Cereal grain flow monitoring method based on cereal grain flow monitoring device

Abstract

The invention relates to a cereal grain flow monitoring method based on a cereal grain flow monitoring device, and the method comprises the steps: obtaining a three-dimensional force vector sequence, the three-dimensional force vector sequence comprises a three-dimensional force vector of each moment in a plurality of moments in a time window used for monitoring the cereal grain flow, and the three-dimensional force vector of each moment in the time window used for monitoring the cereal grain flow is obtained in the cereal grain transportation period; the stress plate is stressed by a force related to cereal grains, so that the magnetoelastomer monomer is stressed, the magnetoelastomer monomer is deformed due to the stress of the magnetoelastomer monomer, the magnetic field of the magnetoelastomer monomer is changed due to the deformation of the magnetoelastomer monomer, and the magnetic field array acquisition circuit outputs a three-dimensional force vector according to the magnetic field change of the magnetoelastomer monomer; for each three-dimensional force vector in the three-dimensional force vector sequence, synthesizing forces in all directions in the three-dimensional force vector into a single-valued impact strength index corresponding to the three-dimensional force vector to obtain a single-valued impact strength index sequence; and determining a grain grain flow prediction result according to the single-valued impact strength index sequence.

Description

Technical Field

[0001] This application belongs to the field of grain flow monitoring technology, specifically relating to a method for monitoring grain flow based on a grain flow monitoring device. Background Technology

[0002] Grain flow rate refers to the mass or volume of grain output from the cleaning system or conveying device per unit time during combine harvester operation. It is a crucial parameter for measuring harvesting efficiency and yield. Real-time monitoring of grain flow rate not only allows for precise tracking of operational progress and crop output, enabling yield monitoring and field variation analysis, but also provides a basis for adaptive adjustment of key parameters such as header height and travel speed, improving operational efficiency and reducing feed losses. Therefore, grain flow rate is essential information that needs to be obtained during combine harvester operations. How to monitor grain flow rate has become a technical problem that needs to be solved. Summary of the Invention

[0003] Based on the above analysis, embodiments of this application provide a method for monitoring grain flow based on a grain flow monitoring device, in order to solve the problem of how to monitor grain flow.

[0004] This application provides a method for monitoring grain flow based on a grain flow monitoring device, the method comprising: A three-dimensional force vector sequence is obtained, which includes the three-dimensional force vector at each of multiple moments within a time window for monitoring grain flow. During the transport of grain, the force plate is subjected to forces related to the grain, causing the magnetoelastic monomer to be stressed. The stress on the magnetoelastic monomer causes deformation of the magnetoelastic monomer, and the deformation of the magnetoelastic monomer causes a change in the magnetic field of the magnetoelastic monomer. The magnetic field array acquisition circuit outputs the three-dimensional force vector based on the change in the magnetic field of the magnetoelastic monomer. For each three-dimensional force vector in the three-dimensional force vector sequence, the forces in all directions of the three-dimensional force vector are combined into a single-value impact strength index corresponding to the three-dimensional force vector to obtain a single-value impact strength index sequence. The predicted grain flow rate is determined based on the single-value impact intensity index sequence.

[0005] Beneficial effects: The grain flow monitoring method provided in this application utilizes a force plate, a magnetoelastic array, and a magnetic field array acquisition circuit to accurately measure the force generated by the grains during transport, represented by a three-dimensional force vector. By using the three-dimensional force vector at each of multiple moments within a time window for monitoring grain flow, the grain flow rate is determined, thus achieving grain flow monitoring. Compared to impulse-based grain flow monitoring, the method provided in this application does not rely on changes in impact force, avoiding the problems of susceptibility to vibration and impurity interference encountered in impulse-based grain flow monitoring. This improves the stability and robustness of grain flow monitoring.

[0006] The grain flow monitoring method provided in this application is based on a grain flow monitoring device with a force plate. This takes into account that in a grain flow consisting of a certain number of grains, some debris is unavoidable. Magnetoelastic monomers have the ability to adsorb small objects in contact with them. If the grain flow monitoring device does not have a force plate, during grain transport, forces related to the grains, such as friction, act directly on the magnetoelastic monomer. Debris from the grain flow will adhere to the magnetoelastic monomer, making the force sensed by the magnetoelastic monomer inaccurate. This leads to inaccurate changes in the magnetic field of the magnetoelastic monomer, resulting in inaccurate force detection and ultimately a decrease in the accuracy of the determined grain flow rate. The grain flow monitoring device features a force-bearing plate positioned above the magnetic elastomer monomer. Forces related to the grains act directly on this plate, preventing grain debris from adhering to the magnetic elastomer monomer and thus improving the accuracy of the determined grain flow rate. Furthermore, the added force-bearing plate allows the device to detect even high-volume grains stacked on top, further enhancing the accuracy of the determined flow rate.

[0007] The grain flow monitoring method provided in this application considers the forces on magnetoelastic monomers at different times within the time window for monitoring grain flow, i.e., the forces represented by three-dimensional force vectors. The forces on magnetoelastic monomers at different times within the time window for monitoring grain flow all participate in the prediction of grain flow. Therefore, a relatively rich set of forces on magnetoelastic monomers can be used to predict grain flow, thereby improving the accuracy of the determined grain flow.

[0008] The grain flow rate monitoring method provided in this application utilizes a three-dimensional force vector to represent forces in three directions. When determining the grain flow rate, it considers the forces acting on the magnetoelastic monomer in multiple directions (i.e., three directions), comprehensively taking into account the force conditions of the magnetoelastic monomer. Finally, it uses a single-value impact strength index, synthesized from the forces in the three-dimensional force vector, to determine the grain flow rate. Therefore, it comprehensively considers the force conditions of the magnetoelastic monomer when determining the grain flow rate, improving the accuracy of the determined grain flow rate.

[0009] The grain flow monitoring method provided in this application is based on a grain flow monitoring device in which the magnetoelastic monomer can include a soft magnetic layer, a permanent magnet layer, and a strain layer. The permanent magnet layer generates a permanent magnet magnetic field, the soft magnetic layer shields external magnetic field interference, prevents the permanent magnet magnetic field from dissipating, and strengthens the permanent magnet magnetic field of the magnetoelastic monomer, and the strain layer receives the force applied to the magnetoelastic monomer. The magnetoelastic monomer possesses soft-magnetic-permanent magnet composite enhancement characteristics and tangential sensitivity. By setting the soft magnetic layer, permanent magnet layer, and strain layer, a magneto-mechanical multi-coupling mechanism can be formed, improving the response sensitivity, measurement accuracy, and stability of the grain flow monitoring device under minute stress changes, and extending the measurement range. The permanent magnet layer can incorporate a controllable magnetization direction and magnetic field distribution design, enabling the magnetoelastic monomer to generate regular magnetic field changes during deformation under stress. This can be combined with a magnetic field array acquisition circuit including a three-dimensional Hall element array to achieve high-precision dynamic monitoring of grain flow. The soft magnetic layer effectively shields external magnetic interference, suppresses magnetic field dissipation, and improves anti-interference performance. The strain layer improves sensitivity under small loads through porous microstructure and can adjust stiffness to achieve customization of range and sensitivity. The grain flow monitoring device has high reliability, real-time performance, and environmental adaptability, improving the applicability of the grain flow monitoring method provided in this application embodiment.

[0010] The grain flow monitoring device based on the grain flow monitoring method provided in this application embodiment can be deployed in any of a variety of ways, thereby covering key flow locations such as screw conveyors and elevators. This makes the grain flow monitoring device based on the grain flow monitoring method provided in this application embodiment easy to apply to combine harvesters, improving the applicability of the grain flow monitoring method provided in this application embodiment. By transmitting the grain-related force to the magnetoelastic array through a force plate, contact-type, high-sensitivity grain flow monitoring is achieved.

[0011] The grain flow monitoring method provided in this application utilizes a random forest model and a long short-term memory (LSTM) network to predict grain flow. Random forests possess the ability to handle high-dimensional inputs, model nonlinear relationships, and perform feature selection, while LSTMs can learn the evolution of signals over time. The combination of these two methods improves the accuracy and robustness of grain flow monitoring, making it suitable for agricultural environments with large signal fluctuations and complex crop types. Furthermore, the random forest and LSTM networks can be designed with lightweight structures to meet real-time requirements, achieving stable, efficient identification and dynamic monitoring of grain flow. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings.

[0013] Figure 1 This is a schematic diagram of an example structure of a magnetoelastic array in a grain flow monitoring device upon which the grain flow monitoring method provided in the embodiments of this application is based; Figure 2 This is a schematic diagram of an example architecture in which the grain flow monitoring method provided in this application is applied when the first device deployment method is adopted; Figure 3 This is a schematic diagram of the square hole in the casing of the screw conveyor when the first device layout is used; Figure 4 This is a schematic diagram of the components related to the grain flow monitoring device on which the grain flow monitoring method provided in this application embodiment is based when the first device deployment method is adopted; Figure 5 This is a schematic diagram of an example architecture in which the grain flow monitoring method provided in this application is applied when the second device deployment method is adopted; Figure 6 This is a schematic diagram of the square hole in the casing of the screw conveyor when the second device layout is adopted; Figure 7 This is a schematic diagram of the components related to the grain flow monitoring device on which the grain flow monitoring method provided in this application embodiment is based when the second device deployment method is adopted; Figure 8 This is a schematic diagram of an example architecture in which the grain flow monitoring method provided in this application is applied when a third device deployment method is adopted; Figure 9This is a schematic diagram of the square opening on the outer casing of the elevator when the third device layout is adopted; Figure 10 This is a schematic diagram of the components related to the grain flow monitoring device on which the grain flow monitoring method provided in this application embodiment is based when the third device deployment method is adopted. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. It should be noted that, unless otherwise specified, the implementation methods and features in the implementation methods in this disclosure can be combined, separated, interchanged, and / or rearranged. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] In the accompanying drawings, the dimensions and relative dimensions of components may be exaggerated for clarity and / or descriptive purposes. When exemplary embodiments can be implemented differently, a specific process sequence may be performed in a different order than that described. For example, two consecutively described processes may be performed substantially simultaneously or in the reverse order of their description. Furthermore, the same reference numerals denote the same components.

[0016] When a component is referred to as being "on" or "above" another component, "connected to," or "joined to" another component, the component may be directly on, directly connected to, or directly joined to the other component, or there may be intermediate components. However, when a component is referred to as being "directly on" another component, "directly connected to," or "directly joined to" another component, there are no intermediate components. Therefore, the term "connection" can refer to a physical connection, an electrical connection, etc., and may or may not have intermediate components.

[0017] For descriptive purposes, this disclosure may use spatial relative terms such as “top,” “bottom,” “below,” “under,” “under,” “below,” “above,” “above,” “higher,” etc., which are relative to components, to describe the relationship between one component and another (other) component as shown in the accompanying drawings.

[0018] The terminology used herein is for the purpose of describing particular embodiments and is not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms “a” and “the” are intended to include the plural forms as well. Furthermore, when the terms “comprising” and / or “including” and variations thereof are used in this specification, it indicates the presence of the stated features, integrals, steps, operations, parts, components, and / or groups thereof, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, parts, components, and / or groups thereof. It should also be noted that, as used herein, the terms “substantially,” “about,” and other similar terms are used as approximate terms rather than as terms of degree, thus explaining the inherent biases in measurements, calculated values, and / or provided values ​​that would be recognized by one of ordinary skill in the art.

[0019] It should be noted that the grain flow monitoring method provided in this application is applied to a combine harvester. The grain flow monitoring device provided in this application is installed on the combine harvester. The grain flow monitoring method provided in this application can be executed by computer equipment on the combine harvester.

[0020] The grain flow monitoring device based on the grain flow monitoring method provided in this application includes: a force plate, a magnetoelastic array, a base plate, and a magnetic field array acquisition circuit. The force plate is disposed above the magnetoelastic array, and the magnetoelastic array is disposed between the force plate and the base plate. The base plate is used to fix the grain flow monitoring device. The magnetoelastic array includes: a magnetoelastic single unit.

[0021] The magnetoelastic array is connected to the load-bearing plate, either directly or via connecting components. The magnetoelastic array is also connected to the base plate, again either directly or via connecting components.

[0022] As an example, the magnetic elastomer array is connected to the load-bearing plate by means of adhesive tape.

[0023] As an example, multiple magnetoelastic monomers in a magnetoelastic array are arranged uniformly in a 3-row × 3-column configuration.

[0024] During the transport of grains, the grains come into contact with a force-bearing plate, subjecting the plate to forces associated with the grains. For example, the forces associated with the grains as they pass over the plate include frictional forces generated by the grains passing over the plate. The force-bearing plate is connected to a magnetoelastic array, and the force applied to the plate causes forces to be applied to the individual magnetoelastic units within the array. Alternatively, the force-bearing plate transmits the applied forces to the individual magnetoelastic units in the array. The individual magnetoelastic units in the array deform due to these forces. In other words, the force applied to the force-bearing plate causes deformation in the individual magnetoelastic units.

[0025] It should be noted that the deformation of a magnetoelastic monomer specifically refers to the tangential deformation of the magnetoelastic monomer.

[0026] It should be noted that the number of magnetoelastic monomers in the magnetoelastic array is denoted as N. During the transport of grains, all N magnetoelastic monomers are subjected to force and undergo deformation; therefore, the magnetoelastic array undergoes deformation.

[0027] The magnetic field array acquisition circuit includes a three-dimensional Hall element array. The three-dimensional Hall elements are used to monitor corresponding magnetic field changes. Each three-dimensional Hall element in the array corresponds to the center of its corresponding magnetoelastic monomer or is positioned with a specific offset.

[0028] For each of the multiple magnetoelastic monomers, the magnetic field changes of the magnetoelastic monomer are monitored using a three-dimensional Hall element corresponding to the magnetoelastic monomer in a three-dimensional Hall element array.

[0029] As an example, for each of a plurality of magnetoelastic monomers, a three-dimensional Hall element in a three-dimensional Hall element array, corresponding to the magnetoelastic monomer, monitors the magnetic flux data of the magnetic field of the magnetoelastic monomer. For each of the plurality of magnetoelastic monomers, a circuit in a magnetic field array acquisition circuit for generating data indicating changes in the magnetic field of the magnetoelastic monomer generates data indicating changes in the magnetic field of the magnetoelastic monomer based on the magnetic flux data of the magnetic field of the magnetoelastic monomer.

[0030] The magnetoelastic array consists of N individual magnetoelastic bodies. When a force plate is applied, all N individual magnetoelastic bodies deform, resulting in changes in their magnetic fields. The magnetic field array acquisition circuit, used to generate a three-dimensional force vector, outputs the three-dimensional force vector based on data indicating the magnetic field changes of the N individual magnetoelastic bodies, a pre-set correlation between this data and the three-dimensional force vector.

[0031] The three-dimensional force vector includes: force in the X direction, force in the Y direction, and force in the Z direction. The X, Y, and Z directions are the directions of the X-axis, Y-axis, and Z-axis in the corresponding coordinate system, respectively. As an example, the corresponding coordinate system can be either the robot coordinate system or the world coordinate system of the combine harvester on which the grain flow monitoring device provided in this application embodiment is installed, based on the grain flow monitoring method.

[0032] In one possible implementation, the magnetoelastic monomer includes: a soft magnetic layer, a permanent magnetic layer, and a strain layer. The permanent magnetic layer is used to generate the magnetic field of the permanent magnetic layer, and the soft magnetic layer is used to shield external magnetic field interference, prevent the magnetic field of the permanent magnetic layer from dissipating, and strengthen the magnetic field of the permanent magnetic layer.

[0033] In this process, the permanent magnet layer in the magnetoelastic monomer can be designed with a specific magnetic field distribution by controlling the direction and magnitude of the magnetizing magnetic field. The introduction of controllable magnetization direction and magnetic field distribution design in the permanent magnet layer enables the magnetoelastic monomer to generate regular magnetic field changes during deformation under stress.

[0034] In one possible implementation, the strain layer employs a porous microstructure, which is used to improve the sensitivity of the magnetoelastic monomer under low pressure and to adjust the stiffness of the strain layer. Adjusting the stiffness of the strain layer is used to adjust the range and sensitivity of the grain flow monitoring device.

[0035] The porous microstructure has pre-designed voids. To achieve the same deformation, a component using a porous microstructure requires less force compared to a solid component.

[0036] In one possible implementation, when the first device layout is adopted, the grain flow monitoring device is installed at the bottom of the screw conveyor; when the first device layout is adopted, the grain is propelled by a conveying auger during the transport of grain, wherein the conveying auger propelling the grain is used to allow the grain to enter the grain bin; when the first device layout is adopted, the force plate is subjected to a force related to the grain as the grain passes through it.

[0037] Specifically, the screw conveyor refers to a grain screw conveyor. Both the screw conveyor and the conveying auger are installed on the combine harvester on which the grain flow monitoring device, upon which the grain flow monitoring method provided in this application is based, is installed. The conveying auger is located inside the screw conveyor.

[0038] When the first device layout method is adopted, the base plate can be outside the outer shell of the screw conveyor.

[0039] When the first device is used, the force related to the grains on the force plate when the grains pass through it includes the frictional force generated when the grains pass through the force plate.

[0040] In one possible implementation, when the second device arrangement is adopted, the grain flow monitoring device is installed at the bottom of the screw conveyor, and a baffle is provided on one side of the force plate. The forces related to the grain include the thrust exerted by the grain on the baffle. When the second device arrangement is adopted, the grain is propelled by a conveying auger during the transport of the grain, wherein the conveying auger propelling the grain is used to allow the grain to enter the grain bin. When the second device arrangement is adopted, the force plate is subjected to forces related to the grain as the grain passes through it.

[0041] When the second device layout method is adopted, the base plate can be located outside the outer shell of the screw conveyor.

[0042] The baffle on one side of the force-bearing plate is part of the force-bearing plate. When the second device arrangement is adopted, when the grain passes through the force-bearing plate, the force on the force-bearing plate related to the grain can include: the pushing force exerted by the grain on the baffle and the frictional force generated when the grain passes through the force-bearing plate.

[0043] In one possible implementation, when the first or second device layout method is adopted, the base plate is made of aluminum, and the center of the base plate has a deep square groove for arranging the magnetoelastic array. The outer shell of the screw conveyor has a square hole, and the front and rear of the square hole of the screw conveyor have circular through holes.

[0044] The base plate is made of aluminum, which allows it to shield against external magnetic fields. This reduces the impact of external magnetic fields on the magnetic field of the individual magnetoelastic elements. The magnetoelastic array can be positioned in a deep square groove at the center of the base plate.

[0045] The statement that the square hole in the outer shell of the screw conveyor has circular through holes at the front and rear refers to the following: in the direction of grain movement, there are circular through holes at the front and rear of the square hole in the outer shell of the screw conveyor. Both the circular through holes at the front and rear of the square hole in the outer shell of the screw conveyor are located on the outer shell of the screw conveyor.

[0046] In either the first or second device layout method, the square opening of the screw conveyor's housing at least accommodates a portion of the magnetoelastic array, and at most accommodates a portion of the force-bearing plate. At least a portion of the force-bearing plate is outside the square opening of the screw conveyor's housing. Alternatively, the magnetoelastic array or the force-bearing plate may protrude from the square opening of the screw conveyor's housing.

[0047] In one possible implementation, when the third device arrangement is adopted, the grain flow monitoring device is installed on the top of the elevator, and the forces related to the grains include: the impact force exerted by the grains on the load-bearing plate; when the third device arrangement is adopted, during the transport of grains, the grains are thrown by a scraper when they reach the top of the elevator, wherein the scraper throws the grains to allow the grains to enter the grain bin; when the third device arrangement is adopted, the grains impact the load-bearing plate during the transport of grains.

[0048] When the third device layout method is adopted, the base plate can be outside the outer shell of the elevator.

[0049] When the third device is used, during the transport of grains, the grains impact the force plate, thereby exerting an impact force on the force plate.

[0050] In one possible implementation, when the third device arrangement is adopted, the base plate is made of aluminum and has a raised ramp at the center; when the third device arrangement is adopted, the raised ramp is used to arrange the magnetoelastic array and to make the magnetoelastic array easy to bear force; when the third device arrangement is adopted, the elevator shell has a square hole and there are circular through holes at the front and back of the square hole on the elevator shell.

[0051] The magnetoelastic array can be mounted on a raised ramp. The raised ramp increases the contact area between the force-bearing plate and the grain, making it easier for the force-bearing plate to bear force. This ease of force bearing on the force-bearing plate, in turn, makes it easier for the magnetoelastic array to bear force.

[0052] The statement that the square hole on the outer shell of the elevator has circular through holes at the front and rear refers to the following: in the direction of grain movement, there are circular through holes at the front and rear of the square hole on the outer shell of the elevator. Both the circular through holes at the front and rear of the square hole on the outer shell of the elevator are located on the outer shell of the elevator.

[0053] In the third arrangement, the square opening of the elevator's outer shell accommodates at least a portion of the magnetoelastic array and at most a portion of the force-bearing plate. At least a portion of the force-bearing plate is outside the square opening of the elevator's outer shell. Alternatively, the magnetoelastic array or the force-bearing plate may protrude from the square opening of the elevator's outer shell.

[0054] refer to Figure 1 It shows a schematic diagram of an example structure of a magnetoelastic array in a grain flow monitoring device on which the grain flow monitoring method provided in the embodiments of this application is based.

[0055] This example structure can be called a magnetoelastic array package 108.

[0056] The magnetoelastic array package 108 includes: an anti-interference circuit package layer 101, a magnetoelastic array 102, a circuit-magnetoelastic unit connection layer 103, and a magnetic field array acquisition circuit 104.

[0057] The magnetoelastic array 102 includes multiple magnetoelastic monomers, each of which includes a soft magnetic layer 105, a permanent magnet layer 106, and a strain layer 107.

[0058] A magnetoelastic array 102 is disposed between an anti-interference circuit encapsulation layer 101 and a circuit-magnetoelastic monomer connection layer 103. The anti-interference circuit encapsulation layer 101 is used to prevent electromagnetic interference. The circuit-magnetoelastic monomer connection layer 103 connects to a magnetic field array acquisition circuit 104. The magnetoelastic array 102 is connected to the magnetic field array acquisition circuit 104 via the circuit-magnetoelastic monomer connection layer 103.

[0059] refer to Figure 2 It shows a schematic diagram of an example architecture in which the grain flow monitoring method provided in the embodiments of this application is applied when the first device deployment method is adopted.

[0060] Figure 2 The example architecture shown includes: a screw conveyor 201, a conveying auger 202, and a grain flow monitoring device 203.

[0061] The screw conveyor 201 and the conveying auger 202 are both installed on the combine harvester where the grain flow monitoring device is installed. The conveying auger 202 is inside the screw conveyor 201.

[0062] refer to Figure 3 It shows a schematic diagram of the square hole in the casing of the screw conveyor when the first device layout is used.

[0063] Figure 3A square hole 301 is shown on the housing of a screw conveyor. The square hole 301 is located near the outlet of the screw conveyor. The square hole 301 is used to secure a grain flow monitoring device to the bottom of the screw conveyor. A magnetoelastic array can protrude from the square hole 301, and a force-bearing plate is located outside the square hole 301.

[0064] refer to Figure 4 It shows a schematic diagram of the components related to the grain flow monitoring device on which the grain flow monitoring method provided in this application embodiment is based when the first device deployment method is adopted.

[0065] Figure 4 The diagram shows a force-bearing plate 401, a magnetoelastic array 402, a base plate 403, an M5 bolt 404, and an M5 nut 405. The base plate 403 is located outside the screw conveyor. Made of aluminum, the base plate 403 shields against external magnetic fields, thus reducing the influence of external magnetic fields on the individual magnetoelastic units. A 1mm deep square groove is cut into the center of the base plate 403 to accommodate the magnetoelastic array 402, in which multiple magnetoelastic units are evenly arranged in a 3x3 row configuration. Above the magnetoelastic array 402 is the force-bearing plate 401, which has chamfered edges on both sides to reduce its impact on the transport of grain kernels. As an example, the force-bearing plate 401, the magnetoelastic array 402, and the base plate 403 are bonded together with liquid adhesive. M5 circular through holes are made at the front and rear of the square hole on the outer shell of the screw conveyor. The base plate 403 has holes corresponding to the M5 circular through holes at the front and rear of the square hole. An M5 bolt 404 passes through the M5 circular through holes and the corresponding holes on the base plate. The portion of the M5 bolt 404 passing through the M5 circular through holes and the corresponding holes on the base plate is coupled to an M5 nut 405, thereby securing the grain flow monitoring device.

[0066] refer to Figure 5 It shows a schematic diagram of an example architecture in which the grain flow monitoring method provided in the embodiments of this application is applied when the second device deployment method is adopted.

[0067] Figure 5 The example architecture shows a screw conveyor 501, a conveying auger 502, and a grain flow monitoring device 503.

[0068] The screw conveyor 501 and the conveying auger 502 are both installed on the combine harvester where the grain flow monitoring device is installed. The conveying auger 502 is inside the screw conveyor 501.

[0069] refer to Figure 6 It shows a schematic diagram of the square hole in the casing of the screw conveyor when the second device layout is adopted.

[0070] Figure 6 A square hole 601 is shown on the housing of a screw conveyor. The square hole 601 is located near the end of the conveying auger. A magnetoelastic array can protrude from the square hole 601, and a force-bearing plate is located outside the square hole 601.

[0071] refer to Figure 7 It shows a schematic diagram of the components related to the grain flow monitoring device on which the grain flow monitoring method provided in this application embodiment is based when the second device deployment method is adopted.

[0072] Figure 7 The diagram shows a force-bearing plate 701, a magnetoelastic array 702, a base plate 703, an M5 bolt 704, and an M5 nut 705. The base plate 703 is made of aluminum, which allows it to shield against external magnetic fields, thus reducing the influence of external magnetic fields on the individual magnetoelastic units. A 1mm deep square groove is cut into the center of the base plate 703 for arranging the magnetoelastic array 702. As an example, multiple magnetoelastic units are evenly arranged in a 3x3 row configuration. The force-bearing plate 701 is located above the magnetoelastic array 702, and its front and rear sides are chamfered to reduce the impact of the force-bearing plate on the transport of grain grains. A baffle is located on one side of the force-bearing plate 701 to receive the thrust generated by the conveyor auger transporting the grain. As an example, the force-bearing plate 701, the magnetoelastic array 702, and the base plate 703 can be bonded together with liquid adhesive. M5 circular through holes are made at the front and rear of the square hole on the outer shell of the screw conveyor. The base plate 703 has holes corresponding to the M5 circular through holes at the front and rear of the square hole. An M5 bolt 704 passes through the M5 circular through holes and the corresponding holes on the base plate. The portion of the M5 bolt 704 passing through the M5 circular through holes and the corresponding holes on the base plate is coupled to an M5 nut 705, thereby securing the grain flow monitoring device.

[0073] refer to Figure 8 It shows a schematic diagram of an example architecture in which the grain flow monitoring method provided in this application embodiment is applied when a third device deployment method is adopted.

[0074] Figure 8 The example architecture shows the elevator 801, scraper 802, and grain flow monitoring device 803.

[0075] refer to Figure 9 It shows a schematic diagram of the square opening in the housing of the elevator when the third device layout is adopted.

[0076] Figure 9 A square hole 901 is shown on the housing of the elevator. The elevator is mounted on a combine harvester on which the grain flow monitoring device, upon which the grain flow monitoring method provided in this embodiment is based, is installed. The square hole 901 is made in the housing of the elevator.

[0077] As an example, a magnetic elastic array can protrude from a square aperture 901, with a force-bearing plate outside the square aperture 901. The square aperture 901 is located at the top of the lift.

[0078] refer to Figure 10 It shows a schematic diagram of the components related to the grain flow monitoring device on which the grain flow monitoring method provided in this application embodiment is based when the third device deployment method is adopted.

[0079] Figure 10 The diagram shows a force-bearing plate 1001, a magnetoelastic array 1002, a base plate 1003, an M5 bolt 1004, and an M5 nut 1005. The base plate 1003 is made of aluminum, which allows it to shield against external magnetic fields, thus reducing the influence of external magnetic fields on the individual magnetoelastic cells. The base plate 1003 has a raised ramp at its center, used to arrange the magnetoelastic array 1002. The ramp ensures a larger contact area between the force-bearing plate 1001 and the grain grains, making it easier for the force-bearing plate 1001 to bear force. This ease of force bearing on the force-bearing plate 1001, in turn, makes it easier for the magnetoelastic array 1002 to bear force. As an example, the multiple magnetoelastic cells in the magnetoelastic array 1002 are evenly arranged in a 3x3 grid. The force-bearing plate 1001 is positioned above the magnetoelastic array 1002. The force-bearing plate 1001 has chamfered edges on all four sides to reduce its impact on the transport of grains. As an example, the force-bearing plate 1001, the magnetoelastic array 1002, and the base plate 1003 are bonded together with liquid adhesive. The base plate 1003 can be located outside the elevator housing. M5 circular through holes are formed at the front and rear of the square hole on the elevator housing. The base plate 1003 has holes corresponding to the M5 circular through holes at the front and rear of the square hole. An M5 bolt 1004 passes through the M5 circular through holes and the corresponding holes on the base plate 1003. The portion of the M5 bolt 1004 passing through the M5 circular through holes and the corresponding holes on the base plate 1003 is coupled to an M5 nut 1005, thereby securing the grain flow monitoring device.

[0080] In this embodiment, a permanent magnet layer in a magnetoelastic monomer can be prepared using a template method. Permanent magnet particle filler, a liquid or viscoelastic flexible substrate, and additives are mixed, with the permanent magnet filler accounting for 30% to 70% by mass. The permanent magnet material composite ink is prepared using physical mixing methods such as mechanical stirring, ultrasonic vibration, or high-pressure averaging. The permanent magnet layer precursor is formed in a mold using methods such as pressure and heating for curing. Magnetization treatment is then performed to distribute the magnetic domains of the permanent magnet layer according to a specific structure, using pulse magnetization, unidirectional magnetization, graded magnetization, or other magnetization methods. In this embodiment, the soft magnetic layer in the magnetoelastic monomer can be prepared using a template method. Soft magnetic particle filler, a liquid or viscoelastic flexible substrate, and additives are mixed, with the soft magnetic filler accounting for 20% to 60% by mass. The soft magnetic material composite ink is prepared using physical mixing methods such as mechanical stirring, ultrasonic vibration, or high-pressure averaging. The soft magnetic layer is then formed in a mold using curing methods such as pressure and heating.

[0081] In this embodiment, the strain layer in the magnetoelastic monomer can be prepared using a sacrificial template method. A porous sacrificial template with a specific structure is prepared using etching, 3D printing, sugar template methods, etc. A flexible strain filler ink is prepared by mixing a liquid or viscoelastic flexible substrate with additives using physical mixing methods such as mechanical stirring, ultrasonic vibration, or high-pressure averaging. The flexible strain filler ink is embedded into the porous sacrificial template using physical methods such as vacuum infusion. The interlocking flexible strain material is formed in a mold by pressing, heating, or other curing methods. The porous sacrificial template is removed using physical methods such as immersion, thermal melting, or ultrasonic vibration, or chemical methods such as solvent immersion and decomposition, to obtain a strain layer with a specific continuous porous structure.

[0082] In this embodiment, the permanent magnet layer, soft magnet layer, and strain layer can be assembled using plasma bonding or coating with a semi-cured flexible substrate ink. The magnetoelastic monomer is then re-formed in a mold using methods such as pressure and heating.

[0083] In this embodiment, the permanent magnet particles can be selected from permanent magnet materials such as AlNiCo alloy powder, IronChromiumCo alloy powder, rare earth permanent magnet powder, and ferrite powder. Preferably, neodymium iron boron powder with a particle size of 5-12 micrometers is selected to obtain a larger and more stable magnetic field. The soft magnetic particles are selected from soft magnetic materials with high permeability and low coercivity, such as iron powder, silicon steel powder, iron-silicon-aluminum powder, iron-cobalt soft magnetic alloy powder, amorphous soft magnetic alloy powder, and permalloy. The flexible substrate can be silicone rubber, such as Ecoflex 00-30, Ecoflex 00-50, polydimethylsiloxane (PDMS), etc.; or thermoplastic elastic materials, such as SEBS, TPU, etc.; or rubber, such as natural rubber, synthetic rubber, etc. Additives include curing agents, crosslinking agents, anti-aging agents, curing or crosslinking catalysts, activators, etc.

[0084] The following describes the grain flow monitoring method provided in the embodiments of this application.

[0085] The grain flow monitoring method provided in this application includes steps S101-S103.

[0086] In step S101, a three-dimensional force vector sequence is obtained. The three-dimensional force vector sequence includes: a three-dimensional force vector at each of multiple moments within a time window for monitoring grain flow. During the transport of grain, the force plate is subjected to a force related to the grain, causing the magnetoelastic monomer to be stressed. The stress on the magnetoelastic monomer causes deformation of the magnetoelastic monomer. The deformation of the magnetoelastic monomer causes a change in the magnetic field of the magnetoelastic monomer. The magnetic field array acquisition circuit outputs the three-dimensional force vector based on the change in the magnetic field of the magnetoelastic monomer.

[0087] In this context, the corresponding moment within a time window used to monitor grain flow can refer to the moment when the three-dimensional force vector for that corresponding moment is generated. The time window used to monitor grain flow refers to the period of time used to monitor grain flow.

[0088] It should be noted that all three-dimensional force vectors in the three-dimensional force vector sequence can be arranged in chronological order according to the time to which the three-dimensional force vectors belong.

[0089] In this embodiment, the three-dimensional force vector includes: force in the X direction, force in the Y direction, and force in the Z direction. The X, Y, and Z directions correspond to the directions of the X-axis, Y-axis, and Z-axis in the coordinate system, respectively. As an example, the corresponding coordinate system can be either the robot coordinate system of the combine harvester or the world coordinate system.

[0090] The three-dimensional force vector at time t within the time window used to monitor grain flow. F ( t ) can be represented as: in, Let represent the force in the X direction of the three-dimensional force vector at time t. Let represent the force in the Y direction of the three-dimensional force vector at time t. This represents the force in the Z direction of the three-dimensional force vector at time t.

[0091] In step S102, for each three-dimensional force vector in the three-dimensional force vector sequence, the forces in all directions of the three-dimensional force vector are combined into a single-value impact strength index corresponding to the three-dimensional force vector, so as to obtain a single-value impact strength index sequence.

[0092] The single-value impact intensity index sequence includes multiple single-value impact intensity indices. Each single-value impact intensity index corresponds to a three-dimensional force vector at a different time.

[0093] All single-value impact intensity indices in the single-value impact intensity index sequence can be arranged in chronological order according to their time values. The time value of a single-value impact intensity index corresponds to the time value of its three-dimensional force vector.

[0094] In step S102, for a three-dimensional force vector, according to the pre-set correlation between the single-value impact strength index and the three-dimensional force vector, the forces in all directions of the three-dimensional force vector are combined into a single-value impact strength index corresponding to the three-dimensional force vector.

[0095] In one possible implementation, the following formula is used to synthesize the forces in all directions of the three-dimensional force vector at time t into a single-valued impact strength index corresponding to the three-dimensional force vector at time t: in, This represents a single-valued impact strength index corresponding to the three-dimensional force vector at time t. Let represent the force in the X direction of the three-dimensional force vector at time t. Let represent the force in the Y direction of the three-dimensional force vector at time t. This represents the force in the Z direction of the three-dimensional force vector at time t.

[0096] In step S103, the predicted grain flow rate is determined based on the single-value impact intensity index sequence.

[0097] Among them, the grain flow prediction result can be: the mass or volume of grain output from the cleaning system or lifting device per unit time during the operation of the combine harvester.

[0098] In one possible implementation, a sequence of single-valued impact intensity indicators is input into a model for predicting grain flow, which then outputs a predicted grain flow result. As an example, the model for predicting grain flow can be one of a decision tree, a support vector machine (SVM), or a neural network. As another example, the neural network can be one of a convolutional neural network or an LSTM. The model for predicting grain flow is trained. It is trained in advance using training data and labels for that training data. The training data includes any type of data from the single-valued impact intensity indicator sequence. The labels on the training data indicate the correct result output by the model when the training data is input into it.

[0099] In one possible implementation, step S103 includes: steps S1031-S1033.

[0100] In step S1031, the first candidate prediction result is generated based on the single-value impact intensity index sequence using the random forest model used to predict the grain flow prediction result. In step S1032, a second candidate prediction result is generated based on the single-value impact intensity index sequence using a long short-term memory (LSTM) network used to predict the grain flow prediction result. In step S1033, the grain flow prediction result is determined based on the first candidate prediction result and the second candidate prediction result.

[0101] In step S1031, statistical features of the preprocessed single-value impact intensity index sequence can be extracted to obtain a statistical feature vector of the preprocessed single-value impact intensity index sequence. This statistical feature vector is then input into a random forest model used to predict grain flow prediction results, yielding the first candidate prediction result output by the random forest model. The preprocessed single-value impact intensity index sequence is obtained by preprocessing the single-value impact intensity index sequence. This preprocessing may include: filtering the single-value impact intensity index sequence to obtain a filtered single-value impact intensity index sequence; and normalizing each item in the filtered single-value impact intensity index sequence to obtain the preprocessed single-value impact intensity index sequence. Filtering is used to remove outliers and items that act as high-frequency noise. The statistical feature vector may include statistical features such as mean, maximum value, and frequency. The mean can refer to the average value of the items in the single-value impact intensity index sequence, and the maximum value can refer to the largest item in the single-value impact intensity index sequence.

[0102] The random forest model used to predict grain flow forecast results includes multiple decision trees. In step S1031, each decision tree in the random forest model can predict a corresponding sub-prediction result based on the statistical feature vector of the preprocessed single-value impact intensity index sequence. A first candidate prediction result can be obtained based on the predicted sub-results from each decision tree. For example, the mean of the predicted sub-results from each decision tree is calculated, and the calculated mean is used as the first candidate prediction result.

[0103] It should be noted that the depth of these decision trees is less than the depth threshold. The random forest model is a lightweight random forest model to reduce computational complexity. As an example, the depth of multiple decision trees is less than 6.

[0104] The random forest model used to predict grain flow forecasts is trained. It is trained in advance using training data and labels for that training data. The training data includes data of any type from the single-valued impact intensity index sequence. The labels on the training data indicate the correct output of the random forest model when the training data is input into it.

[0105] In step S1032, the preprocessed single-valued impact intensity index sequence can be input into the LSTM network used to predict the grain flow rate, resulting in a second candidate prediction output by the LSTM network. The LSTM network used to predict the grain flow rate is trained. The LSTM network is trained in advance using the training data and labels of the training data.

[0106] The training data for the LSTM network used to predict grain flow forecasts includes data of any type from the single-value impact intensity index sequence. The labels on the training data indicate the correct output of the LSTM network when the training data is input into it.

[0107] As an example, the LSTM network used to predict the grain flow forecast is a single-layer LSTM network with 16 hidden layers and less than 5000 model parameters, using a fixed step size sampling window.

[0108] In one possible implementation, in step S1033, the average of the first candidate prediction result and the second candidate prediction result is determined as the grain flow prediction result. The sum of the first candidate prediction result and the second candidate prediction result is calculated, and the sum is divided by 2 to obtain the grain flow prediction result.

[0109] In one possible implementation, in step S1033, a weighted average is performed on the first candidate prediction result and the second candidate prediction result to obtain the grain flow prediction result. This weighted average includes: calculating a weighted sum of the first candidate prediction result and the second candidate prediction result, and dividing this weighted sum by the weighted sum to obtain the grain flow prediction result. The weights of the first candidate prediction result and the second candidate prediction result are preset.

[0110] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0111] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0112] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for monitoring grain flow rate based on a grain flow rate monitoring device, characterized in that: The grain flow monitoring device includes: a force-bearing plate, a magnetoelastic array, a base plate, and a magnetic field array acquisition circuit. The force-bearing plate is positioned above the magnetoelastic array, which is positioned between the force-bearing plate and the base plate. The base plate is used to fix the grain flow monitoring device. The magnetoelastic array includes individual magnetoelastic cells. The method includes: A three-dimensional force vector sequence is obtained, which includes the three-dimensional force vector at each of multiple moments within a time window for monitoring grain flow. During the transport of grain, the force plate is subjected to forces related to the grain, which causes the magnetoelastic monomer to be stressed. The stress on the magnetoelastic monomer causes deformation of the magnetoelastic monomer, which in turn causes changes in the magnetic field of the magnetoelastic monomer. The magnetic field array acquisition circuit outputs the three-dimensional force vector based on the changes in the magnetic field of the magnetoelastic monomer. For each three-dimensional force vector in the three-dimensional force vector sequence, the forces in all directions of the three-dimensional force vector are combined into a single-value impact strength index corresponding to the three-dimensional force vector to obtain a single-value impact strength index sequence. The predicted grain flow rate is determined based on the single-value impact intensity index sequence.

2. The method according to claim 1, characterized in that: When the first device layout is adopted, the grain flow monitoring device is installed at the bottom of the screw conveyor; when the first device layout is adopted, during the transport of grain grains, the grain grains are pushed by the conveying auger, wherein the conveying auger pushes the grain grains to make the grain grains enter the grain bin; when the first device layout is adopted, when the grain grains pass through the force plate, the force plate is subjected to a force related to the grain grains.

3. The method according to claim 1, characterized in that: When the second device layout is adopted, the grain flow monitoring device is installed at the bottom of the screw conveyor, and there is a baffle on one side of the force plate. The forces related to the grain include the thrust exerted by the grain on the baffle. When the second device layout is adopted, the grain is pushed by the conveying auger during the transport of grain, wherein the conveying auger pushes the grain to allow the grain to enter the grain bin. When the second device layout is adopted, the force plate is subjected to the forces related to the grain when the grain passes through it.

4. The method according to claim 1, characterized in that: When the first or second device layout method is adopted, the base plate is made of aluminum, and there is a deep square groove in the center of the base plate. The deep square groove is used to arrange the magnetoelastic array. The outer shell of the screw conveyor has a square hole, and there are circular through holes at the front and back of the square hole of the screw conveyor.

5. The method according to claim 1, characterized in that: When the third device layout is adopted, the grain flow monitoring device is installed on the top of the elevator. The forces related to the grain include: the impact force exerted by the grain on the force plate; when the third device layout is adopted, during the transport of grain, the grain is thrown by a scraper when it reaches the top of the elevator, wherein the scraper throws the grain to allow the grain to enter the grain bin; when the third device layout is adopted, the grain impacts the force plate during the transport of grain.

6. The method according to claim 5, characterized in that: When the third device layout method is adopted, the base plate is made of aluminum, and there is a raised ramp in the center of the base plate; when the third device layout method is adopted, the raised ramp is used to arrange the magnetoelastic array, and the raised ramp is used to make the magnetoelastic array easy to bear force; when the third device layout method is adopted, the outer shell of the elevator has a square hole, and there are circular through holes in front and behind the square hole on the outer shell of the elevator.

7. The method according to claim 1, characterized in that: The magnetoelastic monomer includes: a soft magnetic layer, a permanent magnetic layer, and a strain layer. The permanent magnetic layer is used to generate the magnetic field of the permanent magnetic layer, and the soft magnetic layer is used to shield external magnetic field interference, prevent the magnetic field of the permanent magnetic layer from dissipating, and strengthen the magnetic field of the permanent magnetic layer.

8. The method according to claim 7, characterized in that: The strain layer adopts a porous microstructure, which is used to improve the sensitivity of the magnetoelastic monomer under low pressure and adjust the stiffness of the strain layer. Adjusting the stiffness of the strain layer is used to adjust the range and sensitivity of the grain flow monitoring device.

9. The method according to any one of claims 1-8, characterized in that: Based on the single-value impact intensity index sequence, the predicted results of grain flow rate are determined as follows: Using a random forest model for predicting grain flow forecasts, the first candidate forecast result is generated based on a single-value impact intensity index sequence. Using a Long Short-Term Memory (LSTM) network for predicting grain flow forecasts, a second candidate forecast is generated based on a sequence of single-valued impact intensity indices. Based on the first candidate prediction results and the second candidate prediction results, the grain flow prediction results are determined.

10. The method according to claim 9, characterized in that: Based on the first candidate prediction results and the second candidate prediction results, the predicted grain flow rate is determined as follows: The grain flow prediction result is obtained by weighting the first candidate prediction result and the second candidate prediction result.

Images