A real-time monitoring system and method for total grain loss rate of a combine harvester

Abstract

The application discloses a kind of combine harvester grain total loss rate real-time monitoring system and monitoring method, grain header loss monitoring unit is located in the combine harvester in grain divider, for monitoring header loss amount, grain cleaning loss monitoring unit is located in the combine harvester in vibrating screen one side, for monitoring cleaning loss amount, grain entrainment loss monitoring unit is located in the combine harvester in threshing concave screen one side, for monitoring entrainment loss amount, grain flow monitoring unit is located in the outlet of vertical grain conveying auger in the combine harvester, for monitoring grain flow, signal processing integrated circuit and vehicle-mounted industrial computer are installed in the cab inside in the combine harvester;Signal processing integrated circuit is used to calculate grain total loss rate, and is stored in real time with display on vehicle-mounted industrial computer.The application directly calculates grain total loss rate by fusing cleaning loss, entrainment loss, header loss and grain flow monitoring data, result is more accurate and close to actual, and is not influenced by the speed of advance of combine harvester, swath width and other working condition fluctuation.

Description

Technical Field

[0001] This invention relates to a real-time monitoring system and method for total grain loss rate in a combine harvester. Background Technology

[0002] Total grain loss rate is one of the most important indicators for evaluating the performance of combine harvesters, directly affecting the degree of grain waste and farmers' economic benefits. Grain losses during harvesting mainly include header loss, unthreshed grain loss, entrainment loss, cleaning loss, and missed grain loss. The total loss rate is the percentage of the total mass of grain lost from each of these factors relative to the total mass of harvested grain. Real-time and accurate monitoring of the total grain loss rate not only facilitates farmers and relevant government departments in monitoring and statistically analyzing loss distribution data across different regions, but also provides operational quality feedback for automatic control of combine harvester operating parameters and unmanned driving systems. Currently, the level of informatization in my country's harvesting machinery is low, and grain loss data acquisition often relies on manual labor, which is time-consuming, labor-intensive, and time-consuming. Large foreign agricultural machinery companies such as CLAAS, NEW HOLLAND, and JOHN DEERE have successively developed grain loss monitoring units, continuously improving and validating them in the market. These units are now sold as factory-installed parts or optional accessories for combine harvesters. However, foreign grain loss monitoring units are not well adapted to Chinese crop varieties, and related technologies are strictly kept secret and embargoed. In response to the needs of crop harvesting in my country, domestic scholars have conducted a series of studies on grain loss monitoring methods over the past two decades, such as:

[0003] Chinese invention patent CN100506013C proposes a real-time monitoring method for the random loss rate of a combine harvester. It simultaneously collects signals such as feed rate, grain harvest amount, grain loss, and grain moisture content, and comprehensively calculates the total grain loss rate. The monitored grain loss includes cleaning loss and separation loss. Chinese invention patent CN108370712A discloses an integrated multi-block double-layer cleaning loss sensor with progressive comb teeth. This mechanism solves the problems of traditional sensors being greatly affected by machine vibration and inaccurate sensor counting caused by excessive stalks and other debris. Chinese invention patent CN106376295B relates to a harvester with adaptively adjustable working parameters for the threshing, separation, and cleaning devices, using the monitoring value of the grain entrainment loss sensor as one of the inputs to the adaptive adjustment system. Chinese invention patent CN104737721B relates to an adaptive cleaning control device and method for a combine harvester, using the monitoring value of the grain cleaning loss sensor as one of the feedback conditions of the control system. Chinese invention patent CN106508257A discloses a combine harvester capable of adaptive adjustment and an adaptive adjustment method, using real-time monitored grain entrainment loss rate and grain cleaning loss rate as two input parameters for feedback adjustment. Chinese invention patent CN114451132B designs a real-time monitoring device for corn cleaning loss, including a strip sieve installed below the cleaning and impurity discharge port of the corn harvester and a monitoring system. It establishes corresponding correction relationships through indoor experiments to correct the loss monitoring results, making the measurement results more accurate. Chinese invention patent CN112042371B proposes a corn grain cleaning loss monitoring system and method. Based on the Kalman filter algorithm and large-sample experiments, it determines the threshold for distinguishing the impact force of grains and impurities on the piezoelectric thin-film sensing unit, ensuring the real-time performance and stability of the monitoring system's data transmission.

[0004] However, the inventors of this application have discovered that existing grain loss monitoring technologies, both domestically and internationally, have at least one of the following shortcomings:

[0005] 1. The monitoring scope of grain loss is not comprehensive. The designed monitoring device only targets a certain part of the loss during the harvesting operation, such as cleaning loss and entrainment loss. The total grain loss value is often indirectly reflected or predicted by a single part or a small part of the loss. In addition, no monitoring method for the loss of the header of the combine harvester has been found.

[0006] 2. There is a lack of accurate real-time calculation methods for grain loss rate. A fixed grain flow rate is often indirectly calculated by parameters such as cutting width, feed rate, and grass-to-grain ratio. Then, the real-time loss rate data is obtained by combining the grain loss monitoring value. However, the grain yield fluctuates constantly during the harvesting operation. This calculation method will lead to a certain deviation between the total grain loss rate and the actual value.

[0007] Therefore, there is an urgent need for a total grain loss rate monitoring system for combine harvesters that can accurately monitor the real-time total loss rate during harvesting operations, enable online data viewing and storage, provide reliable operational quality monitoring and feedback data, help domestically produced harvesting machinery move from catching up to leading the way in information technology, and enhance the market competitiveness of products. Summary of the Invention

[0008] The present invention provides a real-time monitoring system and method for total grain loss rate of a combine harvester in order to solve the problems existing in the prior art.

[0009] The technical solutions adopted in this invention are as follows:

[0010] A real-time monitoring system for total grain loss rate in a combine harvester, including

[0011] A grain header loss monitoring unit is installed inside the divider of a combine harvester to monitor the amount of header loss.

[0012] A grain cleaning loss monitoring unit is installed on one side of the vibrating screen in the combine harvester to monitor the amount of cleaning loss.

[0013] A grain entrainment loss monitoring unit is installed on one side of the threshing concave screen in the combine harvester to monitor the amount of entrainment loss.

[0014] A grain flow monitoring unit is installed at the outlet of the vertical grain conveying auger in the combine harvester to monitor the grain flow rate.

[0015] The signal processing integrated circuit and the vehicle-mounted industrial control computer are both installed inside the cab of the combine harvester; the signal processing integrated circuit is used to calculate the total grain loss rate and displays and stores it in real time on the vehicle-mounted industrial control computer.

[0016] Furthermore, the grain header loss monitoring unit includes a first base, a first sensitive plate, and a first piezoelectric ceramic. The first base is located inside the divider, the first sensitive plate is located on the first base, and the first piezoelectric ceramic is attached to the first sensitive plate and electrically connected to the signal processing integrated circuit.

[0017] Furthermore, the grain cleaning loss monitoring unit includes a second piezoelectric ceramic, a second sensitive plate, and a second base. The second base is fixed below the tail of the vibrating screen, the second sensitive plate is disposed on the second base, the second piezoelectric ceramic is attached to the second sensitive plate, and is electrically connected to the signal processing integrated circuit.

[0018] Furthermore, the grain entrainment loss monitoring unit includes a flow guide base, a third sensitive plate, and a third piezoelectric ceramic. The flow guide base is fixed below the tail end of the threshing concave sieve, the third sensitive plate is disposed on the flow guide base, the third piezoelectric ceramic is attached to the third sensitive plate, and is electrically connected to the signal processing integrated circuit.

[0019] Furthermore, the grain flow monitoring unit includes a gantry support, a mounting plate, a pressure sensor, an impact beam, a reference beam, and a universal vibration damping wire. An upper baffle is provided at the outlet of the vertical grain conveying auger, and the gantry support is connected to the upper baffle via the universal vibration damping wire. Both the impact beam and the reference beam are fixed on the mounting plate, which is fixed on the gantry support. One end of the pressure sensor is fixed on the impact beam or the reference beam, and the other end is fixed on the mounting plate. The pressure sensor is electrically connected to the signal processing integrated circuit.

[0020] Furthermore, the signal processing integrated circuit includes an FPGA chip, a charge amplification module, a frequency domain filtering module, an A / D analog-to-digital conversion module, and a data transmission module. The A / D analog-to-digital conversion module, the data transmission module, and the grain flow monitoring unit are all connected to the FPGA chip. The frequency domain filtering module is connected to the A / D analog-to-digital conversion module, and the charge amplification module is connected to the frequency domain filtering module. The charge amplification module is connected to the corresponding grain header loss monitoring unit, grain cleaning loss monitoring unit, or grain entrainment loss monitoring unit.

[0021] This invention also discloses a monitoring method.

[0022] The piezoelectric signals generated by the grain header loss monitoring unit, grain cleaning loss monitoring unit, and grain entrainment loss monitoring unit are sequentially passed through the charge amplification module, frequency domain filtering module, and A / D analog-to-digital conversion module, and finally transmitted to the FPGA chip. The random forest algorithm is used to identify and extract the grain signal from the mixture signal, and to count the number of grain cleaning, entrainment, and header loss monitoring.

[0023] The impact signal and reference signal generated by the grain flow monitoring unit are transmitted to the FPGA chip. Then, the vibration interference signal is filtered out by subtracting the impact signal and the reference signal, and the grain flow is calculated.

[0024] The FPGA chip calculates the total grain loss rate in real time by integrating the number of grains monitored for cleaning / entrainment / cutting table loss and grain flow rate.

[0025] Furthermore, the method for calculating the total seed loss rate in an FPGA chip is as follows:

[0026] Calculate the mass loss of grains per unit time during cleaning:

[0027]

[0028] Calculate the mass of grain entrainment loss per unit time:

[0029]

[0030] Calculate the mass loss of the grain header per unit time:

[0031]

[0032] Where m1, m2, and m3 represent the grain cleaning loss per unit time; n1, n2, and n3 represent the total number of grain header loss monitoring units; z represents the thousand-grain weight of the currently harvested crop; T s The period for loss data statistics; k1, k2, and k3 are the percentages of the total number of grain cleaning losses monitored relative to the actual number of cleaning losses.

[0033] Calculate the grain harvest weight per unit time:

[0034]

[0035] Where q is the grain flow rate; s n For impact signal; s n+1 For the reference signal, n = 1, 3, 5, ...; s n With s n+1 The signals are on the same set of impact beams and reference beams; α is the conversion coefficient between the digital quantity and the total grain mass at the outlet of the vertical grain conveying auger; T q The statistical period for impact signal and reference signal data;

[0036] Calculate the total grain loss rate:

[0037]

[0038] Where L is the total grain loss rate;

[0039] These are m1, m2, m3, and q, respectively, after the data time series has been aligned.

[0040] The present invention has the following beneficial effects:

[0041] 1) The grain loss value monitored in real time by this invention is the sum of cleaning loss, entrainment loss and header loss, which accounts for more than three-quarters of the total grain loss, and can more accurately reflect the actual total loss value.

[0042] 2) This invention directly calculates the total grain loss rate by integrating grain flow monitoring data, resulting in more accurate and realistic results, and is not affected by fluctuations in operating conditions such as combine harvester forward speed and cutting width.

[0043] 3) This invention enables remote display and multi-machine sharing of grain loss data through an IoT cloud platform and remote terminal display software, further improving the informatization level of combine harvesters. Attached Figure Description

[0044] Figure 1 This is a schematic diagram showing the installation location of the real-time monitoring system for total grain loss rate of a combine harvester.

[0045] Figure 2 This is a schematic diagram of the installation of the grain header loss monitoring unit.

[0046] Figure 3 This is a schematic diagram of the installation of the grain cleaning loss monitoring unit.

[0047] Figure 4 This is a schematic diagram of the installation of the grain entrainment loss monitoring unit.

[0048] Figure 5 This is a schematic diagram of the installation of the grain entrainment loss monitoring unit.

[0049] Figure 6 This is a schematic diagram of the modules of a signal processing integrated circuit. Detailed Implementation

[0050] The invention will now be further described with reference to the accompanying drawings.

[0051] like Figure 1 The present invention discloses a real-time monitoring system for total grain loss rate of a combine harvester, comprising a combine harvester body, a grain header loss monitoring unit 2, a grain cleaning loss monitoring unit 4, a grain entrainment loss monitoring unit 5, and a grain flow monitoring unit 8. The combine harvester body is equipped with a divider 1, a vibrating screen 3, a threshing concave screen 6, a vertical grain conveying auger 7, and a driver's cab 10.

[0052] The grain cleaning loss monitoring unit 4 is located below the rear of the vibrating screen 3, the grain entrainment loss monitoring unit 5 is located below the rear of the threshing concave screen 6, the grain header loss monitoring unit 2 is installed inside the divider 1, the grain flow monitoring unit 8 is installed at the outlet of the vertical grain conveying auger 7, the signal processing integrated circuit 9 is installed inside the cab 10, and the vehicle-mounted industrial control computer 11 is installed inside the cab 10.

[0053] like Figure 2 The grain header loss monitoring unit 2 includes a first base 201, a first sensitive plate 202, and a first piezoelectric ceramic 205. The first base 201 is welded inside the divider 1. The first sensitive plate 202 is connected to the first base 201 through a longitudinal rubber vibration damper. The first piezoelectric ceramic 205 is pasted at the center symmetrical position on the back of the first sensitive plate 202. One end of the coaxial cable signal line is welded to the positive and negative poles of the first piezoelectric ceramic 205, and the other end is connected to the signal processing integrated circuit 9.

[0054] The grain header loss monitoring unit 2 can be installed inside a single-sided divider 1 or symmetrically inside both dividers 1. The angle between the first base 201 and the horizontal plane is 30°-60°, and the front of the first base 201 should be hollowed out to prevent material from accumulating on top of it. The top of the first base 201 should be hollowed out at a position perpendicular to the first sensitive plate 202, and the hollowed-out area should be larger than the surface area of ​​the first sensitive plate 202 so that as many lost grains as possible collide with the first sensitive plate 202. In addition, the bottom of the first base 201 should also have a certain hollowed-out area to prevent falling material from accumulating. Using a coaxial cable signal line can significantly reduce the transmission loss of piezoelectric signals and reduce interference from the external environment on the transmission of piezoelectric signals.

[0055] The first sensitive plate 202 is made of stainless steel, aluminum alloy or other metals to increase the amplitude of the impact piezoelectric signal generated by each component of the mixture of grains, stalks and other materials, which helps to increase the signal differentiation. In addition, the lateral width of the first sensitive plate 202 should be consistent with or close to that of the third sensitive plate 503, and its outer side should not extend beyond the top outer side of the divider 1 to prevent unharvested crops from colliding with the first sensitive plate 202.

[0056] The surface area of ​​the first piezoelectric ceramic 205 should be consistent with or close to that of the third piezoelectric ceramic 506, so that the initial piezoelectric signal amplitudes generated by the grain header loss monitoring unit 2 and the grain entrainment loss monitoring unit 5 are not much different, thereby reducing the differences in the parameters of each module in the signal processing integrated circuit 9.

[0057] like Figure 3 The grain cleaning loss monitoring unit 4 includes a second piezoelectric ceramic 401, a second sensitive plate 403, and a second base 404. The second sensitive plate 403 and the second base 404 are connected by a longitudinal rubber vibration damper, and the second base 404 is connected to the side wall 407 of the frame by a transverse rubber vibration damper. The second piezoelectric ceramic 401 is attached to the center symmetrical position on the back of the second sensitive plate 403. One end of the coaxial cable signal line is soldered to the positive / negative pole of the second piezoelectric ceramic 401, and the other end is connected to the signal processing integrated circuit 9.

[0058] The second sensitive plate 403 is made of stainless steel, aluminum alloy, or other metals to increase the amplitude of the impact piezoelectric signals generated by the various components of the mixture of grains and stalks, thereby improving signal discrimination. The second sensitive plate 403 and the second piezoelectric ceramic 401 constitute a monitoring unit. Three to six monitoring units are closely distributed on the second base 404, with a gap of 0.5mm-1.5mm between adjacent monitoring units. The sum of the widths of all monitoring units is greater than or equal to the transverse width of the vibrating screen 3, enabling full-width monitoring of grain loss. This increases the percentage of total monitored grain cleaning loss to actual cleaning loss, reducing statistical errors. Transverse and longitudinal rubber vibration dampers effectively reduce signal interference from vibrations in multiple parts of the combine harvester on the grain cleaning loss monitoring unit 4, reducing grain signal identification and statistical errors.

[0059] like Figure 4 The grain entrainment loss monitoring unit 5 includes a flow guide base 502, a third sensitive plate 503, and a third piezoelectric ceramic 506. The flow guide base 502 is welded to the side wall 407 of the frame. The third sensitive plate 503 is connected to the flow guide base 502 through a longitudinal rubber vibration damper. The third piezoelectric ceramic 506 is pasted on the back of the third sensitive plate 503 at a centrally symmetrical position. One end of the coaxial cable signal line is welded to the positive and negative poles of the third piezoelectric ceramic 506, and the other end is connected to the signal processing integrated circuit 9.

[0060] The third sensitive plate 503 is also made of stainless steel, aluminum alloy and other metals to increase the amplitude of the impact piezoelectric signal generated by each component of the mixture of grains, stalks and other materials, which helps to increase the signal differentiation. In addition, the lateral width of the third sensitive plate 503 should be smaller than that of the second sensitive plate 403, and its outer side should be as close as possible to the rearmost end of the threshing concave screen 6 to reduce the obstruction to the flow of threshing materials.

[0061] The surface area of ​​the third piezoelectric ceramic 506 should be smaller than that of the second piezoelectric ceramic 401. In addition, the ratio of the surface area of ​​the third piezoelectric ceramic 506 to that of the third sensitive plate 503 should be close to that of the ratio of the surface area of ​​the second piezoelectric ceramic 401 to that of the second sensitive plate 403, so that the initial piezoelectric signal amplitudes generated by the seed entrainment loss monitoring unit 5 and the seed cleaning loss monitoring unit 4 are not much different, thereby reducing the differences in the parameters of each module in the signal processing integrated circuit 9.

[0062] The grain entrainment loss monitoring unit 5 can be installed on a single side wall 407 of the frame, or symmetrically on both side walls 407 of the frame. The longitudinal rubber vibration damper can effectively reduce the signal interference of the frame side wall 407 vibration on the grain entrainment loss monitoring unit 5, and reduce grain signal identification and statistical errors.

[0063] The angle between the fixed end face of the third sensitive plate 503 on the guide base 502 and the horizontal plane is 30°-90°. In addition to fixing the third sensitive plate 503, the guide base 502 also guides the flow of the discharged mixture, so that the discharged mixture moves towards the middle of the transverse direction of the vibrating screen 3. This can prevent the discharged mixture from accumulating on the side of the vibrating screen 3, which would lead to a decrease in cleaning efficiency. The guide base 502 also prevents it from accumulating above the third sensitive plate 503, which would lead to poor material discharge.

[0064] like Figure 5 The grain flow monitoring unit 8 includes a gantry support 801, a mounting plate 802, a pressure sensor 804, an impact beam 805, a reference beam 806, and a universal vibration damping wire 807.

[0065] The gantry support 801 is connected to the upper baffle 808 at the outlet of the vertical grain conveying auger 7 via a universal vibration damping steel wire 807, and the mounting plate 802 is fixed on the gantry support 801; one end of the pressure sensor 804 is fixed on the impact beam 805 or the reference beam 806, and the other end is fixed on the mounting plate 802, and the signal line of the pressure sensor 804 is connected to the signal processing integrated circuit 9.

[0066] The impact beam 805 and the reference beam 806 are both made of stainless steel, aluminum alloy and other metals to increase the amplitude of the pressure signal generated by the impact of the grain, which is conducive to signal acquisition and analysis, and at the same time ensures the strength and reliability of the beam structure under strong impact.

[0067] The impact beam 805, the reference beam 806, and the two pressure sensors 804 constitute a monitoring unit. Two to four monitoring units can be distributed on the gantry support 801. The included angle between adjacent monitoring units is 0-30°. Each monitoring unit can be orthogonally arranged with the grain throwing direction at the outlet of the vertical grain conveying auger 7, which can also improve the amplitude of the pressure signal generated by the grain impact, which is beneficial for signal acquisition and analysis.

[0068] The 807 universal vibration damping steel wire can effectively reduce the signal interference of vibration in multiple parts of the combine harvester to the grain flow monitoring unit 8, and reduce the error in grain flow calculation.

[0069] like Figure 6 The signal processing integrated circuit 9 includes an FPGA chip, a charge amplification module, a frequency domain filtering module, an A / D analog-to-digital conversion module, and a data transmission module. The A / D analog-to-digital conversion module, the data transmission module, and the grain flow monitoring unit 8 are all connected to the FPGA chip. The frequency domain filtering module is connected to the A / D analog-to-digital conversion module, and the charge amplification module is connected to the frequency domain filtering module. The charge amplification module is connected to the corresponding grain header loss monitoring unit 2, grain cleaning loss monitoring unit 4, or grain entrainment loss monitoring unit 5.

[0070] The data transmission module includes a CAN bus transceiver module and a 4G wireless transmission module. The CAN bus transceiver module is connected to the CAN port of the vehicle-mounted industrial control computer 11 via two signal lines, and the grain loss rate data can be displayed and stored in real time on the vehicle-mounted industrial control computer 11. The 4G wireless transmission module communicates wirelessly with an IoT cloud platform (such as Alibaba Cloud) in real time. The IoT cloud platform displays the grain loss rate data in real time on mobile phones, tablets, etc., and shares it with multiple devices through remote terminal software.

[0071] The piezoelectric signals (analog signals) generated by the grain cleaning loss monitoring unit 4, the grain entrainment loss monitoring unit 5, and the grain header loss monitoring unit 2 are first transmitted to the charge amplification module, then to the frequency domain filtering module, then to the A / D analog-to-digital conversion module, and finally to the FPGA chip. The random forest algorithm is used to identify and extract the grain signal from the mixture signal, and statistics are performed on the number of grain cleaning, entrainment, and header loss monitoring units. The impact signal (digital signal) and reference signal (digital signal) generated by the grain flow monitoring unit 8 are first transmitted to the FPGA chip. Vibration interference signals are filtered out by subtracting the impact signal and reference signal, and the grain flow rate is calculated. The FPGA chip calculates the total grain loss rate in real time by fusing the number of grain cleaning / entrainment / header loss monitoring units, grain flow rate, and statistical periodic data.

[0072] The method for calculating the total seed loss rate in an FPGA chip is as follows:

[0073] Calculate the mass loss of grains per unit time during cleaning:

[0074]

[0075] Calculate the mass of grain entrainment loss per unit time:

[0076]

[0077] Calculate the mass loss of the grain header per unit time:

[0078]

[0079] Where m1, m2, and m3 represent the mass of grain loss per unit time during cleaning (unit: g / s); n1, n2, and n3 represent the total number of grain header loss monitoring units (unit: units); z represents the thousand-grain weight of the currently harvested crop (unit: g); T s The period for loss data statistics is s; k1, k2 and k3 are the percentages of the total number of grain cleaning losses monitored to the actual number of cleaning losses (based on the average values ​​obtained from loss distribution experiments, with different values ​​corresponding to different harvester models or sensor installation locations).

[0080] Calculate grain flow rate (grain harvest weight per unit time):

[0081]

[0082] Where q is the grain flow rate (unit: g / s); s n For impact signal; s n+1 For the reference signal, n = 1, 3, 5, ...; s n With s n+1 The signals are on the same set of impact beams and reference beams; α is the conversion coefficient between the digital quantity and the total grain mass at the vertical grain conveying auger outlet (based on the average value obtained from grain conveying outlet distribution tests, with different values ​​corresponding to different harvester models or sensor installation positions), g; T q The statistical period (in seconds) for impact signal and reference signal data;

[0083] Calculate the total grain loss rate:

[0084]

[0085] Where L is the total grain loss rate, %. These are m1, m2, m3, and q, respectively, after data time series alignment. This is because during the actual harvesting process, the time when the same batch of grain material flow collides with the sensitive plate of the cleaning loss, entrainment loss, header loss monitoring unit, and flow monitoring unit is not synchronized, and there is a certain time delay between them. Therefore, it is necessary to conduct material flow time measurement experiments according to different harvester models or sensor installation positions, and perform data time series alignment to calculate a more accurate total grain loss rate.

[0086] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be considered within the scope of protection of the present invention.

Claims

1. A method for real-time monitoring of total grain loss rate in a combine harvester, characterized in that: Including a monitoring system, the system includes: The grain header loss monitoring unit is located inside the divider of the combine harvester and is used to monitor the amount of header loss. The grain cleaning loss monitoring unit is located on one side of the vibrating screen in the combine harvester and is used to monitor the amount of cleaning loss. The grain entrainment loss monitoring unit is located on one side of the threshing concave screen in the combine harvester and is used to monitor the amount of entrainment loss. The grain flow monitoring unit is located at the outlet of the vertical grain conveying auger in the combine harvester and is used to monitor the grain flow. The signal processing integrated circuit and the vehicle-mounted industrial control computer are both installed inside the cab of the combine harvester; the signal processing integrated circuit is used to calculate the total grain loss rate and displays and stores it in real time on the vehicle-mounted industrial control computer. The grain header loss monitoring unit includes a first base, a first sensitive plate, and a first piezoelectric ceramic. The first base is located inside the divider, the first sensitive plate is located on the first base, and the first piezoelectric ceramic is attached to the first sensitive plate and electrically connected to the signal processing integrated circuit. The grain cleaning loss monitoring unit includes a second piezoelectric ceramic, a second sensitive plate, and a second base. The second base is fixed below the tail of the vibrating screen, the second sensitive plate is disposed on the second base, the second piezoelectric ceramic is attached to the second sensitive plate, and is electrically connected to the signal processing integrated circuit. The grain entrainment loss monitoring unit includes a flow guide base, a third sensitive plate, and a third piezoelectric ceramic. The flow guide base is fixed below the tail of the threshing concave screen, the third sensitive plate is placed on the flow guide base, and the third piezoelectric ceramic is pasted on the third sensitive plate and electrically connected to the signal processing integrated circuit. The grain flow monitoring unit includes a gantry support, a mounting plate, a pressure sensor, an impact beam, a reference beam, and a universal vibration damping wire. An upper baffle is provided at the outlet of the vertical grain conveying auger. The gantry support and the upper baffle are connected by the universal vibration damping wire. The impact beam and the reference beam are both fixed on the mounting plate, and the mounting plate is fixed on the gantry support. One end of the pressure sensor is fixed on the impact beam or the reference beam, and the other end is fixed on the mounting plate. The pressure sensor is electrically connected to the signal processing integrated circuit (9). The signal processing integrated circuit includes an FPGA chip, a charge amplification module, a frequency domain filtering module, an A / D analog-to-digital conversion module, and a data transmission module. The A / D analog-to-digital conversion module, the data transmission module, and the grain flow monitoring unit are all connected to the FPGA chip. The frequency domain filtering module is connected to the A / D analog-to-digital conversion module, and the charge amplification module is connected to the frequency domain filtering module. The charge amplification module is connected to the corresponding grain header loss monitoring unit, grain cleaning loss monitoring unit, or grain entrainment loss monitoring unit. The method includes: The piezoelectric signals generated by the grain header loss monitoring unit, grain cleaning loss monitoring unit, and grain entrainment loss monitoring unit are sequentially passed through the charge amplification module, frequency domain filtering module, and A / D analog-to-digital conversion module, and finally transmitted to the FPGA chip. The random forest algorithm is used to identify and extract the grain signal from the mixture signal, and to count the number of grain cleaning, entrainment, and header loss monitoring. The impact signal and reference signal generated by the grain flow monitoring unit are transmitted to the FPGA chip. Then, the vibration interference signal is filtered out by subtracting the impact signal and the reference signal, and the grain flow is calculated. The FPGA chip calculates the total grain loss rate in real time by integrating the number of grains monitored for grain cleaning / entrainment / cutting table loss and grain flow rate.

2. The monitoring method as described in claim 1, characterized in that: The method for calculating the total seed loss rate in FPGA chips is as follows: Calculate the mass loss m1 of grain cleaning per unit time: ， Calculate the mass loss of grains carried over per unit time, m2: ， Calculate the mass loss per unit time of the grain header (m3): ， Where n1, n2, and n3 are the total number of grains monitored for loss during grain cleaning, grain entrainment, and grain harvesting, respectively; z is the thousand-grain weight of the currently harvested crop; T s The period for loss data statistics; k1, k2, and k3 are the percentages of the total number of losses monitored from grain cleaning, grain entrainment, and grain header to the actual number of losses from grain cleaning, respectively; Calculate the grain harvest weight per unit time: ， in, For grain flow rate; This is an impact signal; For the reference signal, n=1,3,5,……; and For signals on the same set of impact beams and reference beams; The conversion coefficient between the digital quantity and the total mass of grains at the outlet of the vertical grain conveying auger; The statistical period for impact signal and reference signal data; Calculate the total grain loss rate: ， in, This represents the total grain loss rate. , , , These are m1 and m2 after the data time series are aligned. 2、 m3, q.

Images