Corn ear threshing impact and rubbing collaborative inversion device and method based on space-time collaboration

Abstract

The invention relates to a corn ear threshing impact and rubbing collaborative inversion device and method based on space-time collaboration. The device comprises an IMU module, a power supply module, a flexible piezoelectric film sensor and a corn ear, the IMU module is integrated with an MCU, a wireless communication module and an MEMS sensing module, an accelerometer, a gyroscope and an electronic compass are embedded in a core shaft of the corn ear, the flexible piezoelectric film sensor is arranged on the surface of the core shaft, grains are made of piezoelectric ceramics, and each grain can serve as an independent data source. According to the method, the accuracy is improved by acquiring acceleration, attitude information and pressure data, calculating impact force and rubbing force and adopting a unified space-time reference and data fusion technology, and finally, the BP neural network is utilized to identify a corn threshing motion state, so that collaborative online inversion of impact and rubbing loads is realized, and an optimal design basis is provided for corn ear harvesting elements.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery, and in particular to a device and method for threshing impact and rubbing coordinated inversion of corn ears based on time-space coordination. Background Art

[0002] At present, the mechanized harvesting method of corn in my country is mainly focused on harvesting ears. Although this method meets the current production needs to a certain extent, it still has a significant gap compared with the more advanced mechanized harvesting technology used by other major grain crops such as wheat and rice in terms of operating efficiency, refined management and loss reduction. With the continuous expansion of corn planting area and the continuous growth of output in my country, the demand for the research and development of new corn harvesting technology that can significantly improve operating efficiency, reduce labor intensity and reduce production costs has become more and more urgent. This is not only related to the development prospects of the corn industry, but also directly affects the implementation effect of the national food security strategy.

[0003] The stress conditions of corn in the threshing drum are very complex. Many researchers have conducted extensive research using computer simulation, high-speed photography, power monitoring and other methods, and have achieved certain results. However, due to the extremely complex conditions faced by grain separation machinery in actual operation and the influence of various factors, coupled with the limitations of existing threshing theories, it is difficult for computer simulation results to fully and accurately reflect the actual situation. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a device and method for collaborative inversion of corn cob threshing impact and rubbing based on time-space coordination, so as to more intuitively and accurately understand the stress conditions and posture changes of corn during the threshing process.

[0005] Technical solution: A corn cob threshing impact and rubbing collaborative inversion device based on time-space coordination of the present invention includes an IMU module, a power supply module, a flexible piezoelectric film sensor and a corn cob, the IMU module includes an MCU, a wireless communication module and a MEMS sensor module, the MEMS sensor module includes an accelerometer, a gyroscope, an electronic compass and a flexible piezoelectric film sensor; the corn cob includes a grain and a core shaft, the flexible piezoelectric film sensor is arranged on the surface of the core shaft, the grain is arranged on the flexible piezoelectric film sensor, the grain is made of piezoelectric ceramics, and the accelerometer, gyroscope and electronic compass are arranged in the core shaft.

[0006] Furthermore, the power supply module includes a voltage stabilizing module and a power supply, the voltage stabilizing module includes a linear voltage regulator chip, and the linear voltage regulator chip integrates an overheat protection and current limiting circuit to prevent the circuit from overloading.

[0007] Furthermore, it also includes a sensor housing, which is a cylindrical thin-walled housing, divided into two upper and lower semi-cylindrical housings, and closed with an L-shaped bayonet connection. The sensor housing is provided with other components of the IMU module except the flexible piezoelectric film sensor and a power supply module, which is convenient for implanting the core shaft.

[0008] Furthermore, the accelerometer is a capacitive accelerometer; and the MCU uses an ESP32MC chip as a core MCU.

[0009] A corn ear threshing impact and rubbing collaborative inversion method based on time-space coordination of the present invention uses any of the above-mentioned devices, and comprises the following steps:

[0010] S1: Obtain the acceleration, posture information and pressure data of corn ears through the MEMS sensor module;

[0011] S2: Calculate the impact force and rubbing force on the corn ear based on the acquired data using MCU;

[0012] S3: Adopt a unified space-time reference to achieve time synchronization and unified conversion of space coordinates;

[0013] S4: Integrate data from different sensors through data fusion technology to improve measurement accuracy;

[0014] S5: Use BP neural network to identify the motion state of corn threshing and realize the coordinated online inversion of impact and rubbing loads in the corn threshing process.

[0015] Furthermore, the accelerometer described in S1 is used to accurately measure the acceleration in the three-axis directions, the gyroscope is used to obtain real-time attitude information, including pitch angle, azimuth angle and roll angle, the electronic compass is used to improve the geomagnetic direction perception performance and obtain azimuth angle information, each of the grains is an independent data source, the flexible piezoelectric film sensor converts pressure into an electrical signal, adapts to the curved surface structure of the corn grains, detects pressure changes, and the force between the grains and the flexible piezoelectric film reflects the impact force; the force between the grains reflects the rubbing force exerted on the corn cob.

[0016] Furthermore, the implementation of the unified space-time reference in S3 includes time synchronization and unified conversion of space coordinates. Time synchronization is achieved by means of a network time protocol, and unified conversion of space coordinates is based on an inertial navigation positioning algorithm and quaternion coordinate transformation.

[0017] Furthermore, the data fusion technology in the S4 adopts a particle filter algorithm to effectively integrate data from the accelerometer, gyroscope, electronic compass and flexible piezoelectric film sensor, eliminate noise and improve measurement accuracy.

[0018] Furthermore, the input parameters of the BP neural network in S5 include acceleration, attitude angle and pressure, and the output is the motion state recognition result during the corn threshing process.

[0019] Beneficial effects: Compared with the prior art, the present invention has the following advantages: (1) In the present invention, corn kernels are arranged as independent individuals on the surface of the flexible piezoelectric film sensor in an array manner. Each kernel is an independent data source and can provide unique information about its own state (the kernel is a piezoelectric ceramic, and the degree of deformation of the kernel can reflect the magnitude of the impact force-rubbing force). This layout ensures that each corn kernel can be monitored individually, thereby achieving accurate measurement of the pressure on each corn kernel (the force between the kernel and the flexible piezoelectric film reflects the impact force; the force between the kernel and the kernel reflects the rubbing force on the corn cob); (2) The present invention can obtain acceleration, posture and pressure information, and realize direct observation of the corn harvesting process. Compared with research methods such as computer simulation analysis, high-speed photography and power monitoring, it can more intuitively reveal the force and posture change process of corn during threshing; (3) The present invention has a simple structure, reasonable design, small size, light weight, high precision, and can conveniently and quickly obtain accurate and true information about the corn harvesting process; (4) The present invention uses Wi-Fi as the communication method between the wireless IMU and the host computer, which has the characteristics of high transmission rate, relatively low power consumption, and convenient networking. The ESP32 chip is used as the core MCU, which can achieve high bandwidth, low power consumption, long transmission distance, and both computing and communication functions, saving redundant communication modules and greatly improving the system integration; (5) The present invention uses a MEMS sensor module to take into account the characteristics of large range, low power consumption, high precision, and high integration, and can accurately provide the required acceleration and attitude information. The flexible piezoelectric film sensor used has excellent flexibility and bending performance, can fit the curved surface structure of corn kernels tightly, and is very suitable for use in the complex structure environment of corn. At the same time, due to its light and thin design, this sensor itself has a high sensitivity, can accurately detect tiny pressure, strain or deformation, and is suitable for capturing subtle dynamic signals and environmental changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a cross-sectional view of the device of the present invention;

[0021] Figure 2 It is a schematic diagram of the three-dimensional structure of the device of the present invention;

[0022] Figure 3 Schematic diagram of corn kernels of the present invention

[0023] Figure 4 is a schematic diagram of a sensor module of the present invention;

[0024] Figure 5The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0025] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0026] like Figure 1-2 As shown, a corn ear threshing impact and rubbing coordinated inversion device based on time-space coordination of the present invention includes an IMU module 1, a voltage stabilizing module 2, a battery 3, a flexible piezoelectric film sensor 4, a corn ear 5, and a sensor housing 6. The corn ear 5 includes a kernel and a core shaft. The battery 3 provides a stable voltage to the IMU module 1 through the voltage stabilizing module 2, and the sensor housing 6 is implanted in the core shaft.

[0027] like Figure 4 As shown, the IMU module 1 includes an MCU, a wireless communication module and a MEMS sensor module. The battery 3 provides power for the MCU, the wireless communication module and the MEMS sensor module. The MEMS sensor module includes an accelerometer, a gyroscope, an electronic compass and a flexible piezoelectric film sensor. In some embodiments, the accelerometer is a capacitive accelerometer. When the corn cob is threshed and accelerates in the drum, the spacing or the opposite area between the two poles of the plate or capacitor inside the accelerometer changes, causing the change of the internal capacitance value, thereby reflecting the acceleration change information of the corn cob. The gyroscope can obtain the real-time attitude information of the corn cob, including the pitch angle, the azimuth angle and the roll angle. The electronic compass can improve the sensor's perception performance of the geomagnetic direction and obtain the azimuth angle information of the corn cob with higher accuracy. The acceleration of the corn cob measured by the accelerometer is based on the carrier coordinate system. In order to obtain the force condition of the corn cob in the earth coordinate system, it is necessary to comprehensively calculate the attitude information measured by the gyroscope and the electronic compass. The MCU ensures the performance of the lower computer. In some embodiments, the ESP32MC chip is used as the core MCU. Its sufficient computing speed can integrate the acceleration and posture information measured by the sensor and calculate the impact force of the corn ear during the separation process in real time according to Newton's second law.

[0028] like Figure 2-3The corn kernels shown are arranged as independent individuals on the surface of the flexible piezoelectric film sensor in an array. Each kernel is an independent data source that can provide unique information about its own state (the kernel is a piezoelectric ceramic, and the degree of deformation of the kernel can reflect the impact force-kneading force). This layout ensures that each corn kernel can be monitored individually, thereby achieving accurate measurement of the pressure on each corn kernel (the force between the kernel and the flexible piezoelectric film reflects the impact force; the force between the kernel and the kernel reflects the kneading force on the corn cob). The flexible piezoelectric film sensor has excellent flexibility and bending properties, and can adapt to the pressure detection needs of objects of various shapes and sizes. When the corn kernels are placed on it, a close and comprehensive contact interface is formed between the four sides and the bottom of the kernel and the flexible film. During the corn threshing process, when the threshing element and the concave plate act together on the cob, a series of complex mechanical movements will be generated, including squeezing, kneading and other actions. These actions will cause the corn kernels to separate from the cob while applying different amounts of pressure on different surfaces. Due to the good fit between the corn kernels and the flexible film, each surface of the kernels becomes an effective pressure-sensitive point. As the kernels are subjected to external forces, the flexible film will deform to varying degrees at the corresponding positions. It is by sensing these subtle deformations that the flexible piezoelectric film sensor can accurately capture and measure the specific force conditions experienced by the corn kernels during the entire threshing process.

[0029] In some embodiments, the sensor housing 6 is a cylindrical thin-walled housing, which is divided into two upper and lower semi-cylindrical housings and is closed with an L-shaped bayonet connection for easy disassembly and assembly. The sensor housing contains a sensor and a battery for easy implantation into the corn cob shaft.

[0030] The working process of the present invention is as follows:

[0031] The stress states of corn ears inside the threshing drum mainly include: the impact of the threshing element, free fall motion, the squeezing and kneading of the ears by the threshing element and the concave plate, and the collision between the corn ears and the guide plate. Therefore, this device can realize the online inversion of the impact and kneading loads of the corn ears during the threshing process. The specific workflow is as follows:

[0032] When the corn is impacted, the three-axis accelerometer and three-axis gyroscope in the MEMS sensor module detect the acceleration and posture information of the three axes of the corn respectively. Based on Newton's second law, the above information can be used to solve the actual magnitude and direction of the resultant force on the corn. The specific process is as follows: When the speed of the corn increases, the change in the distance between the plates or capacitors inside the accelerometer or the adjustment of the facing area will cause a corresponding change in the capacitance value. This change reflects the change in acceleration. The initial output data of the IMU accelerometer and the flexible piezoelectric film sensor are based on the coordinate system of the corn itself. In order to obtain the force condition of the corn ear in the geodetic coordinate system, it is necessary to combine the data of the gyroscope in the IMU for calculation and analysis. The positioning algorithm based on inertial navigation, which uses the quaternion Q to calculate the rotation matrix according to the quaternion coordinate transformation. According to the original acceleration vector A and the rotation matrix Calculate the linear acceleration in the carrier coordinate system Linear acceleration and the rotation matrix Calculate the acceleration vector in the geodetic coordinate system According to Newton's second law F=ma, the magnitude and direction of the impact force on the corn can be obtained. The electronic compass improves the gyroscope's perception of the geomagnetic direction, and then based on the three-axis attitude angle and quaternion method, the attitude and position information of the corn in the threshing drum can be calculated in real time.

[0033] Corn kernels are arranged as independent individuals on the surface of the flexible piezoelectric film sensor in an array. Each kernel is an independent data source that can provide unique information about its own status. This layout ensures that each kernel can be monitored individually, thereby achieving accurate measurement of the pressure on each kernel. The flexible piezoelectric film sensor has excellent flexibility and bending properties, and can adapt to the pressure detection needs of objects of various shapes and sizes. When the corn kernels are placed on it, a close and comprehensive contact interface is formed between the four sides and the bottom of the kernel and the flexible film. During the corn threshing process, when the threshing element and the concave plate act together on the ear, a series of complex mechanical movements will be generated, including squeezing, kneading and other actions. These actions will cause the corn kernels to separate from the ear while applying different amounts of pressure on different faces. Due to the good fit between the corn kernels and the flexible film, each face of the kernel becomes an effective pressure sensitive point. As the kernels are subjected to external forces, the flexible film will produce different degrees of deformation at the corresponding positions. It is by sensing these subtle deformations that the flexible piezoelectric film sensor can accurately capture and measure the specific forces experienced by corn kernels during the entire threshing process.

[0034] In the initial state, the flexible piezoelectric film sensor is in a balanced state, no piezoelectric effect is generated, and no voltage signal is output. When the corn kernels are squeezed and rubbed by the threshing element and the concave plate, an external dynamic force acts on the flexible piezoelectric film sensor, and the flexible piezoelectric film sensor will be compressed and deformed. The piezoelectric ceramic material lead zirconate titanate particles will be strained due to stress, and the dipoles in the crystal structure will also change accordingly, generating induced charges on the upper and lower surface electrodes of the flexible piezoelectric film sensor, thereby generating a potential difference between the upper and lower surfaces of the piezoelectric film. In response, the external free charges will move to neutralize the surface potential difference at a certain speed. When the corn is subjected to a fixed rubbing force, the external dynamic force no longer changes, that is, when the flexible film strain remains unchanged, the induced charges in the upper and lower surface electrodes are balanced by the free charges in the external circuit, and the potential difference gradually decreases to zero. When the corn is not subjected to rubbing force, the external force is removed, the deformation of the flexible piezoelectric film sensor is restored, the polarity of the induced charges on the upper and lower surface electrodes changes, and the free charges gradually flow back in the opposite direction, generating opposite electrical signals, reaching the original potential balance, and the flexible piezoelectric film sensor will return to its initial shape.

[0035] MCU ensures the performance of the lower computer, and uses ESP32MC chip as the core MCU. The sufficient computing speed of MCU can integrate the acceleration measured by the sensor, and the posture information calculation is based on Newton's second law, and the impact force and rubbing force of the corn ear in the process of separation can be calculated in real time, and the posture and position information of the corn in the threshing drum can be solved in real time. MCU uses ESP32MC chip to integrate Wi-Fi function, support real-time operating system and Wi-Fi protocol stack, have complete Wi-Fi network function, can be used independently, can ensure the rapid solution of data obtained by each sensor, and send the measured data to the upper computer in real time, so as to realize the coordinated online inversion of impact and rubbing load in the corn threshing process.

[0036] The realization of a unified spatiotemporal reference is crucial, as it provides a common framework for data fusion across multiple sensors and data sources. In this environment, a unified spatiotemporal reference ensures that data from different sensors and measurement devices can be analyzed in the same temporal and spatial framework. Time synchronization can be achieved with the help of the Network Time Protocol (NTP), ensuring that the timestamps of data recorded by all sensors are consistent.

[0037] The first step to achieve a unified spatiotemporal benchmark is time synchronization. First, build a self-built NTP server. Relying on the Network Time Protocol (NTP), the accelerometers, gyroscopes, electronic compasses, and flexible piezoelectric film sensors in MEMS sensors can work under the same time benchmark, ensuring that the sensors in the network can synchronize time with the NTP server so that the timestamps of the captured data are accurate and consistent. This process is crucial for the post-processing of data, especially in the corn threshing environment, where any slight time difference may lead to deviations in the three-dimensional reconstruction results. The following is the formula for the Network Time Protocol (NTP):

[0038] Offset θ:

[0039]

[0040] T1 is the timestamp when the client sends the request, T2 is the timestamp when the MCU receives the request and returns it immediately, T3 is the timestamp when the client receives the response, and T4 is the timestamp recorded by the server before sending the response.

[0041] Round Trip Delay RTT Estimation:

[0042] RTT = (T4-T1)-(T3-T2)

[0043] The drift rate D is:

[0044]

[0045] The interval between two measurements is Δt, and the corresponding offsets are θ1 and θ2 respectively.

[0046] After completing time synchronization, the unified conversion of spatial coordinates is the second step to achieve a unified space-time reference. Different sensors may use different coordinate systems, so the data of each sensor needs to be converted into the same reference coordinate system. This includes calibrating and correcting the accelerometer and flexible piezoelectric film sensor, establishing a mapping relationship between them and the world coordinate system, and converting the coordinates measured by each into a unified coordinate system. The acceleration and pressure of the corn ear measured by the accelerometer and flexible piezoelectric film sensor in the present invention are based on the carrier coordinate system. In order to obtain the force condition of the corn ear in the geodetic coordinate system, it is necessary to comprehensively calculate the attitude information measured by the gyroscope and the electronic compass. A positioning algorithm based on inertial navigation, in which the rotation matrix is ​​calculated using the quaternion Q according to the quaternion coordinate transformation. According to the original acceleration vector A and the rotation matrix Calculate the linear acceleration in the carrier coordinate system Linear acceleration and the rotation matrix Calculate the acceleration vector in the geodetic coordinate system According to the pressure F 2b Rotation Matrix Calculate F in the geodetic coordinate system 2n , thereby converting the acceleration based on the carrier coordinates into the acceleration based on the earth coordinate system, and converting the pressure based on the carrier coordinates into the pressure based on the earth coordinate system.

[0047] The data fusion process is an important method to achieve a unified spatiotemporal benchmark. By using particle filter data fusion technology, data from accelerometers, gyroscopes, electronic compasses, and flexible piezoelectric film sensors in MEMS sensors can be effectively integrated, which helps eliminate noise and improve measurement accuracy. By using a particle filter algorithm, useful information can be effectively extracted from measurement data containing low-frequency vibrations, eliminating or reducing data noise and inconsistencies caused by vibration. The following is the formula for particle filtering:

[0048] Equation of state: x k =[x k ,y k ,z k ,a x ,a y ,a z ,F 22 ,F 2y ,F 2z ,] T

[0049] The state vector x k , which contains the system state, attitude angle, acceleration and pressure at time step k.

[0050] The state equation describes the evolution of the state over time: x k+1 =f(x k ,u k ,w k )

[0051] u k is the control input, w k is zero-mean Gaussian noise.

[0052] Observation equation: z k =h(x k ,v k )

[0053] v k is the observation noise.

[0054] Initialize particles:

[0055] The initial state obeys a prior distribution p(x0).

[0056] Prediction steps:

[0057] are samples sampled from the process noise distribution.

[0058] The weight of each particle:

[0059] Gaussian distribution to model observation noise:

[0060] Σ is the covariance matrix of the observation noise.

[0061] Normalize the weights of all particles so that they sum to 1:

[0062]

[0063] State Estimation:

[0064] Through the above steps, data from different sensors can be effectively integrated, which helps to eliminate noise and improve measurement accuracy. Combined with MCU, the impact force and rubbing force on the corn ear during the detachment process are calculated in real time, and the posture and position information of the corn in the threshing drum are solved in real time. The measured data is sent to the host computer in real time to realize the coordinated online inversion of the impact and rubbing loads in the corn threshing process.

[0065] The motion state of the corn ears in the threshing drum after impact and rubbing is complex and full of uncertainty. It is difficult to distinguish the impact and rubbing force state of the corn ears in a short time through the impact force and pressure curve. Therefore, in order to achieve efficient real-time discrimination of the motion state of the corn ears, strengthen the collaborative online inversion capability of the impact and rubbing loads in the corn threshing process, and observe the corn threshing state intuitively, the BP neural network is used to identify the corn threshing motion state. The data in the corn threshing process, including acceleration, posture information and pressure, are collected. The collected data are preprocessed based on the time-space synergy method to improve the accuracy and generalization ability of the model. The BP neural network consists of an input layer, a hidden layer and an output layer, and the network structure needs to be defined. The number of nodes in the input layer matches the number of input parameters. The input parameters here are the acceleration, posture angle and pressure obtained by the MEMS sensor. The number of nodes in the output layer matches the number of threshing effect indicators to be predicted. The output layer is 1. static, 2. static-free fall, 3. free fall-impact-rubbing, 4. free fall-rubbing-impact. The number of hidden layers and the number of neurons in each layer are adjusted according to the experimental parameters of the actual corn threshing process. The first layer is set to identify the acceleration, attitude angle and pressure curves according to the extracted acceleration, attitude angle and pressure, and the second layer predicts the state of corn under impact and rubbing force. The network weights are iteratively adjusted using the training set data until the model performs well on the training data.

[0066] A large amount of corn threshing process data is read from the sensor in real time, including acceleration, attitude angle and pressure obtained by MEMS sensors, and these data are timestamped through the Network Time Protocol (NTP). The trained model is applied to the actual online inversion of impact and rubbing loads in the corn threshing process, which can effectively identify the corn threshing motion state in real time and enhance the collaborative online inversion capability of impact and rubbing loads in the corn threshing process.

[0067] Therefore, the device and method can realize the coordinated online inversion of impact and rubbing loads in the corn threshing process.

Claims

1. A corn ear threshing impact and rubbing collaborative inversion device based on time-space coordination, characterized in that: The invention comprises an IMU module (1), a power supply module (2, 3), a flexible piezoelectric film sensor (4) and a corn ear (5), wherein the IMU module (1) comprises an MCU, a wireless communication module and a MEMS sensor module, wherein the MEMS sensor module comprises an accelerometer, a gyroscope, an electronic compass and a flexible piezoelectric film sensor (4); the corn ear (5) comprises a kernel and a core shaft, wherein the flexible piezoelectric film sensor (4) is arranged on the surface of the core shaft, and the kernel is arranged on the flexible piezoelectric film sensor (4), wherein the kernel is made of piezoelectric ceramics, and the accelerometer, gyroscope and electronic compass are arranged inside the core shaft.

2. The corn ear threshing impact and rubbing collaborative inversion device based on time-space collaboration according to claim 1 is characterized in that: The power supply module comprises a voltage stabilizing module (2) and a power supply (3); the voltage stabilizing module (2) comprises a linear voltage stabilizing chip; an overheat protection and current limiting circuit are integrated into the linear voltage stabilizing chip to prevent circuit overload.

3. The corn ear threshing impact and rubbing collaborative inversion device based on time-space collaboration according to claim 1 is characterized in that: The sensor housing (6) is a cylindrical thin-walled housing, which is divided into two upper and lower semi-cylindrical housings and is closed by an L-shaped bayonet connection. The sensor housing (6) is provided with other components of the IMU module (1) except the flexible piezoelectric film sensor (4) and a power supply module (2, 3), so as to facilitate the implantation of the core shaft.

4. The corn ear threshing impact and rubbing collaborative inversion device based on time-space collaboration according to claim 1 is characterized in that: The accelerometer is a capacitive accelerometer; the MCU uses the ESP32MC chip as the core MCU.

5. A corn ear threshing impact and rubbing collaborative inversion method based on time-space collaboration, characterized in that: Using the device according to any one of claims 1 to 4, comprising the following steps: S1: Obtain the acceleration, posture information and pressure data of corn ears through the MEMS sensor module; S2: Calculate the impact force and rubbing force on the corn ear based on the acquired data using MCU; S3: Adopt a unified space-time reference to achieve time synchronization and unified conversion of space coordinates; S4: Integrate data from different sensors through data fusion technology to improve measurement accuracy; S5: Use BP neural network to identify the motion state of corn threshing and realize the coordinated online inversion of impact and rubbing loads in the corn threshing process.

6. The corn ear threshing impact and rubbing collaborative inversion method based on time-space collaboration according to claim 5 is characterized in that: The accelerometer described in S1 is used to accurately measure the acceleration in the three-axis directions, the gyroscope is used to obtain real-time attitude information, including pitch angle, azimuth angle and roll angle, the electronic compass is used to improve the geomagnetic direction perception performance and obtain azimuth angle information, each of the grains is an independent data source, the flexible piezoelectric film sensor converts pressure into an electrical signal, adapts to the curved surface structure of the corn grains, detects pressure changes, and the force between the grains and the flexible piezoelectric film reflects the impact force; the force between the grains reflects the rubbing force on the corn cob.

7. The corn ear threshing impact and rubbing collaborative inversion method based on time-space collaboration according to claim 5 is characterized in that: The realization of the unified space-time reference in S3 includes time synchronization and unified conversion of space coordinates. Time synchronization is realized by means of the network time protocol, and the unified conversion of space coordinates is based on the positioning algorithm of inertial navigation and quaternion coordinate transformation.

8. The method for collaborative inversion of corn ear threshing impact and rubbing based on spatiotemporal coordination according to claim 5 is characterized in that: The data fusion technology in the S4 uses a particle filter algorithm to effectively integrate data from the accelerometer, gyroscope, electronic compass and flexible piezoelectric film sensor to eliminate noise and improve measurement accuracy.

9. The corn ear threshing impact and rubbing collaborative inversion method based on time-space collaboration according to claim 5 is characterized in that: The input parameters of the BP neural network in S5 include acceleration, attitude angle and pressure, and the output is the motion state recognition result during the corn threshing process.

Images