threshing separation section, the distance from the mounting plane of the threshing drum top cover to the threshing drum top cover 2 is 230 mm, and the length of the threshing drum top cover in the first threshing separation section is 672 mm; in the second threshing separation section, the distance from the mounting plane of the threshing drum top cover to the threshing drum top cover 2 is 240 mm, and the length of the threshing drum top cover in the second threshing separation section is 675 mm; in the straw discharge section, the distance from the center line of the threshing drum shaft to the threshing drum top cover 2 gradually decreases, reaching 211.50 mm at the very end, and the length of the threshing drum top cover in the straw discharge section is 320 mm. A transition section with a length L0 = 150 mm is designed at the switch between the first and second threshing separation sections.

[0118] The distance from the mounting plane of the threshing drum top cover to the top cover 2 refers to the distance from the mounting plane of the top cover and the frame to the guide strips on the top cover, which varies because there are three guide strips of different sizes. Figure 10 , 11 As shown, there is a change in the distance from the mounting plane of the threshing drum top cover to the threshing drum top cover.

[0119] In one embodiment of the present invention, the number of guide strips 2-1 in the feeding section is set to 5, the spacing between the guide strips is 62mm, and the spiral angle of the guide plate is fixed at 30°; the number of guide strips 2-2 in the first threshing and separation section is set to 9, the spacing between the guide strips is 62mm, and the spiral angle is fixed at 30°; the number of guide strips 2-3 in the second threshing and separation section is set to 6, the spacing between the guide strips is 100mm, and the angle of the guide strips can be adjusted from 20° to 50°.

[0120] In one embodiment of the present invention, preferably, the concave angle of the first concave plate 4 and the second concave plate 5 is 200°, and both the first concave plate 4 and the second concave plate 5 are arranged below the threshing drum 1.

[0121] The spacing of the grid strips 4-1 of the first concave plate 4 along the length of the drum is L2 = 10 mm, and the spacing of the horizontal grid strips 4-2 of the first concave plate along the direction perpendicular to the length of the drum is L1 = 31 mm; the spacing of the grid strips 5-1 of the second concave plate 5 along the length of the drum is L6 = 7 mm, and the spacing of the grid strips 5-2 of the second concave plate 5 along the direction perpendicular to the length of the drum is L5 = 26 mm; the lengths of the first concave plate 4 and the second concave plate 5 along the length of the threshing drum 1 are L3 = L4 = 750 mm; the distance between the mounting surface of the first concave plate 4 and the bottom of the first concave plate 4 is L4 = 465 mm, and the distance between the mounting surface of the second concave plate 5 and the bottom of the second concave plate 5 is L5 = 485 mm; the drop between the first concave plate 4 and the second concave plate 5 is L9 = L10 = 20 mm, in order to increase the threshing and separation space of the second threshing and separation section, loosen the stalk group, accelerate the rapid separation of grains in the stalk group, and reduce entrainment loss. As the crops move along the drum axis, the mature and easily threshed grains are first separated from the first concave sieve 4, which mainly receives the threshing mixture from the short-toothed rods 1-5. Meanwhile, the less mature and difficult-to-thresh crops pass through the second concave sieve 5 and enter the cleaning device under the strong impact of the nail teeth 1-6.

[0122] Figure 2The image shows a harvester, including the threshing and separating device, and also including a stalk crushing device 7, a forward speed monitoring device 8, a feed rate prediction device 11, a conveying trough 13, a forward speed adjustment device 14, a grain entrainment loss monitoring sensor 15, a grain breakage rate monitoring device, a guide bar angle monitoring sensor, a blade gap monitoring sensor, a rotating blade 7-1 fixed shaft speed sensor, a rotating blade 7-1 fixed shaft torque measuring sensor, and a control device;

[0123] The threshing and separating device 10 is connected to the feeding end of the conveying trough 13 and the output end of the threshing and separating device 10 is connected to the stalk crushing device 7. The control device is connected to the stalk crushing device 7, the forward speed monitoring device 8, the threshing and separating device 10, the feeding amount prediction device 11, the forward speed adjustment device 14, the grain entrainment loss monitoring sensor 15, and the grain breakage rate monitoring device.

[0124] The forward speed monitoring device 8 is used to monitor the rotational speed of the track drive wheels of the combine harvester; the feed rate prediction device 11 is equipped with a feed rate prediction model to predict the feed rate prediction value at the next moment; the forward speed adjustment device 14 is used to adjust the forward speed of the combine harvester; the grain entrainment loss monitoring sensor 15 is used to monitor the grain entrainment loss rate; the grain breakage rate monitoring device is used to monitor the grain breakage rate; the guide bar angle monitoring sensor is used to monitor the angle of the guide bar 2-3 in the second threshing and separation section; the blade gap monitoring sensor is used to monitor the gap between the rotating blade 7-1 and the fixed blade 7-6 in the stalk crushing device 7; the speed sensor is used to monitor the rotational speed of the fixed shaft of the rotating blade 7-1 in the stalk crushing device 7; the control device will use the real-time feed rate prediction value at the next moment, the grain breakage rate monitored by the grain breakage rate monitoring device, and the grain entrainment loss rate monitored by the grain entrainment loss monitoring sensor 15, as well as the forward speed of the combine harvester at the current moment, the angle of the guide bar 2-3 in the second threshing and separation section, and the stalk crushing device... The gap between the rotating blade 7-1 and the fixed blade 7-6 in the threshing device 7, and the rotational speed of the fixed shaft of the rotating blade 7-1 are input into the model prediction controller for processing. This yields predicted values ​​for the next moment's forward speed of the combine harvester, the angle of the guide bar 2-3 in the second threshing and separating section of the threshing and separating device 10, the gap between the rotating blade 7-1 and the fixed blade 7-6 in the stalk crushing device 7, and the rotational speed of the fixed shaft of the rotating blade 7-1. The forward speed regulating device 14 is controlled to adjust the forward speed of the combine harvester, the guide bar angle regulating device 6 is controlled to adjust the angle of the guide bar 2-3 in the second threshing and separating section, the linear telescopic electric cylinder 7-3 in the stalk crushing device 7 is controlled to adjust the gap between the rotating blade 7-1 and the fixed blade 7-6, and the hydraulic continuously variable speed motor 7-2 in the stalk crushing device 7 is controlled to adjust the rotational speed of the moving blade shaft. Thus, the forward speed regulating device 14, the angle of the guide bar 2-3 in the second threshing and separating section, the gap between the rotating blade 7-1 and the fixed blade 7-6 in the stalk crushing device 7, and the rotational speed of the fixed shaft of the rotating blade 7-1 are adjusted to the predicted values ​​for the next moment.

[0125] The feed rate prediction device includes an image acquisition unit, a detection unit, and a feed rate prediction unit. The image acquisition unit is used to acquire images of the area to be harvested. The detection unit is used to acquire the forward speed of the combine harvester and the header height. The feed rate prediction unit extracts plant height, plant density, and cutting width from the images acquired by the image acquisition unit, and inputs the plant density, plant height, cutting width, header height, and harvester forward speed of the area to be harvested into the feed rate prediction model constructed based on the RBF-MLP multilayer neural network to predict the feed rate of the combine harvester at the next moment.

[0126] like Figure 21As shown, preferably, the feed rate prediction unit uses the YoloX target detection algorithm to process the RGB images acquired by the image acquisition unit 1 at different times during the operation of the combine harvester to obtain the plant density and feed density, specifically including the following steps:

[0127] S1.1: Collect RGB images of the harvestable area at different times during the actual harvesting process of the combine harvester, and use Labelimg software to annotate each image to mark all rice ears in the image;

[0128] S1.2: After obtaining enough image samples, randomly divide all images, with 80% assigned to the training set and 20% to the test set;

[0129] S1.3: Import the training set into the YOLOX training program for training. After training, a rice panicle detection model is obtained. The rice panicle detection model is tested using a test set to ensure the detection effect of the plant panicle detection model.

[0130] S1.4: Use a checkerboard pattern to calibrate the size of the sampled image and determine the actual size of the camera sampling area;

[0131] S1.5: During actual detection, the RGB image acquired by the image acquisition unit 1 is input into the rice panicle detection model. The rice panicle detection model detects the number of rice panicles in this RGB image and obtains the plant density p based on the actual area of ​​the sampling area.

[0132] S1.6: Collect samples of intact and mature rice plants from areas with uniform rice growth, and analyze the relationship between plant height h and cutting height h using these samples. c The influence of changes on the mass M of a single rice plant was investigated, and a rice plant mass equation was derived to determine the plant height h and the cutting height h. c The mathematical relationship between the rice plant mass M and the weight of a single rice plant is shown below:

[0133]

[0134] Where, m e The average mass of the rice panicle is kg, h is the plant height, and m is the mass per unit length of the stem, kg / m.

[0135] S1.7: Based on the plant density p and the rice plant mass equation M in the sampling area, the feed density ρ (kg / m³) of the combine harvester in the harvesting area is obtained. 2 ).

[0136] ρ=p·M

[0137] In step S1.6, to determine the effect of cutting height on rice plant quality, in one embodiment of the present invention, preferably, 100 complete rice plant samples are taken, and the height of each plant is measured; the plant is cut at a position 5cm above the ground, and the mass of each plant is measured. From the cutting position, 30 1cm long sections of stem are cut upwards from each plant as samples, and the mass of each sample is measured. This yields the rice plant mass at different cutting heights. To reduce errors caused by morphological differences in rice panicles, the mass of the rice panicle of each sample plant is measured, and the average mass is taken as the rice panicle mass m. e The rice plant quality equation was fitted based on the measured data.

[0138] like Figure 22 As shown, in one embodiment of the present invention, preferably, acquiring an image of the area to be harvested and extracting the plant height from the image specifically includes the following steps:

[0139] S2.1: Initialize image acquisition unit 1, acquire depth image and RGB image of the area to be harvested in front of the combine harvester, and perform alignment processing on them;

[0140] S2.2: Set a region of interest of size 200×150 in the middle of the RGB image, perform HSV transformation on the region of interest, extract the pixel region in the range of [10,50,100] and [40,120,250], and project the region onto the corresponding depth image;

[0141] S2.3: Call the get_intrinsics and get_extrinsics_to functions to obtain the camera's intrinsic and extrinsics parameters, and call the rs2_get_depth_scale function to obtain the depth scale of the depth sensor. Obtain the coordinates of the corresponding pixel based on the camera's intrinsic and extrinsics parameters, extract the depth value of the corresponding pixel from the depth image, and save the extracted pixel depth value.

[0142] S2.4: Average the depth values ​​of each saved pixel, multiply the average depth value by the depth scale, and obtain the straight-line distance l from the plant spike to the camera within the region of interest. x The depth camera is installed at the highest point in front of the harvester. Since the image acquisition unit 1 is installed with a certain pitch angle θ, the vertical distance h2 from the image acquisition unit 1 to the crop outline boundary can be obtained by trigonometric function calculation.

[0143] S2.5: The difference between the installation height h1 of the image acquisition unit 1 and the vertical distance h2 from the camera to the crop outline boundary can be used to obtain the plant height h in the harvest area;

[0144] h = h1 - l x sinθ

[0145] Among them, l x sinθ is the height from the camera to the spike layer of the plant.

[0146] The preferred flowchart for obtaining the cutting width of the harvested area is as follows: Figure 23 As shown, it includes the following steps:

[0147] S3.1: Read a frame of depth image from the data stream of the RealSense depth camera, use a linear scanning algorithm to traverse all pixels in the depth image, find the two adjacent pixels with the largest depth difference, and save them to a list;

[0148] S3.2: Call the get_intrinsics and get_extrinsics_to methods to obtain the camera's intrinsic and extrinsic parameters, convert the two adjacent pixels with the largest depth difference into points in the camera coordinate system, convert the pixel coordinates into point coordinates in the camera coordinate system, and extract the coordinate information of all points on the x-axis.

[0149] S3.3: Schematic diagram of the cutting width detection is shown below. Figure 24 As shown, the average of the x-axis coordinates of all the points obtained is the vertical distance from the harvest boundary to the camera. Based on the camera's installation position, the cutting width w of the combine harvester at that moment can be obtained.

[0150] Specifically, to obtain the cutting width of the combine harvester during operation, it is only necessary to convert the extracted boundary pixels into coordinates in the camera coordinate system. The coordinates of the boundary points on the x-axis in the camera coordinate system are the distances of the harvest boundary points relative to the camera. Since the camera's installation position relative to the combine harvester is fixed, adding this distance to the distance of the camera relative to the other side of the header divider gives the current cutting width of the combine harvester.

[0151] like Figure 25 As shown, preferably, establishing the RBF-MLP multilayer neural network includes the following steps:

[0152] In a manner similar to combining perceptrons to form a multilayer perceptron, RBF approximation is incorporated in each hidden layer, aiming to combine the advantages of the nonlinear approximation capability of RBF neural networks with the advantages of multilayer perceptrons in capturing complex features.

[0153] S4.1: Introduce nonlinear transformations in the existing RBF-MLP network: In each hidden layer of the RBF-MLP network except the first layer, each component of the output of the previous hidden layer is processed by a univariate RBF with a scalar center and trainable, and then RBF approximation is performed to generate the output of the current layer, so as to enhance the representation and adaptability of the network.

[0154] Furthermore, in RBF-MLP, the input data undergoes a non-linear transformation via RBF approximation, rather than the linear transformation performed in traditional MLPs. This introduces additional non-linearity into the multilayer perceptron, enabling the use of fewer hidden layers and neurons, thereby reducing training costs.

[0155] Compared to traditional multilayer perceptrons, RBF-MLP uses center- and width-trainable Gaussian functions as activation functions for hidden neurons.

[0156] S4.2: Multiple hidden layers are set between the input and output layers of the RBF-MLP network, which introduces nonlinear transformation, enabling RBF-MLP to have deep learning capabilities;

[0157] S4.3: Calculate the output of the first hidden layer in S4.2; each hidden layer involves two operations, but the operations of the first hidden layer are different from those of subsequent hidden layers:

[0158] S4.3.1: In the first hidden layer, the input variables are first fed into the same Gaussian function a as in the RBF neural network. (1) middle,

[0159]

[0160] The Gaussian function is multivariate, and the dimension of its center is the same as the dimension of the input variable x.

[0161] x is the input variable. It is a multivariate Gaussian function;

[0162] S4.3.2: Perform RBF approximation; that is, apply the linear weighted sum of all Gaussian functions given in the formula of S4.3.1.

[0163] S4.3.3: Calculate the output of this hidden layer:

[0164]

[0165] Among them, W (1) Let z be the weight matrix of the first hidden layer. (1) This is the output of the first hidden layer. It is a multivariate Gaussian function. is the weight node in the first layer of the network, n0 is the number of kernels in the neural network layer, and j and i are identifiers, which are any integers in the range of 1 to n0;

[0166] S4.4: Calculate the outputs of all hidden layers in S4.2 except for the first hidden layer:

[0167] S4.4.1: The j-th feature of layer l-1 It is input into a univariate radial basis function a l middle:

[0168]

[0169] Let j be the j-th feature of the l-1 hidden layer, L be the number of the hidden layers, and l be the nth hidden layer.

[0170] S4.4.2: Perform RBF approximation, i.e., apply the linear weighted sum of all Gaussian functions given in the above formula;

[0171] S4.4.3: Calculate the output z of all hidden layers except the first hidden layer. l .

[0172]

[0173] Finally, the output layer outputs...

[0174]

[0175] Among them, a (L-1) W is the output of the hidden layer neurons in layer L-1. (L) Let b be the weight matrix of the output layer. (L) This is the bias for the output layer.

[0176] The RBF-MLP multilayer neural network utilizes RBF approximation in the hidden layers and simplifies the basis functions by replacing the original multivariate Gaussian functions with univariate radial basis functions. This significantly reduces the dimensionality of the basis functions and the complexity of the network, thereby simplifying the training process.

[0177] Preferably, establishing a combine harvester feed rate prediction model includes the following steps:

[0178] S5.1: Construct the training dataset: The dataset consists of plant height, plant density, and cutting width obtained from image acquisition unit 1 at different times during the operation of the combine harvester, as well as the forward speed and header height of the combine harvester obtained from the detection unit.

[0179] S5.2: Import the dataset into the RBF-MLP multilayer neural network for training to derive the feed rate prediction model: Import the constructed dataset into the established RBF-MLP multilayer neural network for training. The mean squared error (MSE) is used as the loss function during the training process of the feed rate prediction model.

[0180]

[0181] Where n is the number of samples, y i Let i be the true value of the i-th sample. The corresponding predicted value;

[0182] Obtain a batch of training data from the data loader, including input features and corresponding target labels;

[0183] First, the input feature x passes through an RBF layer, which contains multiple radial basis function kernels. This represents the i-th kernel function. Each kernel function calculates the distance between the input data x and the kernel using Euclidean distance d. i :

[0184] d i =||xu i ||2

[0185] Among them, u i It is the center of the i-th kernel, x is the input feature, and d i It is the Euclidean distance;

[0186] Then, the Gaussian kernel function is applied to calculate the output value of each kernel:

[0187]

[0188] Where, σ i It is the standard deviation of the i-th nucleus, RBF i d is the output value of the Gaussian kernel. i It is the Euclidean distance;

[0189] The output of the RBF layer is RBF. i These values ​​become the input to the first layer of the MLP. They pass through a fully connected layer and are processed by the ReLU activation function. Let the first layer of the MLP have N neurons, where z... i Let represent the input of the i-th neuron. Then:

[0190]

[0191] Where M is the number of kernels in the RBF layer, w ij It is the weight connecting the j-th kernel of the RBF layer and the i-th neuron of the first layer of the MLP, b iIt is the deviation of the i-th neuron;

[0192] Next, the ReLU activation function is applied to obtain the intermediate representation h. i :

[0193] h i =max(0, z) i )

[0194] h i h is the input to subsequent layers of the MLP. i It is an intermediate transition function, called an intermediate representation. z i This represents the input to the i-th neuron;

[0195] The intermediate representations of the MLP are mapped to the final output through the second layer of the MLP. The network output is then compared with the actual target label, and the mean squared error (MSE) loss is calculated.

[0196] Calculate the gradient of the loss relative to the model parameters, and propagate the gradient from the loss back to each layer of the network through backpropagation. For example, if the error of the (l+1)th hidden layer has been obtained, then the error of the lth hidden layer is:

[0197]

[0198] δ l Let z be the error of the l-th hidden layer. l For the input of the neurons in the l-th hidden layer, δ l+1 Let z be the error of the (l+1)th hidden layer. l+1 This is the input to the neurons in the (l+1)th hidden layer;

[0199] again

[0200] z l+1 =W l+1 σ(W l a l-1 +b l )+b l+1

[0201] W l+1 Let b be the weight matrix of the (l+1)th hidden layer. l+1 For the bias of the (l+1)th hidden layer, a l-1 This is the output of the result of the (l-1)th hidden layer after passing through the activation function;

[0202] therefore

[0203] δ l =(W l+1 ) T δ l+1 ⊙σ′(W l a l-1 +bl )

[0204] Based on the error δ of the output layer L Using the above formula, δ can be obtained sequentially. L-1 δ L-1 ,...,δ 2 The first layer is the input layer and has no error. After calculating the error of each layer, the gradient of the error function C with respect to the parameters of each layer is calculated:

[0205]

[0206]

[0207] Where T is the transpose symbol;

[0208] Finally, the parameters of each layer are updated using gradient descent:

[0209]

[0210]

[0211] Where η is the learning rate during training, and W l Let b be the weight matrix of the l-th hidden layer. l Let C be the bias of the l-th hidden layer, and C be the mean square error function (MSE).

[0212] The Adam optimizer is used to update the model parameters based on the calculated gradients in order to reduce the value of the loss function;

[0213] Repeat the above steps to iterate through all batches in the dataset until one epoch is completed;

[0214] After each epoch, record the loss value of the last batch in that epoch for subsequent loss curve plotting, such as... Figure 26 , 27 As shown in Figure 28, compared to single RBF-nn and MLP, RBF-MLP converges faster. This means that the model can reach a lower training loss value in a shorter time, saving training time and computational resources. Moreover, faster convergence helps to avoid overfitting on the training data. Figure 28 The curves shown illustrate that the loss value decreases rapidly within the same time period, leading to the conclusion that the RBF-MLP neural network has good approximation accuracy and fast training convergence speed.

[0215] The feed volume prediction model is as follows:

[0216]

[0217] Where Q is the predicted feed amount, v is the forward speed, and h is the forward speed. c h is the height of the cutting platform, w is the plant height, and ρ is the cutting width.

[0218] ρ=p·M

[0219] Where p is the plant density and M is the rice plant mass equation;

[0220]

[0221] Where, m e denoted as , where h is the average mass of the rice panicle; h is the plant height; and m is the mass per unit length of the stem.

[0222] The plant height h in the harvesting area is:

[0223] h = h1 - l x sinθ

[0224] Where h1 is the installation height of the image acquisition unit, l x Let θ be the straight-line distance from the spikelet of the plant within the region of interest to the camera, and l be the pitch angle of the image acquisition unit. x sinθ is the height from the camera to the spike layer of the plant.

[0225] Preferably, the detection unit includes an angle sensor and a Hall sensor; the angle sensor is used to measure the height of the combine harvester's header; the Hall sensor is used to measure the forward speed of the combine harvester.

[0226] Preferably, the feed rate prediction unit includes a Jetson NX embedded processor. The Jetson NX embedded processor is used to predict the feed rate of the combine harvester at the next moment.

[0227] Preferably, the image acquisition unit includes a RealSense depth camera 11.

[0228] The feed rate prediction device obtains the plant density of the harvesting area based on the YoloX target detection algorithm, the plant height and cutting width of the harvesting area based on depth vision perception technology, the header height based on the angle sensor, and the forward speed of the combine harvester based on the Hall sensor. The angle sensor and Hall sensor signals are processed by the CAN communication module and sent to the JetsonNX embedded processor. The JetsonNX embedded processor calls up information such as forward speed, header height, plant height, cutting width, and feed density, and uses the feed rate prediction model based on the RBF-MLP multilayer neural network to predict the feed rate information of the combine harvester at the next moment.

[0229] In one embodiment of the present invention, preferably, it further includes a cutting platform height monitoring mechanism, such as... Figure 20 As shown, the header height monitoring mechanism includes a connector 13-1, a fisheye bearing 13-2, a metal rod 13-3, a fisheye bearing 13-4, a connecting plate 13-5, an angle sensor 13-6, and a connector 13-7. Connector 13-1 is bolted to the side wall of the combine harvester's conveyor trough. The metal rod 13-3 is connected to connector 13-1 and connecting plate 13-5 via fisheye bearings 13-2 and 13-4. The end of connecting plate 13-5 is welded to the angle sensor's rotating shaft. Rotation of the metal rod 13-3 can drive the angle sensor to rotate. The metal rod 13-3 can rotate a certain angle around one end of connector 13-1 and connecting plate 13-5. Connector 13-7 is fixed to the combine harvester frame. When the header height changes, connector 13-1 moves the connecting plate 13-5 upwards or downwards, thereby causing the angle sensor 13-6's rotating shaft to generate an angular displacement α. Based on the calibration results, the header lifting height h of the combine harvester can be obtained. c .

[0230] h c =k·l3sinα

[0231] Where l3 is the length of metal rod 13-3, and k is the proportionality coefficient.

[0232] Preferably, the feed prediction unit uses depth vision perception technology to process the depth images acquired by the image acquisition unit 1 at different times during the operation of the combine harvester to obtain the plant height and cutting width of the area to be harvested.

[0233] The feed rate prediction model is trained using an RBF-MLP multilayer neural network, based on a dataset consisting of feed rate information collected during actual operations, including forward speed, header height, plant height, cutting width, feed density, and manually measured parameters. When calling the feed rate prediction model for prediction, simply input the forward speed, header height, plant height, cutting width, and feed density acquired at the same time into the system, and the system will calculate the corresponding predicted feed rate value.

[0234] The feed rate prediction unit program, plant height acquisition program, and cut width acquisition program run in a Jetson NX embedded processor. Magnets are installed on the pulleys of the main working parts of the combine harvester, and the rotational speed of each working part is obtained through Hall sensor pulse signals.

[0235] The rotational speed and forward speed of the reel are adjusted through the D / A channel output control of the STM32F407 microcontroller.

[0236] like Figure 29As shown, the forward speed adjustment device 14 includes an HST push rod 14-1, an electric cylinder connector 14-2, an electric cylinder 14-3, a displacement sensor 14-4, a bracket 14-5, and a stepper motor controller. The electric cylinder 14-3 and displacement sensor 14-4 are connected to the combine harvester's HST push rod 14-1 via connector 14-2. The electric cylinder 14-3 and displacement sensor 14-4 are mounted in the combine harvester's cab via bracket 14-5. Moving the electric cylinder 14-3 moves the HST push rod 14-1. When the vehicle speed needs to be changed, the controller's D / A channel outputs a control signal. After receiving the corresponding electrical signal, the stepper motor controller drives the electric cylinder 14-3 to move, extending or shortening it by a corresponding distance. The movement distance is monitored by the displacement sensor 14-4 and fed back to the control system. When the corresponding stroke of the HST push rod 14-1 is reached, the electric cylinder 14-3 stops moving until the next control signal is generated.

[0237] When the vehicle speed needs to be changed, the controller outputs a control signal through the D / A channel, and the electric cylinder extends or shortens by the corresponding distance. The movement distance is monitored by the linear displacement sensor and fed back to the control system. When the corresponding HST push rod stroke is reached, the electric cylinder stops moving until the next control signal.

[0238] A control method for the harvester includes the following steps:

[0239] The control device processes the real-time predicted value of the feed amount for the next moment, the grain breakage rate monitored by the grain breakage rate monitoring device, the grain entrainment loss rate monitored by the grain entrainment loss monitoring sensor 15, the current forward speed of the combine harvester, the angle of the guide bar 2-3 of the second threshing and separating section, the gap between the rotating blade 7-1 and the fixed blade 7-6 in the stalk crushing device 7, and the rotation speed of the moving shaft of the stalk crushing device 7 into the input model prediction controller. This results in the predicted values ​​of the forward speed of the combine harvester, the angle of the guide bar 2-3 of the second threshing and separating section in the threshing and separating device 10, the gap between the rotating blade 7-1 and the fixed blade 7-6 in the stalk crushing device 7, and the rotation speed of the fixed shaft of the rotating blade 7-1. This allows the forward speed adjustment device 14, the angle of the guide bar 2-3 of the second threshing and separating section, the gap between the rotating blade 7-11 and the fixed blade 7-6 in the stalk crushing device 7, and the rotation speed of the fixed shaft of the rotating blade 7-1 to be adjusted to the predicted values ​​for the next moment, thereby improving threshing and separating performance and preventing blockage.

[0240] A self-organizing map neural network algorithm is applied in the control device. Based on batch training, a sensor monitoring anomaly data detection program is constructed to promptly identify abnormal data in each monitoring signal. The self-organizing map neural network is set to a 20×10 structure, with 10 coarse training iterations and 20 fine-tuning training iterations, and 3 clusters. When abnormal data is detected in each monitoring signal, it is replaced with previously monitored normal data. Preprocessing, such as filling in missing data from some sensors and data denoising, is then performed to eliminate the impact of random and uncertain factors on subsequent data analysis.

[0241] Preferably, based on the predicted feed amount Q after pretreatment, the monitoring value of the grain entrainment loss monitoring sensor 15, and the monitoring value of the grain breakage rate monitoring device, the following values ​​are calculated: u1(k) is the angle of the guide bar 2-3 in the second threshing and separation section, u2(k) is the forward speed v, u3(k) is the gap between the rotating blade 7-1 and the fixed blade 7-6 of the stalk crushing device 7, and u4(k) is the target value of the rotational speed D of the fixed shaft of the rotating blade 7-1 of the stalk crushing device 7.

[0242] In the development of control devices, model predictive controllers are constructed, which consist of three parts: a state estimator, an MPC controller, and the controlled object. For example... Figure 30 As shown, the MPC controller combines the prediction model, constraints, and objective function optimization to input the optimal control sequence of the controlled object into the controlled platform, and then inputs the current state variable observations. However, the state estimator cannot directly evaluate the observed state; the state estimation needs to be introduced into the MPC controller and further optimized to obtain the control sequence for the next time period. This process is repeated cyclically to achieve a complete control process. Based on the three elements mentioned above, the control principle of the model prediction algorithm is shown in the figure below. During the control process, there exists a desired reference trajectory. In the entire control time domain, let time k be the current time. The controller combines the current system's measurements and the prediction model to predict the output of the system in the future time domain [k, k+N], also known as the prediction time domain. By solving the optimization problem that satisfies the objective function and various constraints, the time period [k, k+N-1] is obtained as the control time domain, which is shorter than the prediction time domain. A series of control variable sequences within the time domain are obtained. The first element in this control sequence is taken as the actual control quantity of the controlled object. When the next time k+1 arrives, the above process is repeated. By continuously completing a series of constrained optimization problems in this way, continuous control over the controlled object can be achieved.

[0243] The application of model predictive control algorithms can be mainly divided into three steps.

[0244] The first step is to measure and read the current system state x(k). If the system state cannot be measured, it needs to be estimated.

[0245] The second step, based on u k u k+1 ,.....u k+N-1 To perform optimization, where u k It is the value at the current moment, u k+1 It is the value at the next moment;

[0246] The third step is to select only u within a prediction interval. k This serves as the control input for this operation.

[0247] The prediction model formula is as follows, given a discrete system described by the state-space method.

[0248] x(k+1)=Ax(k)+Bu(k) (1)

[0249] in, Let x1(k) be the system state variable at time k, x2(k) be the grain entrainment loss and the grain breakage rate, respectively, and u(k) be the system input vector. u1(k) is the angle of the guide bar 2-3 in the second threshing and separation section, u2(k) is the forward speed v, u3(k) is the gap between the rotating blade 7-1 and the fixed blade 7-6 of the stalk crushing device 7, and u4(k) is the rotational speed of the fixed shaft of the rotating blade 7-1 of the stalk crushing device 7. A is the system state matrix, and B is the input matrix.

[0250] N is the prediction interval, and the state vector prediction matrix X(k) and input vector prediction matrix U(k) at time k are:

[0251]

[0252]

[0253] Cost function:

[0254]

[0255] Among them, Q x R and F are weight coefficient matrices.

[0256] When the system's state variable x(k) is measured at time k, the system's state prediction for the entire prediction time domain is as follows:

[0257] x(k|k)=x k

[0258] x(k+1|k)=Ax(k|k)+Bu(k|k)

[0259] x(k+2|k)=Ax(k+1|k)+Bu(k+1|k)=A2 x k +ABu(k|k)+Bu(k+1|k)

[0260] Similarly,

[0261] x(k+N|k)=A N x k +A N-1 Bu(k|k)+Bu(k+N-1|k) (5)

[0262] Therefore, it is permissible.

[0263] X(k)=Mx(k)+Cu(k) (6)

[0264] in,

[0265] M = [I, A, A] 2 A N ] T (7)

[0266]

[0267] I is the identity matrix, A is the system state matrix, B is the input matrix, and N is the prediction interval.

[0268] Next, we will optimize the cost function:

[0269]

[0270] in,

[0271]

[0272] Among them, Q x R, F are weight coefficient matrices, and T represents matrix transpose.

[0273] Next, we further optimize equation (9).

[0274] J = x(k) T Gx(k)+U(k) T HU(k)+2x(k) T EU(k) (10)

[0275] in,

[0276]

[0277] Where T represents the matrix transpose.

[0278] The system should also meet the following constraints:

[0279]

[0280]

[0281] In this control system, the grain entrainment loss x1(k) and the breakage rate x2(k) are used as system state variables. min (k) is the minimum value of the system state variables, x max (k) is the maximum value of the system state variable, u min (k) is the minimum value of the system's input vector, u max (k) is the maximum value of the system's input vector.

[0282] 0≤x1(k)≤Q·P·1%

[0283] 0≤x²(k)≤Q·P·2%

[0284] Where Q is the predicted value of the feed amount of the combine harvester, kg / s, and P is the proportion of grain in the feed amount. P is determined in advance by manual measurement, and the constraint interval of the state variable x(k) at time k can be obtained.

[0285] u(k) is the system input vector, i.e. u1(k) is the angle of the guide bar 2-3 of the second threshing and separation section (unit: degrees), u2(k) is the forward speed v (m / s), u3(k) is the gap (mm) between the rotating blade 7-1 and the fixed blade 7-6 of the stalk crushing device 7, and u4(k) is the rotational speed (rpm) of the fixed shaft of the rotating blade 7-1 of the stalk crushing device 7.

[0286] 20°≤u1(k)≤50°

[0287] 0 m / s ≤ u2(k) ≤ 1.5 m / s

[0288] 5mm≤u3(k)≤20mm

[0289] 2500rpm≤u4(k)≤3000rpm

[0290] In the cost function above, there are two variables: x(k) is the state variable at time k, which has already been calculated; now, only the input vector u exists. k Since J is an unknown quantity, it can be solved by finding the minimum value of J, ultimately yielding the target values ​​of u1(k), u2(k), u3(k), and u4(k). u1(k) is the angle A of the guide bar 2-3 in the second threshing and separation section, u2(k) is the forward velocity v, u3(k) is the gap between the rotating blade 7-1 and the fixed blade 7-6 of the stalk crushing device 7, and u4(k) is the rotational speed of the fixed shaft of the rotating blade 7-1 of the stalk crushing device 7. The state variable x(k) at time k is input into the prediction model to predict the input vector u of the control system at the next time step. kThe target value of the control mechanism is minimized to ensure that the state variable x(k+1) at the next moment is minimized, thereby optimizing grain entrainment loss and grain breakage rate, improving threshing and separation performance, and preventing blockage. This invention has wide applicability and is particularly suitable for harvesting hybrid rice seed production female seeds.

[0291] It should be understood that although this specification is described according to various embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0292] The detailed descriptions listed above are merely specific illustrations of feasible embodiments of the present invention and are not intended to limit the scope of protection of the present invention. All equivalent embodiments or modifications made without departing from the spirit of the present invention should be included within the scope of protection of the present invention.

Claims

1. A threshing and separating device for rice seed harvesting, characterized in that, It includes a threshing and separating device (10); the threshing and separating device (10) includes a threshing drum (1), a threshing drum top cover (2), a stepped concave plate, and a guide bar angle adjustment device (6). The top cover (2) of the threshing drum is located on the upper part of the threshing drum (1), the stepped concave plate is located on the lower part of the threshing drum (1), the top cover (2) of the threshing drum is connected to the stepped concave plate, the threshing drum (1) is connected to the frame of the combine harvester, and the space enclosed between the threshing drum (1), the top cover (2) of the threshing drum, and the stepped concave plate is the threshing separation space, and the threshing space is a gradual threshing space; The top cover (2) of the threshing drum is provided with a guide strip on one side facing the threshing drum (1), and the guide strip angle adjustment device (6) is connected to the guide strip; The threshing drum (1) includes a threshing drum shaft (1-1), a threshing drum spiral feed head (1-2), a threshing drum transition section (1-3), a threshing drum threshing section (1-4), short-patterned bar teeth (1-5), nail teeth (1-6), a threshing drum straw discharge section (1-7), and a threshing drum straw discharge blade (1-8). The threshing drum screw feed head (1-2), threshing drum transition section (1-3), threshing drum threshing section (1-4), and threshing drum straw discharge section (1-7) pass through the threshing drum shaft (1-1) in sequence and are fixed. The threshing drum screw feed head (1-2), threshing drum transition section (1-3), threshing drum threshing section (1-4), and threshing drum straw discharge section (1-7) are connected end to end. The threshing drum spiral feed head (1-2) includes a trapezoidal frustum (1-2-1) and spiral blades (1-2-2), with the spiral blades (1-2-2) spirally arranged along the surface of the trapezoidal frustum (1-2-1); The short-toothed rod plate teeth (1-5) are spirally arranged and fixed in the front section of the threshing transition section (1-3) and the threshing section (1-4) of the threshing drum, while the nail teeth (1-6) are spirally arranged in the middle and rear sections of the threshing section (1-4) of the threshing drum. The straw-discharging blades (1-8) of the threshing drum are spirally arranged along the surface of the straw-discharging section (1-7) of the threshing drum; The cross-sectional area of ​​the threshing drum spiral feed head (1-2) and the threshing drum transition section (1-3) gradually increases along the threshing direction, so that the cross-section of the threshing drum spiral feed head (1-2) and the threshing drum transition section (1-3) forms a trapezoid. The cross-sectional area of ​​the threshing drum straw discharge section (1-7) gradually decreases along the threshing direction, so that the cross-section of the threshing drum straw discharge section (1-7) forms a trapezoid. The threshing space between the threshing drum spiral feed head (1-2), the threshing drum transition section (1-3), the threshing drum straw discharge section (1-7), the threshing drum top cover (2), and the stepped concave plate is gradually changing. The top cover (2) of the threshing drum is provided with a feeding section (2-9), a first threshing separation section (2-10), a transition section (2-11), a second threshing separation section (2-12), and a straw discharge section (2-13) along the length of the threshing drum (1) on the side facing the threshing drum (1). The cross-sectional radii of the feeding section (2-9), the first threshing separation section (2-10), and the second threshing separation section (2-12) increase sequentially, the cross-sectional area of ​​the transition section (2-11) gradually increases along the threshing separation direction, and the cross-sectional area of ​​the straw discharge section (2-13) gradually decreases along the threshing separation direction. The feeding section (2-9) is equipped with a feeding section guide bar (2-1), the first threshing and separation section (2-10) is equipped with a first threshing and separation section guide bar (2-2), the second threshing and separation section (2-12) is equipped with a second threshing and separation section guide bar (2-3), and the second threshing and separation section guide bar (2-3) is connected to the guide bar angle adjustment device (6); The stepped concave plate includes a first concave plate (4) and a second concave plate (5); The first concave plate (4) and the second concave plate (5) are arranged along the length of the threshing drum (1), and the volume of the second concave plate (5) is greater than the volume of the first concave plate (4).

2. The rice seed threshing and separating device according to claim 1, characterized in that, The guide bar angle adjustment device (6) includes a telescopic motor (2-4), a connecting plate (2-5), a push rod (2-6), a guide bar rotation shaft (2-7) for the second threshing and separation section, and a connecting rod (2-8) for the movable end of the guide bar for the second threshing and separation section. One end of the second threshing and separation section guide bar (2-3) has a through hole. The rotating shaft (2-7) passes through the through hole to install the second threshing and separation section guide bar (2-3) on the second threshing and separation section of the threshing drum top cover (2), so that the second threshing and separation section guide bar (2-3) can rotate around the rotating shaft (2-7). The other end of the second threshing and separation section guide bar (2-3) is installed on the connecting plate (2-5) through the connecting rod (2-8), so that each second threshing and separation section guide bar (2-3) can rotate around the rotating shaft (2-7). The section guide bar (2-3) is linked through the connecting plate (2-5). The telescopic motor (2-4) is installed on the top cover (2) of the threshing drum. The electric push rod of the telescopic motor (2-4) is connected to the connecting rod (2-8) through the push rod (2-6). When the electric push rod of the telescopic motor (2-4) is working, it drives the connecting plate (2-5) to move through the push rod (2-6), thereby driving the second threshing and separation section guide bar (2-3) to rotate around the rotating shaft (2-7).

3. A harvester, characterized in that, Includes the threshing and separation device as described in claim 1 or 2.

4. The harvester according to claim 3, characterized in that, It also includes a stalk crushing device (7), a forward speed monitoring device (8), a feed amount prediction device (11), a conveying trough (13), a forward speed adjustment device (14), a grain entrainment loss monitoring sensor (15), a grain breakage rate monitoring device, a guide bar angle monitoring sensor, a blade gap monitoring sensor, a rotating blade fixed shaft speed monitoring sensor and a control device for the stalk crushing device (7). The threshing and separating device (10) has its feeding end connected to the conveying trough (13) and its output end connected to the stalk crushing device (7). The control device is connected to the stalk crushing device (7), the forward speed monitoring device (8), the threshing and separating device (10), the feeding amount prediction device (11), the forward speed adjustment device (14), the grain entrainment loss monitoring sensor (15), and the grain breakage rate monitoring device, respectively. The forward speed monitoring device (8) is used to monitor the rotational speed of the track drive wheel of the combine harvester; the feed amount prediction device (11) is equipped with a feed amount prediction model to predict the feed amount at the next moment; the forward speed adjustment device (14) is used to adjust the forward speed of the combine harvester; the grain entrainment loss monitoring sensor (15) is used to monitor the grain entrainment loss rate; the grain breakage rate monitoring device is used to monitor the grain breakage rate; the guide bar angle monitoring sensor is used to monitor the angle of the guide bar (2-3) of the second threshing and separation section; the blade gap monitoring sensor is used to monitor the gap between the rotating blade (7-1) and the fixed blade (7-6) in the stalk crushing device (7); the stalk crushing device (7) moving blade shaft rotation sensor is used to monitor the rotation of the stalk crushing device (7) moving blade shaft. The control device processes the real-time predicted feed rate for the next moment, the grain breakage rate monitored by the grain breakage rate monitoring device and the grain entrainment loss rate monitored by the grain entrainment loss monitoring sensor (15), as well as the current forward speed of the combine harvester, the angle of the guide bar (2-3) of the second threshing and separating section, the gap between the rotating blade (7-1) and the fixed blade (7-6) in the stalk crushing device (7), and the rotational speed of the fixed shaft of the rotating blade of the stalk crushing device (7) into the model prediction controller to obtain the next moment's forward speed of the combine harvester, the angle of the guide bar (2-3) of the second threshing and separating section in the threshing and separating device (10), the gap between the rotating blade (7-1) and the fixed blade (7-6) in the stalk crushing device (7), and the rotational speed of the fixed shaft of the rotating blade of the stalk crushing device (7). The predicted value of the rotational speed of the fixed shaft of the rotating blade is obtained, and the forward speed adjustment device (14) is used to adjust the forward speed of the combine harvester, the guide bar angle adjustment device (6) is used to adjust the angle of the guide bar (2-3) of the second threshing and separating section, the linear telescopic electric cylinder (7-3) of the stalk crushing device (7) is used to adjust the gap between the rotating blade (7-1) and the fixed blade (7-6), and the hydraulic stepless speed regulating motor (7-2) of the stalk crushing device (7) is used to adjust the rotational speed of the fixed shaft of the rotating blade, thereby adjusting the forward speed adjustment device (14), the angle of the guide bar (2-3) of the second threshing and separating section, the gap between the rotating blade (7-1) and the fixed blade (7-6) in the stalk crushing device (7), and the rotational speed of the fixed shaft of the rotating blade of the stalk crushing device (7) to the predicted value of the next moment.

5. A control method for a harvester according to claim 4, characterized in that, Includes the following steps: The control device processes the following data in real time: the feed prediction value for the next moment obtained from the feed prediction model of the feed prediction device (11); the grain breakage rate monitored by the grain breakage rate monitoring device and the grain entrainment loss rate monitored by the grain entrainment loss monitoring sensor (15); the forward speed of the combine harvester at the current moment; the angle of the guide strip (2-3) of the second threshing and separating section; the gap between the rotating blade (7-1) and the fixed blade (7-6) in the stalk crushing device (7); and the rotational speed of the fixed shaft of the rotating blade of the stalk crushing device (7). This data is then used to obtain the next moment's feed prediction value. The forward speed of the harvester, the angle of the guide bar (2-3) of the second threshing and separating section in the threshing and separating device (10), the gap between the rotating blade (7-1) and the fixed blade (7-6) in the stalk crushing device (7), and the predicted value of the rotation speed of the fixed shaft of the rotating blade in the stalk crushing device (7) are used to adjust the forward speed regulating device (14), the angle of the guide bar (2-3) of the second threshing and separating section, the gap between the rotating blade (7-1) and the fixed blade (7-6) in the stalk crushing device (7), and the rotation speed of the moving blade shaft of the stalk crushing device (7) to the predicted value of the next moment, thereby improving the threshing and separating performance and preventing blockage.

6. The control method for the harvester according to claim 5, characterized in that, The feed volume prediction model is as follows: Where Q is the predicted feed amount, For forward speed, For the height of the cutting platform, Where w is the plant height and w is the cutting width. Feed density; Where p is the plant density and M is the rice plant mass equation; in, The average weight of the rice panicle; Plant height Mass per unit length of stem; Plant height in the harvesting area for: in, The installation height of the image acquisition unit. The straight-line distance from the spikelet of the plant to the camera within the region of interest. The pitch angle for installing the image acquisition unit. The height of the camera from the spike layer of the plant.

7. The control method for the harvester according to claim 5, characterized in that, The model prediction controller is an MPC controller, and the following prediction model is provided: in, Let x1(k) be the system state variable at time k, x2(k) be the grain entrainment loss, x2(k) be the grain breakage rate, and u(k) be the system input vector. , It is the angle A of the guide bar (2-3) in the second threshing and separation section. It is the forward velocity v. It is the gap between the rotating blade (7-1) and the fixed blade (7-6) of the stalk crushing device (7). The rotating blade of the stalk crushing device (7) has a fixed shaft rotation speed, y(k) is the output of the system, A is the system state matrix, and B is the input matrix; and The following constraints must be met: in, It is the minimum value of the system state variables. It is the maximum value of the system state variables. It is the minimum value of the system's input vector. It is the maximum value of the system's input vector; make , The following constraints must be met: Among them, grain entrainment loss breakage rate As a system state variable, Q is the predicted value of the combine harvester's feed amount in kg / s, and P is the proportion of grain in the feed amount; make , , The following constraints must be met: 0mm

Images